This item is only available as the following downloads:
LEARNING UNDER STRESS: SEPARATING THE EFFECTS OF ALLOPREGNANOLONE AND FLUOXETINE IN CARASSIUS AURATUS BY JEREMY DAVID EVANS A Thesis Submitted to the Division of Natural Sciences New College of Florida in partial fulfillment of the requirements for the degree Bachelor of Arts Under the sponsorship of Alfred Beulig, PhD Sarasota, Florida May, 2012
ii Dedication to Dinah, Devin, Donald: My caryatids. Foremost, to my gold fish. I'd a monument construe to the folks I knew but words are hardly true, frenetically flew from my mind the moment it was you anew. Lit like passions in somber passing stations, love like roses in spring cessation: you may be mine to root attestations.
iii Contents Dedication ................................ ................................ ................................ ................................ ........ ii Contents ................................ ................................ ................................ ................................ .......... iii Abstract ................................ ................................ ................................ ................................ ............ v Introduction ................................ ................................ ................................ ................................ ..... 1 Stress, the Endo crine System, and the Brain ................................ ................................ ............... 3 Defining Stress ................................ ................................ ................................ ......................... 3 Acute Allostatic Induction ................................ ................................ ................................ ........ 8 HPA/HPI Regulation and Dysregulation ................................ ................................ ................. 16 Cortisol in Humans and Fish ................................ ................................ ................................ ... 26 Allopregnanlone as a Mediator of the HPA/HPI ................................ ................................ .... 32 GABA A Receptors: Structure and Function ................................ ................................ ............ 34 Ve rtebrate Neurosteroids ................................ ................................ ................................ ...... 39 The Case for Telencephalic Homologies from Teleosts to Tetrapods ................................ ....... 52 Eversion vs. Evagination ................................ ................................ ................................ ......... 53 Tele ostean Dorsal Pallium ................................ ................................ ................................ ...... 56 Behavioral and Lesion Studies ................................ ................................ ............................... 59 Rationale for the Present Study ................................ ................................ ................................ 61 Hypotheses ................................ ................................ ................................ ............................ 62 Methods ................................ ................................ ................................ ................................ ......... 63 Subjects ................................ ................................ ................................ ................................ ...... 63 Husbandry and Aquaria ................................ ................................ ................................ ............. 64 Location ................................ ................................ ................................ ................................ ...... 65 Apparatus ................................ ................................ ................................ ................................ ... 65 Data Collection ................................ ................................ ................................ ........................... 66 Scototaxis ................................ ................................ ................................ ............................... 66 Delay Learning Task ................................ ................................ ................................ ............... 67 Trace Le arning Task ................................ ................................ ................................ ................ 68 Experimental Design ................................ ................................ ................................ .................. 68 Treatment Preparation ................................ ................................ ................................ .......... 69 Results ................................ ................................ ................................ ................................ ............ 71
iv Scototaxis ................................ ................................ ................................ ................................ ... 71 Cognition ................................ ................................ ................................ ................................ .... 73 Delay Learning ................................ ................................ ................................ ........................ 74 Trace Learning ................................ ................................ ................................ ........................ 81 Observations ................................ ................................ ................................ .............................. 88 Discussion ................................ ................................ ................................ ................................ ...... 89 Interpretations ................................ ................................ ................................ ........................... 91 Possible Confounds ................................ ................................ ................................ .................... 96 Depression and SSRI Efficacy ................................ ................................ ................................ ..... 97 References ................................ ................................ ................................ ................................ ... 103 APPENDIX I: ................................ ................................ ................................ ................................ .. 142 Additional Graphs and Tables ................................ ................................ ................................ .. 142 APPENDIX II: ................................ ................................ ................................ ................................ 150 Actinopterygii ................................ ................................ ................................ ....................... 150 Chondichthyes ................................ ................................ ................................ ..................... 152 Coelacanths ................................ ................................ ................................ .......................... 154 Anatomical and Hodological Models of Pallial Arrangemen ................................ ............... 157 Embryonic Origins of Pallial Zones ................................ ................................ ....................... 163
v Abstract This thesis compares the cognitive and affective effects of two anxiolytic compounds in the gol dfish, Carassius auratus: fluoxetine (Prozac), an antidepressant, and allopregnanolone, a neur osteroid native to the brain of vertebrates. There is a literature revi ew on the role of the intera ctions of stress, the endocrine system, and the brain in both goldfish and humans to stage the hypotheses in question, ultimately predicting a decline in cognition in fish exposed to the anxi olytics unless they are also presente d with exogenous stress. The results of the scototaxic trials indicate a significant increase in exploratory behavior in only the allopregnanolone group over the control. Cortisol treated groups exhibited greater initial avoidance behavior, but this effect diminished in latter trials, resulting in significant impairments by the final block. Those fish treated with anxiolytics and cortisol presented a similar pattern, but with only allopregnan olone+cortisol demonstrating hippocampal dependent avoidance behav ior on par with the co ntrol group by the end of the experiment. Throughout the trials, all fish treated with fluoxetine demonstrated substantial impairments, especially in tests that required the retention of a trace stimulus. In conclusion: it is apparent that fluoxetine impedes temporal cognition in goldfish, even when compensating for hypercortisolemia, and that allopregnanolone is capable of am eliorating the cognitive deficits imposed by chronic stress on the hippocampal homologue in gol dfish. Additiona lly, the acute anxiolytic effects of allopregnanolone are likely conserved in the goldfish.
1 Introduction Modern psychopharmaceutical therapies are notorious for their inefficacy relative to other medical treatments and often toxic side effects (Leucht et al, 2012). Given that psychiatric disorders are generally only loosely correlated with biological findings and that the understan ding of causality is still a distant goal, it is unsurprising that clinical psychiatry has not yet been realized as clinical neuroscience (Reynolds et al, 2009; Insel and Quirion, 2012; Kraemer, Schultz, and Arndt, 2002). Instead, po sitive results in clinical trials lead to the adoption of treatment pa radigms for complex disorders without a full understanding of the mechanisms involved. An e xample of this reasoning is the establishment the Serotonin Hypothesis in developing the sele ct ive serotonin reuptake inhibitor (SSRI) Prozac as a treatment for major depression This a pproach, as will be addressed later, is highly problematic. Unfortunately, coping with the co mplexities of diseases that manifest subtle behavioral symptoms isprofou ndly difficult and defin itive etiology is hard to come by The careful control of confounding variables may be better managed in animal models, though this solution requires the effective demonstration of anal ogous behaviors between species composed of hom ologous organs in order to translate effects to other species (Green, Gabrielsson, and Fone 2011). This experiment involves the use of an animal model to manipulate possible confoun ding variables in an attempt to disentangle their possible effects. In any comparison of behavioral effects, there is an array of known and unknown factors that may interact to affect experimental results, especially complicating the investigation of the profile of various treatments. The use of a fish model in the laborator y per mits an exacting level of control of and the administration of experimental treatments that would not be feasible with human su bjects. However, extrapolation to the human condition must, as with any model, be exercised
2 with care. To provide a basis for possible analogies to function and homologies to structure, a review of previous literature on the physiological systems of vertebrates that regulate the stress response and its effects on learning is carried out. Ultimately, this t hesis aims to explore the relative effects of the popular pharmaceutical fluoxetine ( Prozac ) and the neurosteroid allopregnanolone, on c ognition in goldfish. Since the former substance has been correlated to lowered performance in shuttle box avoidance lear ning (Beulig et Fowler, 2008) and the latter has been shown to attenuate performance on similar tasks in rats (Johanssonet al, 2002) and humans (Kask et al, 2007), it is predicted that both a nxi o lytics may have negative effects for the acquisition of o perant conditioning To dissect these possible cognitive deficits from their value as antidepressants, the performance of the fish a dministered these treatments will be compared to a parallel cohort with the same dosing sche dule plus cortisol, the principl e regulator of chronic stress in fish and humans. It is hypothesized that the negative effects on learning in the non stressed paradigm will be overpowered by the benefits provided by these antidepressa nts in this cortisol implanted model of chronic stress resulting in greater avoidance behavior over the cortisol alone group. By initially investigating the compounds this design aims to explore the acute affective changes produced by these treatments, which may be pr edic tive of the long term affective profiles of these compounds. It is predicted that both fluox etine and allopregnanolone without cortisol will promote significantly greater exploratory beha vior given the anxiolytic profiles of these drugs. Cortisol should be roughly the same as the co ntrol group, as the hyperarousal of this corticosteroid competes with its regulation of anxiety r esponse, likely summing to produce no overall effect on exploratory behavior.
3 The differ e netial performance of these animals in a t ask with and without a temporal delay between the presentation of the conditional and unconditional stimuli (CS and UCS) will be utilized to determine the effects of the treatments on hippocampal dependent behavior I mprovements in avoidance performance in fish that must retain memory of the preceding stim ulus over a delay during which the require use of the lateral zone of the dorasalis telencephali, whereas those animals with this region ablated perform ide ntically to controls in non delayed learning paradigms (Portavella et al, 2004). This region is pu rported ly ho mologous to the hippocampal formation in mammals a region essential to memory function and implicated in the development of pathologies such as depression (No rthcutt, 2008; Sheline et al, 1996). Cortisol a glucocorticoid of the endocrine stress axis in humans and fish but not rats, has been shown to lead to atrophy of the hippocampus when present in excess (Zhang et al, 2006). By demonstrating the selective e ffects of substances with similar antidepressant properties in hypercortisolemic ani mals versus controls, it may be possible to understand the actions of the chemicals in a context resembling major depression in humans In major depre ssion cognition is si gnificantly impaired largely through inhibition and atrophy of the hippoca mpus (Lupien et al, 1998; Sapols ky, 2000). T he results of this experiment suggest the efficacy of allopregnanolone compared to the well researched Prozac as a treatment for depres si on that may have greater efficacy and fewer side effects. Stress, the Endocrine System, and the Brain Defining Stress Everybody knows what stress is and nobody knows what it is. The word stress, like success, failure, or happiness, means different things to di fferent people and, except for a few specialized scientists, no one has really tried to define it although it has become part of our daily vocabulary. Is it e ffort, fatigue, pain, fear, the need for concentration, t he humiliation of ce n-
4 sure, loss of blood, or even an unexpected success that requires complete r eformulation of one's life? The answer is yes and no. That is what makes the de finition of stress so difficult. Every one of these conditions can p roduce stress, all other as well. (Selye, 1976) Stress has been a topic surrounded by controversy even before the first rodents were i njected with topsoil by the famous Hans Selye. Walter Cannon was the first to employ the term itions that focused on emotional upset as well as a physiological phenomenon (Cannon, 1935). Cannon u rged his peers in medicine to address psychological stress as an underlying issue for their disturbances, and should have, therefore, a natural interest in the effects of emotional the definition of stress and advise a practical role for stress research in medicine, he was rel atively unheeded due to the absence of publis hed findings and a lack of pre established models of stress research. Hans Selye an endocrinologist and most memorable researcher in this field who elaborated the theory of General Adaptation Syndrome, observed a series of common p athologies that resulte (Selye, 1946). This led Selye to investigate the roles of the various hormones and organs i nvolved in stress, essentially the modern HPA axis, and he began to recognize a pattern of arousal of non specific stress that we now term HPA axis activation. Measuring the short term cons equences of extreme stress, such as the atrophy of the liver and thymus after the injection of su b lethal doses Selye contrasts this generalized demand with th e specific effects of particular
5 T ABLE 1 : HPA/HPI H ORMONES 1 Iino et al, 1987; 2 Vaughan et al, 1995; 3 Bern, 1985; 4 Conlon, 2007; 5 Lederis, 1985; 6 Bernier, 1999; 7 Bernier et al, 2001; 8 Lenz et al, 1985 ; 9 Chatzaki et al, 2003; 10 Lederis et al, 1982; 11 Fryer et al, 1983; 12 Mommsen et al, 1999; 13 Herbert et al, 2006; 14 Sangalang et al, 1972; 15 Kuhn et al, 2004; 16 Aguilera, 1994; 17 Ball and Olivereau, 1965; 18 Rae et al, 1979; 19 Jessop, 1999; 20 Huising et al, 2004; 21 Alderman et al, 2007; 22 Holsboer, 1999; 23 To et al, 1999; 24 Merali et al, 2004; 25 Mazeaud et al, 1977; 26 Stone et al, 2003; 27 Wortsman, 2002; 28 Sorrels and Sapolsky, 2007; 29 Floras et al, 1988; 3 0 Kohm and Sanders, 2000; 31 ,Aston Jones et al, 2009; 32 Arana et al, 1985
6 in his earliest studies these specific effects were the actions of various drugs and, later, the b ehaviors aimed at avoiding the noxious stimuli in the future. (Mason, 1975) Only after the discovery of the role of glucocorticoids as a useful hormonal indicator by the Nobel Laureates Reichstein, Ke ndall, and Hench in 1944 did the science of stress truly esc alate. It was quickly realized that elevations of glucocorticoids followed a diverse array of stimuli which range from mild (e.g., exposure to novel objects) to moderate (e.g., restraint ) to seve re (e.g., prolonged social isolation). Selye went further to characterize stressors into two general was capable of surmounting the difficulties presented or not, respectively. This prima facie pi cture of stress and glucocorticoids became muddled as glucocorticoids were found to have a mu ltiphasic daily regulation that corresponded not only with acute stress but also the circadian c yF IGURE 1 : R ELATIVE ELEVATIONS OF CORTICOSTERONE ( SUMMED A REA U NDER C URVE ) IN THE SAME RATS EXPOSED VA RIOUS STIMULI ; T ESTS WERE ADMINISTER ED IN SERIES WITH BASELINE SAMPLE S FOLLOWED BY 15 MINUTES OF EXPOSURE AND THEN ON E HOUR OF REC OVERY ; CONTROL ESTABLISHED FROM HOME CAGE SAMPLE DUR ING REST HALF WAY BETWEEN TOTAL TESTING (A DAPTED FROM K OOLHAASA ET AL 2011)
7 cle (Buckley and Schatzberg, 2005). Furthermore, glucocorticoids are also mobilized in response to seemingly appetitive scenarios, such as during social victory an d sexual activity Koolhaasa and colleagues, perturbed by these inconsistencies, call the traditional model of stress into question: they insist that stress is be st defined as an unpredictable and uncontrollable stimulus and response (2011). They cite that the sympathetic adrenomedullary (SAM) system and hypothalamic pituatary adrenal (HPA) axis, while indisputably activated during stress reactions, also serve to f acilitate general beha viors that cannot be defined as stress, such as the mobilization of glucose and localization of ox ygen during bouts of increased metabolism. When one defines stress as Cannon and Selye do, a the reality that most actions result in alterations of homeostasis, from affective shifts to essential and stereotyped behaviors. Elaborating on this model, Koolhaasa et al. review a number of recent studies on r odents. They assert that most models of stre ss research suffer from diminishing magnitude of effect as the animals attenuate their responses to the increasingly predictable stress; it has not been established whether or not this represents a n inherent physiological habituation process or a cognitive recognition of patterns that leads to the termination of an anxiety response (Gri ssom and Bhatnagar, 2009). Investigators originally employed chronic, mild stress (CMS) to i nduce chronic hypercoticosteronemia, though it was found that with modifications t o vary the daily stressors (e.g., shaking the home cages one day and exposure to loud noise the next) or randomizing intervals of exposure that it was possible to induce a state considered akin to d epression (Willner, 2005). Recently, it has even been fou nd that the use of predictable CMS (PCMS) induces the opposite changes in rodents, improving affect, memory, and hippocampal volume (Parihar et al, 2011).
8 Controllability refers to the capability of the organism to mitigate the stressor. In the case of lea rned swimming, rats demonstrate consistently lessened circulating adrenaline, gl ucose, and corticosterone, simultaneous with increased free fatty acids and noradrenaline after learning the activity (Scheurink et al, 1999). The lessening of the glucocortico id may be seen as adaptive as the role of elevated corticosteroids is considered hazardous to long term health (S a polsky et al, 2002; Bodo, 2010); it is likewise useful to the organism to limit elevations in blood sugar, instead relying on free fatty acids given that elevations in glucose contribute to (Halliwell et Gutteridge, 1986; Holloszy et Coyle, 1984). Controllability in a psychological context is demonstrab le in the paradigm of learned helplessness, in which an experimental subject is stimulus (Abramson et al, 1978; Maier et al, 1976). The effects of this paradigm in rats typically results outcomes traditiona l l y associated with prolonged distress: less exploratory behavior, decreased body weight, loss of sucrose preference, impairment in spatial learning, declining hippocampal volume/function, chronic elevation of corticosterone, and decreased expre ssion of brain derived neurotrophic factor (BDNF) and cAMP response element binding protein (CREB) (Raizer et al, 2003; Songa et al, 2006; Malberg et al, 2003). Acute Allostatic Induction Koolhasa and colleagues sug gest that t he combination of unpredictabi lity and uncontro llability, allows for a more specific model of stress that is defined by the induction of alternate Dr. Bruce McEwen. McEwen describes allostasis as profile generally to the benefit of the organism (2004). The induction of the SAM system is one of the first responders to stress, followed by the more
9 gradual release o f corticosteroids and cytokines, eventually returning to a less allostatic state over time assuming that the stressor attenuates. In the CNS, the actions of catecholamines p otentiate the release of cAMP and promote LTP in the limbic structures whereas the effect of gl ucocorticoids depends on the amplitude and duration of their exposure, sometimes encouraging LTP and under other conditions inhibiting it severely (McEwen and Sapolsky, 2002; McEwen, 2007; Ryan et al, 2010). During states of acute allostasis associated with HPA/HPI and SAM activation, there is a common constellation of physiological changes that vary little across vertebrates (Denver, 2009) When an organism is exposed to a predator, for example, the uncontrollable and unpr edictable situation is transduced via the senses and transmitted via various thalamic and a few extrathalamic tracts to the brain stem nuclei (specifically the locus coeruleus and raphe nuclei), amygdala, and the bed nucleus of the stria terminalis (BST) (Herman and Cullinan, 1997). The locus coeruleus, the source of norepinephrine in the CNS which projects broadly, connects d irectly to the paraventricular nucleus (PVN) of the hypothalamus, the convergent point of HPA/HPI neurosecretion for all acu te stressors. Upon stimulatio n, 1 2 rece ptors of the PVN respond with cAMP elevation which increases the synthesis of CRH and potent iates release of CRH via Ca 2+ intake (Plotsky et al, 1989). The central effects of catecholamine are further enhanced by their systemic release into the periphery via CRH/ACTH where they act to raise blood sugar, mobilize free fatty acids, initiate proteolysis, increase cardiac output, enhance absorption of glucose into skeletal muscle, inhibit smooth muscle contraction, dilate airways, i ncrease blood pressure, and attenuate nociceptive pathways (Wingfield and Romero, 2011; Wortsman, 2002). Furthermore, epinephrine and particularly norepinephrine have been shown 2 1 ad renergic rece ptors which synergistically increase cAMP and thus CREB, a process which has been shown to be
10 both essential to encode emotional memory in aversively motivated tasks and beneficial to co nsolidating many other kinds of memory (Campolongo and Ro ozendaal, 2011; Tronsen et al, 2012; Poulace et al, 2011; Blass et al, 2012). The amygdala is well known for its role in human emotion, a demonstrative case being the patient known as SP who presents with bilateral amygdala r calcification due to a rare gen e tic disorder, Urbach Wiethe disease (Feinstein et al, 2011) The authors report that SP demo nstrates intense curiosity and risk taking in a manner not unlike Kl ver Bucy syndrome, demo nstrating entirely normal emotions aside from the utter absence of fear Interestingly, SP does not demonstrate profound issues of memory or emotional reporting despite that this has been shown in others with amygdalar damage accrued later in life (Squire and Zola Morgan, 1991) I t must be noted however, that such comparison s are inherently fraught with difficulties due to confluent lesions. Lesions to the hippocampi of humans have also been shown to produce pr ofound memory deficits ( Corkin, 2002; Zola Morgan et al, 1986) and some have suggested that the dynamic interplay bet ween the amygdala and hippocampus permits recovery from amygd andition and 20 others with unilateral amygdalar lesions, Anderson and Phelps suggest that amy gdalar damage does not alter emotional self reporting on an acute or 30 day basis though the authors concur in a previous study that similar lesions do impair the perception of emotionality (2002; 2001). Lesions to the central amygdaloid nucleus (CeA) produce pro found deficits in CRH circ ulation and, consequently, diminish the release of circulating glucocorticoids (Gray, 2007) While most of the efferents from CR H neurons of the CeA converge with other tracts into the BST which in turn connects directly to the PV N, there are several second order means by which the
11 CeA modulates the PVN Some of the projections from the CeA emit CRH directly onto the la teral hypothalamus, a center which when stimulated electrically elicits alertness mediated via orexin, increased a ppetite, and attenuated nociception ( Hagan et al, 1999 ; Delgado and Anand,1952 ; Naleway, 2011) The action of catecholamine on the lateral hypothalamus has been shown induce anorexia (Leibowitz, 1975; Berridge and Valenstein, 1991). Other projections of the CRF positive CeA include the medial and lateral parabrachial nuclei (mPBN and lPBN) mesencephalic nuclei of the trigeminal, the dorsal vagal complex, mesencaphalic reticular fo rmation, and the periaqueductal gray (Gray, 2007); functionally, these area s represent the tran sduction of taste, the coordination of mastication, the parasympathetic innervation for the th oracic cavity, a center essential for maintaining general arousal, and a complex center for regula ting nociception and stereotyped defensive b ehaviors. The CeA receives afferents from CRH ne urons in the hypothalamus and throughout the midbrain, dopamine (DA) tracts serotonin (5HT) from their respective nuclei, and substance P innervation from the brainstem (Gray, 2007) su ggesting roles for HPA feedback loops, satiety regulation, environmental monitoring, and noc iception. Most authors are careful to delineate at least three differentiable sources of HPA/HPI activation : 1) stressors which elicit pain/discomfort, 2) those which alter direct l y physiolog ical conditions, and 3) those which necessitate the assembly of multiple sensory experiences and rely on memory to associate the stimulus as a threat. The first pathway is necessarily interconnected by the input from the anterolateral sy stem and, indeed, th e PVN receives afferents from the spinotha lamic and spinoreticular tracts,
12 also sending analgesic signals to the dorsal column upon electrical or glutamate stimulation ( Palkovitz et al, 1995; Rojas Piloni et al, 2008) This signal may be further modulated by the amygdalar connections to the trigeminal ganglion CGRP neurons ( ess ential in the transmission of nociception from the facial nerves) as well as extensive innervations with dynoprhinergic ne urons (an endogenous source of opioids) but is important to note that experimental pain still effects sympathetic c hanges in rats with ablated CeA (Gray, 2007) Thus limbic input is not ne cessary for basic allostatic adaption to nociceptive stimuli but instead merely modifies it (Gray, 2007 ; Herman and Cullinan, 2007 ) The second source of activation comprises brainstem tracts to the hypothala mus suff icient to create a full fledged neuroendocrine response, as evidenced by various lesion studies ( Herman and Cullinan, 1997; Buijs and Eden, 2000; Van de Kar and Blai r, 1999 ). Regardle ss, it has been shown that limbic regions, such as the amygdala and neocortex play a significant role in modulating this response, especial ly in attenuating HPA /HPI activation over time in accor dance with stressor magnitude (Gray, 2006 ; S nyder et al, 2011 ). The primordial necessity for the regulation of the internal physiological state is intimately connected to the HPA/HPI axes, as is evidenced by the architecture of this ancient pathway. The third sourc e of activation is due to the integ ration of sensory signals and the distr ibution of integrated to retention centers for storage and subsequent retri e val. This may lead to response s to complex stressors such as psychosocial interactions, changes in environmental conditions, perceived prospe cts, and prediction of future outcomes. The production of s uch complex stressors relies on telencephalic limbic structures such as the hippocampus and amy gdala in order to integrate various inputs often associating the noxious and benign through pa ttern r ecogni tion as is accomplished in instrumental conditioning ( Van de Kar and Blair, 1999 ).
13 Lesion studies reveal that the BLA of the amygdala and CA1 area of the hippocampus are esse ntial for successful avoidance conditioning as well as a special reliance on the prefrontal cortex ( PFC ) in mammals ( Herman and Cullinan, 1997 ; Kim et al, 2006). Saha and Datta further affirm the roles of these limbic structures in learning by demonstrating a profound increase in CREB in the hippocampus, amygdala, and hypothalamus upon successful avoidance training (2007) Sim ilarly there is significant evidence that acute HPA activation is reciprocally engaged with hipp ocampal neurogenesis (Brown et al, 2007; Jacobsen and Sapolsky, 1991; Snyder et al, 2007), though there is confli cting evidence about the role of adult neurogenesis in learning (Leuner et al, 2006). Altogether, the CeA clearly plays a paramount role in the assimilation of various se nsory data as well as the recall of past experiences, whereas the BLA proves essential for consol idating such memories for later reference and preventing inappropriate excitation of the CeA via GABAergic inhibtion (Poulos et al, 2009) In turn, the hippocampus dually serves to mitigate HPA/HPI as well as other specific functions of memory a nd recall. It is apparent that these limbic organs are critical for both acquisition and extinction of the allostatic state due to complex stressors (Tronson et al, 2002; Feinstein et al, 2011) The effect of acute corticosteroid elevation on cognition is complex, but appears to be strongly dependent upon dose and duration (Lupien and McEwen, 1996). The two receptors of glucocorticoids present in the CNS of mammals, the glucocorticoid receptor (GR) and mineral ocorticoid receptor (MR), have different affini ties and different distributions. The MR, structurally similar to the MR present in the kidneys that binds aldosterone, has an approximately tenfold higher propensity to bind cortisol/corticos t erone than the GR and in the CNS, is limited to the hippocampu s, amygdala, and lateral septum, with smaller distributions throughout. The GR is predominantly located in the hippocampus but finds residence fairly ubiquitously in the CNS of all vertebrates (Denver, 2009) The effects of binding these receptors are bot h immediate and
14 long lasting, leading to genomic changes after a GR is brought into the nucleus and interacts (typically promoting) genes proximal to a glucocorticoid response element (GRE ). (Herbert et al, 2006) Glucocorticoids, in addition to all other n eurosteroids (Majewska, 1992), allosterically modulate GABA A Rs in both a positive or negative manner, dependent on their sulfation and co ncentration (Majewska 1985 ). During positive modulation, this is essentially the same mech anism of action as benzodiaz epines and an immediate anxiolytic phenotype does indeed occur with the exogenous administration of various corticosteroids in fish and mammals (Cerda Reverter and Canosa, 2009; Denver, 2009; Maninger et al, 2009; Lambert et al, 2009; Mellon and Griffin, 2 002). This topic is covered in greater depth in the coming section regarding the acute actions of allopregnanolone in the CNS.
15 F IGURE 2 : T HE HPA/HPI AXES ME: MEDIAN EMINENCE NPO: NUCLEUS PREOPTICUS PVN: PARAVENTRICULAR NUCL EUS CBG: CORTISOL BINDING GLOBULIN
16 HPA /HPI Regulation and Dysregulation During periods of homeostasis there are multitudinous regulators that hold tight rein over the HPA/HPI/SAM sys tems given their functional overlap with daily activities of both b eha v ior and metabolism. Circadian rhythms operate both independently of and are modified by HPA/HPI diel cycles (Buckley and Schatzberg, 2005 ; Azpeleta, 2010 ). It has been shown in rats th though the SCN itself lacks glucocorticoid receptors (Herbert et al, 20 06). Feeding has also been shown to raise glucocorticoid levels in many animals especially fish and corticosteroids in sy nergistally increase insulin secretion and potency (Van der Boon et al, 1991; Dallman et al, 1993). Various disorders can modulate HPA activity as well. Sleep disorders and obesity are the central points for modern researchers (Buckley and Schatzberg, 2005; Madden and Morrison, 2009) It has also been demonstrated that epigenetic factors, such as early maternal care, can produce lifel ong alterations to the typical MR/GR profile and lead to predispositions toward d epression and anxiety disorders ( Herbert et al, 2006; Lemaire et al, 2000 ) It has been repeatedly shown that many individuals suffering from major depression and dementia exh ibit an endurin gly suboptimal elimination of corticosteroids from circulation, especially when challenged by an acute stressor such as dexamethasone (Bhagwagar et al, 2002; Barden et al, 1995; 1993 ; Pariante and Lightman, 2008 ) Indeed, even gender is complicit in corticosteroid regulation (Elakovic et al, 2011 ; Ku msta et al, 2007 ), with estrogen at the forefront of negative feedback antagonism through the ER of the hypothalamus (Weiser and Handa, 2009) This reveals the innate importance of sex ste roid regulation of the HPA/HPI axes and elucidates part of the role of metabolic syndrome in producing persistent feed forward derailment given the strong correlation between hyper e-
17 strogenism and obesity (Mattson et al, 2009; Cooper and Steward, 20 09; Madden and Morrison, 2009 ). Bruce McEwen considers prolonged periods of allostasis to be at the root of various ps ychiatric conditions, particularly those of chronic anxiety and depression, citing the hypertrophic pattern in the amygdala and trend towa rd smaller hippocampal and to a lesser extent, pre frontal cort ical volumes of those suffering post traumatic stress disorder (PTSD), major unipolar depression (MD), and chronic anxiety (McEwen, 2004). McEwen, originally famous for his co ntributions to th e Glucocorticoid Cascade hypothesis of accelerated aging (Sapolsky et al, 1986), posits glucocorticoid excess to significantly contribute to cellular damage and a specific pattern of metabolic influence that is defined by the upregulation of inflammatory c ytokines, ox idative stress, and SAM activation leading to a manifestation of a portrait of chronic inflammation, me tabolic syndrome, hypertension, immunosuppression, failure to thrive, and loss of fertility (McEwen, 2009 ). These changes are very similar to a Cushingoid state, a preclinical constellation of findings related to chronic hypercorticosteroidemia though rarely a precursor to the actual Cushing s syn drome which is typically precipitated by a ACTH hypers e cretory adenoma (Wo lkowitz et al, 2009) T standing as many HPA axis abnormalities are observed in these patients but the data is imprecise and ps ychiatry is far from employing biomarkers for diagnosis ( Kraemer et al, 2002; O ) dementia MD, anorexia nervosa, and PTSD, there is no clear correlation for any specific b iomarker and the prevalence or severity of these disorders (Kaye et al, 1987; Gold and Chrousos, 2002; Brien et al, 1993 ) It is proposed by Schuder and his colleagues that varying degrees of
18 adrenal insufficiency affect many psychiatric patients, particularly those who have suffered traumatic episodes, some of whom he has treated effectively with regular hydrocortisone treatments (2005). At least some subsets of PTSD and MD have been shown to respond to r epla cement hydrocortisone therapy ( Holsboer, 2001), but it must be noted that hypercort iso lemia is not always comorbid with these disorders (Penninx et al, 2007; Pariante et Lightman, 2008) nor that cortisol is necessarily the best candidate measure for hyperactivity of the HPA axis given its inhibitory effect on the release of CRH and ACTH in a normal organism and its phy siological role as a blunting effect to the acute stress response (Herbert et al, 2006). Others suggest that CRH is a better indicator of HPA impairment The exogenous admi nistration of CRH p e r cipitates depressive symptoms and there has been some success in the use of CRH 1 R receptor antagonists in pilot studies on rats (Holsboer et al, 2008) O ther s advocate the use of the dexamethasone suppression test (DST) relative to plasma CRH as a useful predictor of modern SSRI response (Ising et al, 2005). Florian Holsboer is the leading proponent behind the development of antidepressive and anxiolytic drugs modeled after modulation of the H PA axis hormones in the CNS as well as the among the first to utilize the DST in depressive patien ts (Holsboer, 200 1). H e postulates that a combination of impaired CRH 1 R sensitivity, mineralcort icoid (MR) and glucocorticoid (GR) mediated HPA inhibition, and generalized in ability to atten uate HPA activity is central to the etiology of many diverse psychiatric disorders The sharpened edge of his argument lies on the diverse nature of antidepressant drugs that are only united by a fairly consistent clinical profile rather than receptor specificity or biological activity; he posits HPA dysfunction as the underlying cause in many diverse psychiatric conditions and HPA mod ulation to be the overarching modus operandi of modern antidepressants.
19 While the results of these DST are fascinating, their conclusions are com plicated by the facilitation of multi drug resistance (mdr) by p glycoprotein (Pgp), a n exporter of exogenous and some lipophilic endogenous compounds back into central circulation from the rather selective blood brain barrier ( BBB), which rapidly removes DEX, other steroids such as DHEA, and even endogenous cortisol in the human and rodent brain (Karssen et al, 2001 ; Asaba et al, 2000 ). Since brain concentrations of cortisone and 11 dehydrocorticosterone considered to be the i nactive 11 keto steroids of cortisol and corticosterone respectively, are not affected by the pre sence of a Pgp inhibitor, this raises the possibility that the neural actions of glucocorticoids are regulated by the presence of the enzymatic conversion of these metabolites at the CNS loc i of their activity (Karssen et al, 2001). This hypothesis is supported by the widespread CNS distrib ution of 11 hydroxys teroid dehydrogenase ( 11 HSD) type I particularly in the hippocampus, cerebellum, and cortex of mammals (McEwen, 2007; Herber t et al, 2006) Other investigators have traced similar distrubtions of this enzyme in zebrafish, coupled with a similar system for the regulation for the BBB (Sakamoto et al, 2001; Miller et al, 2002; Eliceiri et al, 2011). These findings are intriguing, but offer no hint toward the etiology of HPA dysfunction: does psychiatric disease precede or follow HPA dysregulation? The best evidence comes from F IGURE 3 : DEX/ACTH C HA LLENGE D IAGNOSTIC T HE ADMI NISTRATION OF DEX, A POTENT GLUC OCO RTICOID WITH HIGHER MR/GR AFFI NITY THAN CORTISOL BLUNTS THE R ESPONSE TO CRH IN HEALTHY VOLU NTEERS BUT DEPRESS ED PATIENTS EXHIBIT EXAGGERATED ACTH RELEASE LIKELY DUE TO EITHER CRH 1 R DENSENSITIZATION AND / OR DYSFUNCTIONAL REGULA TION OF THE NEGATIVE CORTICOS TEROID FEEDBACK EFFECS OF DEX. F URTHER FINDINGS DO IN FACT REVEAL AN IM PARIMENT OF DEX AND EXOGENOUS CORTIS OL RESPONSE ONLY IN DEPRESSED PA TIENTS (A DAPTED FROM H OLSBOER 2001)
20 recent observations regarding the feedback mechanisms of the HPA/SAM system and their changes over time du ring prolonged states of allostasis. Various changes to gross and ultrastructural anatomy of the hippocampus as well as electrophysiolo gy have been observed : decreased hippocampal volume, significant loss of sy naptic spines in all but the CA2 region of the hippocam pus substantial loss of hippocampal fun ctions such as memory and spatial learning, potentiated connections to the lateral habenula from the ventral tegmental area (VTA) and the downregulation of GRs in the hippocampus and hyp othalamus ( Li et al, 2011; Hajszana et al, 2008; Song et al, 2006 ; Buwalda et al, 2005 ; Buwalda et al, 2000 ). Benedict and colleagues also observed persistent inhibition (averaging four weeks) of LTP in the hippocampus in a model of learned helplessness, inescapable foot sho ck (2009). This effect was not affected by genetic strains supposed to be predisposed to depression nor was the helpless behavior displayed correlated with the magnitude of the s uppression of normal LTP ( Ryan et al, 2010 ) S imilar designs have shown that t his effect is prevented via the administr ation of allopregnanolone, an endogenous neurosteroid that acts centrally as an anxiolytic and neurogenic agent directly to the hippocampus or amygdala (Shirayama et al, 2010 ). Other e xperiments have shown the atte nuation of hippocampal neurogenesis due to inescapable stress to be averted with a dosing schedule of fluoxetine (Malberg et al, 2003) The potentiation of the VTA lateral habenular connection has been correlated with i ncreased helplessness, anxiety, stres s, pain, depression, and acquisition of avoidance learning ( Hikosaka et al, 2008; Thornton and Bradbury, 1983; Sutherland, 1980). The lateral habenula is thought to be the functional center of aversive motivation, though its ablation only produces significant deficits in fear conditioning in scenarios of high stress (Thronton and Bradbury, 1983). The electrical stimulation of this region profoundly suppresses the raphe nuclei via a G A-
21 response (Wang and Aghajanian, 1977; Hong and Hikosaka, 2008). Lateral habenular activation also acts through GABAergic transmission to utterly (98%) and transiently (~85ms) terminate dopam in ergic response at the loss of an exp ected reward (Huifang and Shepard, 2007 ; Hong and Hikosaka, 2008 ; Matsumoto and Hikosaka, 2007 ) The vast majority of lateral habenular affe rents are from the VTA and approximately 50% of the se neurons are dopaminergic, perhaps ind icating a role in the tem porary suppression of 5HT and NE (Gruber et al, 2007) I ndeed, dop aminergic stimulation of the habenula proper transitively suppresses 5HT release in the cat (Reisine et al, 1982). The purported role of lateral habenular signalling in the recognition of activates the lateral habenula it inhibits 5HT NE and DA release, and thus limits plasticity and the esta blish ment of the present behavioral paradigm. The inhibition of t he dopaminergic lateral haben ula via tryptophan depletion in rats and deep brain stimulation in humans has been shown to amelio rate the s ymptoms of treatment resistant depression (Sartorius and Henn, 2007; Sartior ius et al, 2010). Two entrepeneuring scient ists have even gone so far as to patent the use of zebrafish for screening cerebral drug efflux from the BBB due to the extensive similarities to the human system (Goldsmith and Fleming, 2007). Since the mechanisms of inhibition are GABAergic in nature, th is implies that negative modulators of the GABA A R may act as ant idepressants. DHEA, one neurosteroid with such activ ity, has been demonstrated to be an effective therapy for some subsets of MD (Wolkowitz et al, 1997) It should be noted that the opposite effect of GABA A R allosteric potentiation may also improve depressive symptoms, as is evidenced by the effects of potent GABA A R modulators such as benzodiazepines a nd allopregnanolone (Zhao et al, 2011; Pinna, 2010). Th ese paradoxical findings suggest that GABA A R modification does not contribute significantly to the antidepre s-
22 sant activities of these drugs. Instead, it is possible that genomic modficiation mediated through neurosteroids that alter hippocampal dynamics ma y be responsible for recovery from the state of chronic stress I t is well known that many drugs known to improve depressive symptoms u pregulate the peripheral benzodiazepine receptor ( TSPO ) which in turn increases the synthesis of neurosteroids (Pinna, 20 10) Both allopregnanolone and DHEA have been demonstrated to have genomic effects in the hippocampus altering axon guidance, neurite outgrowth, neuronal cell cycle, neurodegeneration, inflammation, steroid receptor synthesis, and neurogenesis (Rebala, 201 1 ; Mo et al, 2009 ) A recent hypothesis has been forwarded regarding the regulation of glucocorticoid r eceptors in the hippocampus during times of allostasis ( He rbert et al, 20 06 ). T he extraordinarily high density of MRs and GRs has been cal culated to be in at least mammals according to a recent in situ analysis (McAuley et al, 2009) In other words, all somatic cells express MR/GRs and their binding leads to the removal of circulating glucocorticoids but the hippocampus holds a unique place in this regulatory schema due to its presence in the CNS, its high metabolic rate, its preemin ent density of corticosteroid receptors, and its capacity for adult neurogenesis. Since prolonged exposure to allostatically elevated gl ucocorticoids has been shown to prevent mitosis of neurogenic cells, halt maturation, and induce atrophy the hippocampu s seems unusually vulnerable to circulating glucocorticoids and this mechanism of atrophy is not ostensibly adaptive (Kassahn et al, 2009; McEwen, 2009), There is also increasing evidence that adult hippocampal neurogenesis plays less of a role in learnin g than was once thought (Kerr et al, 2010; Snyder et al, 2011 ; Leuner et al, 2006 ). This is concurrent with the finding that most antidepressants increase mRNA levels of both MRs and GR, increase neurogenesis, and upregulate the production of neurogenic/ne uroprotective
23 steroids (Barden and Reul and Holsboer, 1995; Jacobs et al, 2000 ; Pinna et al, 2009 ) Santerelli and collegues have demonstrated that both the 5HT1A receptor and hippocampal neurogenesis are essential for antidepressant response (2003). In a brilliant recent study by Snyder and his compatriots, glial fibrillary acidic protein (GFAP) promoter was used to produce herpes simplex thymidine kinase in the radial glia precu rsors to adult neurogenesis in mice who were then treated with the antiviral valganciclovir an acyclic nucleotide analogue that destroys only mitotic cells expressing viral thymidine k inase (2011) These transfected mice were no different than controls in any measure unless they were treated with the antiviral, which was demonstra ted to abolish adult neurogenesis in the hippocampus via doublecortin histology. A series of commo n tests for depressive behavior and anxiety were performed ( novelty suppressed feeding, forced swim, sucrose preference eleva ted plus maze ); those animals wi thout neurogenic hippo campi exhibited greater a n hedonia, d epression, acute elevations of corticosteorids but did not show increased anxiety as indicated by the elevated plus maze Finally, a DST as well as exposure to a restraint stressor demonstrated tha t the neurogenically impaired mice were much slower to eliminate their plasma cort icosterone These results were repeated in another model of annihilation of hippocampal neur ogenesis employing X radiation over a limited portion of the telencephalon designe d to only a blate the dentate gyrus; very similar methods in another radio ablation study verify these results, strongly affirming that hippocampal neurogenesis plays a central role in attenuating the long term effects of HPA activation through the endocytosis and of circulating glucocorticoid via e ndocytosis and degradation (Santarelli et al, 2003) These findings corresponds well with the notion that hippocampal neurogenesis may be implicated as the modus operandi of antidepressant drugs. Since it is well established that signi f-
24 icant HPA axis dysregulation occurs in depressive illness (Barden et al, 1995); that neurogenesis plays a central role as a negative feedback (Snyder et al, 2011); that hippocampal function is o ften impaired by depression chronic stress, and glucocorticoid elevation ( Bhagwagar et al, 2002; Burke et al, 2005; Het et al, 2005; Holsboer, 2001 ); it is unsurprising that the mice require both the 5HT1A receptor (which stimulates neuroproliferation) and intact hippocampi to experi ence the antidepressant effects of fluoxetine (Santarelli et al, 2003). Furthermore, the antidepres sant response has been correlated with return to baseline neurogenesis across many different drugs (Malberg et al, 2000; Duman et al, 2001; Malberg, 2004 ). Graziano Pinna began to recognize the importance of hormonal signals for antidepre ssant effects due to drugs (e.g., desipramine, lithium, paroxetine) and sleep deprivation, identif ying increased glial prevalence of T 3 in treated rats and a strong upregulat ion brain derived T 3 during HPA allostatic induction (Pinna et al, 2003; Baumgartner et al, 1998). Pinna was the first note that the clinical doses of fluoxetine almost certainly lead to plasma concentrations of the drug that are inadequate to modify the s erotonnin reuptake transporter (SERT) but are suff icient to stereospecifically and positively modulate the 3 steroid dehydrogenase (3 HSD) to upregulate the production of allopregnanolone from progesterone (200 5 ). Impressively, he shows that only the s e nantiomers have significant action on 3 HSD, with s norfluoxetine b eing the most effective at decreasing aggressive responses and increasing allopregnanolone, d espite the non stereospecific regulation of SERT modulation. Later work correlates the normaliz ation of the allopregnanolone synthesis after excessive androgenic steroids use or models of PTSD is correlated with both behavioral recovery and the upregulation BDNF (Matrisciano et al, 2010; Nin et al, 2012; Pinna, 2010). Pinna has demonstrated that the infusion of s norfluoxetine into the rat BLA is sufficient to increase local allopregnanolone synthesis and attenuate aggre ssion in socially iso lated mice, though this effect did not occur in microinfusions into the PFC or
25 nucleus accumbens (Nelson and Pi nna, 2010). These results suggest the sustained GABAergic modulation via allopregnanolone in BLA may attenuate activation of CeA /VTA during recalled arousal due to complex stressors, a mechanism of anxiolysis independent of hippocampal fun ction. Waner Schm idt and Duman have also found evidence for the importance of growth hormones in neurogenic based antidepressant response, but instead identify VEGF as an esse ntial signal in two different models of depression, learned helplessness and UCMS (2006). The auth ors conclude that these behavioral effects likely through the VEGF Flk 1 cascade due to their complete prevention and/or extinction of fluoxetine or des ip ramine response by the Flk 1 antagonist SU5416 Furthermore, they indentify an in vitro response in subventricular zone (SVZ) neural prgenitors. The genomic actions of the steroids may explain the upregulation of growth signals like VEGF and BDNF (Rebala, 2011). The ratio of MRs to GRs both within the hippocampus has been shown to be corre lated with stress coping styles, depressive/anxiety symptomology, early childhood experiences, and impaired glucocorticoid disposal (Veenema et al, 2003; Lopez et al, 1997; Rozeboom et al, 2007; Oitzl et al, 2010; Herman et al, 2006 ). Sapolsky proposes in the classical Glucocorticoid Cascade hypothesis that the downregulation of M Rs in the hippocampus following prolonged or exce ssive glucocorticoid exposure leads to the degenerative changes and thus a higher MR/GR ratio is protective against stress (Sapolsk y et al, 2002). It has since been shown that mice transgenically modified to express a mild upregulation of MRs in the forebrain do indeed exhibit less anxiety like behavior, express fewer GRs in the hippocampus, and display additional limbic 5HT 1A Rs (Roze boom et al, 2007). In rats exposed to unpredictable CMS and in the post mortem brains of suicide victims, the opposite pattern is ex pressed: the decreased binding and expression of
26 5HT 1A Rs, hi gher basal corticosterone and the downregulation of MRs in the h ippocampus (Lopez et al, 1997). While imipramine the first antidepressant drug, prevented these effects in rats, fluoxetine was shown to actually increase corticosterone levels over that of the control while simultanously improving 5HT 1A response (Lopez e t al, 1997). It has also been observed that the MR and GR can heterodimerize intracellularly, with the effect of inhibiting the transpritional a ctivity of GR; this represents a direct method for the suppression of GR induced transcriptional modification wh ile MRs are not fully saturated, though this is only possible in cells which express both MRs and GRs (Liu et al, 1995). In the common carp, the downregulation of the MR due to chronic hypercortisolemia is concomitant with the equivalent downregulation of GR mRNA (). The dynamics between neurogenesis, the MR/GR ratio, and 5HT 1A receptors in the hi ppocampus all participated in common negative feedback loop for the HPA/HPI axis, though it is also clear that this inhibition becomes overridden over time s of prolonged allostasis Cortisol in Human s and Fish Cortisol represents a unique point of convergence for most teleosts and most mammals : while there are some exceptions (e.g., rodentia), both employ cortisol as the principle regulator of glucocorticoid mediated HPA/HPI activation (Mommsen et al, 1999). These functions do overlap substantially, but the additional importance of cortisol to as an osmoreg ulator akin to aldosterone in mammals is still poorly understood, notably because it is a less attractive theory to hypothesize about than to pander to the strong motivations to employ fish as experimental models of neuroimmunoendocrinology (Blaser et al, 2010; Steenbergen et al, 2011; Green et al, 2011; Popesku et al, 2008). This review will inevitably reflect this bias in research volume. Cortisol is the glucocorticoid secreted by the adrenal glands of mammals and the interrenal glands of teleosts in res ponse to stress, of any of the previous three varieties
27 discussed The effects are pronounce d and universal: every somatic cell in the body expresses MR s /GRs and the resulting interactions result in increased energy mobilization (e.g., gluconeogensis and l ipolysis) and increased metabolic rate (favoring catabolism) as part of overall sympathetic activation. It is the culmination of sympathetic response and direct activation of MR s /GRs in the amygdal a and hippocampus that causes a complex profile of ef fects on memory, learning, and cognition (Buijs and Van Eden, 2000) These chang es, which have been shown to have to act as pronounced modulators of cognitive function, are mediated by variables including the length of cortisol exposure, titer of free circulatin g cortisol versus bound cortisol and the measure of cognition; this section will tease apart the abundance of data on cortisol in both humans and teleosts with a special focus on the fear conditioning or avoidance paradigm Cortisol is the often consider ed the utmost hormone responsible for stress adaptation and, less exclusively, wakefulness and attention The free circulating end product of the HPA /HPI axis, cortisol is elevated as the result of a cascade of interactions stemming from either a stressful event or as a diel process which climaxes during peak wakefulness and postprandially. When stress is tranduced by either the limbic system, in cases of social or emotional di stress, or brain stem nuclei, when the stress is of immediate threat to homeostasis such as thermoregulatory challenges, an afferent arrives at the PVN of the hypothalamus which stimulates the secretion of corticotropin releasing hormone (CRH) and arginine vasopressin ( A VP ) into the pituitary portal in mammals or portal vei n fish The cells of the anterior pituitary then secrete adrenocorticotrophic hormone ( ACTH ) in response, which circulates until it acts on the adrenal glands to stimulate cortisol release in a pulsile manner Cortisol may then act on any somatic cell in t he body when it not bound to a corticosterone binding protein (CBP) as previously described, and provides its own negative feedback to the hypothalamus and pituitary
28 gland, eventually terminating its elevation and returning to homeostasis after the stress has been resolved. Other negative feedback is provided by the hippocampus, the BST, and the preoptic area, which innervate the PVN with GABAergic synapses. (Herman et Cullinan, 1997) In medicine, t his steroid has been of particular interest because o f its gross malfunctioning, its involvement in aging or stress related general changes, and its role in mental health. Chronic hypercortisolemia, also known as Cushing's disease, is indicated by hypertension, fatigue, hyperglycemia, increased liquid intake irritability, loss of libido, depression, and severe memory deficits. In over 70% of the cases, this extreme malignancy manifests itself after the formation of a pituitary adenoma, a neoplasm derived from the cells of the anterior pituitary that secrete ACTH (Loughlin et al, 2005). The oppos ing syndrome, hypocortisolemia, adrenal insufficiency is less common and less studied. Even subclinical chronic hypercortisolemia is associated with poor prognoses: severe deficits to memory, i ncreased oxidative stress, and widespread atrophy of muscle are some of the principle elements of stress induced aging (Liu et Mori, 1999 ; McEwen, 2009 ). The discussion of cortisol as an aging accelerant has been ongoing since the inception of the free rad ical theory of aging and has in fact generated its own unique theory of aging that relates the decreased capacity of geriatrics to recover homeostasis from challenging events (Holliwell et Gutteridge, 1986; Sapolsky et al, 2002). Studies in the elderly su ggest that lower cortisol is not only correlated with longer life, but healthier neurological outlooks: aerobic fitness, the exertion of which is well established as cortisol depleting, was shown to correspond with significantly larger bilateral hippocampa l volume and increased spatial skills (Erickson et al, 2009). L ater, it was found that higher levels of BDNF, a neurotrophic factor with a diel cycle similar to cortisol but with significant decline as one ages, correlates to a higher preservation of hippocampa l
29 volume (Erickson et al, 2010). A nother longitudinal study with elderly participants concluded that higher cortisol salivary titers predicted decreased declarative memory function, which is intimately linked to the hippocampus, and poorer execut ive abilities after three years (Li et al, 2006). The study of hippocampal volume does not necessarily correlate to function, however, and a recent study does well to expound the high rate of variability in both young and old populations, casting a shad ow of doubt on the often accepted theory of corticosteroid induced aging (Lupien et al, 2007). No doubt, much of the research on cortisol has been focused on the degenerative effects its chronic elevation on the hippocampus, a brain structure essential fo r spatial and temporal cognition, declarative memory, and normal affect (Kandel et al, 2000). Aside from the integral role of the hypothalamus and pituitary gland in the glucocorticoid cascade, the hippocampus was the first brain region with observed pheno type variation due to stress as it contains the highest concentration of adrenoglucocorticoid receptors of the high affinity variety (MR ) in the entire CNS in higher vertebrates (McEwen, 2007). The hippocampus is thus the most cortisol and, more importantly, its activity acts to attenuate the release of CRH and A VP in the PVN (Herman et Cullinan, 1997). Interestingly, the decline in hippocampal function serves as a feed forward mechanism in cases of chronic hypercortisolemia, o ften considered to render the disease d state. The mechanism responsible for this has been dually implicated as direct neurotoxicity to pyramidal cells during periods of high concentrations and chronic inhibition due to milder doses leading to gradual apopt osis of inactive cells (Wolkowitz et al, 2009). Recent interdisciplinary work has produced a mathematical model of the HPA axis and its probable effects on the hippocampal CA1 region, which is particularly prone to atrophy when inhibited by cortisol; the m odel ultimately concluded an approximate four fold decrease in hippocampal activity due to chronic stress
30 compared to an identical aging profile with non elevated cortisol (McAuley et al, 2009). The decline in hippocampal size and function observed in many cases of chronic hypercortisolemia has important implications for mood, memory consolidation and cognition as the hippocampus plays an essential role in all of these disparate processes. The relationship between psychiatric disorders chronic stress an d hippocampal function has been a burgeoning field in human subjects given the relative ease of assigning tasks that noninvasively measure spatial and/or temporal learning yielding an abundance of evidence on the long term ramifications of stressful event s and the potential mechanisms via which such complex psychological trauma can alter physiological systems. In a study of 24 police officers who had all undergone similar trauma but only 50% of which developed symptoms of post traumatic stress disorder (PTSD) it was found that the PTSD group had significantly higher salivary cortisol upon waking and billaterally smaller hippocampi according to fMRI imaging (Lindauer et al, 2006). Furthermore the PTSD sufferers performed significantly worse on verbal rec all (traditionally correlated with hippocampal function) while this did not necessarily associate with the volume of their hippocampi, suggesting another factor in their impairment which the authors' believe may be attributed to the selective inhibition of the BLA Cortisol is known to exert some effect on memory recall, manifesting at its most severe as retrograde amnesia, through the inhibition of the BLA, though this effect has not been directly reproduced in humans undergoing a chronic stress state suc h as the members of the PTSD trial (Wolkowitz et al, 2009; Het et al, 2005). It has been shown that even in non clinical conditions such as personality is relatable to hippocampal volume, with those possessing lower self esteem and external loci of control having significantly smaller hippocampi (Pruessner et al, 2005).
31 Because of the hippocampus's role in cognition, it is possible to study its functionality indirectly through behavioral outcomes such as skill acquisition (O'Keefe et Nadel, 1978). The sh uttle box paradigm, in which an aversive stimulus (the unconditioned stimulus or UCS) is paired with an otherwise neutral stimulus (the conditioned stimulus or CS), presents as a well accepted task which most experimental animals readily learn under normal circumstances. When rats have been treated with cortisol and subjected to this test in the past, however, results have been inconclusive but suggestive of a non linear, dose dependent curve that presents the poorest cognitive outcomes at very low and very high doses (Lupien et McEwen, 1996). HSD inhibitor, or given tested later in the afternoon when cortisol is naturally very low, the administration of cortisol improved the dela yed recall (hippocampal dependent) task performance; when cortisol was given during a time of circadian regulated peak concentration (i.e., the morning), exogenous significantly impaired performance on this same task (2001). This is consistent with the mul titude of effects cortisol has on the telencephalon, notably the improvement of memory consolidation due to BLA interactions and the inhibition of retrieval via the hypopolarization of the hippocampus both effects relying heavily on dose ( Het et al, 2005) If this conclusion were true, it would cement the portrayal of cortisol as an adaptive compound to improve learning during times of relative stress. The use of teleost models, specifically the goldfish and zebrafish, has become increasingly popular for studies involving steroids and even psychopharmeceuticals (Popesku et al, 2008 ; Egan et al, 2009 ). The strong conservation of the steroidal system and homologies in the telencephalon make teleosts desirable models for preliminary drug trials, as these syst ems are in some ways more related than that of mouse and human (Salas et al, 2006): specifically cortisol takes on the same role in fish but not rodents (Bernier et Peter, 2001). Additionally, the
32 teleost forebrain presents two regions akin morphologicall y as well as functionally similar to the hippocampus and the amygdala in higher vertebrates: the lateral and medial pallia, respectively (Flood et al, 1976; Wullimann et Rink, 2002; Portavella et al, 2004; Salas et al, 2006; Northcutt, 2006). Such homologi es have also benefited research due in part to the availability of teleost subjects and several established paradigms, including shuttle box avoidance. The goldfish model also benefits from a vast body of literature regarding cortisol and stress. Since co rtisol circulation raises during peak activity in most animals in a diel cycle, the concentrations in goldfish, a notable exception, have been of interest (Bernier et al, 2001; Butler, 2005). It has shown that goldfish in fact have two near equivalent peak s and an intermediate, lesser trough in their maximal serum cortisol (Singley et Chaven, 1973). These (Noeske et Spieler, 1983). This is further complicated by t he many feedback regulators of cortisol: melatonin decreases cortisol, various hormones released while feeding increase it while starvation has no effect, infections paradoxically increase cortisol release, and sexual cycles also have dichotomous, drastic effects (Azpeleta et al, 2010; Bernier, 1988; Garina et al, 2007; Li et al, 2008; Butler et al, 2005; Fryer et al, 1992). Thus, like cortisol experiments in humans and rodents, it is important to control for variables such as the time of day, the time of feeding, and the temperature. Allopregnanlone as a M ediator of the HPA/HPI For a brief time in the hey days of steroid syntheses and mammalian sterol research, found (at first) only in the corpus luteum of ovulating females (Jensen, 1935). By 1938 allopre gnanolone was isolated from the adrenal cortex and suggestions were made regarding its assoc iations with progesterone, assertions of a possible role in gestational activity given its metabolic
33 and physiologic proximity to progesterone (Beall et Reichstein, 1938). It has since been demo nstrated that allopregnanolone increases its concentration line arly with increases in serum pr ogesterone and that hypothalamic concentrations significantly decrease gonadotrophin produ ction (Frye et al, 2011; Laconi et al, 2001). Another study found that rats injected with an all opregnanolone antiserum showed more pro nounced lordosis behavior and increased the number of oocytes released upon ovulation (Genazzani, 1995). The role of allopregnanolone in the e strous cycle is still mildly controversial, but most agree that allopregnanolone plays an important role in atten orating the depressive effects of progesterone withdrawal (Lambert et al, 2009). The principal support for this theory arises from the study of premenstrual dysphoric disorder ( PMDD), an affective disorder characterized by a unipolar depressive symptomatology and progesterone withdrawl the morbidity of which is consistently exhibited with lower serum and cerebrospinal fluid concentrations of allopregnanolone (Girdler et al, 2001 ; Rapkin et al, 1997). The hypoth esized justification for these differences bridges two disparate theories of depressive affective dysregulation: the neurogenesis theory of depression and the progesterone withdrawal model of post partum depression (Stoffel and Craft, 2004; Jacobs et al, 2000 ) Allopregnanolone has been described as the most potent endogenous facilitator of the GABA A receptor, the ligand inhibitory circuits (Brot et al, 1997; Bitran et al, 1999; Majewska et al, 1992). There have been at least two binding sites for allopregnanolone identified thus far, with binding to either site elici ting a full potentiation of chloride influx (Bracamontes et al, 2011); this is somewhat debatable, however, as some argue that the steroid is highly lipophilic and prefers a transmembrane site for GABA A R modulation (Chisari et al, 2010). Allopregnanolone and its related neurosteroids are synthesized in both glial and neuronal cells, with rece nt research implicating neurons as its pr i-
34 mary source in the CNS (Agis Balboa, 2006 ; Tsutsui et al, 2006 ). It has been shown that allopre gnanolone is regularly secreted from these cells, acting on the GABA A receptors in both a tonic (i.e., synaptic) and ph asic (i.e., extrasynaptic) manner, with different subunit compositions re ndering different sensitivities to this effect (Mukai, 2007; Turkman et al, 2011). The actions of a llopregnanolone on the GABA A R, while the most potent endogenous allosteric modulator of this promiscuous receptor, not unique; all neurosteroids enact allosteric effects upon GABA A Rs (Majewska, 1986). The hippocampus is perhaps the central organ of action where it has been shown to have neuroprotective effects in response to acute glucocorticoid release as well as exhibit the gr eatest neuroprotective activity ( Lennartsson et al, 2012 ). Work in ovariect omized rats reveals that allopregnenolone improves memory in inhibitory avoidance retention but impairs acquisit ion when microinfused into the dorsal hippocampus, whereas oestradiol plus progesterone significantly impaired learning and retention (Escudero et al, 2011). The amygdala also responds to and produces allopregananolone and it has been shown that the a ntid epressant response triggerd by the neurosteroid is partly attributable to amygdalar effect, likely through a GABAergic mechanism (Shirayama et al, 2010; Nelson and Pinna, 2010 ). Understanding the phenotypic shift from the a baseline state to stress induced allostasis necessitates understanding not only the direct effects of allopregnanolone but also the role of the GABA A receptors, which are extensively versatile and ubiquitous inhibitory force s in the CNS of all vertebrates. GABA A Receptors: Structure and Function GABA has long been considered the primary inhibitory neurotransmitter of the CNS (Kandel et al, 2000 ) Its actions in neurons and some glia are regulated by interactions with the ionotropic GABA A receptor (GABA A R) family C receptor family; slower metabotropic responses to GABA are the result of GABA B receptor activation (Jacob et al, 2008).
35 While there are other fast acting inhibitory receptors, specifically the glycine receptors, GABAergic systems outn umbe r these other receptors by at least two to one in the rat (Kandel et al 2000). Modern predictions claim that GABAergic synapses account for approximately 30% of the total number of synapses in the human brain (Mhler, 2006). m, its receptors have been isolated from virtually every part of the mammalian brain, albeit in different isoforms. GABA A Rs, the most abundant and best studied of the three classes of GABA receptors, have a total of 19 A R belongs to the superfamily of ligand gated ionopores it shares a common structure with other receptors such as the 5HT 3 R, nAChR, and GlyR: it exists as a pentamer with subunits of 4 transmembrane domains (TMs), with a T ABLE 2 :GABA A R I SOFORM DISTRIBUTION IN THE R AT B RAIN (M C K ERNAN AND W HITING 1996)
36 and the TM2 portion likely lining the membrane. The structure of the nAChR is canonical of this superfamily and, unfortunately, much of what is inferred about the structure of GABA A R is derived from data on this relative while no direct characterization has yet been performed using imaging techniques (Jacob et al, 2008; Olsen et Sieghart, 2008) Despite the number of theoretical combinations (totaling 1889568, assuming that order is sig nificant, that repeated receptors are permissible, and not taking into account the several known posttranscriptional modifications), certain GABA A R isoforms are more prevalent often in definable anatomical regions. These populations are summarized in Tabl e 2 (McKernan et Whiting, 1996). evidence suggests that this single subtype might even reflect over 60% of the total GABA A R population in the human brain (Mhler, 2006). This part icular combination is composed of two hippocampus. It has been shown through synthesis and implementation/combination of subunit trimers and dimers in the membrane of Xenopus oocytes that the functional configuration of these subunits is in the order
37 This receptor, as do most subunit expressing receptors, possesses a benzodiazepine binding site that allosterically enhances activation to strengthen the flow of chloride ions intracellularly. This site has been the topic of much speculation as it is of clinical importance and not always present in receptor s with the appropriate subunits. R ecent theory concludes that the site is in fact interfacial, formed at the and Buhr, 2005). Recent evidence has shown that a codes a critical site that, when altered, eliminates benzodiazepine activity completely in nearly all subtypes (Ramerstorfer et al, 2010). Many other endogenous ligands have been discovered to modulate GABA A Rs. Zinc has also been shown to bind interfacially attenuate Cl current as long as zinc concentrations are constant, the biochemical mechanism of which is now fully elucidated (Trudell et al, 2008); it has also been shown that zinc ligands play an important role in regulating cortical plasti city, not in small part due to their effect on the GABA A R which participates in the phasic control of pyramidal output (Nakashima, 2009). There are binding sites for protons, which respond to changes in intracellular pH to maintain homeostasis, and protein kinases, which play an F IGURE 4 GABA A R S TRUCTURE ( FROM M C K ERNAN ET W HI TING 1997)
38 important role in receptor regulation (McKernan et Whiting, 1996). New ligands for GABA A Rs are constantly being tested, given their incredible pharmacological value; for example, valerenic acid (a constituent of valerian, a common h erbal sleep aid) has been shown at low doses to modulate nonspecific GABA A Rs likely at the loreclezole (an anticonvulsant) binding site; ultimately, valerenate increases chloride currents into the cell by five to tenfold and synthesized analogs have maxim ally increased this activation to a staggering 100 fold (Sascha et al, 2010). Categorizing GABA A Rs based on subunit motifs has become a useful means of characterizing the behavioral effects of different subclasses. After an extraord inary review of pharmaco logy, Hanns that target any one of the 28 known wildtype receptors (Mhler, 2006). H e states that activation have an important role in catalepsy; sensitivity and respond predominantly to neurosteroidal modulation They feature prominently in extrasynaptic parts of pyramidal cells in the hippocamp us, facilitating LTP The differences between recep tors include not only di fferent binding sites, but also operate via that the former receptors are both faster to inactivate and exhibit stronger desensitation of their post act ivation effects (Barberis et al, 2007). A specialized form of GABAergic neuron, the neurogliaform cell, is present in the cortex and hippocampus and emits GABA at an extremely slow (10 30ms) rate to extrasynaptic targets along with THDOC (a neurosteroid cl osely related to allopregnanolone) to highly sensitive GABA A paracrine, neuromodulatory IPSP s (Szabadics et al, 2007).
39 Vertebrate Neurosteroids The modulation of GABA A Rs with neurosteroids is of particular int erest given both 1. Role in regulating stress homeostasis 2. Protective role in preventing excitotoxic episodes 3. Enhancement of hippocampal and cortical plasticity These effects are accomplished by both the immediate actions of the neurosteroids as a llosteric modulators of the GABA A Rs as well as the genomic modifications due to t heir actions as ste roids. Since the GABA A R is ubiquitous in the CNS and its structure has been strongly con served in vertebrates this receptor will be the predominant topic i n the coming section. Unfortunately, the transcriptional activities of neurosteroids remain to be classified in depth and this public ation bias will define the review.
40 F IGURE 5 S TEROIDOGENESIS : THE PRODUCTION OF MA JOR STEROIDS AND THE IR CATEGORIES ( COURTESY OF W IKIPEDIA : HTTP :// EN WIKIPEDIA ORG / WIKI /F ILE :S TEROIDOGENESIS SVG ) ; NOTE : THP = ALLOPREGNANOLONE
41 The synthesis of neurosteroids is, by definition, from within the central nervous system, typically from n onspecialized (e.g., oligodendrocy tes) glial and neuronal cells, though it is important to note that many other steroids also influence brain function. Al l neurosteroids originate from pre g nenolone, which in turn is converted from cholesterol, the origin of all steroids. Pregnenolone and p rogesterone, which is a single enzymatic step away, are converted into an array of different neurosteroids which all fal l under the three categories: the glucocorticoids/mineralglucocorticoids, which are responsible for regulating stress, glucose mobilization, and ionic homeostasis; androgens, the male sex hormones, often act as selective growth factors in the CNS; and estr ogens, the female sex hormones, have anxiolytic and glucose consolidating effects. The neurosteroids discussed in the work known to positively modulate the GABA A R are allopregnanolone (also known as tetrahydoprogesterone or THP), tetrahydrodeoxycorticoster one (THDOC), and androsterone; these steroids particularly act bimodally to enhance Cl burst duration at low doses (20 behaving as a direct agonist at only slightly higher doses. Neurosteroid antagonists include pregnenolone sulfate (PS) and dihydroepiandrosterone sulfate (DHEAS). Cortisol, a glucocortic oid produced predominantly in the adrenal cortex but also locally in the hippocampus and cortex via hydroxysteroid dehydrogenase ( ) also contributes a unique role in the modulation of the GABA A R according to the same hormetic pattern (Majewska et al, 1991) As noted before, many important steroids that act on the brain originate from other sources, typically the adrenal cortex or primary sex organs, and act on the brain or undergo further processing This is a minor pathway for allopregnanlone but much more significant for cortisol despite circulating cortisol being actively exported from the brain, favoring the glucocorticoid produced locally from serum cortisone (Auchus, 2009). The sulfated forms of the
42 antagonistic hormones as well as corti sol are hydrophilic, produced in the adrenal cortex, and transported throughout the body, but the blood brain barrier effectively eliminates their transport into the brain; while this may occur to a marginal extent, it is with the aid of organic anion tran sporting peptides which have been demonstrated, for at least DHEAS, to efflux at a rate more than ten times higher than influx (Asaba et al, 2000). For the non sulfated neurosteroids, transport into the brain is less inhibited due to their lipophilic natur e (Majewska et al, 1991) T steroid dehydrogenase ( HSD) is in high concentrations in the CNS and their precursors are in fact the GABA A R antagonistic modulators implying that the local production of these s teroids is essential (Celotti et al, 1992). The binding site for all of these neurosteroids is considered to be one and the same, or at least very similar in localization and effect (Belelli and Lambert, 2005). Research done on allopregnanolone has revealed that a single asparag ine residue present on M1 of the 1 5 determines the ability for the neurosteroid to achieve modulation of the GABA A current. Given that replacing this residue with another, shorter chain nucleophile (ser ine or thr eonine ) leads to somewhat compromised potentiation, it has been shown that this residue, subunit receptor subtypes, need bind only one molecule of neurosteroid to bring about its maximal effect. (Hosie et al, 2008 )
43 F IGURE 7 : C OMMON NEUROSTEROIDS ORIGINATING FROM 3 HSD. (A DAPTED FROM M EJEWSK ET AL 1991)
44 While the binding site has been fully elucidated and neurosteroids have been shown to subunit of the GABA A R, different subunit isotypes confer different as its relatives and this insensitivity has rendered the extrasynaptic (tonic) inhibition in the CA1 reg ion of the hippocampus significantly less responsive to neurosteroids than other nearby brain areas considered very sensitive to neurosteroidal modulation, tend to express extrasynaptic GABA A the Cl influx due to neurosteroid potentiation. This effect has been determined to be a matter F IGURE 8 T HE CONSERVATION OF T HE ACTIVE BINDING RE SIDUES IN THE GABAAR SUBUNITS ( FROM H OSIE ET AL 2008)
45 of enhanced transduction rather than alternative/further binding indicating that the inclusion of the smaller size and different shape lends itself to allosteric sliding due to interstitial binding (Ooishi et al, 2007). Another contributing factor to neur osteroidal insensitivity is the activity of protein kinase C (PKC), which has been shown to inhibit allopregnanolone modulation when active and permit it when inhibited. PKC particularly has shown to produce greater susceptibility to ethanol, barbiturates and benzodiazepines and a less anxious phenotype upon genetic knock out, matching the effects of high concentrations of allopregnanolone in the WT (Vergnano et al, 2009). (Belelli et Lambert, 2005) The modulation of these receptors in vivo occurs in when local neurosteroids are upregulated due either to: 1) stress, 2) global hormonal changes, 3) drug induced upregulation of the 18kDa translocator protein ( TSPO 1 ). It has been demonstrated in many species that social isolation, defeat, a nd co rtisol exposure lead to en hanced levels of brain allopregnanolone (Zimmerberg et al, 1994). Pregnancy, pubescence, menopause, andropause, development, and menstruation are but a few of the recognized hormonal paradigms of the body in which neurosteroids ar e well represented; particularly, serum increases in progesterone can be imported to local sites of neurost er oidal synthesis and this increases output substantially (Maguire et al, 2007). Allopregnanolone has been shown to protect against ischemia and con tusive injury in rat models (Djeballi et al, 2005; Morali et al, 2011 ; Sayeed et al, 2009 ) and the TSPO has also been shown to be upregulated during traumatic brain injury oxidative stress, and neurotoxins ( Papadopoulos and Lecanu, 2009 ; Martin et al, 2009 ). The activation of TSPO a translocator protein responsible for transporting ch olesterol into the mitochondria, is considered the rate limiting step in neurosteroid synthesis. While new ligands for this receptor are only recently being developed, the primary endogenous ligand s are 1 18kD TSPO is the more modern name of the protein formerly named the periperhal benzodiaezipine receptor (TSPO) and the mitochondrial benzodiazepine receptor (TSPO)
46 not yet fully understood (Corsi et al, 2008). M any dr ugs, such as ethanol, caffeine, barbiturates, and fluoxetine, have been shown to activate PBE and increase allopregnanolone concentrations through increased availability of pregnenolone to the subsequent enzymes, 3 HSD (pregnenolone to progesterone) and 3 HSD (progesterone to allopregnanolone) (Asaba et al, 2000). Natural elevations of TSPO are especially common the hippocampus, piriform cortex, cerebellum, and cortex when mice are exposed to neurotoxins and inflammatory cytokines on an acute or chronic basis ( Chen and Guilarter, 2008 ). Drugs that agonize the TSPO prove to be potent anxiolytics with very favorable side effect profiles compared to modern benzodiazepine s (Nothdrufter et al, 2012). O thers have suggested these ligands as candidates for the treatment for affective disorders (Rupprecht et al, 2010). Wehrenberg Prange Kiel, and Rune have identified extensive colocalization of 3 HSD, TSPO and Aromatase in the hippocampi of marmosets and rats indicating a central role of TSPO in local neurosteroid synthesis (2001) Researche r s have also shown the widespread distribution of these enzymes in the CNS of zebrafish particularly correlating with zones of higher neurogenesis (Sakamoto et al, 2001). The re are multiple effects of neurosteroid s independent of the GABA A R. In the short term affected neurons display increased inhibition and on a greater timescale, trophic and neuro plastic changes are rendered to the cellular architecture (Tsutsui et al, 2011; Chisari et al, 2010; Ioannis et al, 2008 ; Ooishi et al, 2007; Mellon et al, 2002 ) The mechanism for either of these effects is not fully elaborated. Some speculate that the interplay between the GABA A R and VDAC2 channel, a voltage g ated anion channel, which when tubulin may play a role in instigating long term effects (Belilli et Lambert, 2005 ). Others seek more conventional explanations for plasticity through the effects of PKC, which frequently but contextually re nder
47 the GABA A R s insensitive to neurosteroids (Mukai et al, 2007). For example, in the supraoptic nucleus, magnocellular oxytocin secreting cells are inhibited by progesterone and allopregnanolone; d uring parturition, allopregnanolone unavailability couple d with the phosphorylation of GABA A Rs via PKC render them suddenly uninhibited and capable of releasing oxytocin during birth and indefinitely after, as long as circulating progesterone remains low (Belilli et Lambert, 2005 ; Frye et al, 2011 ). Both cortiso l and corticosterone are glucocorticoids implicated in the modulation of the GABA A R in a biphasic manner: at low concentrations they do not potentiate nor attenuate the actions of the receptor until a threshold, then increase the chloride influx in a posit ive manner and then, as concentrations continue to increase, the effects drop off again, ultimately i nhibiting the current slightly This parabolic response is common to many drugs and plays an important part in the equilibrium of alertness and anxiolysis that is controlled dynamically by the steroidal system during times of chronic stress. (Majewska, 1985) A R is especially valuable during periods of allostasis mediated by elevated glu cocorticoids. G iven excitotoxic apoptosis in hippocampal cells at high concentrations spare hippocampal cells during acute stress ( Gravanis and Mellon, 2010 ) Allopregnanolone and THDOC particularly have been implicated in this compensatory neuroprotective effect (Schumaker et al, 2004). Slightly modified versions of allopregnanolone have also been demonstrated to potently antagonize HSD type II, suggesting a direct effect on cortisol production as well (Latif et al, 2005). The net effect of progesterone (and its metabolites) during
48 homeostasis leads to slight neuronal inhibition, the secretion of myel i nating s ignals from juxtacrin ic neurons and astroglia, and long term neuro plastic inhibition (Jacob et al, 2008). One last important role for the actions of the paracrinic neurosteroids is their participation in development. There are distinct changes in neurosteroidal sensitivity (controlled partly via PKC phosphorylation, described above) as an organism ages; a mou se at 10 days, for example, is roughly threefold more sensitive to positive modulation of the GABA A R than at 20 days (Belilli et Lamber, 2005). Furthermore, GABA is essential in development as a neurotrophic and chemotaxic signal, where it may eit her depol arize or hyperpolarize depending on the cell type and maturity (Manent et al, 2005; Behar et al, 1998). It participates in the developmental processes of cell migration, patterning, and fate determination and is a principal regulator of F IGURE 9 THE I NVERTED J SHAPED EFFECTS OF GL UCOCORTICOID MODULAT ION OF THE GABA A R ON MUSCIMOL (GABA AGONIST) B INDING (FROM MAJEWSKA, 1985)
49 early migratory sig nals by means of its ability to influence calcium transients (Komuro et al, 2005). Fetal alcohol syndrome can be partly exp lained by the disruption of the GABA A R in embryogenesis with the exogenous agonist ethanol (Kandel et al, 2000). Many other psychiatr ic and neurologic disorders are associated with altered levels of neurosteroids in a similar way, due to long term shifts in neuronal plasticity and function. Other symptomatic displays of neuro steroidal abnormalities include epilepsy, schizophrenia, unipo lar and bi polar depression, Conservation of Vertebrate Neurosteroid s between Teleosts and Humans It has been shown previously that goldfish are well quipped with the enzymes necessary to produce neurosteroids such as cortisol and allopregnanolone (Popesku, 2008; see above r eview). Functionally, these neurosteroids may have dive rgent yet overlapping roles in both tel e osts and humans. Cortisol is in fact an ess ential hormone as it do ubles as both a mineralcorticoid and glucocorticoid in tel eosts, who m lack al dosterone. The dual control by cortisol as the most selecteive ligand for MRs and GRs in teleosts is indicative of an overlapping function of osmoregulatio n and stress regulation; indeed, increased cortisol is pr oduces copious urine extretion (Mommsen et al, 1999) The transition to more hypertonic waters, easily definable as a challenge to physiologic homeostasis, represents an F IGURE 10 : T HE ENZYMATIC PATHWAY FOR CORTISOL PRODUCT ION IN TEL EOSTS T HOSE ENZYMES WITH TH E GREY ELLIPSE ARE MITOCHON DIRAL ( A DAPTED FROM M OMMSEN ET AL 1999)
50 essential regulation of stres s in fish which increases circ u lating cortisol (Porthe Nibelle and Lahlou, 1974 ). Fish have been shown to possess an MR despite previous doubts (Asterb e ry et al, 2011). The teleost MR displays greater affinity for 11 deoxycorticosterone than cortisol, thou gh both GRs and and the MR of the carp have equivalent affinity for cortisol (Stolte et al, 2005). MR transcription is significantly higher in the CNS (especially in the Dl, CRH cells, and the NPO, pitu itary pars distalis) over the peripheral organs (e.g., the kidneys, gills, skin) in at least the rainbow trout and carp T his implies either a greater role of the carp MR in HPI regulation over osmore gulatory functions. During periods of chronic hypercortisolemia both GRs and MR mRNA is downregulated at the same ratio; chronically hypocortisolemic fish express one variant of the GR about twice as much as the other in Dl region of the dorsal pallium a region analogous to the hippocampus of tetrapods (Sturm et al, 2005; Stiolte et al, 2008). This theory of MR/ GR 1 /GR 2 ratio regulation of HPI allostasis has not yet mounted sufficient evidence to rival the human equivalent, but is suggestive of a similar mechanism of CNS cortisol regulation as has been pr eviously described in the human condition. It has been shown that the lone GR in most teleosts operates genomically through intr anuclear homodimer formation which binds GREs, identical to the human condition (Aluru and Vijayan, 200 8). The conservation of GREs in vertebrates is well documented ( Denv er, 2009; Chang and Hsu, 2004 ), though it is important to note teleost s alone appeared to have a whole genome duplication event some 230 mya, leading to both new ligand/ receptor types and GRE sequen ces ( Santini et al, 2008; Alsop and Vijayan, 2009 ; Asterbe ry et al, 2011 ). The full implic ation of this is far from characterized, though the preservation of HPA/HPI axis consituents is strong and many such duplicated genes have been deleted in different species ( Stolte et al, 2008 ; Alsop and Vijayan, 2008 ). For e xample, while the common carp possesses two slightly di fferent ( 95 % AA identity ) CRH hormones, the zebrafish has been shown to have only a single CRH
51 sequence ( Stolte et al, 2008; Flik et al, 2005). The conservation of HPI/HPA function is largely supported by the similar effects of exogenous cortisol on teleosts as in mammals (Van der Boon et al, 2003; Alsop and Vijayan, 2008; Aluru and Vijayan, 2009 ). There are accumulating diffe rences between human and teleost cortisol function as the topic becomes better researched (Lin et al, 2011; Asterbery et al, 2011; Flores, 2011). It must also be noted that cortisol plays a major in development of teleost fishes, with the formation of the function HPI system within hours of hatching (Roussea et al, 1999; Flik et al, 2005; Alsop and Vijayan, 2008 ). Goldfish possess a bimodal crest of heightened cortisol activity during their diel regul ation wh ich is intimately tied to their time of feeding (Noeske and Spieler, 2006; Singley and Chavin, 1973 ). An interaction between se x steroids and cortisol has also been suggested due to a potent pulse of cortisol four to eight hours prior to ovulation (Cook and Peter, 1979). While teleosts have not been shown to possess a CBG protein that regulates available cortisol in the blood, the CRH binding protein (CRH BP) is fairly ubiquitous and has been shown to be regulated through stressor exposure and steroid effects (Flik et al, 2005). DHEAS has similarly been shown to be produced in the head kidneys of fish a neurosteroid known to modulate glucocorticoid systems in other vertebrats (Bentley, 2001). The effects of cortisol or other neurosteroids on the GABA A R have not been documented in fish, though this thesis predicts an acute GABAergic function of the neurosteroid allopre gnanolone due to the remarkable conservation of the GABA A R. Its predecessor has even b een localized even in Porifera ( Chondrilla nuclea ) a species which has been shown to have all of the workings of a GABAergic nervous system: glutamate decarboxylase, GABA transpo rters, and the metabotropic GABA B receptor (Ramoino et al, 2007). The first appearance of the actual receptor is in cnidarians (including Nematostella vectensis) who share a modern version of the GABA A R subunit with organisms as distant as h umans (Hemmr ich et Bosch, 2008). This extensive co n-
52 servation is likely due to the essentiality of these genes and the receptor in development, as e videnced by the high rate of retention of all LIM homeobox genes (Srivastata et al, 2010). T he complexity of the GABA A R a nd its many facets of modulation are made sensible in light of the 9 million years of evolution that it took to achieve (Ramoino et al, 2007) The neurogenic, neuroprotective, and neuroplastic effects of allopregnanolone through genomic action may conserve d, but there is no research into this topic to date. Allopregnan olone has been shown to play important roles in development in taxa as diverse as amphibians and African lungfish (Hollis et al, 2004; Math ieu et al, 2001; Rebali, 2011), though its role in sp ecies prior to the tetrapodian actinopterygian split is presently unknown. It is well documented that fish experience life long neurogenesis in many parts of the brain (Zupanc, 2008; Cayre et al, 2002; ). Social stress or hierachal status has been shown to significantly attenuate overall neur ogenesis in the teleostean brain ( Maruska et al, 2012; Sorensen et al, 2007 ). Interestingly, social ization via electrocommunication in the electric fish Apteronotus leptorhynchus has been shown to be suffi cient to stimulate neurogenesis, not unlike the effects of ECT in humans (Dunlap et al, 2008; Scott et al, 2000). The Case for Telencephalic Homologies from Teleosts to Tetrapods The conservation of functional and structural components of the HPI and HPA are suff icient to draw many interesting parallels from studies on stress in fish to other vertebrates, but the targets of their action on cognition is complicated by intricate differences in co m parative neuroanatomy (R o driguez et al, 2007). The r e gions of interest in humans, the hippocampus and amy g dala, likely predate the split from the actinopterygii tetrapods but the morphology of the Actinopterygian dorsal telencephalon has proven to be exceptionally deviant from the preceding or following taxa, making assertions of homology particularly difficult ( Nieuwenhuys, 2009 )
53 Eversion vs. Evagination To appreciate the di stinctiveness of the everted forebrain organization, it b ecomes necessary to elucidate the how the actual mech anism unfurls in contrast to the more typical process of evag ination. Prior to either of these processes, the neural tube bulges at the future site s of the three primary divisions (forebrain, midbrain, and hindbrain) with an additional point lateral thickening to begin the formation of the diencephalon. In evagination, the telencephalon then proceeds to form the lateral ventricles through ventromedia l growth followed by dorsocaudal movement, ultimately resolved in animals with especially large telencephalons (e.g., cetaceans and hominidae) with a caudolateral shift that forms the temporal lobes. In eversion this process is initially reversed: the comm on ventricle, covered only by a thin tela chorioidea, expands dorsolaterally while the telencephalon grows ventromedially, becoming largely enveloped by the ventricle and narrowly separated from the contralateral hemisphere, joined by a pr onounced anterior commissure. These processes are juxtaposed in transverse illustrations in Fi gures 13 and 15 Recent hodological evidence indicated that there is possibly an additional ca udolateral migration with a following partial rostral component in the periventricula r zones in teleosts with large telencephalons, though this is controverted by more recent evidence (Yamamoto et al, 2007; Mueller et al, 2011). Altogether, even a simple eversion completely r everses the typical layout of the four unit pallium, jumbling the ventral pallium (VP), medial pall iFigure 11 Contrast of Teleost Eversion vs Mammalian Evagination. All four pallial regions are well defined in typical vertebrates, where as there are only hypotheses regarding their location in the A ctinopterygian associated with a new nomenclature representing the anatomical relativity of histolog ically similar regions. (Adapted from from Ito et al, 2009)
54 um (MP), dorsal pallium (DP), and lateral pallium (LP) into positions where their names no lon g er anatomically represent them. (Northcutt, 2008) The question of why the A ctinopterygii have such unique forebrains has not gone unnoticed. According to Georg Striedter and Glenn Northcutt, both prominent evolutionary neurosc ientists and the latter an expert in this field specifically, it seems likely that eversion was a solution to a morpholo gical problem posed by selectiv e pre ssures: [No one has] proposed any adult functional significance of tele ncephalic ever sion versus evagin ation. Therefore, we propose that telencephalic eversion in the ray a(Gould and Lewontin 1979) or feature that evolved not because of its own funct ional significance but because it was a mechanically expedient solution to constraints imposed by other changes in the species. F IGURE 12 A) T HE RELATIVE BODY LEN GTHS (BL) OF A REPR ESENTATIVE SAMPLE OF EACH TAXON ; B) M ATURE BL VS L ARVAL BL, NOTE THAT ONLY A CTINOPTERYGII TEND T O HAVE SIMILARLY SMA LL LARVAL BL REGARDLESS OF ADULT BL; C) A SAM PLE OF EXCLUSIVELY A CTINOPTERYGII REVEAL S A STARK NEGATIVE C ORRELATION OF I NCREASING BRAIN AND B ODY RATIO AND INCREA SING BODY WEIGHT ( ADAPTED FORM S TREIDTER ET N ORTHCUTT 2006)
55 Northcutt and Striedter survey a number of reasons for this hypothesis: the Actinopterygii u nderwent a dramatic decrease in bod y size from previous groups, approximately five times smal ler on average compared to the massive placoderms and even tinier than the more modest Sa rcoptergyii; a paradigm shift in reproductive strategy, focusing on the expulsion of high quantities of larva e without parental support, contemporaneous with the novel trend toward a consisten tly small size of juvenile stages that were no longer correlated to adult body size; lastly, it is well known that vertebrate brain body mass ratios generally increase as bo dy size decreases and the hasty shrinking of Actinopterygii body size may have resulted in a cramped neurocranium. In an analysis of 64 catfish and 21 shark embryos as well as a qualitative analysis of another five sp ecies, the investigators determined tha the volume of the olfactory or optic organs. While this does not prove their hypothesis directly, it d oes provide substantial support for this theory. (2006) Given the relationship of this unique morphology to its predecessors and the substantial evidence indicating conservation of major developmental systems with better studied animals, it is tempting to assert homologous structures. This has proven to be exceptionally difficult The Teleostei have recently been of particular interest to modern researchers for their potential as model animals for pharmacology and endocrine analysis due to the remarkable s imilarity of their neuroendocrine system to higher vertebrates (Popesku et al, 2008; Hammerschmidt, 2011; Steenberger et al, 2010; Renshaw and Trede, 2012; Gestri et al, 2011) and this has sparked d ebate over the relevancy of such a comparison, especially in preliminary drug trials with beha vioral measures (Maximino, 2010). The truth is that the exact nature of these homologies r emains uncertain, but the long disputed identity of the dorsal pallia of teleosts is advancing rapi d-
56 ly over the past year with the aid of new molecular markers after a stalemate of theorizing spanning over five decades. Teleostean Dorsal Pallium In 1963, Nieuwenhuys proposed the common neutral nomenclature for the basic nissl stain contrasting cell groups, an array of terms still in use today. The regions of common histol ogy were named purely by anatomical location: D represents the area dorsalis telencephali (the entire pallium) with its total of (at most five) major subdivisions, pars centralis (Dc), pars ventr alis (Dv), pars dorsa lis (Dd), pars m e dialis (Dm), and pars posterioralis (Dp). Further analysis has led to the christening of many additional cell u lar regions, all of which are much more controve rsial and none pertain to the interpretation of the primary pallial divisions. Th e controversy over the placement of these regions has arisen from va r ious disparate a ttempts to characte r ize this region from phylogenetic, hodological, embryogological, topological, behavioral, and genetic approaches: all of these lines of inquiry have r esulted in unique pos itions regarding the correspondence to homologous pallial divisions (the MP, LP, DP, and VP). F IGURE 13 C OMPARING SIMPLE EVAG INATION ( INVERSION ), A, TO SIMPLE EVERSION B, TO COMPLEX EVERSION C
57 This paper does not presume homology to tetrapods, but acknowledges the possibility of their existence as do the preeminent sc ientists in this field (Nieuwenh uys, 2009; Northcutt, 2006 ). The most recent evidence of four distinct pall ial divisions has been found in the zebrafish, by far the most common teleost model (Mueller et al, 2011). By infusing markers of NOS, GABA signaling, and mitosis into the m ature zebrafish telencephalon, Mueller has confirmed that Dp is derived from migratory neurons from Dl, that Dc in fact originates from a histogen ic site near the sulcus ypsiloniformi that migrates caudoventrally, and surmises that Dd likely does not exist at all in the zebrafish, instead representing the heterogenous distribution of Dc cells that cling to the paraventricular space prior to migrati on. The first finding confirms long held hypothe sis by Northcutt, Nieuwenhuys, and Gans based on the mutual connectivity of the lateral olfactory tract to the Dp and Dl of basal ray finned fishes. The latter findings are encouraging, but Northcutt expres ses concern over genera lizing from the zebrafish to the teleost forebrain in general and Mueller subsequently agrees that there is work to due in establishing these regions in more basal actinopterygians in order to co nF IG URE 14 : T HE FOUR MOST INFLUEN TIAL THEORIES FOR HO MOL OGOUS PALLIAL DIVISIO NS STAND IN DIRECT C ONTRADICTION A) P R OVIDES LEFT SIDED REFERENCE TO T HE N IEUWENHUYS NOMENCLATURE A LL OF THESE ILLUSTRA TIONS REPRESENT THE MOST RECENT PROPOSALS FROM EACH GROUP EXCE PTING B (W ULLIMANN ET M UELLER 2004), WHICH WAS DRAMATICAL LY ALTERED IN THE MO RE RECENT PAPER BY M UELLER AND A G ERMAN COLLABORATIVE TEAM IN M ARCH 2011. (A DAPTED FROM Y AMAMOTO 2009)
58 Figure 15 A new model of partial eversion based on embryonic and genetic evidence. A) Three ontogenetic stages of the teleostean forebrain; B) Adult zebrafish brain at the level of the nucleus ementia; C) Adult mouse brain at the level of the basolateral n ucleus of the amygdala (From Mueller et al 2011) fidently assert such as homologies (N orthcutt, 2011; Mueller, 2011). There is still significant doubt as to whether or not teleosts possess a region equivalent to the mammalian neoc ortex, ultimately derived from the DP. These discoveries suggest the likelihood of a simple eversion process in which DP is pushed ventrally by the overgrown VP and MP. Indeed, this suggestion is supported by the ana logous functions of these regions to the amygdala and hippocampus of other vertebrates, r espectively. (Northcutt, 2011). (Appendix II further characterizes the controversy and history of this topic essential to the establishment of fish as relevant model organisms for psychoactive compounds.)
59 Behavioral and Lesion Studie s A group from the University of Seville, Spain have been working to supplement the ho dological evidence with behavioral studies of ablation of certain regions of the pallium. S etting out to prove the analogous functions of the teleostean Dl and Dm with th e tetrapodian hipp ocam pus and amygdala, the work of these individuals has contributed greatly to the modern e nthusiasm for the use of zebrafish as model animals in behavioral studies. A contrasted summary of their work to date in zebrafish and goldfish com pared to various studies in rodents are illu strated below F IGURE 16 LEFT: LOCALIZATION OF LESI ONS IN P ORTAVELLA AND V ARGAS 200 5 ; THE BLACK ZONE REPRE SENTS THE SMALLEST LESION WHER EAS THE GREY DEMONST RATES THE BOUNDARIES OF T HE FURTHEST LESIONS RIGHT: R EVIEW OF LESION STUDIES I N T ELEOSTEI COMPARED TO SIMILAR OUTCOMES IN R ODENTIA ; H = HIPPOCAMPUS A = AMY GDALA (A DAPTED FROM P ORTAVELLA ET AL 2004; B ROGLIO ET AL 2006)
60 A representative experiment performed by Portavella and colleagues tested the effect way active avoidance paradigm, a task which is better known as fear conditioning in part due to the necessary ac tivation of the amy gdala in mammals (Portavella, Torres, and Salas, 2004). They demonstrated that all groups pe rformed equally well upon making a criterion level of accuracy and sustained at least 75% avoi dance of the noxious stimuli (electric shock) for a t least six sessions before undergoing surgery. Given that the homologous medial and lateral pallial divisions are not yet resolved, the scientists took the terms as anatomically representative: in the lateral lesion they ablated all of Dp and Dl, marginal ly nicking the side Dc in the process and in the medial surgery they removed about 50% iommissure; n = 8), sham operated (n = 8), and control (n = 8) groups were subjected to an avoidance paradigm du ring which they participated in 10 trials a day with an intertrial interval averaging about 90 s econds with a 15 second presentation of the CS, a n overlapping 5 second exposure to the UCS, and a 10 second rest in the event of a failure; a measure of behavior very akin to the Delay co ndition utilized in this experiment (see Methods). In a second experiment, the training procedure was identical with the exception of the CS terminating 5 seconds prior to the onset of the UCS, sed in perfo rmance, dropping to about 10% in avoidance that gradually increased over six sessions to about 30% less than the other groups, strongly performing at 80%. In the Trace condition, it was st as poorly as the other two groups in the previous experiment, with all three damaged groups performing avoidance behaviors about 20% of the time. The author concludes that this is further grounds for homology between these
61 structures, but the criticism stands that behavioral evidence such as this allows only for asse rtions of analogy or the similarity in function across taxa, and that the lesions employed in this study were not precise for deliberate statements regarding the function of such structures In another study the Spanish group demonstrated that Dd of goldfish may be selectively et al, 2009). Rodriguez and colleagues determined that very similar l earning reversal of trained turtles and goldfish in a closed cross maze can be effected by damage to the medial cortex and Dl, respectively; furthermore, this is consistent with previous experiments of rodents with hi ppocampal damage (2002). In an open fie ld task, modeled after the rodent standard, it was found that lesions to Dl and Dp, as compared to lesions to Dm or Dm and Dd, produced significant de ficits in performance in finding the way to an open gate, an effect further potentiated when the position of the goal was reversed (Broglio et al, 2010). Altogether, coupled with many more studies of a similar nature, the Spanish group has strongly elucidated the function of the Dm and Dl regions of the teleostean brain as functionally analogous to the amygdal ar and hippocampal formations in other taxa, respectively. Rationale for the Present Study Goldfish provide an excellent model for studies on neurogenesis and its effects on cogn ition as the majority of the fish brain undergoes extensive neural turnover th roughout adulthood (Zupanc, 2009) and the majority of trophic and hormonal factors are shared across vertebrates (Popesku et al, 2008). Since the HPI axis is regulated via a similar hormonal and anatomical system as the HPA loop, chronic cortisol treatment memory and cognitive deficits largely related to hippocampal atrophy and dysfunction. The e f-
62 fect of antidepressants has been shown to be dependent upon the presence of the hippoca mpus and improv ed neurogenesis. It is known that fluoxetine increases neurogenesis through the 5HT 1A R and the upregul ation of allopregnanolone, but still has a negative effect on fear conditioning aquisition in gol dfish. This may be due to the inadequate experience of fe ar or other memory impairments caused by fluoxetine. Allopregnanolone has been shown to acutely attenuate the acquisition of certain tasks, including those that are hippocampal dependent, likely through GABA A R modulation. The long term effects of allopreg nanolone are poorly characterized, but known to effect transcription to increase synaptic plasticity, escalate neurogenesis, and promote neuron survival. There is su bstantial evidence that allopregnanolone is produced in abundance in the hippocampus when a functional HPA axis encounters a stressor or the CNS is exposed to a toxic agent. Previous studies on goldfish demonstrate the hippocampal dependency of Trace avoi dance conditioning (Portavella, Torres, and Salas, 2004) and others demonstrate the utilit y of scototaxis as an anxiety like behavior (Maximino et al, 2010; Maximino et al, 2010). By testing the acute effects of both fluoxetine and allopregnanolone on affect it can be established which has the greater acute anxiolytic profile, potentially impli cating or ruling out this effect as a means of cognitive impairment. Through the use of both Delay and Trace operant learning par adigms, it is possible to localize general effects of cognition necessary for avoidance training compared to the specific effec ts imposed on the hippocampus. Hypotheses Scototaxis 1. Allopregnanolone will decrease scototaxis 2. Fluoxetine will decrease scototaxis 3. Cortisol will not effect on scototaxis 4. Those fish treated with cortisol and an anxiolytic will display decreased scot ota x is over those who received cortisol alone
63 Delay Conditioning 5. Cortisol treatment will effect fewer avoidances, greater escapes and failures 6. Fluoxetine will produce a deficit in avoidance acquisition 7. Allopregnanolone will delay avoidance acquisition 8. Cortisol p lus anxiolytic groups will have avoidance behavior on par with the co ntrol Trace Conditioning 9. Cortisol treatment will produce fewer avoidances, more escapes and failures 10. Fluoxetine will attenuate avoidance behavior 11. Allopregnanolone will delay avoidance acquisition 12. Cortisol plus anxiolytic groups will achieve control levels of avoidance 13. Allopregnanolone+cortisol will improve avoidance learning over fluox etine+cortisol Methods This experiment was carried out under IUCUC protocol # R 3893 in Pritzker Mar ine L aboratory at New College of Florida under the supervision of Dr. Alfred Beulig. Subjects Ninety six subjects of the species Carassius auratus auratus were acquired from Se ascape Aquaria, a local retailer. The goldfish were weighed, measured, and acclimated to labor aAll subjects were treated in accordance with IUCUC protocol #R3893, which, among other mandates, requires that all subjects receive appropriate medical care for any distress. There were several days in which the fish were not tested but instead allowed to rest and ind ividually treated with either antibiotics (Erythromycin + Muracin), copper (CupriSol ), or sea salt to manage infections such as Saprolegnia fun gus, Ichthyophthirius multifiliis, and Flexibacter columnaris. These days were instrumental in maintaining the health of these immunocompr omised animals and many exhibited increased performance on the subsequent testing days. Du ring all trials a small dose of antibiotics was used in the shuttle box itself to help control infection,
64 prevent cross contamination between tanks, and prevent the annihilation of the bacterial biofi lter in their home tanks. Husbandry and Aquaria Paramount to the success of this exp eriment was the maintenance of the health of the subjects in experimental conditions, particularly in the treatment groups in which all subjects were administered a high dose of cortisol and thus immunocompromised. Immediately upon receiving the goldfish t hey were housed in two large communal tanks and treated aggressively with a combination of erythromycin, sea salt, chloride, and heat. While many of the fish had no apparent symptoms of bacterial or parasitic infection and this treatment was thus exercised hthyophthirius multifiliis, as directed by the manufa cturer, a full four day treatment, and salt was administered at a dose of one teaspoon per 10 ga llons per day for four days. Heat was applied conservatively during water changes by replacing nd replacing it with heated, chlorinated tap water, ge ntly bringing the tanks to a temperature of approximately 33C for less than half an hour. This combination of treatments effectively removed all signs of disease in every fish, which were i ndividually examined prior to injections to ensure their health. When not participating in experimental trials, the fish were housed in eighteen 75 liter ( 20 gallon) tanks with two partitions in each, making for 48 separate compartments. The part itions are perforated enough to allow visual and chemical contact between the fish, preventing stress due to social isolation. Their feedings were scheduled daily and, on a testing day, 30 le aquarium heater placed in the central compartment of each tank. All natural lighting was barred
65 from the laboratory housing the fish and three full spectrum, high wattage bulbs were automa tically timed to set a 10:14 hour light dark photoperiod. Regular water changes were mandated to maintain the health of the fish given limited filtration, with a 25% exchange with lukewarm chlorinated tap water every eight days and a 50% exchange every eight days on a schedule so that there was a water change every four days of testing (see the experimental log). Subject group assignment was accomplished with a pseudorandom number generator inherent to behavioral tracking software. The groups were generally housed together based on drug assignment, as is necessary due to the water solubility of fluoxetine. In order to avoid hou s ing other groups in a tank with fluoxetine it was necessary to fill one additional 7 gallon tank with a single goldfish on the same dosing schedule as the other members of its group; while this sub ject did not have chemosensory communications with his compatriots, his tank was placed within inches of a non experimental holding tank, the inhabitants of which he oste nsibly interacted with visually. Location The experiment was conducted at Pritzker Ma rine Laboratory at New College of Florida. Rd., Sarasota, FL, 34243. Apparat us Eight aquatic shuttle boxes were enclosed within isolation chambers ( box The shuttle box has two chambers of equal size, separated by a hurdle that comes up to six ce ntimeters below the top edge of the box Both sides provide AC current and the fish cannot r emain on the hurdle. Two 28V lights were suspended a few inches a bove the top of the tank over each compartment and the electric shock is delivered by a metal mesh electrode connected to a stimulator set to discharge 3.8V for 500ms per second during the appropriate stage. With two
66 exceptions, the device is as described in Brush and Knaff's 1959 publication for an automated shuttle box: two lights were placed overhead, rather than on each side, and the central hurdle contained no plateau and only one sensor. Coulbourn Graphic State interfacing software (Coulbourn Instrum ents, Allentown, PA) was used to regulate the stimuli and their timing. It was also employed for data collection from the photosensor. G*Power 3 was employed for the a priori power tests; SAS 9.3 was used to generate statistical analyses. Hydrocortisone, allopregnanolone, and fluoxetine were acquired from Sigma Aldrich. Data Collection The order by which the fish are nominated for their trials on test days was determined cot otaxis trials were performed the day before dosing and two days after. The Trace and Non trace Cognitive trials were contemporaneousl y carried out over a course of sixteen days at twenty trial s per day immediately after the Scototaxis trials concluded. The water was changed after each trial for each shuttle box in order to renew oxygen sa turation and drain off pheromonal by a carbon filtration system before returning to the stock used to fill the shuttle boxes. Scototaxis recent protocol publication in Nature by Maximino et al (2010). F ish were placed in a half dark and half light tank with the central hurdle allowing passage between these chambers. After five minutes of habituation, the number of passes through from one chamber to the other is recor ded as well as the total duration in each side. Goldfish have been previously demonstrated to
67 exhibit a strong dark sided preference during the day to permit crypsis and the cumulative time spent in the light side as well as moving from one side to the other are considered indicators of fearlessness explorative tendencies, and positive affect. Crypsis, simply a te rm for hiding beha vior, plays a significant role in the defensive repertoire of goldfish in their native ecology as a means of avoiding predation. F i sh were placed in the modified boxes with one light bulb removed and an opaque pla stic sheet divding the chamber in half, one side light and the other dark The fish were all o bserved to be on the light side as the boxes were closed, allowing one to discern the position of the fish based on the number of photocell breaks. This protocol was employed before an y other in two trials (pre and post treatment a dministration) I n order to avoid the obvious confounds that would result from the use of the same skinner box and expos ure to the same stimuli (light), scototaxic trials were only performed before the intiat ion of avoidance training. Delay Learning Task This protocol is designed to test an association between an unconditioned stimulus (UCS) and a conditioned stimulus (CS) as is typical in classical conditioning. The UCS is a mild electric shock that is always preceded and accompanied by the presentation of light, the CS. Nearly all subjects learn this association over time, but their performance in avoiding the UCS is affected by a confluence of fa ctors including affect, recall and learning A fish was placed in an unmodified shuttle box and the door shut, sealing off all light and most noise. The experimental protocol proceeds in four stages: in S1, there is no stimulus for an interval of 20 seconds with a 52% chance of proceeding to S2; in S2, the lights turn on for 10 s econds without any shock at which point the fish may cross the hurdle to register an Avoid and
68 return to S1, otherwise proceeding to S3; in S3 the light and a 3.5V shock alternate for half a second at a time for 5 seconds, during which the fish may cross the hurdle to complete an E scape or fail to do so and proceed to S4; in S4 there is no stimulus and that state counts as a Fai lure, with 3 seconds elapsing before the return to S1. This routine was repeated 20 times per testing day per subject, which occurred for 16 consecutive days with several interruptions as described previously. Trace Learning Task The trace cognitive task is designed to measure the ability of the subjects to associate the CS and UCS across a small degree of temporal separa tion, employing a process presumably Since the CS is ostensibly uncoupled from the UCS in this experiment (no light is present during shocks nor does it imme diately precede S3), successful mastery of this task requires temporal awareness that has been shown in other studies to be dependent upon the lateral aspect of the dorsal pallium, a region not essential to the Delay paradigm. All of the state conditions a re ide ntical to the non trace task with the addition of a 5 second stage (S5) without light or shock b ethen S4. These trials ran concurrently with the dela y learning task on the same 16 days at the same times at twenty trials a day. Experimental Design The ninety six fish wer e divided into twelve groups of eight individuals each All groups received injections of either vehicle (coconut oil) or one of five solutions of experimental co mpounds. The fish were randomly selected from a large holding tank and assigned to a group
69 trials; an example of such a name would be CON3T representing control #3 in the trace cond ition. The fish were first tested via the scototaxis paradigm once, injected after completing the task, given one day to rest, and then tested identically again the next day with no further inje ction s. The scototaxis test was thus only sensitive to the acute anxiolytic effects of drugs, such as the benzodiazepine like activity of allopregnanolone or the actions of SERT inhibition by fluox etine. It is notable that the antidepressive effects of fluoxeti ne at clinical and experimental doses onset only after a latency period in the order of weeks, due to unknown factors that may be a ttributed to hippocampal neurogenesis ( Tome et al, 199 7; Jacobs et al, 2000 ) The same day that the scototaxis trials conclud ed the fish began avoidance train ing. Treatment Preparation All experimental compounds, excluding fluoxetine, are lipid soluble and relatively i mmiscible in water, thus oil based preparations were necessitated. According to previous work with various teleosts, coconut oil provides an excellent lipid substrate for the i.p. delivery of F IGURE 17 : T HE EXPERIMENTAL SCHE DULE E ACH SOLID BLACK LINE INDICATES A TESTING DAY WHEREAS EACH GRAY LINE IS DA Y OF TREATMENT AND R EST
70 drugs, e xtending their metabolism and prolonging duration from a few days to several weeks (Vijayan et al, 1994; Vijayan et al, 1996; Vijayan et al, 1991; Reddy et al, 1995). Since Vijayan and colleagues assert that they find intact coconut oil implants in their c old water fish up to three weeks after injection, the release of the drugs may be dependent on the melting of the vehicle which would place the expected time to be significantly shorter considering the higher holding temperatures of the goldfish in this st udy (18 C vs 23 C) and virgin coconut oil's total melting point at 24C The solutions were heated until the coconut oil was sufficiently liquefied and vo rtexed until solution was ostensibly homogenous with no visible solute. The fish were weighed prior to injection and then allowed to rest for 36 hours prior to further experimentation. The treatment groups received the following compounds, providing 16 subjects per treatment: allopregnanolone, fluoxetine, vehicle only, allopregnanolone plus cortisol, fluox etine plus cortisol, and cortisol. The fish were further divided by condition (Delay or Trace) into groups of eight each. The dose of cortisol, 250 g/g, was chosen in light of a pilot study which revealed initial diminishing returns in avoidance condition ing level and enhancing effects at lower doses. The intention was to create deficits in avoidance conditioning due to chronic HPI activation that may in turn be treated with antidepressive and/or anxiolytic effects. The dose of allopregnanolone was chosen based on previous studies with rats (e.g., Turkman et al, 2004) at 8 mg/kg. This dose proved insufficient to produce lasting anesthetic e ffects. No prior data on the use of allopregnanolone in teleosts exists. Fluoxetine was dissolved in the water of the h ome aquaria at a concentration of 81 g/L, consistent with previous work (Beulig and Fowler, 2009). The metabolism of fluoxetine in tel eosts has shown to produce the same active metabolite (norfluoxetine) as it does in human albeit
71 in lesser concentrations (Smith et al, 2010), assuring that the effects of fluoxetine continued to be experienced even after initial metabolism. Fluoxetine was replaced during each water change at a rate to maintain an equivalent concentration to the initial dose. Results Scototax is The scototaxic data is presented with asterisks indicating significant differences from the control and black lines which overlap those groups which displayed no significant differences between them. F IGURE 18 : N UMBER OF T RANSITIONS P RE TREATMENT MEANS MINUS POST TREATMENT MEANS E RROR BARS INDICATE ST ANDARD DEVIATIONS BOTTOM LINES DEMONST RATE T UKEY S GROUPING The data was checked for normality, equality and homogeneity of variance via Ma uANOVA was then employed to compare the means for both the nu mber of transitions and the duration of time spent in the light. The main effect of treatment was significant in the difference (pre injection minus post injection) of transitions, F(5, 96) = 6.60 p < M = 33.62, SD 33.62 18.31 2.18 0.94 3.12 8.06 -60 -50 -40 -30 -20 -10 0 10 Allo Flu Con AlloCort FluCort Cort
72 = 18.33) alone exhibited significance over the control ( M = 2.18, SD = 5.23) and all other groups excluding fluoxetine ( M = 18.31, SD = 14.80). Given th at the allopregnanolone treated fish exhibited greater explorative behavior, which is repressed in more fearful fish, it is concluded that allopregnanolone produces anxiolytic in goldfish whereas no other compound achieved significance over the control. F IGURE 19 : T IME S PENT IN THE L IGHT P RE TREATMENT MEANS MINU S POST TREATMENT MEANS E RROR BARS INDICATE ST ANDARD DEVIATIONS TOP LINES DEMONSTRAT E T UKEY S GROUPING The main effect of treatment was significant when evaluated in te rms of the mean di fference of the duration of time spent in the light ed portion of the tank, F(5, 96) = 6.88, p < M = 5864, SD = 2636) alone exhibited signif icantly greater time in the lighted portion of th e tank than the control ( M = 1137, SD = 6237) and all other groups excluding fluoxetine ( M = 2907, SD = 2775). High variability in all but the allopregnanolone groups is revealed in the above graph. 5864 2907 1137 938 540 395 -2000 0 2000 4000 6000 8000 10000 Allo Flu Con AlloCort FluCort Cort
73 Since allopregnanolone treated fish spent significantly more time in the light chamber than the control group they displayed a less er tendency to hide in the dark indicating decreased anxiety over controls. Cognition The data was checked for n ormality, equality and homogeneity of variance via Ma umeans for the avoidances, escapes, and failures in each trial according their corresponding comprising four consecutive days. The procedure for analysis was identical for both the Delay and Trace data. All data on tests of cognition are presented in bar graphs with color indicating group, e rror bars displaying the standard deviation, asterisks revealing significant difference from co ntrols, and lettering to demonstrate pairwise comparison. Those groups with significant differences between them but not over the control do not have asterisks but are instead grouped exclusively. F are but not significantly different than those also in or those gro ups inclusive of the same group such as B olone
74 Delay Learning Avoidance Behavior F IGURE 20 A VOIDANCE M EANS D ELAY C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AND COMPARED TO THE CONTROL FOR SIGN IFICANCE WITHIN EACH BLOCK INDICATED BY ASTE RISK P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BARS SIGNIFY STANDARD DEVIATION There was a significant overall effect for treatment for avoid ance frequency, F(5, 42) = 6.72, p = .0002; block, F(3, 42) = 72.19, p < .0001; and an interaction between block and trea tment, F(15, 42) = 5.27, p < .0001. Testing with repeated measures reveals that block 1 exhibits significance when evaluated by treatme nt, F(5, 42) = 4.28, p = .0037; block 2 is significant, F(5, 42) = 8.50, p < .0001; block three is significant, F(5, 42) = 3.20, p = .0171; and block 4 is also si gnificant, F(5, 42) = 10.51, p < .0001. All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort Block 1 Block 2 Block 3 Block 4 Avoids 4.6 4.1 4.4 12 20 17 7.9 1.3 14 21 26 24 24 13 27 31 24 29 40 34 34 38 33 22 A A AB B AB AB BC C ABC AB A A AB B AB AB A A A A A A B 0 5 10 15 20 25 30 35 40 45 50 Mean # of Avoids per Block A
75 Significant Pairwise Comparisons that none of the results proved significant compared to the sham group excluding cortisol in the fourth block ( M = 21.57, SD = 5.91) versus control ( M = 36.71, SD = 2.93). Again, no further significance was found between the positive control, the cortisol group, excluding the comparison between cortisol and all other groups in block four. Nonsignificant Pairwise Comparisons While certain trends evade significance against the control groups, there are various groupings depicted by the lettering schema on the above graph. For example, the fluoxetine group performed consistently lower than all others prior to block four: there is a very modest mean in block one ( M = 4.14, SD = 2.59), the lowest mean of all avoidances in block two ( M = 1.28, SD = 1.07), and a re latively very small average in block three ( M = 13.14, SD = 3.13). While none of these effects reach significance over the control, the block one fluoxetine group M = 4.14, SD = 2.59) is significantly worse than the fluoxetine cortisol group of the same bl ock ( M = 20.29, SD = 15.06), the highest block one mean, which was also significant compared to all opregnanolone and all other groups that did not receive cortisol. In block two, fluoxetine ( M = 1.28, SD = 1.07) was also significantly worse than all of the groups which received cortisol. In block three, fluoxetine ( M = 13.14, SD = 3.13) subjects performed significantly worse than cort isol ( M = 28.71, SD = 13.25) and allopregnanolone cortisol ( M = 31.43, SD = 16.06). Fluoxetine ( M = 33.86, SD = 1.069) perfor med as well as the other groups in the final block. Trends Despite a strong trend toward improvement of avoidance behavior over time, the Delay condition presents two decreases in mean avoidance: fluoxetine from block one to block two ( M = 4.14, SD = 2.59 vs M = 1.29, SD = 1.06) and cortisol from block three to four ( M = 28.714, SD = 13.25 vs M = 21.57, SD = 5.91). Additionally, there is an abrupt rise in the average performance
76 of the fluoxetine group from block three to four ( M = 13.14, SD = 3.13 vs M = 3 3.857, SD = 3.132). Fluoxetine in block two presents with the overall lowest avoidance mean ( M = 1.28, SD = 1.07) and allopregnanolone in block four ( M = 40, SD = 6.06) exhibits the highest overall avoi dance mean in both Delay and Trace experiments. Conclu sions I t may be stated that that no treatment produced effects in Delay avoidance learning over the control, with the exception of cortisol over a period of 12 days. Since those fish treated with cortisol exhibited decreased avoidance but those who receive d an anxiolytic in addition to the cortisol implant were on par with the control, it is concluded that anxiolytic treatment had a significant role in ameliorating the cognitive deficit produced by chronic hypercortisolemia.
77 Escape Behavior F IGURE 21 E SCAPE M EANS D ELAY C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AND COMPARED TO THE CONTROL FOR SIGNIFIC ANCE WITHIN EACH BLO CK INDICATED BY ASTERIS K P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BAR S SIGNIFY STANDARD DEVIATION There was also a significant overall effect for treatment on escape frequency, F(5, 42) = 12.20, p < .0001; block, F(3, 42) = 21.69, p < .0001; and an interaction between block and trea tment, F(15, 42) = 2.70, p = .0016. Testi ng with repeated measures reveals that block 1 exhibits significance when evaluated by treatment, F(5, 42) = 9.56, p < .0001; block 2 is significant, F(5, 42) = 3.65, p = .0009; block three is not significant, F(5, 42) = 1.58, p = .1904; and block 4 is si gni f icant, F(5, 42) = 3.86, p = .0066. All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort Block 1 Block 2 Block 3 Block 4 Avoids 26 27 25 43 33 47 44 48 38 51 40 49 38 43 36 31 46 40 31 37 33 38 38 43 C C C AB BC A AB AB B A AB AB A A A A A B AB B AB AB A 0 10 20 30 40 50 60 70 Mean # of Escapes per Block A *
78 Significant Pairwise Comparisons cortisol ( M = 43.14, SD = 9.27) and cortisol ( M = 46.63, SD =9.86) both significantly differed from the control ( M = 24.63, SD = 4 .95). In block two, only allopregnanolone cortisol ( M = 51.14, SD = 4.94) showed a significant deviation from the control ( M = 38.14, SD = 4.18). In block four, cortisol ( M = 43.00, SD = 7.26) proved to be significantly higher than the control ( M = 32.71, SD = 6.34). Nonsignificant Pairwise Comparisons pronounced in the non cortisol groups. Allopregnanolone, fluoxetine, and the control groups steadily decline in the n umber of escapes from blocks two to four. Cortisol shows a slight drop in mean escapes with a concomitant decrease in variability from blocks two to three ( M = 49.29, SD = 13.93 to M = 40.43, SD = 9.14), followed by a marginal increase in mean and decrease in standard deviation from blocks three to four ( M = 40.43, SD = 9.14 to M = 43.00, SD = 7.26). A llopregnanolone cortisol and fluoxetine cortisol show similar nonsignificant fluctuations from blocks two to four, defying the consistent reduction of escapes observed in the anxiolytic and control groups. Trends The escape data presents much higher variability than the cortisol implanted groups e xhibiting consistently higher variance than the control or the anxiolytic only groups. While block one escapes are e specially low in the control, allopregnanolone, and fluoxetine cohorts co mpared to the cortisol containing groups, this effect did not prove significant and these values largely normalize by block two.
79 Conclusions Since cortisol in block four produced significantly mo re escape behavior than the co ntrol it is concluded that the deficit to Delay avoidance learning is likely not due to motor system complications associated with the drug. Failure Behavior F IGURE 22 F AILURE M EANS D ELAY C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AND COMPARED TO THE CONTROL FOR SIGNIFIC ANCE WITHIN EACH BLO CK INDICATED BY ASTERIS K P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BARS SIGNIFY ST ANDARD DEVIATION Ther e was a significant overall effect for treatment on failure frequency, F(5, 42) = 27.21, p < .0001; block, F(3, 42) = 138.53, p < .0001; and an interaction between block and All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort All Flu Con AllC FluC Cort Block 1 Block 2 Block 3 Block 4 Avoids 50 49 51 25 29 18 29 31 28 7.6 14 6.3 18 23 18 5.7 9.7 11 9 9.4 11 5.3 9.4 15 A A A BC B BC A A A B B B AB A AB BC BC AB AB AB B AB A 0 10 20 30 40 50 60 Mean # of Failures per Block C *
80 treatment, F(15, 42) = 8.78, p < .0001. Testing with repeated measures reveals tha t block 1 e xhibits significance when evaluated by treatment, F(5, 42) = 21.80, p < .0001; block 2 is signif icant, F(5, 42) = 25.94, p < .0001; block three is significant, F(5, 42) = 6.09, p = .0004; and block 4 is not significant, F(5, 42) = 2.08, p = 09 12. Significant Pairwise Comparisons M = 51.00, SD = 4.93) and all opregnanolone cortisol ( M = 24.50, SD = 9.10), fluoxetine cortisol ( M = 28.15, SD = 15.02), and cortisol ( M = 18.25, SD = 10.25) in block one. Again, allopregnanolone cortisol ( M = 7.57, SD = 7.44), fluoxetine cortisol ( M = 14.29, SD = 8.34), and cortisol ( M = 6.29, SD = 5.19) proved to be significantly lower than the control ( M = 27.88, SD = 4.45) in block two By block three, only all opregnanolone cortisol ( M = 5.71, SD = 4.82) presents as significantly lower than the control ( M = 17.57, SD = 6.05) but not cortisol ( M = 10.86, SD = 12.59). Trends There is a strong trend in all groups to decline in failures fro m block one to four at a d ecelerating rate suggestive of a floor effect. The only exception is the increasing rate of failures and oscilating variability in the cortisol group from block two ( M = 6.29, SD = 5.187) to block three ( M = 10.86, SD = 12.59) to block four ( M = 15.43, SD = 9.57). It should also be noted that the last group to lose significance over control is the allopregnanolone cortisol cohort, ultimat ely presenting with the fewest failures by block four ( M = 5.29, SD = 2.06). Conclusions While all of the cortisol treated groups exhibited significantly more failures in blocks one and two, they did not display significantly less avoidance nor escape behavior in the same blocks. This suggests perhaps helpless behavior in these earlier blocks that was not due to motor
81 inhibition, since many of the fish demonstrated significantly higher escapes over the same period. Trace Learning F IGURE 23 A VOIDANCE M EANS T RACE C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AN D COMPARED TO THE CO NTROL FOR SIGNIFICAN CE WITHIN EACH BLOCK INDICATED BY ASTE RISK P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BARS SIGNIFY STANDARD DEVIATION There was a significant overall effect of treatment on avoidance in the Trace condition, F(5, 45) = 9.05, p < .0001; block, F(3, 45) = 118.12, p < .0001; and an interaction between block and treatment, F(15, 45) = 5.90, p < .0001. Testing with repeated measures reveals that block 1 does not show significance when evaluated by treatment, F(5, 45) = 1.59, p = .1892; block 2 is All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t Block 1 Block 2 Block 3 Block 4 Avoids 4 4 4 8 3 10 8 3 12 14 11 16 26 14 25 17 13 23 33 32 32 23 12 21 A A A A A A AB B A A AB A A C AB ABC BC ABC A A A A C B 0 5 10 15 20 25 30 35 40 Mean # of Avoids per Block *
82 significant, F(5, 45) = 5.48, p = .0007; block three is significant, F(5, 45) = 4.48, p = .0025; and block 4 is also significant, F(5, 45) = 21.23, p < .0001. Significant Pairwise Comparisons eveals significant differences from the control in blocks two, three, and four. In block two, fluoxetine ( M = 3.00, SD = 3.28) was significantly lower than the sham trea tment ( M = 11.5, SD = .92). In block three, fluoxetine ( M = 14.00, = SD 4.29) was again lower than the control ( M = 24.62, SD = 3.00). In block four, fluoxetine cortisol ( M = 31.57, SD = 4.76) and cortisol ( M = 21, SD = 5.606) were both lower than the control and cortisol also exhibited a si gnificantly lower mea n than fluoxetine cortisol. No comparisons reveal significant differences between the positive control and any other groups aside from those in block four, described already. Nonsignificant Pairwise Comparisons Cortisol exhibits a curious profile over time Initially presenting the highest mean in block one and the greatest variation of all groups throughout in the Trace condition ( M = 1.28, SD = 12.68), cortisol also overtakes the other groups in block two while maintaining the most variance in this block ( M = 15.63, SD = 8.14). In block three, cortisol ( M = 23.00, SD = 9.26) ranks above the other two cortisol containing treatments and comparable to allopregnanolone ( M = 26.63, SD = 3.40) and the control ( M = 24.63, SD = 2.97) but, by block four, falls sign ificantly b elow the allopregnanolone cortisol ( M = 17.25, SD = 12.94) group and all other non cortisol groups. Trends Interestingly, from block three to block four, cortisol ( M = 23.00, SD = 9.26 vs M = 21, SD = 5.61) and fluoxetine cortisol ( M = 13.43, SD = 6.92 vs M = 11.71, SD = 5.77) both decrease slightly despite a strong trend in all other groups towards improvement over all blocks. The only
83 other violation of this trend occurs from block one to block two with fluoxetine treated fish ( M = 4.43, SD = 1.31 vs M = 3.00, SD = 3.28). Another notable change is the abrupt improvement of the fluoxetine group from block three to four ( M = 14.00, SD = 4.29 vs M = 31.57, SD = 4.76), resolving the significantly lower performance compared to control fro m blocks two and three. Conclusions Allopreganolone cortisol displays significant improvement in block four over both fluo xetine cortisol and cortisol, indicating its superior role in the retention of Trace avoidance lear ning in conditions of chronic hyper cortisolemia. The significantly lower performance of fluox etine cortisol over the positive control suggests that fish who were treated with fluoxetine cortisol exhibit impairments to hippocampal function that are greater than treatment with cort isol alone, though fluoxetine treatment did rescure avoidance learning in the Delay condition. The significant deficits produced by fluoxetine in blocks two and three are suggestive of a speci fic negative influence on temporal memory.
84 Escape Behavior F IGURE 24 E SCAPE M EANS T RACE C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AND COMPARED TO THE CONTROL FOR SIGNIFIC ANCE WITHIN EACH BLO CK INDICATED BY ASTERIS K P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BARS SIGNIFY STANDARD DEVIATION There was a significant overall effect of treatment on Trace escape performance, F(5, 45) = 44.54, p < .0001; block, F(3, 45) = 21.87, p < .0001; and an interaction between block and treatment, F(15, 45) = 4.69, p < .0001. Test ing with repeated measures reveals that block one is significant when evaluated by treatment, F(5, 45) = 49.30, p < .0001; block two is significant, F(5, 45) = 8.79, p < .0001; block three is significant, F(5, 45) = 5.28, p = .0008; and block four is also significant, F(5, 45) = 11.76, p < .0001. All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t Block 1 Block 2 Block 3 Block 4 Avoids 25 26 25 56 58 46 41 47 42 54 61 55 37 44 40 50 59 43 34 35 33 37 50 49 C C C A A B C BC C AB A AB B B B AB A B B B B B A A 0 10 20 30 40 50 60 70 80 Mean # of Escapes per Block * *
85 Significant Pairwise Comparisons M = 24.63, SD = 2.92) in the number of escapes in block one by allopregnanolone cortisol ( M = 56.13, SD = 5.11), fluoxetine cortisol ( M = 57.50, SD = 8.37), and cortisol ( M = 45.88, SD = 10.08); cortisol is also significantly lower than both allopregnanolone cortisol and fluoxetine cortisol. In block two, allopregnan olone cortisol ( M = 54.25, SD = 10.22), fluoxetine cortisol ( M = 61.29, SD = 10.18), and cortisol ( M = 55.38, SD = 8.37) remain significantly higher than the control, though cortisol is no longer st atistically significant when compared to other cortisol inclusive treatments. By block three only fluoxetine cortisol ( M = 59.29, SD = 7.11) possesses significantly greater escapes than the co ntrol ( M = 40.13, SD = 5.49). In the final block, both cortisol ( M = 49.38, SD = 5.71) and fluoxetine cortisol ( M = 49.86, SD = 5.55) produced signific antly more escape behavior than the all other groups. Significant Pairwise Comparisons Initially all of the hypercortisolemic fish demonstrate significantly higher escapes and this trend is attenuated as the other treatment groups increase in escape behavi or and all but fluoxetine cortisol decline in escape reactions. By block four, allopregnanolone cortisol ( M = 36.75, SD = 10.81) is grouped with the control group ( M = 33.25, SD = 3.96) and significantly lower than both fluoxetine cortisol ( M = 49.86, SD = 5.55) and cortisol ( M = 49.375, SD = 5.71). It is also fascinating that cortisol ( M = 42.88, SD = 10.06) is not significantly deviant from the co ntrol group ( M = 40.13, SD = 5.49) in block three and significantly lower than fluoxetine cortisol ( M = 59.29, SD = 7.11) and then rises to significantly eclipse the control ( M = 36.75, SD = 3.96) in block four. This increase is stark in light of the general trend toward decreased escape after block two, defied by both this increase and fluoxetine y constant escape perfo rmance which is both the highest in each block and significant over the control at all levels.
86 Conclusions Since nearly all cortisol treated groups displayed significantly higher avoidance over the contr ol in nearly every block, this is strong evidence to rule out changes due to motor impai rment or lethargy. Failure Behavior F IGURE 25 F AILURE M EANS T RACE C ONDITIONING T REATMENT GROUP MEANS ARE DIVIDED BY B LOCK AND COMPARED TO THE CONTROL FOR SIGNIFIC ANCE W ITHIN EACH BLOCK INDICATED BY ASTERIS K P AIRWISE COMPARISONS ARE REPRESENTED VIA LETTERED GROUPINGS E RROR BARS SIGNIFY ST ANDARD DEVIATION There was a significant overall effect of treatment on Trace failures, F(5, 45) = 72.78, p < .0001; block, F(3, 45) = 98.05, p < .0001; and an interaction between block and treatment, F(15, 45) = 10.04, p < .0001. Testing with repeated measures reveals that block one shows signif iAll Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t All Flu Con AllC Flu C Cor t Block 1 Block 2 Block 3 Block 4 Avoids 51 50 51 16 19 24 30 31 27 10 8 9 17 22 15 4 7 14 13 15 15 14 18 10 A A A B B B A A A B B B A A AB C BC AB AB AB AB AB A B 0 10 20 30 40 50 60 Mean # of Failures per Block *
87 cance, F(5, 45) = 36.22 p < .0001; block 2 is significant, F(5, 45) = 23.41, p < .0001; b lock three is significant, F(5, 45) = 9.24, p < .0001; and block 4 is not significant, F(5, 45) = 2.09, p = .0879. Significant Pairwise Comparisons cortisol ( M = 15.50, SD = 10.98), fluoxetine cortisol ( M = 19.25, SD = 10.15), and cortisol ( M = 24, SD = 10.28) statistically significant co mpared to the control in block one. In block two, allopregnanolone cortisol ( M = 9.75, SD = 6.65), fluoxetine cortisol ( M = 8.00, SD = 7.70), and cortisol ( M = 9.00, SD = 7.25) remain significantly lower than the control ( M = 26.88, SD = 3.98) and all groups that did not receive a glucocort icoid. By block three, allopregnanolone cortisol ( M = 3.50, SD = 1.07) drops to the minimum va lue for all mean failures in both conditio ns, significantly lower than the control ( M = 15.25, SD = 5.37), cortisol ( M = 14.13, SD = 8.81), and the anxiolytic groups. In block four no treatments si gnificantly differed from the control ( M = 14.88, SD = 2.91). Nonsignificant Pairwise Comparisons The re is a very high rate of failure in the control and anxiolytic only groups in block one that decreases steadily over time; a similar decline occurs from blocks one to three in all other treatments. By block four there is not only the absence of significan t deviation from the control ( M = 14.88, SD = 3.91), but all groups present equivovally excepting the significant difference between the greater rate of failure in fluoxetine cortisol ( M = 18.43, SD = 6.63) than cortisol ( M = 9.63, SD = 3.74). Conclusions The Trace condition exhibits opposite effects of failures as observed from the Delay condition, with significanty fewer failures in those fish who were trated with cortisol versus those who were not This is coupled with stronger findings indicating increa sed escape behavior over the other groups and relatively equivalent avoidance performance. This suggests that the
88 cortisol implanted fish were energetic but incapable of establishing the CS UCS association until later blocks. Observations Notably, all of the cortisol only fish (16 of 16) remained chronically infected throughout all trials whereas a minority of fluoxetine cortisol (5 of 16) and almost none of allopregnan olone cortisol (2 of 16) contracted disease. While some of the 48 fish of the non cortis ol receiving groups presented with illness, the ailments were mild and most (45 of 48) required no further treatment than was given during daily testing. Regardless of group treatment, the rates of mo rtality throughout the entire experiment were nearly equ ivalent : one from each drug treatment group in the Delay condition and three from different treatments the Trace condition (in the Allo, Flu, and Allo+Cort schedules) expired b efore the completion of block 4. Observation of the fish in their home tank reve aled no evidence of enduring hyperactiv ity or hypoactivity. Immediately after injections there was a profound sedative action of all opregnanolone injection that bordered on anesthesia, which has been accomplished at much higher doses in rats (Singh et al, 2010), and a mild, inconsistent (five of sixteen subjects) i ncrease in locomotive activity following cortisol only injections. Both of these effects were trans ient and the latter may be attributable to injection/handling/restraint stress, but the former w as consistent (fourteen of sixteen exhibited strong sedation) and persisted for approximately five minutes after injections, after which time the fish proceeded to swim and eat per baseline. Since the scototaxic procedure was carried out the day after inj ections, the fish were allowed ample time to rest before experimentation; it is worth considering that this possible locomotive i msignificantly higher number of bou ndary crosses compared to all but the fluoxetine group.
89 Discussion The results of the present experiment further support the findings of previous work in our laboratory indicating significant impairment of avoidance learning in goldfish by water borne flu oxetine administration (Beulig and Fowler, 2006). Though these trends were present in both conditions, they only achieved significance in the Trace subset; this is likely due to the smaller group size used in this experiment compared to the previous invest igation, rendering the present study less sensitive to this effect. No significant affective changes in fish treated with fluoxetine nor fluoxetine plus cortisol were observed, though fluoxetine was the only group to exhibit an effect similar to allopregna nolone in the tests for scototaxic behavior T his conflicts with previous studies in rats and humans, though the scototaxic trials only tested acute affective differences and there is a several week delay before antidepressant changes appear in other organ isms (Zhang et al, 2000; Wilde and Benfield, 1998 ; Tome et al, 1997 ) The learning impairment, spatial cognition deficit, and conditioned place aversion caused by allopregnanolone in rodents and memory dysfunction in humans was not observed in the goldfish (Silvers et al, 2003; Johansson et al, 2002; Beauchamp et al, 2000; Kask et al, 2008 ). In the present experiment, the allopregnanolone group alone exhibited no significant difference from the control nor did the allopregnanolone cortisol cohort. I n fact, the allopregnanolone group held the highest mean avoidances in both conditions by the end of the study and all opregnanolone cortisol demonstrated significant ly fewer f ailure s in both conditions The signif icant effect of allopregnanolone in both th e difference in duration of time spent in the light chamber and the number of transitions made during the scototaxis procedure supports previous findings indicating its ef feciveness as an acute anxiolytic ( Engin and Treit, 2007 ) It may be su ggested that t he persistence of this effect from humans to rodents to teleost s supports its modus operandi via the highly conserved GABA A R (Brot et al, 1997).
90 C ortisol produced no significant effect in avoidance conditioning as is suggested by pr evious literature (Beatt y and Beatty, 1970) despite the expectation that cortisol might enhance emotional memory consolidation via BLA activation (Buchanan and Lovallo, 2001 ) or that it may lead to deficits in declarative mem ory via hippocampal dysfunction (Kirschbaum et al, 1996 ) The net effects of cortisol on cognition are still not resolved in many taxa (Het et al, 2005) After a period of 16 days t he fish which received cortisol or fluoxetine cortisol performed significantly worse at Trace avoidance and the cortisol group proved to be the only significant difference from control in the Delay condition for avoidance behavior Indeed, for cortisol explicitly there was a significant decline in avoidance behavior in the final block of conditions that was almost exclusively associated with an increase in failures in the Delay condition and with escapes in the Trace paradigm. (These differential effects may be d ue to the greater innate stress of the Delay condition, which may require an action to avoid/escape as often as every 25 seconds, as co mpared to the Trace condition which permits additional five seconds per round, generally increa sing the total run time fr om two to four minutes over the Delay condition. The additional stress of this paradigm is evidenced by the increased deaths in the Delay condition, in which six su bjects died while only three expired during the Trace training ) The result s of the Trace co ndition support the hypothesis that th e long term effects of cortisol impair spatial/temporal learning possibly due to hippocampal dysfunction as is well represented in previous work in mammals ( Lupien et al, 1998; Starkman et al, 1992 ). The d ecreased per formance of in Trace avoidance in all cortisol regimens but the allopregnanolone cortisol group implies that allopregnanolone successfully rescued avoidance aquisition in this condition.
91 While cortisol exhibited no significant difference in the scototaxis procedure, it exhibited the highest variance of all groups, implying differential effects which may be related to sex or age differences in the subjects (Wolf et al, 2001; Buchanan and Tranel, 2007). Interpretations Within the five days following injections no significant effects are reported for any of acute be nzodiazepine ctivity at high doses (Riddle et al, 1990 ), but in line with behavioral observations which suggested no such effects except transiently after injections While it is worth noting that those groups which received cortisol performed better in general than those who di d not in the first two blocks, there is also a startling degree of variabi lity that precludes a stronger effect. Given that it is necessary that the formation of the UCS CS association depends on the serendipitous crossing of the hurdle, hyperactivity or g reater arousal might be considered advantageous (Carpenter and Summers, 2008; Brown, Laland, and Krause, 2007) ; this is very possibly the reason for the higher early performance of the cortisol groups as well as the higher variability, indicating perhaps t hat an association was not made but rather a This, in turn, could lead to sooner association of the stimuli, perhaps enhance d by the consolid ating effects of HPI arousal via the BLA. This is supported in the Delay condition in which there are a higher number of initial avoidances than in the non cortisol groups for blocks one and two, with all of the groups barring cortisol exhibiting nonsignificantly differences by block four The final block revealed the onl iance decreased while their mean also fell from the previous block. This is the predicted effect of long term cortisol elevation: impaired memory and increased helpless behavior. Given that the
92 c ortisol anxiolytic groups rated very similarly to one another and the non cortisol groups in the delay condition it might be suggested that both allopregnanolone and fluoxetine proved effe ctive in ameliorating the exogenous stress imposed by the cortisol implant. The Trace condition, designed to demonstrate the coordination of hippocampal and amygdalar function in the fish, shows significant impairment of acquisition of the fluoxetine in blocks two and three compared to the control group. This is paradoxic al given that fluoxetine has been shown to improve hippocampal function in patients suffering from depression as well as laboratory mammals: instead, the treatment including both fluoxetine and cortisol performed the worst in the final block when the compr omising effects on the hippocampus should be most pronounced. The consistent and profound impairment of shuttle box avoidance training through fluoxetine administration has been demonstrated previously several times in our labo ratory with both goldfish an d pinfish (Beulig and Fowler, 2008; others unpublished) though the Trace task had not yet been utilized for goldfish dosed with fluoxetine tremendous improvement from block three to four may be compared to the fluoxetine cortisol s contemporaneous failure to improve : perhaps the drug was thoroughly metabolized by the end of block three. If this were the case, the low mean, low variation fluoxetine group may have been liberated of cognitive impairme nt or affective helplessness allowing more optimal performance on the already acquired task ; indeed, up to par with their conspecifics. The mode rately low mean, high variance fluoxetine cortisol group might have been benefitting from the hyperarousal provi ded by the cortisol without actually successfully establishing the UCS/CS co nnection due to the handicap posed by fluoxetine ; as the active influence of the fluoxetine waned, the variance diminished but without a larger mean which in turn was dwarfed by t he improvement of all other groups.
93 In the first ever analysis of fluoxetine metabolism in fish hepatocytes, Smith and co lleagues document the clearance rates of four young goldfish hepatic microsomes when incuba ted with either 45 g/L or 80 g/L for eith er 30 or 60 minutes, respectively (2010) Using their limited data, it is calculated that would be possible for a single young goldfish liver to metabolize about 1092 g/L/day with substantial variance (SD = 1401) that limits the applicability of this dat a While this in vitro measure reveals itself to be rather inconsistent, it does establish that Carassius is capable of digesting fluoxetine to a small extent to its active metabolite norfluox etine though this value is much lower than it is in mammals (ab out 2 % total metabolism versus nearly 100% in clinical applications). The induction of the CYP family of hepatic enzymes was also observed, as is consistent in humans, and this may significantly increase the metabolism during chronic treatment. While fluox etine has been found to be especially hepatotoxic to rai nbow trout hepatocytes the study in question employed what would be nearly lethal concentr ations. This does not permit easy comparison due to the in vitro methodology, and employs a different teleost species (Laville et al, 2003); regardless, it is possible that hepatotoxicity ha mpered the ability of the fish to metabolize fluoxetine and possibly generated complicating fa ctors that impaired avoidance performance. Other explanations for the profound de ficits in avoidance performance exhibited by the fluoxetine groups may arise from a mechanism of action specific to fish. Mennigen and co lleagues have performed numerous studies to elucidate the effect of fluoxetine on fish due to the increasing ecological presence of the drug from wastewater effluent; indeed, some surveys have indicated that fluoxetine is present in rivers in the USA and Canada at levels up .54 /L Metcalfe et al, 2003 ; Brooks et al, 2003 ) What is more alarming is that fluoxetine bioaccu m ulates to a moderate extent in the nervous tissues, skeletomusculature, and liver (Nakamura et al, 2008; Mennigen et al, 2011); in a study utilizing only slightly less fluoxetine than the present,
94 .64 /L over seven days with 21 days of fresh water to de toxify, found a half life of 9.4 1.1 days with possibly longer accumulation of norfluoxetine in the Japanese medaka, Oryzias latipes (Paterson and Metcalfe, 2008). This indicates the potential fo r long lasting effects in our expe rimental animals, though the cognitive effects may normalize over time or diminish due to d ecli n ing environmental concentrations It has also been demonstrated that significant dysfunction of the male and female r eproductive axes in goldfish is induced by low doses of f luoxetine ( Mennigen et al, 2010 ), a fasc ial, 1988). Fluoxetine may also effect nitrogen disposal in low doses to marine, the possibility of which suggests interaction wit h HPI axis, though the study was improperly designed to arrive at such a conclusion (Morando et al, 2009). Fluoxetine has consistently deprecated avoidance performance in teleosts, but has been considered to improve cognitive function in rodents and peopl e ( Flood and Cherkin, 1987; Nowakowska et al, 1996 ) While older antidepressants deliver their benefits with significant i mpairments to memory, fluoxetine and other SSRIs were the first to ameliorate this side effect which was thought to be associate d with anticholinergic effects (Kumar and Kulkarni, 1996). Since it well known that fluoxetine increases synaptic plasticity in diverse areas of the brain in both neurons (Stewart and Reid, 1999; Vetencourt et al, 2008; Ampuero et al, 2010 ; Alme et al, 2007) and glia (Czeh et al, 2005), it is not su rprising that fluoxetine would modulate certain forms of learning and those who find it reasonable that hippocampal neurogenesis is requisite for learning would find fluoxetine to be an attractive agent for improving h ippocampal function (Li et al, 2009) Despite expectations, it has been found more consistently t hat fluoxetine hi nders spatial and temporal tasks in rodents and that hippocampal neurogenesis does not provide
95 neural substrates for spatial learning (Satvat et al, 2011 ; Spankswick et al, 2007 ; Majlessi and Naghdi, 2002 ) arise from models of traumatic injury, a very different circumstance than MD illness or chronic hypercortis olemia. Fluoxetine has also been shown to negatively affect spatial learning and i ncrease anxiety irreparably in adolescent rats (Sass and Wortwein, 2011); in adolescent humans, fluoxetine treatment has been correlated with enduring memory problems (Bangs et al, 1994). Spec ific to avoidance conditioning fluoxetine has been shown to hinder acquisition in rats ind ependent of sensory alterations (Nelson et al, 1997 ). Fluoxetine has also been shown to suppress circulating estrogen, perhaps through the redist ribution of steroidogenic products toward neurosteroids such as allopregnanolone, which suppressed the recovery of ovariectomized rats with estrogen replacement on a measure of spatial memory (Taylor et al, 2004). Both sex and age have are known to be impo rtant predi ctors for SSRI response, likely due to the variability of glucocorticoid receptor profiles (Joyce et al, 2003). Fluoxetine has also been shown to affect the binding of MRs but not GRs in the rat hi ppocampus, pituatry, hypothalamus, and prefront al cortex, thus altering the MR/GR ratio (El akovic et al, 2011); it is possible that these sexually dimorphic findings contributed to variable findings in the present subjects who were not sexed prior to the experiment. Cortisol produced a deficit only ov er chronic exposure in both Delay and Trace cond itions. All groups other exhibited a significant improvement in avoidance performance over both fluoxetine cortisol and cortisol alone in the final block of the Trace condition. F luoxetine in turn proved to b e significantly worse than allopregnanolone in blocks two and three of the trace task ultimately worse than no anxiolytic treatment to ame liorate the degenerative effects of cortisol on shuttle box avoidance. This data, coupled with modern research on the se compounds, is
96 response to allostatic states regulat ed by chronic hypercortisolemia Indeed, it has been shown that previous use of anxiolytic drugs (including S SRIs) is the greatest predictor of relapse in MD (Claxton et al, 2000). SSRI withdrawl or discontinuation sy ndrome, is a challenging and well known phenomenon with risks often greater than the initial depressive episode ( Haddad, 1998 ). Some have even sugg ested fluoxetine as a treatment for SSRI withdraw a l syndrome given the relatively mild severity of discontinuation compared to other SSRI withdraw a l symptoms (Bennzi, 2008), implying, perhaps, even more severe hipp ocampal dysfunction is associated with the use of other SSRIs during periods of chronic allostasis. Possible Confounds Goldfish have been shown via miRNA analysis to exhibit extensive neuroendocrine changes based on breeding status and seasonality (Zhang et al, 2009). While the issue of se asonality is controlled due to contemporaneous testing, the fish were not evaluated for maturity though the similar size affords some estimation of aging status. The former issue may lead to issues of replication whereas the latter was likely accounted for during randomization. As noted above, the fish were not sexed prior to experimentation. The significance of sex in cortisol, allopregnanolone, and fluoxetine response has been noted already. These effects no doubt translate to the fish model as well as the utilization of these compounds corresponds to their mammalian counterparts, which raises concern over disproportionate response within groups which may have had an imbalance of one sex over the other. This is a legitimate concern given the small group siz es, though the differential response to cortisol and allopregnanolone are likely to be less significant at the high exogenous dosing schedule employed in this study. It has been shown that there is a sex dependent effect of stress on MR/GR ratio in rats (K itraki et al, 2004). This is at least partly mediated by changes in the concentration of circulating CBG due
97 to the the actions of estrogens and the glucocorticoid antagonizing effects of progesterone, which have been documented in both rats and humans (Yo ung, 2008); the absence of an obv ious CBG in teleosts implies that the mechanisms of sex specific control may be regulated ot herwise or to less an extent. anxiolytic dose divergence across sex is likely not to be the product of GABA A R mo dification or regulation (Zimmerberg et al, 1999) indicating that the most likely conserved mode of action for fish to mammals is unaffected and implying similar acute effects as has been observed in this experiment Illness in the cortisol group remained a consistent problem throughout testing, leading ntion may have hampered rtisol group ex hibited consistently high avoidances in both Delay and Trace paradigms until block four, by which the fish were in excellent relative condition. Furthermore, the cortisol group co nsistently had high er escapes and fewer failures: raking across conditions, t he cortisol infused fish scored significantly higher than the control in five blocks of escapes and significantly lower in four blocks of failures The other cortisol implanted groups were substantially more robust than their cortisol only counterparts and exhibited similar trends. come under recent controversy due to the analysis of previous unpublished negative findings. Depression and SSRI Efficacy In a groundbr eaking meta analysis by Kirsch and Sapirstein published in Public Library of Science in 2008, the authors concluded that the trials submitted to the FDA for the antidepressant drugs fluoxetine (a SSRI), venlafaxine (a SNRI), nefazodone (a weak SNDRI), and paroxetine (a SSRI) showed insignificant impro vements in treating depression when the
98 prior to their initiation, a practice in integrity that has yet to be implemented by the scientific community at larg into the pending approval process despite all such registered trials remaining accessible to researchers. Not only did Kirsch and his contemporaries demonstrate that the s e dr ugs failed to produce significant decreases relative to baseline levels of depression in a within subjects design but that they also fail to improve Hamiliton Rating Score of Depression (HRSD) scores over placebo in all but the most extreme severity of ba seline depression. Furthermore, these changes in severely depressed patients correlated more with declining response to placebo therapy than to changes in drug efficacy. Fluoxetine (Prozac) was once psychiatric condit ions, to date including six FDA approved diagnoses and numerous off the label uses; it single handedly reformed the cultural view of depression from a devastating illness that required similarly dramatic treatments (e.g., ECT, tricyclics, and insulin indu ced comas) to a commonplace disease with excellent prognoses. Prozac quickly found itself embedded into the American lifestyle, and resulted in major depression disorder (MDD) diagnoses skyrocketing and encouraging psychiatrists to write scripts for a rela effectiveness. Even nine years after Prozac's FDA approval in 1987, antidepressant use is still growing : from 1996 to 2005, total lifetime antidepressant use, the vast majority of which was with SSRIs, increased f rom 5.84% of the nation to 10.12% (Olfson et Marcus, 2009). Prozac specifically garnered approximately forty million patients worldwide by 2002 and sales grossed over 22 billion dollars in the United States alone before Eli Lilly, its father company, lost its patent exclusivity (Wong et al, 2005). These numbers have doubtlessly grown as the generic version of fluoxetine has made a cheap drug even cheaper, allowing for even greater accessibility.
99 The implications of any drug so widely used are obviously mag nified. Resultantly, many controversial analyses of fluoxetine's efficacy and side effects have arisen from the era even before its lan d m ark release in 1987 (Wernicke et al, 1986; Benfielf et al, 1986). Research findings have been variable but the most re cent meta analyses conclude that the majority of trials have produced either modestly significant advantage over placebo in cases of especially results, indicates an overall lack of efficacy over placebo (Kirsch et al, 2008; Turner et al, 2008 ; Barbui et al, 2011 ). Those meta analyses which do not arrive at such conclusion s are often very limited in scope and still maintain only detect advantages over other therapi es (Arroll, 2005). Fluoxetine in particular, due in part to its pioneering role in the field, has been shown to be one of the most tested, yet least effective and most rejected of the modern SSRIs for MDD (Cipriani et al, 2009). Several scientists have eve n gone so far as to suggest unethical research practices and suppression of data as the cause of the the initial affirmative results and Phase III trials (Ioannidis, 2008). While this point is certainly disputable, it may be stated incontrovertibly that fl uoxetine is not a particularly effective drug for MDD: at best, meta analysis has found that there is a mere 8% difference over placebo and, at worst, there is no advantage to fluoxetine therapy at all (Kirsch et al, 2008). The mechanism of action of fluo xetine in ameliorating depressive symptoms remains to additional 5HT in the synaptic cleft (Fuller et al, 1991) While serotonin abnormalities have been observe d in both humans and animal models of depression, there is no present in depressed brains (Tanti et Belzung, 2010). Alternate mechanisms have been proposed, usually focusing on the modulation of trophic factors such as BDNF and allopregnanolone to provide anxioslysis or stimulate
100 neurogenesis (Schmidt, 2010; Alme et al, 2007; al, 2010; Rodriquez Landa et al, 2009; Pibiri et al, 2008; Uzunova et al, 2004; Ugale et al, 2004; Akwa et a l, 1999; Griffen and Mellon, 1999 ). The most conclusive evidence to date, however, reveals that only one of fluoxetine's enantiomers possesses the ability to improve symptoms of depression in rats, though both increase 5HT in the synaptic cleft; further r esearch by Pinna and his colleagues also revealed that subclinical (i.e., too low to register a measurable change in synaptic 5HT) doses of fluoxetine can treat symptoms of depression in rats as well as clinical traditional SSRI for his alternate theory of fluoxetine action involving the stereospecific modulation of Selective Brain Steroids Stimulation ( 2009). According to Pinna and his research team, brain steroids are responsible for the improvement of depress ion in Prozac treatment, as well as potentially other SSRIs. One particular steroid that has bee n identified, allopregnanolone as a neuroprotectant and potential therapeutic agent. Allopregnanolone is decreased in rats exposed to chronic stress (i.e., the depression model) and is likewise returned to baseline levels when these animals are treated with SSRIs. The model Pinna proposes is that allopregnenolone regulation is modified via steroid dehydrogenase, an enz yme that produces allopregnanolone in the brain. While this may eventually provide an explicit mechanism for fluoxetine therapies, it is emphasized that this would also explicate the indirect nature of the drug and account for its relative inefficacy compa red to, for example, beta blockers or any other non empirically formulated drug therapy. (Pinna et al, 2006) The premise guiding the development of fluoxetine was based on the empirical findings supporting anxiolyitc qualities of adequate serotonin transm ission and the, at that time, the presumed involvement of 5HT in mood disorders (Wong et al, 2005). While this is excellent
101 incentive for frontier research and, indeed, fluoxetine presents as an unprecedented tool for studying 5HT's role in vivo, this is a sticky situation for the clinician responsible for dispensing this drug without an adequate profile of its biological outcome. In pursuit of faster acting pharmaceuti cals, much of the research in M D is focued on means to expedite the two to eight week del ay minimize some of the side effects (Machado Vieira et al, 2008). Shortcomings as a pharma ceutical aside, fluoxetine at the very least replicates the effect of a placebo and it would A common problem in the employment this therapeutic drugs in psychiatry is the management of side effect profiles. Some of the many side effects that have arisen from the practice of empirical drug discovery have revealed themselves to be present in fluoxetine therapy: nausea, anxiety, decreased libido, dry mouth, dizziness, d iarrhea, other flu like symptoms, rhinorrhea, insomnia, and weakness are the FDA required minor side effects listed on fluoxetine scripts; anaphylaxis, bizarre behavior, gastrointestinal bleeding, decreased coordination, hallucinations, suicidality, and ag gressiveness are only a handful of the 36 less effects (Drugs.com, 2011). One particular side effect of interest caused a sizable controversy surrounding Prozac in the early 2000's: pediatric suicide. A series of well reported stories of adolescent violence, suicide, and bizzarre behavior led to the investigation of this potential side effect in patients under 25 years of age; an excellent resource documenting this hype exists at www.SSRIStor ies.com which provides citations to over 4,300 such reports in the media. This press, coupled with a series of investigations which returned with conclusive evidence of doubled suicidal ideation and attempt behavior compared to an inert drug, prompted the FDA to append a Black Box Warning on all fluoxetine prescriptions in 2004 (Simon et al, 2004). This necessitated a reevaluation of the drug in pediatric practices: the consensus of several respected publications is that SSRIs carry increased risks for chi ldren, but remain a viable treatment (Cohen, 2007). Precautions are merited, as described by Daniel Pine's guidelines for
102 SSRI therapy in children prior to the FDA acknowledgment, because of the increased vulnerability of children to side effects and their increased response to placebo: the advantage over placebo is weak, at best, and the risk reward ratio is correspondingly bleak for pediatric treatment (2002). reduce the negative side effects of existing treatments for depression such as MAOI inhibitors, and to reduce broad spectrum effects of others. The development of understanding the ctiveness and reducing side effects.
103 References Abramson LY, Seligman ME, Teasdale JD. 1978. Learned helplessness in humans: Critique and reformulation. Journal of Abnormal Psychology; Journal of Abnormal Psychology 87(1):49. Ags Balboa RC. 2006. Characterization of brain neurons that express enzymes mediating neur osteroid biosynthesis. Proc Natl Acad Sci U S A 103(39):14602. Aguilera G. 1994. Regulation of pituitary ACTH secretion during chronic stress. Front Neuroe ndocrinol 15(4):321 50. Akwa Y, Purdy RH, Koob GF, Britton KT. 1999. The amygdala mediates the anxiolytic like effect of the neurosteroid allopregnanolone in rat. Behav Brain Res 106(1 2):119 25. Alme MN, Wibrand K, Dagestad G, Bramham CR, Luca MD. 2007. Chronic fluoxetine treatment indu ces brain region specific upregulation of genes associated with BDNF induced long term potentiation. Neural Plast 2007:63. Alsop D and Vijayan M. 2009. The zebrafish stress axis: Molecular fallout from the teleost specific genome duplication event. Gen Com p Endocrinol 161(1):62 6. Alsop D and Vijayan MM. 2009. Molecular programming of the corticosteroid stress axis during zebrafish development. Comparative Biochemistry and Physiology Part A: Molecular & Integr ative Physiology 153(1):49 54. Aluru N and Vijay an MM. 2009. Stress transcriptomics in fish: A role for genomic cortisol signa ling. Gen Comp Endocrinol 164(2 3):142 50. Ampuero E, Rubio F, Falcon R, Sandoval M, Diaz Veliz G, Gonzalez R, Earle N, Dagnino Subiabre A, Aboitiz F, Orrego F. 2010. Chronic flu oxetine treatment induces structural plasticity and sele ctive changes in glutamate receptor subunits in the rat cerebral cortex. Neuroscience 169(1):98 108. Anderson GM, Cook EH Jr, Blakely RD. 2009. Serotonin rising. comment on: N engl J med. 2009 mar 5;3 60(10):957 9. N Engl J Med 360(10):2581. Anderson AK and Phelps EA. 2002. Is the human amygdala critical for the subjective experience of emotion? evidence of intact dispositional affect in patients with amygdala lesions. J Cogn Neurosci 14(5):709 20. Ande rson AK and Phelps EA. 2001. Lesions of the human amygdala impair enhanced perception of emotionally salient events. Nature 411(6835):305 9. Anderson LB, Anderson PB, Anderson TB, Bishop A, Anderson J. 2009. Effects of selective ser otonin reuptake inhibito rson motor neuron survival. International Journal of General Medicine Antonuccio DO. 2000. Raising questions about antidepressants. Psychother Psychosom 68(1):3.
104 Arana GW, Baldessarini RJ, Ornsteen M. 1985. The dexamethasone suppression test for diagn osi s and prognosis in psychiatry: Commentary and review. Arch Gen Psychiatry 42(12):1193. Arterbery AS, Fergus DJ, Fogarty EA, Mayberry J, Deitcher DL, Kraus WL, Bass AH. 2011. Evol ution of ligand specificity in vertebrate corticosteroid receptors. BMC Evolut ionary Biology 11(1):14. Asaba H, Hosoya K Takanaga H, Ohtsuki S, Tamura E, Takizawa T, Terasaki T. 2000. Blood?brain barrier is involved in the efflux transport of a neuroactive steroid, dehydroepiandrosterone su lfate, via organic anion transporting pol ypeptide 2. J Neurochem 75(5):1907 16. Aston Jones G, Meijas Aponte C, Waterhouse B, Coeruleus L. 2009. Norepinephrine: CNS pat hways and neurophysiology. Stress Science: Neuroendocrinology :388. Auchus R. 2009. Non traditional metabolic pathways of adrenal steroids. Reviews in Endocrine and Metabolic Disorders 10(1):27 32. Azpeleta C. 2010. Melatonin reduces locomotor activity and circulating cortisol in goldfish. Horm Behav 57(3):323. Bangs ME, Petti TA, Janus MD. 1994. Fluoxetine induced memory impairment in an adolescent. Journal of the American Academy of Child & Adolescent Psychiatry 33(9):1303 6. Barberis A, Mozrzymas JW, Ortinski PI, Vicini S. 2007. Desensitization and binding properties d etermine distinct ?1?2?2 and ?3?2?2 GABAA receptor channel kine tic behavior. Eur J Neurosci 25(9):2726 40. Barbui C, Cipriani A, Patel V, Ayuso Mateos JL, van Ommeren M. 2011. Efficacy of antidepre ssants and benzodiazepines in minor depression: Systematic review and meta analysis. The Bri tish Journal of Psychiatry 198 (1):11 6. Barden N, Reul JMHM, Holsboer F. 1995. Do antidepressants stabilize mood through actions on the hypothalamic pituitary adrenocortical system? Trends Neurosci 18(1):6 11. Bass DI, Partain KN, Manns JR. 2012. Event specific enhancement of memory vi a brief electrical stimulation to the basolateral complex of the amygdala in rats. Behav Neurosci 126(1):204. olo g ical Chemistry 277(48):46020. Baumgartner A, Hiedra L, Pi Iodothyronine deiodinase activity is extremely sensitive to stress. J Neurochem 71(2):817 26. Beatty PA, Beatty WW, Bowman RE, Gilchrist JC. 1970. The effects of ACTH, adrenalectomy and dex amethasone on the acquisition of an avoidance response in rats. Physiol Behav 5(8):939 44.
105 Beauchamp MH, Ormerod BK, Jhamandas K, Boegman RJ, Beninger RJ. 2000. Neurosteroids and reward: Allopregnanolone produces a conditioned place aversion in rats. Pharm acology Bi ochemistry and Behavior 67(1):29 35. Begliuomini S, Lenzi E, Ninni F, Casarosa E, Merlini S, Pluchino N, Valentino V, Luisi S, Luisi M, Genazzani AR. 2008. Plasma brain derived neurotrophic factor daily variations in men: Correl ation with cortisol circadian rhythm. J Endocrinol 197(2):429 35. Behar TN, Schaffner AE, Scott CA, O'Connell C, Barker JL. 1998. Differential response of cortical plate and ventricular zone cells to GABA as a migration stimulus. J Neurosci 18(16):6378 87. Belelli D. 2005. Neurosteroids: Endogenous regulators of the GABAA receptor. Nature R eviews.Neuroscience 6(7):565. Benazzi F. 2008. Fluoxetine for the treatment of SSRI discontinuation syndrome. The Intern ational Journal of Neuropsychopharmacology 11(05):725 6. Bent ley P. 2001. Sex hormones in vertebrates. Bern HA. 1985. The elusive urophysis twenty five years in pursuit of caudal neurohormones. Am Zool 25(3):763 70. Bernier NJ and Peter RE. 2001. Appetite suppressing effects of urotensin I and corticotropin releas ing hormone in goldfish< i>(carassius auratus). Neuroendocrinology 73(4):248 60. Bernier NJ, Lin X, Peter RE. 1999. Differential expression of corticotropin releasing factor (CRF) and urotensin I precursor genes, and evidence of CRF gene expression reg ulated by cortisol in goldfish brain. Gen Comp Endocrinol 116(3):461 77. Bernier NJ and Peter RE. 2001. The hypothalamic pituitary interrenal axis and the control of food intake in teleost fish. Comparative Biochemistry and Physiology Part B: Biochemistry and Molecular Biology 129(2 3):639 44. Beulig A and Fowler J. 2008. Fish on prozac: Effect of serotonin reuptake inhibitors on cognition in goldfish. Behav Neurosci 122(2):426 32. Bhagwagar Z, Hafizi S, Cowen PJ. 2002. Cortisol modulation of 5 HT mediated growth hormone release in recovered depressed patients. J Affect Disord 72(3):249 55. Bitran D, Shiekh M, McLeod M. 1995. Anxiolytic effect of progesterone is mediated by the ne urosteroid allopregnanolone at brain GABAA receptors. J Neuroendocrinol 7(3):17 1 7. Bitterman ME. 1964. Classical conditioning in the goldfish as a function of the CS UCS interval. J Comp Physiol Psychol 58(3):359 66. Blaser RE, Chadwick L, McGinnis GC. 2010. Behavioral measures of anxiety in zebrafish (danio rerio). Behav Brain Res 208(1):56 62.
106 Bodo C. 2010. Stress: Tough on the inside. Nature Reviews.Neuroscience 11(7):455. Bodo C. 2010. Affective disorders: A faster way to happiness. Nature Reviews.Drug Discovery 9(3):192. Boshra H, Li J, Sunyer J. 2006. Recent advances on the com plement system of teleost fish. Fish Shellfish Immunol 20(2):239 62. Bracamontes J, McCollum M, Esch C, Li P, Ann J, Steinbach JH, Akk G. 2011. Occupation of either site for the neurosteroid allopregnanolone potentiates the opening of the GABAA receptor i nduced from either transmitter binding site. Mol Pharmacol Bradley AJ. 2011. Animal studies in stress research: The brain, accelerated ageing and impaired control of stress homones. 3rd world conference of the IMTL society for traumatic stress stu dies. 53 p. Brambilla F. 2004. Plasma concentrations of anxiolytic neurosteroids in men with normal anxiety scores: A correlation analysis. Neuropsychobiology 50(1):6. Bremner J, Scott T, Delaney R, Southwick S, Mason J, Johnson D, Innis R, McCarthy G, Charney D. 1993. Deficits in short term memory in posttraumatic stress disorder. Am J Psychiatry 150(7):1015 9. Brezden BL, Fenwick JC, Moon TW. 1975. The effects of acclimation temperature and conditio ning temperature on the learning rate of the goldfish carassius a uratus. Comparative Biochemi stry and Physiology Part A: Physiology 50(2):373 7. Briley M and Moret C. Neurobiological mechanisms involved in antidepressant therapies. [r evised October 1993;. Brinton D. 2006. Therapeutic potential of neurogenesis for preven tion and recovery from al zheimers disease: Allopregnanolone as a proof of concept neurogenic agent. Current Alzheimer Research 3(3):185. Broglio C and Salas C. 2010. Lateral but not medial telencephalic pallium ablation impairs the use of goldfish spatial allocentric strategies in a. Behav Brain Res 214(2):480 7. Broglio C, Rodrguez F, Salas C. 2003. Spatial cognition and its neural basis in teleost fishes. Fish Fish 4(3):247 55. Broglio C, Gmez A, Durn E, Salas C, Rodrguez F. 2011. Brain and cognition in teleost fish. Fish Cognition and Behavior :325 58. Brooks BW, Foran CM, Richards SM, Weston J, Turner PK, Stanley JK, Solomon KR, Slattery M, La Point TW. 2003. Aquatic ecotoxicology of fluoxetine. Toxicol Lett 142(3):169 83. Brot MD, Akwa Y, Purdy RH, Koob GF, Britton KT. 1997. The anxiolytic like effects of the neur osteroid allopregnanolone: Interactions with GABAA receptors. Eur J Pharmacol 325(1):1 7.
107 Bruce Overmier J and Flood NB. 1969. Passive avoidance in forebrain ablated teleost fish, carassius auratus. Physiol Behav 4(5):791 4. Bruce Jones PN. 1994. Indomethacin and cognitive function in healthy elderly volunteers. Br J Clin Pharmacol 38(1):45. Brush FR and Knaff PR. 1959. A device for detecting and controlling automatic programming of avoidance conditioning in a shuttle box. Am J Psychol 72(2):pp. 275 278. Buchanan TW and Lovallo WR. 2001. Enhanced memory for emotional material following stress level cortisol treatment in humans. Psychoneuroendocrinology 26(3):307 17. Buchanan TW and Tranel D. 2 008. Stress and emotional memory retrieval: Effects of sex and cortisol response. Neurobiol Learn Mem 89(2):134 41. Buijs RM and Van Eden CG. 2000. The integration of stress by the hypothalamus, amygdala and prefrontal cortex: Balance between the autonomic nervous system and the neuroendocrine sy stem. Prog Brain Res 126:117 32. Burke HM, Davis MC, Otte C, Mohr DC. 2005. Depression and cortisol responses to psychological stress: A meta analysis. Psychoneuroendocrinology 30(9):846 56. Bury NR and Sturm A. 200 7. Evolution of the corticosteroid receptor signalling pathway in fish. Gen Comp Endocrinol 153(1):47 56. Butler AB and Hodos W. 2005. Comparative vertebrate neuroanatomy: Evolution and adaption. Hoboken, NJ: John Wiley & Sons, Inc. Buwalda B, Kole MHP, Ve enema AH, Huininga M, de Boer SF, Korte SM, Koolhaas JM. 2005. Long term effects of social stress on brain and behavior: A focus on hippocampal functioning. Neuroscience & Biobehavioral Reviews 29(1):83 97. Buwalda B, Felszeghy K, Horvth KM, Nyakas C, de Boer SF, Bohus B, Koolhaas JM. 2001. Te mporal and spatial dynamics of corticosteroid receptor down regulation in rat brain following s ocial defeat. Physiol Behav 72(3):349 54. Cahn C. 2006. Roland kuhn, 1912 2005. Neuropsychopharmacology 31(5):1096 Campo longo P and Roozendaal B. 2011. Acute glucocorticoids interact with arousal state in reg u lating Long term memory formation. The Handbook of Stress :179 200. Cannon WB. 1935. Stresses and strains of homeostasis. Am J Med Sci 189(1):13. Carpenter RE, Korzan WJ, Bockholt C, Watt MJ, Forster GL, Renner KJ, Summers CH. 2009. Cort icotropin releasing factor influences aggression and monoamines: Modulation of attacks and r etreats. Neuroscience 158(2):412 25.
108 Carpenter RE and Summers CH. 2009. Learning strategies du ring fear conditioning. Neurobiol Learn Mem 91(4):415 23. Cayre M, Malaterre J, Scotto Lomassese S, Strambi C, Strambi A. 2002. The common properties of neurogenesis in the adult brain: From invertebrates to vertebrates. Comparative Biochemistry and Physio logy Part B: Biochemistry and Molecular Biology 132(1):1 15. Celotti F, Melcangi R, Martini L. 1992. The 5 alpha reductase in the brain: Molecular aspects and relation to brain function. Front Neuroendocrinol 13(2):163. Cerd Reverter JM and Canosa LF. 200 9. Chapter 1 neuroendocrine systems of the fish brain. In: Fish physiology. Dr. Nicholas J. Bernier, Dr. Glen Van Der Kraak,Dr.Anthony P.Farrell and Dr.Colin J.Brauner, editor. Academic Press. 3 p. Chandroo K, Duncan I, Moccia R. 2004. Can fish suffer?: Pe rspectives on sentience, pain, fear and stress. Appl Anim Behav Sci 86(3 4):225 50. Chang CL and Hsu SYT. 2004. Ancient evolution of stress regulating peptides in vertebrates. Pe ptides 25(10):1681 8. Charalampopoulos I, Remboutsika E, Margioris AN, Gravani s A. 2008. Neurosteroids as modul ators of neurogenesis and neuronal survival. Trends in Endocrinology & Metabolism 19(8):300 7. Chatzaki E, Charalampopoulos I, Leontidis C, Mouzas I, Tzardi M, Tsatsanis C, Margioris A, Gravanis A. 2003. Urocortin in human gastric mucosa: Relationship to inflammatory activity. Journal of Clinical Endocrinology & Metabolism 88(1):478 83. Chen CC and Fernald RD. 2008. Sequences, expression patterns and regulation of the corticotr opin releasing factor system in a teleost. Gen C omp Endocrinol 157(2):148 55. Chen MK and Guilarte TR. 2008. Translocator protein 18 kDa (TSPO): Molecular sensor of brain injury and repair. Pharmacol Ther 118(1):1 17. CHENEY DL. 1995. Pregnenolone sulfate antagonizes dizocilpine amnesia: Role for allopr egnan olone. Neuroreport 6(12):1697. Chisari M, Eisenman LN, Covey DF, Mennerick S, Zorumski CF. 2010. The sticky issue of neur osteroids and GABAA receptors. Trends Neurosci 33(7):299 306. Chu S and Metcalfe CD. 2007. Analysis of paroxetine, fluoxetine and norfluoxetine in fish tissues using pressurized liquid extraction, mixed mode solid phase extraction cleanup and liquid chr omatography tandem mass spectrometry. Journal of Chromatography A 1163(1 2):112 8. Claxton AJ, Li Z, McKENDRICK J. 2000. Selective se rotonin reuptake inhibitor treatment in the UK: Risk of relapse or recurrence of depression. The British Journal of Psychiatry 177(2):163 8. Clunie M. 2003. Psychiatric side effects of indomethacin in parturients. Canadian Journal of A nesthesia 50(6):586.
109 Cohen MJ and Strumwasser F, editors. 1985. Comparative neurobiology: Modes of communic ation in the nervous system. Collins LE and Waldeck RF. 2006. Telencephalic ablation results in decreased startle response in goldfish. Brain Res 1111(1):162 5. Colwill RM and Creton R. 2011. Imaging escape and avoidance behavior in zebrafish larvae. Rev Neurosci 22(1):63 73. Peptides 29(5):651 7. Cook AF, Stacey NE, Peter R E. 1980. Periovulatory changes in serum cortisol levels in the gol dfish, carassius auratus. Gen Comp Endocrinol 40(4):507 10. hydroxysteroid dehydrogenase type 1 and its role in the hypothalamus pituitary adrenal axis, m etabolic syndrome, and inflammation. Journal of Clinical Endocrinology & Metabolism 94(12):4645 54. Corkin S. 2002. What's new with the amnesic patient HM? Nature Reviews Neuroscience 3(2):153 60. Corsi L, Geminiani E, Baraldi M. 2008. Peripheral benzodiaz epine receptor (PBR) new insight in cell proliferation and cell differentiation review. Current Clinical Pharmacology 3(1):38 45. Cowen PJ. 2008. Serotonin and depression: Pathophysiological mechanism or marketing myth? Trends Pharmacol Sci 29(9):433 6. Cr immins E, Vasunilashorn S, Kim JK, Alley D. 2008. Chapter 5 biomarkers related to aging in human populations. In: Advances in clinical chemistry. Gregory S. Makowski, editor. Elsevier. 161 p. Dalsgaard MK, Madsen FF, Secher NH, Laursen H, Quistorff B. 2011 High glycogen levels in the hippocampus of patients with epilepsy J Cereb Blood Flow Metab 31(2):792. Danscher G and Fjerdingstad EJ. 1975. Diethyldithiocarbamate (antabuse): Decrease of brain heavy metal staining pattern and improved consilidation of sh uttle avoidance in goldfish. Brain Res 83(1):143 55. D'Aquila PS, Canu S, Sardella M, Spanu C, Serra G, Franconi F. 2010. Dopamine is involved in the antidepressant like effect of allopregnanolone in the forced swimming test in female rats. Behav Pharmac ol 21(1):21. Delgado JMR and Anand BK. 1952. Increase of food intake induced by electrical stimulation of the lateral hypothalamus. American Journal of Physiology -Legacy Content 172(1):162 8. Demski LS. 1984. Can parcellation account for the evolution of behavioral plasticity associated with large brains? Behav Brain Sci 7(03):335.
110 Denver RJ. 2009. Structural and functional evolution of vertebrate neuroendocrine stress sy stems. Ann N Y Acad Sci 1163(1):1 16. Diamond J. 1973. Analysis of mauthner cell respo nses to iontophoretically delivered pulses of GABA, glycine and L glutamate. J Physiol (Lond ) 232(1):113. Diamond J. 1971. The mauthner cell. In: Fish physiology. W.S. Hoar and D.J. Randall, editor. Ac ademic Press. 265 p. Diamond J. 1968. The activation a nd distribution of GABA and L glutamate receptors on goldfish mauthner neurones: An analysis of dendritic remote inhibition. J Physiol (Lond ) 194(3):669. Diorio D, Viau V, Meaney M. 1993. The role of the medial prefrontal cortex (cingulate gyrus) in the r egulation of hypothalamic pituitary adrenal responses to stress. J Neurosci 13(9):3839 47. Diotel N, Do Rego JL, Anglade I, Vaillant C, Pellegrini E, Vaudry H, Kah O. 2011. The brain of tel eost fish, a source, and a target of sexual steroids. Frontiers in Neuroscience 5. Diotel N, Page YL, Mouriec K, Tong S, Pellegrini E, Vaillant C, Anglade I, Brion F, Pakdel F, Chung B, et al. 2010. Aromatase in the brain of teleost fish: Expression, regulation and putative fun ctions. Front Neuroendocrinol 31(2):172 92. D jebaili M, Guo Q, Pettus EH, Hoffman SW, Stein DG. 2005. The neurosteroids progesterone and allopregnanolone reduce cell death, gliosis, and functional deficits after traumatic brain injury in rats. J Neurotrauma 22(1):106 18. Doering DD, Steckelbroeck S, reductase type 1 and type 2 activity. Steroids 67(10):859 67. Duman RS, Nakagawa S, Malberg J. 2001. Regulation of adult neurogenesis by antidepressant treatment. Neuropsychopharmacology 25(6 ):836 44. Dunlap KD, McCarthy EA, Jashari D. 2008. Electrocommunication signals alone are sufficient to increase neurogenesis in the brain of adult electric fish, apteronotus leptorhynchus. Develo pmental Neurobiology 68(12):1420 8. Dunlop R, Millsopp S, La ming P. 2006. Avoidance learning in goldfish (carassius auratus) and trout (oncorhynchus mykiss) and implications for pain perception. Appl Anim Behav Sci 97(2 4):255 71. Durn E, Ocaa FM, Broglio C, Rodrguez F, Salas C. 2010. Lateral but not medial tele ncephalic pallium ablation impairs the use of goldfish spatial allocentric strategies in a. Behav Brain Res 214(2):480 7. Durn E, Ocaa FM, Gmez A, Jimnez Moya F, Broglio C, Rodrguez F, Salas C. 2008. Telencep halon ablation impairs goldfish allocentric Exp 68:519 25.
111 Eaton RC. 1988. Flexible body dynamics of the goldfish C start: Implications for reticulospinal command mechanisms. The Journal of Neuroscience 8(8):2758. Eaton RC. 1977. The mauthner initiated startle response in teleost fish. J Exp Biol 66:65. Egan RJ, Bergner CL, Hart PC, Cachat JM, Canavello PR, Elegante MF, Elkhayat SI, Bartels BK, Tien AK, Tien DH, et al. 2009. Understanding behavioral and physiological phenotypes of stress and an xiety in zebrafish. Behav Brain Res 205(1):38 44. Elakovic I, Djordjevic A, Adzic M, Djordjevic J, Radojcic M, Matic G. 2011. Gender specific r esponse of brain corticosteroid receptors to stress and fluoxetine Brain Res Eliceiri BP, Gonzalez AM, Baird A. 2011. Zebrafish model of the blood brain barrier: Morpholog ical and permeability studies. Methods Mol Biol 686:371 8. Emmanuvel Rajan K, Ganesh A, Dharaneedharan S, Radhakrishnan K. 2011. Spatial learning induced egr 1 expression in telencephalon of gold fish carassius auratus. Fish Physiol Biochem 37(1):153 9. Engin E and Treit D. 2007. The anxiolytic like effects of allopregnanolone vary as a function of intracerebral microinfusion site: The amygdala, medial prefrontal cortex, or hippocampus. B ehav Pharm acol 18(5 6):461. Erickson KI, Prakash RS, Voss MW, Chaddock L, Hu L, Morris KS, White SM, Wjcicki TR, McAuley E, Kramer AF. 2009. Aerobic fitness is associated with hippocampal volume in elderly humans. Hippocampus 19(10):1030 9. Erickson KI, Prakash RS, Voss MW, Chaddock L, Heo S, McLaren M, Pence BD, Martin SA, Vieira VJ, Woods JA, et al. 2010. Brain derived neurotrophic factor is associated with age related d ecline in hippocampal volume. J Neurosci 30(15):5368 75. Escudero C, Casas S, Giuliani F, Bazzo cchini V, Garca S, Yunes R, Cabrera R. 2011. Allopregnan olone prevents memory impairment: Effect on mRNA expression and enzymatic activity of hipp ocampal 3 reductase. Brain Res Bull Espallergues J, Givalois L, Temsamani J, Laruell e C, Maurice T. 2009. The 3 [beta] hydroxysteroid dehydrogenase inhibitor trilostane shows antidepressant properties in mice. Psychoneuroend ocrinology 34(5):644 59. Falls W, Miserendino M, Davis M. 1992. Extinction of fear potentiated startle: Blockade by inf usion of an NMDA antagonist into the amygdala. The Journal of Neuroscience 12(3):854 63. Fangmann P, Assion HJ, Juckel G, Gonzlez C, Lpez Muoz F. 2008. Half a century of antid epressant drugs: On the clinical introduction of monoamine oxidase inhibit ors, tricyclics, and te tracyclics. part II: Tricyclics and tetracyclics. J Clin Psychopharmacol 28(1):1.
112 Feinstein JS, Adolphs R, Damasio A, Tranel D. 2011. The human amygdala and the induction and experience of fear. Current Biology 21(1):34 8. Fernandes MC, Alvares EP, Gama P, Silveira M. 2003. Serotonin in the nervous system of the head region of the land planarian bipalium kewense. Tissue and Cell 35(6):479 86. Fernndez Guasti A and Picazo O. 1999. Sexual differentiation modifies the allopregnanolone a nxiolytic actions in rats. Psychoneuroendocrinology 24(3):251 67. Fernndez Guasti A and Picazo O. 1995. Flumazenil blocks the anxiolytic action of allopregnan olone. Eur J Pharmacol 281(1):113 5. Flik G, Klaren PHM, Van den Burg EH, Metz JR, Huising MO. 20 06. CRF and stress in fish. Gen Comp Endocrinol 146(1):36 44. Flink EB. 1961. Adenohypophysis and adrenal cortex. Annu Rev Physiol 23(1):229 62. Flood J and Cherkin A. 1987. Fluoxetine enhances memory processing in mice. Psychopharm acology (Berl ) 93(1):36 43. Flood NC, Overmier JB, Savage GE. 1976. Teleost telencephalon and learning: An interpretive review of data and hypotheses. Physiol Behav 16(6):783 98. Floras JS, Aylward PE, Victor RG, Mark AL, Abboud FM. 1988. Epinephrine facilitates neurogenic vasoc onstriction in humans. J Clin Invest 81(4):1265. Flores AM. 2011. Does the Mineralocorticoid Receptor have a Functional Role in the Teleost Gill? Flores CE. 2010. Variability of distribution of Ca2 /Calmodulin dependent kinase II at mixed sy napses on the mauthner cell: Colocalization and association with connexin 35. The Journal of Neuroscience 30(28):9488. Flores CE. 2008. An immunochemical marker for goldfish mauthner cells. J Neurosci Methods 175(1):64. Frye CA. 2009. Progesterone reduces depressive be haviour of young ovariectomized, aged pr ogestin receptor knockout, and aged wild type mice in the tail suspension test. Journal of Psych opharmacology Frye CA, Hirst JJ, Brunton PJ, Russell JA. 2011. Neurosteroids for a successful pregnancy. Stress 14(1): 1 5. FRYER J, LEDERIS K, RIVIER J. 1983. UROTENSIN I, A CRF LIKE NEUROPEPTIDE, STIMULATES ACTH. RELEASE FROM THE TELEOST PITUITARY. Endocrinology 113(6):2308 10. Fryer JN and Leung E. 1982. Neurohypophysial hormonal control of cortisol secretion in the te leost carassius auratus. Gen Comp Endocrinol 48(4):425 31.
113 Fujii K. 1988. Phylogenetic detection of serotonin immunoreactive cells in the central nervous system of invertebrates. Comparative Biochemistry and Physiology.C, Comparative Pharmacol ogy and Toxico logy 89(2):233. Fuller RW, Wong DT, Robertson DW. 1991. Fluoxetine, a selective inhibitor of serotonin uptake. Med Res Rev 11(1):17 34. Furshpan EJ. 1964. "Electrical transmission" at an excitatory synapse in a vertebrate brain. Sc ience 144(3620):878 80. G allon RL. 1974. Spatial location of a visual signal and shuttle box avoidance acquisition by gol dfish (carassius auratus). J Comp Physiol Psychol 86(2):316 21. Garcia A, Marti O, Valles A, Dal Zotto S, Armario A. 2000. Recovery of the hypothalamic pituitar y adrenal response to stress. Neuroendocrinology 72(2):114 25. Garina DV, Kuz'mina VV, Gerasimov YV. 2007. The effect of epinephrine on feeding and motion patterns in goldfish carassius auratus (L.). Comparative Biochemistry and Physiology Part A: Molecu lar & Integrative Physiology 148(3):544 9. Gaspar P. 2003. The developmental role of serotonin: News from mouse molecular genetics. Nature Reviews.Neuroscience 4(12):1002. Genazzani AR. 1998. Circulating levels of allopregnanolone in humans: Gender, age, a nd end ocrineinfluences. J Clin Endocrinol Metab 83(6):2099. George O, Valle M, Vitiello S, Moal ML, Piazza P, Mayo W. Low brain allopregnanolone levels mediate flattened circadian activity associated with memory impairments in aged rats. Biol Ps ychiatry I n Press, Corrected Proof. Geven EJW, Verkaar F, Flik G, Klaren PHM. 2006. Experimental hyperthyroidism and central m ediators of stress axis and thyroid axis activity in common carp (cyprinus carpio L.). J Mol End ocrinol 37(3):443. Gibbs DM. 1983. Effect of the serotonin reuptake inhibitor fluoxetine on corticotropin releasing factor and vasopressin secretion into hypophysial portal blood. Brain Res 280(1):176. Giraldez Perez RM, Gaytan SP, Torres B, Pasaro R. 2009. Co localization of nitric oxide synthase a nd choline acetyltransferase in the brain of the goldfish (carassius auratus). J Chem Neuroanat 37(1):1 17. Girdler SS, Straneva PA, Light KC, Pedersen CA, Morrow AL. 2001. Allopregnanolone levels and reactivity to mental stress in premenstrual dysphoric d isorder. Biol Psychiatry 49(9):788 97. Gold P and Chrousos G. 2002. Organization of the stress system and its dysregulation in mela ncholic and atypical depression: High vs low CRH/NE states. Mol Psychiatry 7(3):254 75.
114 GOLDSMITH P and FLEMING A. 2007. Scre ening methods employing zebrafish and the blood brain barrier. Screening Methods Employing Zebrafish and the Blood Brain Barrier Gotz V. 1978. Paranoid psychosis with indomethacin. Br Med J 1(6104):49. Gravanis AG and Mellon SH. 2010. Hormones in neurode generation, neuroprotection and ne urogenesis. Recherche 67:02. Gray T. 1993. Amygdaloid CRF pathways: Role in autonomic, neuroendocrine, and behavioral responses to stress. Ann N Y Acad Sci 697(1):53 60. Green A, Gabrielsson J, Fone K. 2011. Translational neuropharmacology and the appropriate and effective use of animal models. Br J Pharmacol 164(4):1041 3. Griffin LD and Mellon SH. 1999. Selective serotonin reuptake inhibitors directly alter activity of neurosteroidogenic enzymes. Proc Natl Acad Sci U S A 96(23):pp. 13512 13517. Grissom N and Bhatnagar S. 2009. Habituation to repeated stress: Get used to it. Neurobiol Learn Mem 92(2):215 24. Gruber C, Kahl A, Lebenheim L, Kowski A, Dittgen A, Veh RW. 2007. Dopaminergic projections from the VTA substantially contribute to the mesohabenular pathway in the rat. Neurosci Lett 427(3):165 70. Grupp L, Wolburg H, Mack AF. 2010. Astroglial structures in the zebrafish brain. J Comp Neurol 518(21):4277 87. Guidotti A, Dong E, Matsumoto K, Pinna G, Rasmusson AM, Costa E. 2001. The socially isolated dihydroprogesterone in psychiatric disorders. Brain Res Rev 37(1 3):110 5. Gulinello M, Orman R, Smith SS. 2003. Sex differences in anxiety, sensorimotor ga ting and e xpression of the ?4 subunit of the GABAA receptor in the amygdala after progesterone withdra wal. Eur J Neurosci 17(3):641 8. Haddad P. 1998. The SSRI discontinuation syndrome. Journal of Psychopharmacology 12(3):305 13. Hagan JJ, Leslie RA, Patel S, Evans ML, Wattam TA, Holmes S, Benham CD, Taylor SG, Routledge C, Hemmati P. 1999. Orexin A activates locus coeruleus cell firing and increases arousal in the rat. Proceedings of the National Academy of Sciences 96(19):10911. Hajszan T, Dow A, Warner S chmidt JL, Szigeti Buck K, Sallam NL, Parducz A, Leranth C, Duman RS. 2009. Remodeling of hippocampal spine synapses in the rat learned helplessness model of depression. Biol Psychiatry 65(5):392 400. Hale RL. 1990. Indomethacin does not antagonize the anxiolytic action of ethanol in the elevated plus maze. Psychopharmacology (Berl ) 101(2):203.
115 Halliwell B and Gutteridge JMC. 1986. Free radicals in biology and medicine. Oxford; New York: Clarendon Press ; Oxford University Press. Hampson RE, Miller F, P alchik G, Deadwyler SA. 2011. Cannabinoid receptor activation modifies NMDA receptor mediated release of intracellular calcium: Implications for endocannabinoid control of hippocampal neural plasticity Neuropharmacology Hemmrich G and Bosch TC. 2008. Com pagen, a comparative genomics platform for early branc hing metazoan animals, reveals early origins of genes regulating stem cell differentiation. B ioessays 30(10):1010 8. Herd MB, Belelli D, Lambert JJ. 2007. Neurosteroid modulation of synaptic and extrasy naptic GABAA receptors. Pharmacol Ther 116(1):20 34. Herman JP and Cullinan WE. 1997. Neurocircuitry of stress: Central control of the hypothalamo pituitary adrenocortical axis. Trends Neurosci 20(2):78 84. Het S, Ramlow G, Wolf OT. 2005. A meta analytic r eview of the effects of acute cortisol admi nistration on human memory. Psychoneuroendocrinology 30(8):771 84. Hikosaka O, Sesack SR, Lecourtier L, Shepard PD. 2008. Habenula: Crossroad between the basal ganglia and the limbic system. The Journal of Neurosc ience 28(46):11825 9. Hodes GE, Hill Smith TE, Suckow RF, Cooper TB, Lucki I. 2010. Sex specific effects of chronic fluoxetine treatment on neuroplasticity and pharmacokinetics in mice. Journal of Pharmacology and Experimental Therapeutics 332(1):266 73. H ollis D, Goetz F, Roberts S, Boyd S. 2004. Acute neurosteroid modulation and subunit isolation of the gamma aminobutyric acidA receptor in the bullfrog, rana catesbeiana. J Mol Endocrinol 32(3):921 34. Holloszy JO and Coyle EF. 1984. Adaptations of skeleta l muscle to endurance exercise and their metabolic consequences. J Appl Physiol 56(4):831 8. Holsboer F. 2001. Stress, hypercortisolism and corticosteroid receptors in depression: Implic atons for therapy. J Affect Disord 62(1 2):77 91. Holsboer F and Ising M. 2008. Central CRH system in depression and anxiety -evidence from cli nical studies with CRH1 receptor antagonists. Eur J Pharmacol 583(2 3):350 7. Hong S and Hikosaka O. 2008. The globus pallidus sends reward related signals to the lateral habenula. Ne uron 60(4):720 9. Horner JL. 1961. A shuttle box for fish and a control circuit of general applicability. Am J Psychol 74(1):114. Hornung J. 2003. The human raphe nuclei and the serotonergic system. J Chem Neuroanat 26(4):331 43.
116 Hosie AM. 2009. Conserved site for neurosteroid modulation of GABAA receptors. Neuropha rmacology 56(1):149. Hosie AM, Wilkins ME, Smart TG. 2007. Neurosteroid binding sites on GABAA receptors. Pha rmacol Ther 116(1):7 19. Huising M, Metz J, Van Schooten C, Taverne Thiele A, Hermsen T, Verburg van Kemenade B, Flik G. 2004. Structural characterisation of a cyprinid (cyprinus carpio L.) CRH, CRH BP and CRH R1, and the role of these proteins in the acute stress response. J Mol Endocrinol 32(3):627 48. Hurtado Parrado C. 2010. Neuronal me chanisms of learning in teleost fish. Universitas Psych ologica (3):663 78. Idler D and Freeman H. 1968. Binding of testosterone, 1 [alpha] hydroxycorticosterone and co rtisol by plasma proteins of fish. Gen Comp Endocrinol 11(2):366 72. Iino K, Sasano H, Ok i Y, Andoh N, Shin RW, Kitamoto T, Totsune K, Takahashi K, Suzuki H, Nagura H. 1997. Urocortin expression in human pituitary gland and pituitary adenoma. Journal of Clin ical Endocrinology & Metabolism 82(11):3842 50. Imagawa T, Kitagawa H, Uehara M. 1996. The innervation of the chromaffin cells in the head kidney of the carp, cyprinus carpio; regional differences of the connections between nerve en dings and chromaffin cells. J Anat 188(Pt 1):149. Iovieno N, van Nieuwenhuizen A, Clain A, Baer L, Nierenberg A A. 2011. Residual symptoms after remission of major depressive disorder with fluoxetine and risk of relapse Depress Anxiety 28(2):137 44. Ishrat T, Sayeed I, Atif F, Hua F, Stein DG. 2010. Progesterone and allopregnanolone attenuate blood brain barrier dys function following permanent focal ischemia by regulating the expression of matrix metalloproteinases. Exp Neurol 226(1):183 90. Ising M, Knzel HE, Binder EB, Nickel T, Modell S, Holsboer F. 2005. The combined dexam ethasone/CRH test as a potential surroga te marker in depression. Prog Neuro Psychopharmacol Biol Psychiatry 29(6):1085 93. Jacob TC. 2008. GABAA receptor trafficking and its role in the dynamic modulation of neuronal inhibition. Nature Reviews.Neuroscience 9(5):331. Jacobs BL, van Praag H, Gage FH. 2000. Adult brain neurogenesis and psychiatry: A novel theory of depression. Mol Psychiatry 5(3):262 9. Jain NS, Hirani K, Chopde CT. 2005. Reversal of caffeine induced anxiety by neurosteroid 3 alpha hydroxy 5 alpha pregnane 20 one in rats. Neuropharm acology 48(5):627 38. Jazayeri S, Tehrani Doost M, Keshavarz SA, Hosseini M, Djazayery A, Amini H, Jalali M, Peet M. 2008. Comparison of therapeutic effects of omega 3 fatty acid eicosapentaenoic acid and fluo x-
117 etine, separately and in combination, in major depressive disorder. Aust N Z J Psychiatry 42(3):192 8. Ji H and Shepard PD. 2007. Lateral habenula stimulation inhibits rat midbrain dopamine neurons through a GABAA receptor mediated mechanism. The Journal of Neuroscience 27(26):6923 30. Johansson I, Bi rzniece V, Lindblad C, Olsson T, Bckstrm T. 2002. Allopregnanolone inhibits learning in the morris water maze. Brain Res 934(2):125 31. Johnson BT and Kirsch I. 2008. Do antidepressants work? statistical significance versus clinical benefits. Significanc e 5(2):54 8. Johnson TP. 1998. Partitioning the effects of temperature and kinematic viscosity on the C start performance of adult fishes. J Exp Biol 201:2045. Jonnakuty C. 2008. What do we know about serotonin? J Cell Physiol 217(2):301. Joyce P, Mulder R Luty S, McKenzie J, Rae A. 2003. A differential response to nortriptyline and fluoxetine in melancholic depression: The importance of age and gender. Acta Psychiatr Scand 108(1):20 3. Kandel ER, Schwartz JH, Jessel TM, editors. 2000. Principles of neural science. 4th ed. New York, NY: McGraw Hill. 1414 p. Kandel ER. 1965. Mechanism of heterosynaptic facilitation in the giant cell of the abdominal ganglion of aplysia depilans. J Physiol (Lond ) 181(1):28. Karssen A, Meijer O, Van Der Sandt I, Lucassen P, D e Lange E, De Boer A, De Kloet E. 2001. Mu ltidrug resistance P glycoprotein hampers the access of cortisol but not of corticosterone to mouse and human brain. Endocrinology 142(6):2686 94. Kask K, Bckstrm T, Nilsson LG, Sundstrm Poromaa I. 2008. Allopre gnanolone impairs episodic memory in healthy women. Psychopharmacology (Berl ) 199(2):161 8. Kassahn KS, Crozier RH, Prtner HO, Caley MJ. 2009. Animal performance and stress: Responses and tolerance limits at different levels of biological organisation. B iological Reviews 84(2):277 92. Kato N, Kamimura S, Uematsu K. 2008. Central nervous system underlying fish swimming [A r eview]. In: Bio mechanisms of swimming and flying. Springer Japan. KAYE WH, GWIRTSMAN HE, GEORGE DT, EBERT MH, JIMERSON DC, TOMAI TP, C HROUSOS GP, GOLD PW. 1987. Elevated cerebrospinal fluid levels of immunoreactive corticotropin releasing hormone in anorexia nervosa: Relation to state of nutrition, adrenal function, and intensity of depression. Journal of Clinical Endocrinology & Metabol ism 64(2):203 8. Kempermann G and Kronenberg G. 2003. Depressed new neurons? Adult hippocampal neur ogenesis and a cellular plasticity hypothesis of major depression. Biol Psychiatry 54(5):499 503.
118 Kerr AL, Steuer EL, Pochtarev V, Swain RA. 2010. Angiogenes is but not neurogenesis is critical for normal learning and memory acquisition Neuroscience 171(1):214 26. Kim H, Shimojo S, O'Doherty J,P. 2006. Is avoiding an aversive outcome rewarding? neural su bstrates of avoidance learning in the human brain. PLoS Bi ol 4(8):e233. Kim YS, Stumpf WE, Sar M. 1978. Topography of estrogen target cell in the forebrain of goldfish, carassius auratus. J Comp Neurol 182(4):611 20. King SR. 2008. Emerging roles for neurosteroids in sexual behavior and function. J Androl 29(5):5 24 33. Kirsch I, Deacon BJ, Huedo Medina T, Scoboria A, Moore TJ, Johnson BT. 2008. Initial severity and antidepressant benefits: A meta analysis of data submitted to the food and drug admi nistration. PLoS Med 5(2):e45. Kirschbaum C, Wolf O, May M, Wippich W, Hellhammer D. 1996. Stress and treatment induced elevations of cortisol levels associated with impaired declarative memory in healthy adults. Life Sci 58(17):1475 83. Kitraki E, Kremmyda O, Youlatos D, Alexis MN, Kittas C. 2004. Gender dependent altera tions in corticosteroid receptor status and spatial performance following 21 days of restraint stress. Neuroscience 125(1):47 55. Koenig E. 1979. Ribosomal RNA in mauthner axon: Implications for a protein synthesizing m achinery in the myelinated axon. Brai n Res 174(1):95 107. Kohm AP and Sanders VM. 2000. Norepinephrine: A messenger from the brain to the immune system. Immunol Today 21(11):539 42. Koolhaas JM, Bartolomucci A, Buwalda B, de Boer SF, Flgge G, Korte SM, Meerlo P, Murison R, Olivier B, Palanza P, et al. 2011. Stress revisited: A critical evaluation of the stress concept. Ne uroscience & Biobehavioral Reviews 35(5):1291 301. Kopp S, Baur R, Sigel E, Mhler H, Altmann K 2010. Highly potent modulation of GABAA rece ptors by valerenic acid derivati ves. ChemMedChem 5(5):678 81. Kraemer HC, Schultz SK, Arndt S. 2002. Biomarkers in psychiatry: Methodological issues. Amer ican Journal of Geriatric Psych 10(6):653. Kreke N. 2008. Physiological endpoints for potential SSRI interactions in fish. Crit Rev To xicol 38(3):215. Kuhn E, Corneillie S, Ollevier F. 1986. Circadian variations in plasma osmolality, electrolytes, gl ucose, and cortisol in carp (cyprinus carpio). Gen Comp Endocrinol 61(3):459 68.
119 Kumsta R, Entringer S, Koper JW, van Rossum EFC, Hellhammer DH, Wst S. 2007. Sex specific associations between common glucocorticoid receptor gene variants and hypothalamus pituitary adrenal axis responses to psychosocial stress. Biol Psychiatry 62(8):863 9. Lacasse JR and Leo J. 2005. Serotonin and depression: A disconnect between the advertisements and the scientific literature. PLoS Medicine 2(12):1211. Laconi MR, Casteller G, Gargiulo PA, Bregonzio C, Cabrera RJ. 2001. The anxiolytic effect of all opregnanolone is associated with gonadal hormonal status in fema le rats. Eur J Pharmacol 417(1 2):111 6. Lambert JJ, Cooper MA, Simmons RDJ, Weir CJ, Belelli D. 2009. Neurosteroids: Endogenous all osteric modulators of GABAA receptors. Psychoneuroendocrinology 34(Supplement 1):S48 58. Laming PR. 1980. Physiological chan ges observed in the goldfish (carassius auratus) during b ehavioral arousal and fright. Behav Neural Biol 29(2):255. Larsson LE. 1956. The relation between the startle reaction and the non specific EEG response to sudden stimuli with a discussion on the mec hanism of arousal. Electroencephalogr Clin Ne urophysiol 8(4):631. hydroxysteroid dehydrogenase isoforms 1 and 2 of adrenal origin. Mol Cell Endocrinol 243(1):43 50. Leder is K, Letter A, McMaster D, Moore G, Schlesinger D. 1982. Complete amino acid sequence of urotensin I, a hypotensive and corticotropin releasing neuropeptide from catostomus. Sc ience 218(4568):162 5. Lederis K, Fryer J, Rivier J, MacCannell KL, Kobayashi Y Woo N, Wong KL. 1985. Neurohormones from fish tails. II: Actions of urotensin I in mammals and fishes. Recent Prog Horm Res 41:553 76. Leibowitz SF. 1975. Catecholaminergic mechanisms of the lateral hypothalamus: Their role in the mediation of amphetamin e anorexia. Brain Res 98(3):529 45. Lemaire V, Koehl M, Le Moal M, Abrous D. 2000. Prenatal stress produces learning deficits ass ociated with an inhibition of neurogenesis in the hippocampus. Proceedings of the National Academy of Sciences 97(20):11032. Le nnartsson AK, Kushnir MM, Bergquist J, Billig H, Jonsdottir IH. 2012. Sex steroid levels temp orarily increase in response to acute psychosocial stress in healthy men and women. International Journal of Psychophysiology Lenz HJ, Fisher LA, Vale WW, Brown MR. 1985. Corticotropin releasing factor, sauvagine, and urotensin I: Effects on blood flow. American Journal of Physiology Regulatory, Integrative and Comparative Physiology 249(1):R85 90.
120 Levenson JL. 1993. Fluoxetine treatment of depression caused by in terferon nterol 88(5):760. Li B, Piriz J, Mirrione M, Chung CH, Proulx CD, Schulz D, Henn F, Malinow R. 2011. Synaptic p otentiation onto habenula neurons in the learned helplessness model of depression. Nature 470(7335):535 9. Li WL, Cai HH, Wang B, Chen L, Zhou QG, Luo CX, Liu N, Ding XS, Zhu DY. 2009. Chronic fluoxetine treatment improves ischemia induced spatial cognitive deficits through increasing hippoca mpal neurogenesis after stroke. J Neurosci Res 87(1):112 22. Li D, Xie P, Zhang X. 2 008. Changes in plasma thyroid hormones and cortisol levels in crucian carp (carassius auratus) exposed to the extracted microcystins. Chemosphere 74(1):13 8. Li G, Cherrier MM, Tsuang DW, Petrie EC, Colasurdo EA, Craft S, Schellenberg GD, Peskind ER, Rask ind MA, Wilkinson CW. 2006. Salivary cortisol and memory function in human aging. Ne urobiol Aging 27(11):1705 14. Lin CH, Tsai IL, Su CH, Tseng DY, Hwang PP. 2011. Reverse effect of mammalian hypocalcemic cortisol in fish: Cortisol stimulates Ca2 uptake vi a glucocorticoid receptor mediated vitamin D3 metabolism. PloS One 6(8):e23689. Lin DD. 2005. Diagnosis and management of surgical adrenal diseases. Urology Lindauer RJL. 2006. Cortisol, learning, memory, and attention in relation to smaller hippocampal volume in police officers with posttraumatic stress disorder. Biol Psychiatry 59(2):171. Liu W, Wang J, Sauter NK, Pearce D. 1995. Steroid receptor heterodimerization demonstrated in vitro and in vivo. Proceedings of the National Academy of Sciences 92(26) :12480. Liu J. 1999. Stress, aging, and brain oxidative damage. Neurochem Res 24(11):1479. Lommatzsch M, Hornych K, Zingler C, Schuff Werner P, Hppner J, Virchow JC. 2006. Maternal serum concentrations of BDNF and depression in the perinatal period. Psych oneuroendocrino logy 31(3):388 94. Lpez JF, Chalmers DT, Little KY, Watson SJ. 1998. Regulation of serotonin< sub> 1A, gl ucocorticoid, and mineralocorticoid receptor in rat and human hippocampus: Implications for the neurobiology of depression. Biol Psychiatry 43(8):547 73. Lopez Munoz F and Alamo C. 2009. Monoaminergic neurotransmission: The history of the di scovery of antidepressants from 1950s until today. Curr Pharm Des 15(14):1563 86. Lovely RH, Pagano RR, Paolino RM. 1972. Shuttle box avoidance performance and basal cort icosterone levels as a function of duration of individual housing in rats. J Comp Physiol Psychol 81(2):331 5.
121 Lupien SJ, Wilkinson CW, Brire S, Mnard C, Ng Ying Kin N, Nair N. 2002. The modulatory e ffects of corticosteroids on cognition: Studies in young human populations. Psychoneuroendocr inology 27(3):401 16. Lupien S. 1994. Basal cortisol levels and cognitive deficits in human aging. The Journal of Neur oscience 14(5):2893. Lupien SJ, Evans A, Lord C, Miles J, Pruessner M, Pike B, Pruessner JC. 2007. Hippocampal vo lume is as variable in young as in older adults: Implications for the notion of hippocampal atr ophy in humans. Neuroimage 34(2):479 85. Lupien SJ and McEwen BS. 1997. The acute effects of corticosteroids on cogniti on: Integration of animal and human model studies. Brain Res Rev 24(1):1 27. Macdonald RL and Olsen RW. 1994. GABAA receptor channels. Annu Rev Neurosci 17:569 602. Madden CJ and Morrison SF. 2009. Neurons in the paraventricular nucleus of the hypothalamus inhibit sympathetic outflow to brown adipose tissue. American Journal of Physiology Regulatory, Integrative and Comparative Physiology 296(3):R831 43. Maguire J. 2007. Neurosteroid synthesis mediated regulation of GABAA receptors: Relevance to the ovarian cycle and stress. The Journal of Neuroscience 27(9):2155. Maier SF and Seligman ME. 1976. Learned helplessness: Theory and evidence. Journal of Expe rimental Psychology: General; Journal of Experimental Psychology: General 105(1):3. Majewska MD. 1990. The neurosteroid dehydroepiandrosterone sulfate is an allosteric antag onist of the GABAA receptor. Brain Res 526(1):143. Majewska MD. 1986. Steroid hormone metabolites are barbiturate like modulators of the GABA receptor. Science 232(4753):1004. Majewska MD, B isserbe J, Eskay RL. 1985. Glucocorticoids are modulators of GABAA receptors in brain. Brain Res 339(1):178 82. Majewska MD. 1992. Neurosteroids: Endogenous bimodal modulators of the GABAA receptor mechanism of action and physiological significance. Prog N eurobiol 38(4):379 94. Majlessi N and Naghdi N. 2002. Impaired spatial learning in the morris water maze induced by serotonin reuptake inhibitors in rats. Behav Pharmacol 13(3):237. Malberg JE. 2004. Implications of adult hippocampal neurogenesis in antide pressant action. Journal of Psychiatry and Neuroscience 29(3):196. Malberg JE, Eisch AJ, Nestler EJ, Duman RS. 2000. Chronic antidepressant treatment increases neurogenesis in adult rat hippocampus. The Journal of Neuroscience 20(24):9104 10.
122 Malberg JE. 2 003. Cell proliferation in adult hippocampus is decreased by inescapable stress: Reversal by fluoxetine treatment. Neuropsychopharmacology Manent J, Demarque M, Jorquera I, Pellegrino C, Ben Ari Y, Aniksztejn L, Represa A. 2005. A noncanonical release of GABA and glutamate modulates neuronal migration. J Neurosci 25(19):4755 65. Maninger N, Wolkowitz OM, Reus VI, Epel ES, Mellon SH. 2009. Neurobiological and neurops ychiatric effects of dehydroepiandrosterone (DHEA) and DHEA sulfate (DHEAS). Front Neuroe nd ocrinol 30(1):65 91. Marchant NJ, Hamlin AS, McNally GP. 2009. Lateral hypothalamus is required for context induced reinstatement of extinguished reward seeking. The Journal of Neuroscience 29(5):1331 42. Martn A, Boisgard R, Thz B, Van Camp N, Kuhnast B, Damont A, Kassiou M, Doll F, Tavitian B. 2009. Evaluation of the PBR/TSPO radioligand [ 18F] DPA 714 in a rat model of focal cerebral ischemia. Journal of Cerebral Blood Flow & Metabolism 30(1):230 41. Martn I, Gmez A, Salas C, Puerto A, Ro drguez F. 2011. Dorsomedial pallium lesions impair taste aversion learning in goldfish. Neurobiol Learn Mem Martnez lvarez R, Volkoff H, Muoz Cueto J, Delgado M. 2009. Effect of calcitonin gene related peptide (CGRP), adrenomedullin and adrenomedulli n 2/intermedin on food intake in goldfish (carassius auratus). Peptides 30(4):803 7. Maruska KP, Carpenter RE, Fernald RD. 2012. Characterization of cell proliferation throughout the brain of the african cichlid fish astatotilapia burtoni and its regulatio n by social status. J Comp Neurol Mason JW. 1975. A historical view of the stress field. Journal of Human Stress ( 1): 6. Mathieu M, Mensah Nyagan AG, Vallarino M, Do Rgo J Beaujean D, Vaudry D, Luu The V, Pelletier G, Vaudry H. 2001. Immunohistoc hemical localization of 3? hydroxysteroid dehydr ogenase and 5? reductase in the brain of the african lungfish protopterus annectens. J Comp Neurol 438(2):123 35. Matrisciano F, Modafferi AME, Togna GI, Barone Y, Pinna G, Nicoletti F, Scaccianoce S. 2010. R epeated anabolic androgenic steroid treatment causes antidepressant reversible alterations of the hypothalamic pituitary adrenal axis, BDNF levels and behavior. Neuropharmacology 58(7):1078 84. Matsumoto M and Hikosaka O. 2007. Lateral habenula as a source of negative reward signals in dopamine neurons. Nature 447(7148):1111 5.
123 Matsumoto K. 2007. GABAA receptor neurotransmission dysfunction in a mouse model of social isolation induced stress: Possible insights into a non serotonergic mechanism of action of SSRIs in mood and anxiety disorders. Stress 10(1):3. Mattsson C, Reynolds RM, Simonyte K, Olsson T, Walker BR. 2009. Combined receptor antag onist stimulation of the hypothalamic pituitary adrenal axis test identifies impaired negative feedback sensitivity to cortisol in obese men. Journal of Clinical Endocrinology & Metabolism 94(4):1347 52. Mauthner L. 1859. Untersuchungen ber den bau des rckenmarkes der fische. Sitz Ber Kgl Preuss Wiss Maximino C. 2008. A cladistic and comparative analysis of kinemati c components of the fast start of fishes, with a note on body size constraints. Journal of Comparative Physiology.A, Ne uroethology, Sensory, Neural, and Behavioral Physiology 194(11):939 44. Maximino C, Marques dB, Dias,Claudio Alberto Gellis de Mattos, Gouveia A, Morato S. 2010. Scototaxis as anxiety like behavior in fish. Nat Protocols 5(2):209 16. Maximino C, de Brito TM, da Silva Batista AW, Herculano AM, Morato S, Gouveia Jr. A. 2010. Measuring anxiety in zebrafish: A critical review. Behav Brain Res 214(2):157 71. Mazeaud MM, Mazeaud F, Donaldson EM. 1977. Primary and secondary effects of stress in fish: Some new data with a general review. Trans Am Fish Soc 106(3):201 12. McAuley M, Kenny R, Kirkwood T, Wilkinson D, Jones J, Miller V. 2009. A mathem atical model of aging related and cortisol induced hippocampal dysfunction. BMC Neuroscience 10(1):26. McEwen BS. 2009. The brain is the central organ of stress and adaptation. Neuroimage 47(3):911. McEwen BS. 2007. Physiology and neurobiology of stress an d adaptation: Central role of the brain. Physiol Rev 87(3):873 904. MCEWEN BS. 2004. Protection and damage from acute and chronic stress: Allostasis and all ostatic overload and relevance to the pathophysiology of psychiatric disorders. Ann N Y Acad Sci 103 2(1):1 7. McEwen BS and Sapolsky RM. 1995. Stress and cognitive function. Curr Opin Neurobiol 5(2):205 16. McKernan RM and Whiting PJ. 1996. Which GABAA receptor subtypes really occur in the brain? Trends Neurosci 19(4):139 43. Mellon SH and Griffin LD. 20 02. Neurosteroids: Biochemistry and clinical significance. Trends in Endocrinology and Metabolism 13(1):35 43.
124 Meneses A. 1999. 5 HT system and cognition. Neuroscience & Biobehavioral Reviews 23(8):1111 25. Mennigen JA, Sassine J, Trudeau VL, Moon TW. 2010 Waterborne fluoxetine disrupts feeding and energy metabolism in the goldfish carassius auratus. Aquatic Toxicology 100(1):128 37. Mennigen JA, Stroud P, Zamora JM, Moon TW, Trudeau VL. 2011. Pharmaceuticals as neuroe ndocrine disruptors: Lessons learned f rom fish on prozac. J Toxicol Environ Health, Pt B 14(5 7):387 412. Mennigen JA, Harris EA, Chang JP, Moon TW, Trudeau VL. 2009. Fluoxetine affects weight gain and expression of feeding peptides in the female goldfish brain. Regul Pept 155(1 3):99 104. Men nigen JA, Martyniuk CJ, Crump K, Xiong H, Zhao E, Popesku J, Anisman H, Cossins AR, Xia X, Trudeau VL. 2008. Effects of fluoxetine on the reproductive axis of female goldfish (carassius auratus). Physiological Genomics 35(3):273 82. Merali Z, Khan S, Micha ud D, Shippy S, Anisman H. 2004. Does amygdaloid corticotropin releasing hormone (CRH) mediate anxiety like behaviors? dissociation of anxiogenic effects and CRH release. Eur J Neurosci 20(1):229 39. Metcalfe CD, Chu S, Judt C, Li H, Oakes KD, Servos MR, A ndrews DM. 2010. Antidepressants and their metabolites in municipal wastewater, and downstream exposure in an urban watershed. Environmental Toxicology and Chemistry 29(1):79 89. es of mauthner neurons with the preference in the goldfish to turn rightwards or leftwards]. Morfologi 127(2):16 9. Miller DS, Graeff C, Droulle L, Fricker S, Fricker G. 2002. Xenobiotic efflux pumps in isolated fish brain capillaries. American Journal of Physiology Regulatory, Integrative and Comparative Phys iology 282(1):R191 8. Mo Q, Lu S, Garippa C, Brownstein MJ, Simon NG. 2009. Genome wide analysis of DHEA and DHT induced gene expression in mouse hypothalamus and hippocampus. J Steroid Biochem Mol Bi ol 114(3):135 43. Mhler H. 2006. GABA A receptor diversity and pharmacology. Cell Tissue Res 326(2):505. Mommsen TP, Vijayan MM, Moon TW. 1999. Cortisol in teleosts: Dynamics, mechanisms of a ction, and metabolic regulation. Rev Fish Biol Fish 9(3):211 68. MONCRIEFF J. 2002. The antidepressant debate. The British Journal of Psychiatry 180(3):193 4. Moral G, Montes P, Hernndez Morales L, Monfil T, Espinosa Garca C, Cervantes M. 2011. Ne uroprotective effects of progesterone and allopregnanolone on long ter m cognitive outcome a fter global cerebral ischemia. Restorative Neurol Neurosci 29(1):1 15.
125 Morando MB, Medeiros LR, McDonald MD. 2009. Fluoxetine treatment affects nitrogen waste excretion and osmoregulation in a marine teleost fish. Aquatic Toxicology 95 (2):164 71. Morse DE. 1979. GABA induces behavioral and developmental metamorphosis in planktonic molluscan larvae. Brain Res Bull 4(5):704. Motzo C. 1996. Inhibition of basal and stress induced dopamine release in the cerebral cortex and nucleus accumbens of freely moving rats by the neurosteroid allopregnanolone. Journal of Psychopharmacology 10(4):266. Mueller T. 2012. What is the thalamus in zebrafish. Front.Neurosci 6:64. Mueller T. 2011. The conserved bauplan of the teleostean telencephalon. Brain Beh av Evol 78(4):259 60. Mueller T, Dong Z, Berberoglu MA, Guo S. 2011. The dorsal pallium in zebrafish, danio rerio (c yprinidae, teleostei). Brain Res Mukai H. 2007. Local neurosteroid production in the hippocampus: Influence on synaptic plasti city of memory. Neuroendocrinology 84(4):255. Nakamura Y, Yamamoto H, Sekizawa J, Kondo T, Hirai N, Tatarazako N. 2008. The effects of pH on fluoxetine in japanese medaka (oryzias latipes): Acute toxicity in fish larvae and bioaccumul ation in juvenile fish. Chemos phere 70(5):865 73. Nakashima AS. 2009. Zinc and cortical plasticity. Brain Res Rev 59(2):347. Naleway E and Holden J. 2011. Stimulation of the lateral hypothalamus produces analgesia m ediated by the A7 catecholamine cell group in the pons. Naughton M. 2 000. A review of the role of serotonin receptors in psychiatric disorders. Human Psychopharmacology 15(6):397. Nechmad A, Maayan R, Spivak B, Ramadan E, Poyurovsky M, Weizman A. 2003. Brain neuroste roid changes after paroxetine administration in mice. Euro pean Neuropsychopharmacology 13(5):327 32. Nelson CJ, Jordan WP, Bohan RT. 1997. Daily fluoxetine administration impairs avoidance lear ning in the rat without altering sensory thresholds. Prog Neuro Psychopharmacol Biol Psychiatry 21(6):1043 57. Nelson M a nd Pinna G. 2011. S norfluoxetine microinfused into the basolateral amygdala i ncreases allopregnanolone levels and reduces aggression in socially isolated mice. Neuropharm acology 60(7):1154 9. Nelson M and Pinna G. S norfluoxetine microinfused into the bas olateral amygdala increases allopregnanolone levels and reduces aggression in socially isolated mice. Neuropharmacology In Press, Corrected Proof.
126 Nieuwenhuys R. 2011. The development and general morphology of the telencephalon of a ctinopterygian fishes: S ynopsis, documentation and commentary. Brain Structure and Function 215(3):141 57. Nikitina N, Sauka Spengler T, Bronner Fraser M. 2009. Chapter 1 gene regulatory networks in neural crest development and evolution. In: Current topics in developmental biolo gy. William R. Jeffery, editor. Academic Press. 1 p. Nin MS, Martinez L, Pibiri F, Nelson M, Pinna G. 2011. Frontiers: Neurosteroids reduce social isolation induced behavioral deficits: A proposed link with neurosteroid mediated upregulation of BDNF expres sion. Frontiers in Neuroendocrine Science 2. Noeske TA and Spieler RE. 1983. Photoperiod and diel variations of serum cortisol, thyroxine and protein in goldfish, carassius auratus L. J Fish Biol 23(6):705 10. Northcutt RG. 2011. Do teleost fishes possess a homolog of mammalian isocortex? Brain Behav Evol 78(2):136 8. Northcutt RG. 1995. The forebrain of gnathostomes: In search of a morphotype. Brain Behav Evol 46(4 5):275 318. Nothdurfter C, Rammes G, Baghai TC, Schle C, Schumacher M, Papadopoulos V, Rupp recht R. 2012. Translocator protein (18 kDa) as a target for novel anxiolytics with a favourable Side Effect profile. J Neuroendocrinol 24(1):82 92. Nowakowska E, Chodera A, Kus K. 1996. Anxiolytic and memory improving activity of fluoxetine. Pol J Pharmac ol 48(3):255. O'Brien JT, Ames D, Schweitzer I. 1993. HPA axis function in depression and dementia: A review. Int J Geriatr Psychiatry 8(11):887 98. Oitzl MS, Champagne DL, Van Der Veen R, De Kloet ER. 2010. Brain development under stress: Hypotheses of gl ucocorticoid actions revisited. Neuroscience & Biobehavioral Reviews 34(6):853 66. O'Keefe J and Nadel L. 1978. Chapter 8: Aversively motivated behavior. In: The hipppocampus as a cognitive map. Oxford, England: Oxford University Press. Olsen RW. 2008. Int aminobutyric acidA receptors: Classification on the basis of subunit composition, pharmacology, and function. u pdate. Pharmacol Rev 60(3):243. Olsen RW and Sieghart W. 2009. GABAA receptors: Subtypes pro vide diversity of function and pharmacology. Neuropharmacology 56(1):141 8. Ooishi Y. 2007. Neurosteroids rapidly modulate synaptic plasticity of hippocampus at CA1 r egion. Neurosci Res 58:S191.
127 Overli O, Sorensen C, Pulman KGT, Pottinger TG, Korzan W, Sum mers CH, Nilsson GE. 2007. Ev olutionary background for stress coping styles: Relationships between physiological, behavioral, and cognitive traits in non mammalian vertebrates. Neuroscience & Biobehavioral Reviews 31(3):396 412. verli Winberg S, Pottin ger TG. 2005. Behavioral and neuroendocrine correlates of selection for stress responsiveness in rainbow trout a review. Integrative and Comparative Biology 45(3):463. Overmier JB and Savage GE. 1974. Effects of telencephalic ablation on trace classical co nditio ning of heart rate in goldfish. Exp Neurol 42(2):339 46. Palkovits M, Baffi JS, Dvori S. 1995. Neuronal organization of stress response. Ann N Y Acad Sci 771(1):313 26. Papadopoulos V and Lecanu L. 2009. Translocator protein (18 kDa) TSPO: An emergin g ther apeutic target in neurotrauma. Exp Neurol 219(1):53 7. Pariante CM and Lightman SL. 2008. The HPA axis in major depression: Classical theories and new developments. Trends Neurosci 31(9):464 8. Parihar VK, Hattiangady B, Kuruba R, Shuai B, Shetty AK. 2009. Predictable chronic mild stress improves mood, hippocampal neurogenesis and memory. Mol Psychiatry 16(2):171 83. Park H, Choi I, Nakamura M, Cho J, Lee M, Jang I. 2011. Multiple effects of allopregnanolone on GABAergic responses in single hippocampa l CA3 pyramidal neurons. Eur J Pharmacol 652(1 3):46 54. Patchev VK, Bachurin SO, Albers M, Fritzemeier KH, Papadopoulos V. 2008. Neurotrophic estr ogens: Essential profile and endpoints for drug discovery< sup> . Drug Discov Today 13(17):734 47. Patc hev V, Shoaib M, Holsboer F, Almeida O. 1994. The neurosteroid tetrahydroprogesterone counteracts corticotropin releasing hormone induced anxiety and alters the release and gene expression of corticotropin releasing hormone in the rat hypothalamus. Neurosc ience 62(1):265 71. Penninx BWJH. 2007. Late life depressive symptoms are associated with both hyperactivity and hypoactivity of the hypothalamo pituitary adrenal axis. The American Journal of Geriatric Ps ychiatry 15(6):522. Perovic S. 1999. Origin of neur onal like receptors in metazoa: Cloning of a metabotropic glut amate/GABA like receptor from the marine sponge geodia cydonium. Cell Tissue Res 296(2):395. Pibiri F, Nelson M, Carboni G, Pinna G. 2006. Neurosteroids regulate mouse aggression induced by anab olic androgenic steroids. Neuroreport 17(14):1537 41.
128 Pibiri F, Nelson M, Guidotti A, Costa E, Pinna G. 2008. Decreased corticolimbic allopregnanolone expression during social isolation enhances contextual fear: A model relevant for posttraumatic stress di sorder. Proceedings of the National Academy of Sciences 105(14):5567 72. Pickering A and Pottinger T. 1995. Biochemical effects of stress. Biochemistry and Molecular B iology of Fishes 5:349 79. Pinna G. 2010. In a mouse model relevant for PTSD, selective b rain steroidogenic stimulants (SBSSs) improve behavioral deficits by normalizing allopregnanolone biosynthesis. Behav Pha rmacol 21(5 6):438. Pinna G, Broedel O, Eravci M, Stoltenburg Didinger G, Plueckhan H, Fuxius S, Meinhold H, Bau mgartner A. 2003. Thyro id hormones in the rat amygdala as common targets for antidepressant drugs, mood stabilizers, and sleep deprivation. Biol Psychiatry 54(10):1049 59. Pinna G. 2006. Fluoxetine and norfluoxetine stereospecifically and selectively increase brain neurosteroid content at doses that are inactive on 5 HT reuptake. Psychopharmacology (Berl ) 186(3):362. Pinna G, Costa E, Guidotti A. 2009. SSRIs act as selective brain steroidogenic stimulants (SBSSs) at low doses that are inactive on 5 HT reuptake. Current Opinion i n Pharmacology 9(1):24 30. Pinna G, Costa E, Guidotti A. 2004. Fluoxetine and norfluoxetine stereospecifically facilitate pe ntobarbital sedation by increasing neurosteroids. Proceedings of the National Academy of Scien ces of the United States of America 10 1(16):6222 5. Pinna G, Dong E, Matsumoto K, Costa E, Guidotti A. 2003. In socially isolated mice, the reversal of brain allopregnanolone down regulation mediates the anti aggressive action of fluoxetine. Proceedings of the National Academy of Sciences of t he United States of America 100(4):2035 40. Piront M and Schmidt R. 1988. Inhibition of long term memory formation by anti ependymin antisera after active shock avoidance learning in goldfish. Brain Res 442(1):53 62. PLOTSKY PM, CUNNINGHAM Jr ET, WIDMAIER EP. 1989. Catecholaminergic modulation of cort icotropin releasing factor and adrenocorticotropin secretion. Endocr Rev 10(4):437 58. Popesku JT, Martyniuk CJ, Mennigen J, Xiong H, Zhang D, Xia X, Cossins AR, Trudeau VL. 2008. The goldfish (carassius auratu s) as a model for neuroendocrine signaling. Mol Cell Endocrinol 293(1 2):43 56. Portavella M, Vargas J, Torres B, Salas C. 2002. The effects of telencephalic pallial lesions on sp atial, temporal, and emotional learning in goldfish. Brain Res Bull 57(3 4):3 97 9. Portavella M and Vargas JP. 2005. Emotional and spatial learning in goldfish is dependent on different telencephalic pallial systems. European Journal of Neuroscience 21(10):2800 6.
129 Portavella M, Torres B, Salas C. 2004. Avoidance response in goldfis h: Emotional and temporal involvement of medial and lateral telencephalic pallium. The Journal of Neuroscience 24(9):2335 42. Portavella M, Torres B, Salas C, Papini MR. 2004. Lesions of the medial pallium, but not of the lateral pallium, disrupt spaced tr ial avoidance learning in goldfish (carassius auratus). Neurosci Lett 362(2):75 8. Portavella M, Salas C, Vargas JP, Papini MR. 2003. Involvement of the telencephalon in spaced trial avoidance learning in the goldfish (carassius auratus). Physiol Behav 80( 1):49 56. PORTHE NIBELLE J and LAHLOU B. 1974. PLASMA CONCENTRATIONS OF CORTISOL IN H YPOPHYSECTOMIZED AND SODIUM CHLORIDE ADAPTED GOLDFISH (CARASSIUS AURATUS L.). J Endocrinol 63(2):377 87. Portz DE, Woodley CM, Cech JJ. 2006. Stress associated impacts of short term holding on fishes. Rev Fish Biol Fish 16(2):125 70. Poulos AM, Li V, Sterlace SS, Tokushige F, Ponnusamy R, Fanselow MS. 2009. Persistence of fear memory across time requires the basolateral amygdala complex. Proceedings of the National Academy of Sciences 106(28):11737 41. Pritchett DB, Sontheimer H, Shivers BD, Ymer S, Kettenmann H, Schofield PR, Seeburg PH. 1989. Importance of a novel GABAA receptor subunit for benzodiazepine pharmacology. Nature 338(6216):582 5. Pruessner JC, Baldwin MW, Dedo vic K, Renwick R, Mahani NK, Lord C, Meaney M, Lupien S. 2005. Self esteem, locus of control, hippocampal volume, and cortisol regulation in young and old adulthood. Neuroimage 28(4):815 26. Pushchina EV, Varaksin AA, Kalinina GG. 2007. [Cytoarchitectonic and neurochemical properties 49(6):460 70. Radwanska K, Nikolaev E, Kaczmarek L. 2010. Central noradrenergic lesion induced by DSP 4 i mpairs the acquisition of avoidance reactions and prevents molecular ch anges in the amygdala. Neurobiol Learn Mem 94(3):303 11. Rae P, Gutmann N, Tsao J, Schimmer B. 1979. Mutations in cyclic AMP dependent protein kinase and corticotropin (ACTH) sensitive adenylate cyclase affect adrenal steroidogenesis. Proceedings of the Na tional Academy of Sciences 76(4):1896. changes the potency and efficacy of benzodiazepine site ligands in different GABAA receptor subtypes. Eur J Pharmacol 636(1 3) :18 27.
130 Ramirez OA and Carrer HF. 1989. Correlation between threshold to induce long term potenti ation in the hippocampus and performance in a shuttle box avoidance response in rats. Neurosci Lett 104(1 2):152 6. Ramoino P, Gallus L, Paluzzi S, Raiteri L, Diaspro A, Fato M, Bonanno G, Tagliafierro G, Ferretti C, Manconi R. 2007. The GABAergic like system in the marine demosponge chondrilla nucula. M icrosc Res Tech 70(11):944 51. Rapkin AJ, Morgan M, Goldman L, Brann DW, Simone D, Mahesh VB. 1997. Progestero ne m etabolite allopregnanolone in women with premenstrual syndrome. Obstetrics & Gynecology 90(5):709 14. Rasmusson AM, Pinna G, Paliwal P, Weisman D, Gottschalk C, Charney D, Krystal J, Guidotti A. 2006. Decreased cerebrospinal fluid allopregnanolone leve ls in women with posttraumatic stress disorder. Biol Psychiatry 60(7):704 13. Rebala M. 2011. Allopregnanolone: Synthesis, genomic mechanism of action and its role in i ntracellular signaling mechanisms of synaptic plasticity in the adult brain of xenopus l aevis. Th esis, University of Notre Dame Reddy P, Vijayan M, Leatherland J, Moon T. 1995. Does RU486 modify hormonal responses to handling stressor and cortisol treatment in fed and fasted rainbow trout? J Fish Biol 46(2):341 59. Reisine TD, Soubrie P, Ar taud F, Glowinski J. 1982. Involvement of lateral habenula dorsal raphe neurons in the differential regulation of striatal and nigral serotonergic transmission cats. The Journal of Neuroscience 2(8):1062 71. Riddle MA, King RA, Hardin MT, Scahill L, Ort SI Chappell P, Rasmusson A, Leckman JF. 1990. Behavioral side effects of fluoxetine in children and adolescents. J Child Adolesc Psychopharm acol 1(3):193 8. Riess D. 1972. Unsignalled avoidance in a shuttlebox: A rapid acquisition, high efficiency par adigm. J Exp Anal Behav 18(1):169. Robert Brush F and Levine S. 1966. Adrenocortical activity and avoidance learning as a function of time after fear conditioning. Physiol Behav 1(3):309 11. Robertson JD. 1963. The ultrastructure of mauthner cell synapses and no des in goldfish brains. J Cell Biol 19(1):159. Rodgers RJ and Johnson NJT. 1998. Behaviorally selective effects of neuroactive steroids on plus maze anxiety in mice. Pharmacology Biochemistry and Behavior 59(1):221 32. Rodriguez F, Duran E, Gomez A, Ocana F, Alvarez E, Jimnez Moya F, Broglio C, Salas C. 2005. Cognitive and emotional functions of the teleost fish cerebellum. Brain Res Bull 66(4):365 70.
131 Rodrguez F, Broglio C, Durn E, Gmez A, Salas C. 2007. Neural mechanisms of learning in te leost fish. :243 -277. Rodrguez Landa JF, Contreras CM, Garca Ros RI. 2009. Allopregnanolone microinjected into the lateral septum or dorsal hippocampus reduces immobility in the forced swim test: Particip ation of the GABAA receptor. Behav Pharmacol 20(7):614. Ro gers J, Kirby LC, Hempelman SR, Berry DL, McGeer PL, Kaszniak AW, Zalinski J, Cofield M, Mansukhani L, Willson P, et al. 1993. Clinical trial of indomethacin in alzheimer's disease. Ne urology 43(8):1609. Rojas Vega S, Strder HK, Vera Wahrmann B, Schmidt A Bloch W, Hollmann W. 2006. Acute BDNF and cortisol response to low intensity exercise and following ramp incremental exercise to exhaustion in humans. Brain Res 1121(1):59 65. Rojas Piloni G, Martnez Lorenzana G, DelaTorre S, Conds Lara M. 2008. Nocice ptive sp inothalamic tract and postsynaptic dorsal column neurons are modulated by paraventricular h ypothalamic activation. Eur J Neurosci 28(3):546 58. Romaguera F and Mattioli R. 2008. Chlorpheniramine impairs spatial choice learning in tele ncephalon abla ted fish. Biol Res 41(3):341 8. steroid redu chydroxysteroid dehydrogenases in hypothalamus and other brain tissues of the male rat. Biochimica Et Biophysica Acta ( BBA) Lipids and Lipid Metabolism 248(3):489 502. Rosen CJ. 2009. Serotonin rising -the bone, brain, bowel connection. New England Journal of Medicine, the 360(10):957. Rousseau K, Le Belle N, Marchelidon J, Dufour S. 1999. Evidence that corticotropin rel easing hormone acts as a growth hormone releasing factor in a primitive teleost, the european eel (a nguilla anguilla). J Neuroendocrinol 11(5):385 92. Rozeboom AM, Akil H, Seasholtz AF. 2007. Mineralocorticoid receptor overexpression in for ebrain decreases anxiety like behavior and alters the stress response in mice. Proceedings of the National Academy of Sciences 104(11):4688. Rupprecht R, Papadopoulos V, Rammes G, Baghai TC, Fan J, Akula N, Groyer G, Adams D, Sch umacher M. 2010. Translocator protein (18 k Da)(TSPO) as a therapeutic target for neurological and psychiatric disorders. Nature Reviews Drug Discovery 9(12):971 88. Ryan BK, Vollmayr B, Klyubin I, Gass P, Rowan MJ. 2010. Persistent inhibition of hippocampal long term potentiation in vivo by learned helplessness stress. Hippocampus 20(6):758 67. Saha S and Datta S. 2005. Two way active avoidance training specific increases in phosphor ylated cAMP response element binding protein in the dorsal hippocampus, amygdala, and hyp othalamus. Eur J Neurosci 21(12):3403 14.
132 hydroxysteroid dehydr oisomerase in the zebrafish central nervous system. J Comp Neurol 439(3):291 305. Salas C, Broglio C, Rodrguez F, Lpez JC, Portavella M, Torres B. 1996. Telencephalic ablation in goldfish impairs performance in a  spatial constancy'problem but not in a cued one. Behav Brain Res 79(1 2):193 200. Salas C, Broglio C, Durn E, Gmez A, Ocaa F, Jimnez Moya F, Rodrguez F. 2006. Neuropsychology of learning and memory in teleost fish. Zebrafish 3(2):157 71. Sangalang G, Truscott B, Idler D. 1972. A comparison of steroidogenesis in vitro in two teleosts, the marine herring, clupea, and the freshwater atlantic salmon, salmo. J Endocrinol 53(3):433. Santarelli L, Saxe M, Gross C, Surget A, Battaglia F, Dulawa S, Weisstaub N, Lee J, Duman R, Arancio O, et al. 2003. Requirement of hippocampal neurogenesis for the behavioral effects of antidepressants. Science 301(5634):805 9. Santini F, Harmon L, Carne vale G, Alfaro M. 2009. Did genome duplication drive the origin of teleosts? A comparative study of diversification in ray finned fishes. BMC Evolutionary Biology 9(1):194. Sapolsky RM. 1992. Stress, the aging brain, and the mechanisms of neuron death. Ric hmond.: Sapolsky RM. 2001. Depression, antidepressants, and the shrinking hippocampus. Proceedings of the National Academy of Sciences of the United States of America 98(22):12320 2. Sapolsky RM. 2000. Glucocorticoids and hippocampal atrophy in neuropsyc hiatric disorders. Arch Gen Psychiatry 57(10):925 35. Sapolsky RM, Krey LC, McEwen BS. 2002. The neuroendocrinology of stress and aging: The gl ucocorticoid cascade hypothesis. Sci Aging Knowl Environ 2002(38):cp21. Sartorius A and Henn FA. 2007. Deep brain stimulation of the lateral habenula in treatment r esistant major depression. Med Hypotheses 69(6):1305 8. Sartorius A, Kiening KL, Kirsch P, Von Gall CC, Haberkorn U, Unterberg AW, Henn FA, Meyer Lindenberg A. 2010. Remission of major depression under dee p brain stimulation of the lateral habenula in a therapy refractory patient. Biol Psychiatry 67(2):e9 e11. Sass A and Wrtwein G. 2011. The effect of subchronic fluoxetine treatment on learning and memory in adolescent rats. Behav Brain Res Satvat E, Ghe idi A, Voll S, Odintsova IV, Marrone DF. 2011. Location is everything: Neurons born during fluoxetine treatment accumulate in regions that do not support spatial learning. Neur opharmacology
133 Scheurink AJW, Ammar AA, Benthem B, Van Dijk G, Sdersten PAT. 1 999. Exercise and the reg ulation of energy intake. Int J Obes 23:1 6. Schlosser G. 2010. Making senses: Development of vertebrate cranial placodes. In: International review of cell and molecular biology. Kwang Jeon, editor. Academic Press. 129 p. Schmidt H D. 2010. Peripheral BDNF produces antidepressant like effects in cellular and beha vioral models. Neuropsychopharmacology Schofield PR, Darlison MG, Fujita N, Burt DR, Stephenson FA, Rodriguez H, Rhee LM, Ramacha ndran J, Reale V, Glencorse TA. 1987. Seque nce and functional expression of the GABA A receptor shows a ligand gated receptor super family. Nature 328(6127):221 7. Schreck CB, Contreras Sanchez W, Fitzpatrick MS. 2001. Effects of stress on fish reproduction, gamete quality, and progeny* 1. Aquacult ure 197(1 4):3 24. SCHUDER SE. 2005. Stress induced hypocortisolemia diagnosed as psychiatric disorders respo nsive to hydrocortisone replacement. Ann N Y Acad Sci 1057(1):466 78. Schle C, Eser D, Baghai TC, Nothdurfter C, Kessler JS, Rupprecht R. Neuroact ive steroids in a ffective disorders: Target for novel antidepressant or anxiolytic drugs? Neuroscience In Press, Uncorrected Proof. Schumacher M, Guennoun R, Robert F, Carelli C, Gago N, Ghoumari A, Gonzalez Deniselle MC, Gonzalez SL, Ibanez C, Labombarda F, et al. 2004. Local synthesis and dual actions of progeste rone in the nervous system: Neuroprotection and myelination. Growth Hormone & IGF Research 14(Supplement 1):18 33. Schuster S. 2010. Big decisions by small networks. Bioessays 32(8):727 35. SCHWAR TZ JI. 1983. Severe depersonalization and anxiety associated with indomethacin. South Med J 76(5):679. Scobie SR. 1970. The response shock shock shock interval and unsignalled avoidance in goldfish. J Exp Anal Behav 14(2):219 24. Scott BW, Wojtowicz JM, Bu rnham WMI. 2000. Neurogenesis in the dentate gyrus of the rat fo llowing electroconvulsive shock seizures. Exp Neurol 165(2):231 6. SELIGMAN ME and MAIER SF. 1967. FAILURE TO ESCAPE TRAUMATIC SHOCK. J Exp Psychol 74(1):1 9. Selye H. 1977. Stress without dis tress. School Guidance Worker 32(5):5 13. Selye H. 1973. The evolution of the stress concept: The originator of the concept traces its d evelopment from the discovery in 1936 of the alarm reaction to modern therapeutic applications of syntoxic and catatoxic hormones. Am Sci 61(6):692 9.
134 Selye H. 1946. The general adaptation syndrome and the diseases of adaptation. The Journal of Clinical Endocrinology 6(2):117 230. Selye H. 1936. A syndrome produced by diverse nocuous agents. Nature; Nature Shirayama Y, Mu neoka K, Fukumoto M, Tadokoro S, Fukami G, Hashimoto K, Iyo M. 2010. Inf usions of allopregnanolone into the hippocampus and amygdala, but not into the nucleus accu mbens and medial prefrontal cortex, produce antidepressant effects on the learned helplessnes s rats. Hippocampus :n/a,n/a. Shishimi A. 1985. Latent inhibition experiments with goldfish (carassius auratus). Journal of Comparative Psychology 99(3):316 27. Sigel E and Buhr A. 1997. The benzodiazepine binding site of GABAA receptors. Trends Pharm acol Sci 18(4):425 9. Silvers JM, Tokunaga S, Berry RB, White AM, Matthews DB. 2003. Impairments in spatial lear ning and memory: Ethanol, allopregnanolone, and the hippocampus. Brain Res Rev 43(3):275 84. Singh A and Kumar A. 2008. Possible GABAergic modulation in the protective effect of allopre gnanolone on sleep deprivation induced anxiety like behavior and oxidative damage in mice. Methods Find Exp Clin Pharmacol 30(9):681 9. Singh C, Liu L, Wang JM, Irwin RW, Yao J, Chen S, Henry S, Thompson RF, Brinton RD. Allopre gnanolone restores hippocampal dependent learning and memory and neural progenitor survival in aging 3xTgAD and nonTg mice. Neurobiol Aging (0). Singh S, Hota D, Prakash A, Khanduja KL, Arora SK, Chakrabarti A. 2010. Allopregnanolone, the active met abolite of progesterone protects against neuronal damage in picrotoxin induced se izure model in mice. Pharmacology Biochemistry and Behavior 94(3):416 22. Singley JA and Chavin W. 1975. Serum cortisol in normal goldfish (carassius auratus L.). Compa rative Biochemistry and Physiology Part A: Physiology 50(1):77 82. Smeets WJAJ and Gonzlez A. 2000. Catecholamine systems in the brain of vertebrates: New perspectives through a comparative approach. Brain Res Rev 33(2 3):308 79. Smith EM, Chu S, Paterson G, Met calfe CD, Wilson JY. 2010. Cross species comparison of fluox etine metabolism with fish liver microsomes. Chemosphere 79(1):26 32. Smith ME. 2004. Noise induced stress response and hearing loss in goldfish (carassius auratus). J Exp Biol 207(3):427. Snyder JS, Soumier A, Brewer M, Pickel J, Cameron HA. 2011. Adult hippocampal neurogenesis buffers stress responses and depressive behaviour. Nature 476(7361):458 61. Somoza GM, Yu KL, Peter RE. 1988. Serotonin stimulates gonadotropin release in female and male g oldfish, carassius auratus L. Gen Comp Endocrinol 72(3):374 82.
135 Song L, Che W, Min Wei W, Murakami Y, Matsumoto K. 2006. Impairment of the spatial learning and memory induced by learned helplessness and chronic mild stress. Pharmacology Biochemi stry and Be havior 83(2):186 93. Srensen C, verli Summers CH, Nilsson GE. 2007. Social regulation of neurogenesis in tel eosts. Brain Behav Evol 70(4):239 46. Srensen C, Nilsson GE, Summers CH, verli Social stress reduces forebrain cell proliferation in rainb ow trout (oncorhynchus mykiss). Behav Brain Res In Press, Corrected Proof. Sorrells SF and Sapolsky RM. 2007. An inflammatory review of glucocorticoid actions in the CNS. Brain Behav Immun 21(3):259 72. Spanswick SC, Epp JR, Keith JR, Sutherland RJ. 2007. Adrenalectomy induced granule cell d egeneration in the hippocampus causes spatial memory deficits that are not reversed by chronic treatment with corticosterone or fluoxetine. Hippocampus 17(2):137 46. Spieler RE and Noeske TA. 1981. Timing of a single dai ly meal and diel variations of serum th yroxine, triiodothyronine and cortisol in goldfish. Life Sci 28(26):2939 44. Srivastava M, Larroux C, Lu D, Mohanty K, Chapman J, Degnan B, Rokhsar D. 2010. Early evol ution of the LIM homeobox gene family. BMC Biology 8(1):4. Starkman MN, Gebarski SS, Berent S, Schteingart DE. 1992. Hippocampal formation volume, memory dysfunction, and cortisol levels in patients with cushing's syndrome. Biol Psychiatry 32(9):756 65. Starkman MN, Giordani B, Gebarski SS, Berent S, Scho rk MA, Schteingart DE. 1999. Decrease in Psychiatry 46(12):1595 602. Steenbergen PJ, Richardson MK, Champagne DL. The use of the zebrafish model in stress r esearch. Prog Neuro Psychopharmacol Biol Psychiatry In Press, Corrected Proof. Sternbach H. 2003. Are antidepressants carcinogenic? A review of preclinical and clinical studies. J Clin Psychiatry 64(10):1153 62. Stoffel EC and Craft RM. 2004. Ovarian hormone withdr awal rats. Physiol Behav 83(3):505 13. Stolte EH, de Mazon AF, Leon Koosterziel KM, Jesiak M, Bury NR, Sturm A, Savelkoul HFJ, van Kemenade B, Flik G. 2008. Corticosteroid receptors involved in stress regulation in common car p, cyprinus carpio. J Endocrinol 198(2):403. Stone EA, Grunewald GL, Lin Y, Ahsan R, Rosengarten H, Kramer HK, Quartermain D. 2003. Role adrenoceptors in motor activity in mice. Synapse 49(1):67 76.
136 Sturm A, Bury N, Dengreville L, Fagart J, Flouriot G, Rafestin Oblin M, Prunet P. 2005. 11 deoxycorticosterone is a potent agonist of the rainbow trout (oncorhynchus mykiss) mineral ocorticoid receptor. Endocrinology 146(1):47 55. Suda Y, Kurokawa D, Takeuchi M, Kajikawa E, Kuratani S, Amemiya C, Aizawa S. 2009. Evolution of otx paralogue usages in early patterning of the vertebrate head. Dev Biol 325(1):282 95. Sutherland RJ. 1982. The dorsal diencephalic conduction system: A review of the anatomy and functions of the haben ular complex. Neuroscience & Biobehavioral Reviews 6(1):1 13. Szabadics J. 2007. Different transmitter transients underlie presynaptic cell type specificity of GABAA, slow and GABAA, fast. Proc Natl Acad Sci U S A 104(37):14831. Szabo TM, Brookings T, Preu ss T, Faber DS. 2008. Effects of temperature acclimation on a central neural circuit and its behavioral output. J Neurophysiol 100(6):2997 3008. Taher L, McGaughey DM, Maragh S, Aneas I, Bessling SL, Miller W, Nobrega MA, McCallion AS, Ovcharenko I. 2011. Genome wide identification of conserved regulatory function in diverged sequences. Genome Research Tanaka S, Kondo H, Kanda K, Ashino T, Nakamachi T, Sekikawa K, Iwakura Y, Shioda S, Numaz awa S, Yoshida T. 2011. Involvement of interleukin 1 in lipopolysa ccaride induced microglial act ivation and learning and memory deficits J Neurosci Res Tanti A and Belzung C. 2010. REVIEW: Open questions in current models of antidepressant a ction. Br J Pharmacol 159(6):1187 200. Tavolga W. 1977. Behavioural thresholds for diffuse illumination in the goldfish. J Exp Biol 67(1):89 96. Teitsma CA, Anglade I, Lethimonier C, Le Dran G, Saligaut D, Ducouret B, Kah O. 1999. Gluc ocorticoid receptor immunoreactivity in neurons and pituitary cells implicated in reproductive func tions in rainbow trout: A double immunohistochemical study. Biol Reprod 60(3):642. Thornton EW and Bradbury GE. 1989. Effort and stress influence the effect of lesion of the habenula complex in one way active avoidance learning. Physiol Behav 45(5):929 35. To CT, Anheuer ZE, Bagdy G. 1999. Effects of acute and chronic fluoxetine treatment on CRH induced anxiety. Neuroreport 10(3):553. Tome MB, Cloninger CR, Watson JP, Isaac MT. 1997. Serotonergic autoreceptor blockade in the reduction of antidepressant late ncy: Personality variables and response to paroxetine and pi ndolol. J Affect Disord 44(2 3):101 9. Tramontina AC, Tramontina F, Bobermin LD, Zanotto C, Souza DF, Leite MC, Nardin P, Gottfried C, Gonalves C. 2008. Secretion of S100B, an astrocyte derived n eurotrophic protein, is stim u-
137 lated by fluoxetine via a mechanism independent of serotonin. Prog Neuro Psychopharmacol Biol Psychiatry 32(6):1580 3. Tronson NC, Wiseman SL, Neve RL, Nestler EJ, Olausson P, Taylor JR. 2012. Distinctive roles for amygdalar CR EB in reconsolidation and extinction of fear memory. Learning & Memory 19(5):178 81. Trudell JR. 2008. Molecular modeling and mutagenesis reveals a tetradentate binding site for Zn2 in GABAA [alpha][beta] receptors and provides a structural basis for the m odulating effect of the [gamma] subunit. Journal of Chemical Information and Modeling 48(2):344. Tsutsui K, Ukena K, Sakamoto H, Okuyama S, Haraguchi S. 2011. Frontiers: Biosynthesis, mode of action, and functional significance of neurosteroids in the purk inje cell. Frontiers in Neuroend ocrine Science 2. Turkmen S, Backstrom T, Wahlstrom G, Andreen L, Johansson I. 2011. Tolerance to allopre gnanolone with focus on the GABA A receptor. Br J Pharmacol 162(2):311 27. Turkmen S, Lundgren P, Birzniece V, Zingmark E, Backstrom T, Johansson I. 2004. 3? 20? dihydroxy 5? pregnane (UC1011) antagonism of the GABA potentiation and the learning i mpairment induced in rats by allopregnanolone. Eur J Neurosci 20(6):1604 12. Ugale RR, Sharma AN, Kokare DM, Hirani K, Subhedar NK, Chopde CT. 2007. Neurosteroid all opregnanolone mediates anxiolytic effect of etifoxine in rats. Brain Res 1184:193 201. Ugale RR, Mittal N, Hirani K, Chopde CT. 2004. Essentiality of central GABAergic neuroactive steroid allopregnanolone for anticonvul sant action of fluoxetine against pentylenetetrazole induced seizures in mice. Brain Res 1023(1):102 11. Ullmann JFP, Cowin G, Kurniawan ND, Collin SP. 2010. A three dimensional digital atlas of the zebrafish brain. Neuroimage 51(1):76 82. Umminger BL and Gist DH. 1973. Effects of thermal acclimation on physiological responses to handling stress, cortisol and aldosterone injections in the goldfish, carassius auratus. Compar ative Biochemistry and Physiology Part A: Physiology 44(3):967 77. Ungar F and Halber g F. 1962. Circadian rhythm in the in vitro response of mouse adrenal to adrenocorticotropic hormone. Science 137(3535):1058 60. Uzunova V, Sheline Y, Davis JM, Rasmusson A, Uzunov DP, Costa E, Guidotti A. 1998. Increase in the cerebrospinal fluid content of neurosteroids in patients with unipolar major depression who are receiving fluoxetine or fluvoxamine. Proceedings of the National Academy of Sciences of the United States of America 95(6):3239 44. Uzunova V, Wrynn AS, Kinnunen A, Ceci M, Kohler C, Uzuno v DP. 2004. Chronic antidepressants reverse cerebrocortical allopregnanolone decline in the olfactory bulbectomized rat. Eur J Pharmacol 486(1):31 4.
138 Valluzzi JA and Chan KH. 2007. Effects of fluoxetine on hippocampal dependent and hippoca mpal independent learning tasks. Behav Pharmacol 18(5 6):507. Van de Kar LD and Blair ML. 1999. Forebrain pathways mediating stress induced hormone secr etion. Front Neuroendocrinol 20(1):1 48. Van den Burg E, Metz J, Spanings F, Wendelaar Bonga S, Flik G. 2005. Plasma [alpha] MSH and acetylated [beta] endorphin levels following stress vary according to CRH sensitivity of the pitu itary melanotropes in common carp, cyprinus carpio. Gen Comp Endocrinol 140(3):210 21. Van Der Boon J, Van Den Thillart GE, Addink ADF. 1991. T he effects of cortisol administration on intermediary metabolism in teleost fish. Comparative Biochemistry and Physiology Part A: Phy siology 100(1):47 53. Van der Borght K, Meerlo P, M. Luiten PG, Eggen BJL, Zee EAVd. 2005. Effects of active shock avoidanc e learning on hippocampal neurogenesis and plasma levels of corticosterone. Behav Brain Res 157(1):23 30. Vargas JP, Bingman VP, Portavella M, Lpez JC. 2006. Telencephalon and geometric space in goldfish. Eur J Neurosci 24(10):2870 8. Vargas JP, Lpez JC, Portavella M. 2009. What are the functions of fish brain pallium? Brain Res Bull 79(6):436 40. Vaughan J, Donaldson C, Bittencourt J, Perrin MH, Lewis K, Sutton S, Chan R, Turnbull AV, Lovejoy D, Rivier C. 1995. Urocortin, a mammalian neuropeptide related to fish urotensin I and to corticotropin releasing factor. Nature 378(6554):287 92. Veenema A, Meijer O, De Kloet E, Koolhaas J. 2003. Genetic selection for coping style predicts stressor susceptibility. J Neuroendocrinol 15(3):256 67. Venard C, Boujedain i N, Belon P, Mensah Nyagan A, Patte Mensah C. 2008. Regulation of neur osteroid allopregnanolone biosynthesis in the rat spinal cord by glycine and the alkaloidal analogs strychnine and gelsemine. Neuroscience 153(1):154 61. Vergnano AM. 2007. PKC activati on sets an upper limit to the functional plasticity of GABAergic transmission induced by endogenous neurosteroids. Eur J Neurosci 26(5):1173. Vetencourt JFM, Sale A, Viegi A, Baroncelli L, De Pasquale R, O'Leary OF, Castrn E, Maffei L. 2008. The antidepre ssant fluoxetine restores plasticity in the adult visual cortex. Science 320(5874):385 8. Vialou V, Robison AJ, LaPlant QC, Covington HE, Dietz DM, Ohnishi YN, Mouzon E, Rush AJ, Watts EL, Wallace DL, et al. 2010. [Delta]FosB in brain reward circuits media tes resilience to stress and antidepressant responses. Nat Neurosci 13(6):745 52.
139 Vijayan M, Ballantyne J, Leatherland J. 1991. Cortisol induced changes in some aspects of the intermediary metabolism of< i> salvelinus fontinalis. Gen Comp Endocrinol 82 (3):476 86. Vijayan M, Reddy P, Leatherland J, Moon T. 1994. The effects of cortisol on hepatocyte metab olism in rainbow trout: A study using the steroid analogue RU486. Gen Comp Endocrinol 96(1):75 84. Wang RY and Aghajanian GK. 1977. Physiological eviden ce for habenula as major link between forebrain and midbrain raphe. Science 197(4298):89 91. Wang WB, Wang JG, Li AH, Cai TZ. 2004. Changes of cortisol and lysozyme levels in carassius a uratus blood after handling stress. Acta Hydrobiologica Sinica 28(6):6 82 4. Wang JM, Liu L, Irwin RW, Chen S, Brinton RD. 2008. Regenerative potential of allopregnan olone. Brain Res Rev 57(2):398 409. Warner Schmidt JL and Duman RS. 2007. VEGF is an essential mediator of the neurogenic and behavioral actions of antidepressan ts. Science's STKE 104(11):4647. Warner Schmidt JL, Vanover KE, Chen EY, Marshall JJ, Greengard P. 2011. Antidepressant effects of selective serotonin reuptake inhibitors (SSRIs) are attenuated by antiinflammatory drugs in mice and humans. Proceedings of t he National Academy of Sciences Weber III ES. 2011. Fish analgesia: Pain, stress, fear aversion, or nociception? Veterinary Clinics of North America: Exotic Animal Practice 14(1):21 32. Weiger WA. 1997. Serotonergic modulation of behaviour: A phylogeneti c overview. Biological Reviews 72(1):61. Weiser M and Handa R. 2009. Estrogen impairs glucocorticoid dependent negative feedback on the hypothalamic pituitary adrenal axis via estrogen receptor alpha within the hypothalamus. Neuroscience 159(2):883 95. Wei ss S, Zottoli S, Do S, Faber D, Preuss T. 2006. Correlation of C start behaviors with neural a ctivity recorded from the hindbrain in free swimming goldfish (carassius auratus). J Exp Biol 209(23):4788 801. Wilde MI and Benfield P. 1998. Fluoxetine: A pharm acoeconomic review of its use in depression. Pharmacoeconomics 13(5):543 61. Willner P. 2005. Chronic mild stress (CMS) revisited: Consistency and behavioural neurobiological concordance in the effects of CMS. Neuropsychobiology 52(2):90 110. Winberg S and Lepage O. 1998. Elevation of brain 5 HT activity, POMC expression, and plasma cortisol in socially subordinate rainbow trout. American Journal of Physiology Regulatory, Int egrative and Comparative Physiology 274(3):R645 54.
140 Winberg S, Nilsson A, Hylland P Sderstm V, Nilsson GE. 1997. Serotonin as a regulator of h ypothalamic pituitary interrenal activity in teleost fish. Neurosci Lett 230(2):113 6. Wingfield JC and Romero LM. 2001. Adrenocortical responses to stress and their modulation in free living ve rtebrates. Comprehensive Physiology Wolkowitz OM, Reus VI, Roberts E, Manfredi F, Chan T, Raum WJ, Ormiston S, Johnson R, Canick J, Brizendine L. 1997. Dehydroepiandrosterone (DHEA) treatment of depression. Biol Psychiatry 41(3):311 8. Wolkowitz OM, Burk e H, Epel ES, Reus VI. 2009. Glucocorticoids. Ann N Y Acad Sci 1179(1):19 40. Wong DT, Perry KW, Bymaster FP. 2005. The discovery of fluoxetine hydrochloride (prozac). N ature Reviews.Drug Discovery 4(9):764. Woodruff Pak DS and Disterhoft JF. 2008. Where is the trace in trace conditioning? Trends Ne urosci 31(2):105 12. Wortsman J. 2002. Role of epinephrine in acute stress. Endocrinol Metab Clin North Am 31(1):79. Wullimann MF and Mueller T. 2004. Teleostean and mammalian forebrains contrasted: Evidence fro m genes to behavior. J Comp Neurol 475(2):143 62. Wullimann MF and Rink E. 2002. The teleostean forebrain: A comparative and developmental view based on early proliferation, Pax6 activity and catecholaminergic organization. Brain Res Bull 57(3 4):363 70. Y oung EA. 1995. Glucocorticoid cascade hypothesis revisited: Role of gonadal steroids. Depre ssion 3(1 2):20 7. Zhang D, Xiong H, Mennigen JA, Popesku JT, Marlatt VL, Martyniuk CJ, Crump K, Cossins AR, Xia X, Trudeau VL. 2009. Defining global neuroendocrine gene expression patterns associated with reproductive seasonality in fish. PloS One 4(6):e5816. Zhang Y, Raap DK, Garcia F, Serres F, Ma Q, Battaglia G, Van de Kar LD. 2000. Long term fluox etine produces behavioral anxiolytic effects without inhibiting neu roendocrine responses to co nditioned stress in rats. Brain Res 855(1):58 66. Zhang L, Zhou R, Li X, Ursano RJ, Li H. 2006. Stress induced change of mitochondria membrane potential regulated by genomic and non genomic GR signaling: A possible mechanism for hipp ocampus atrophy in PTSD. Med Hypotheses 66(6):1205 8. Zhao Y, Wang Z, Dai J, Chen L, Huang Y, Zhan Z. 2011. Beneficial effects of benzodiazepine diaz epam on chronic stress induced impairment of hippocampal structural plasticity and depression like beha vior in mice. Behav Brain Res
141 Ziegler E. 2009. Interaction of androsterone and progesterone with inhibitory ligand gated ion channels: A patch clamp study. Naunyn Schmiedebergs Arch Pharmacol 380(4):277. Zimmerberg B, Rackow SH, George Friedman KP. 1999. Sex dependent behavioral effects of the THP) in neonatal and adult rats after postnatal stress. Pharmacology Biochemistry and Behavior 64(4):717 24. Zimmerberg B and Kajunski EW. 2004. Sexually dimorphic effects of po stnatal allopregnanolone on the development of anxiety behavior after early deprivation. Pharmacology Biochemistry and Behavior 78(3):465 71. Zimmerberg B, Brunelli SA, Hofer MA. 1994. Reduction of rat pup ultrasonic vocalizations by the neuroactive steroi d allopregnanolone. Pharmacology Biochemistry and Behavior 47(3):735 8. Zimmerberg B, Brunelli SA, Fluty AJ, Frye CA. 2005. Differences in affective behaviors and hipp ocampal allopregnanolone levels in adult rats of lines selectively bred for infantile voc alizations. Behav Brain Res 159(2):301 11. Zola Morgan S, Squire LR, Amaral D. 1986. Human amnesia and the medial temporal region: E nduring memory impairment following a bilateral lesion limited to field CA1 of the hippocampus. The Journal of Neuroscience 6(10):2950 67. Zottoli SJ. 2000. The mauthner cell: What has it taught us? Neuroscientist 6(1):25. Zottoli SJ. 1999. Decrease in occurrence of fast startle responses after selective mauthner cell ablation in goldfish (carassius auratus). Journal of Compara tive Physiology.B, Biochemical, Sy stemic, and Environmental Physiology 184(2):207. Zupanc GKH. 2009. Towards brain repair: Insights from teleost fish. Semin Cell Dev Biol 20(6):683 90.
142 APPENDIX I: Additional Graphs and Tables A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct A F C AC FC Ct Delay Trace Delay Trace Delay Trace Delay Trace Block 1 Block 2 Block 3 Block 4 Avoids 5 4 6 1 2 1 4 4 4 8 3 1 6 1 1 2 2 2 8 3 1 1 1 1 2 1 2 3 2 2 2 1 2 1 1 2 4 3 3 3 3 2 3 3 3 2 1 2 0 5 10 15 20 25 30 35 40 45 50 # of Avoids per Block Avoidance Means: Both Conditions
143 0 5 10 15 20 25 30 35 40 45 Delay Delay Trace Trace Delay Delay Trace Trace Delay Delay Trace Trace Delay Delay Trace Trace Delay Delay Trace Trace Delay Delay Trace Trace 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 1 2 3 4 A A A A A A A A F F F F F F F F C C C C C C C C AC AC AC AC AC AC AC AC FC FC FC FC FC FC FC FC Ct Ct Ct Ct Ct Ct Ct Ct Avoidance, by Treatment
150 A PPENDIX II: The Teleostean Telencephalic Controversy The vertebrate nervous system has many conserved features from hagfish through to the most complex manifestations in cetaceans and primates. In all vertebrates the neural crest develops to form three pri ncipal components of the brain: the prosencephalon, mesencephalon, and rhombencephalon. During embryogenesis these structures differentiate from the fledgling neural tube, generally through hypertrophy near areas of similar genetic expression and coord inat ed fate maps. The telencephalon particularly folds via evagination compared to the simple burgeoning of the diencephalon, leading not only to preliminary divisions but also the formation of the lateral ventricles. Interestingly, species of the class Actino pterygii do not exhibit evagin ation, but rather invaginate (or evert) the neural tube resulting in a unique configuration of the common neuroanatomical features typically associated with phylogenetically precedent and subsequent vertebrates. This distincti ve arrangement leads to a number of important questions regarding the extent of this divergence from prior neuroanatomical morphologies, which may be extended to ask about the behavioral capabilities of this class compared to their near evol utionary cousin s. These topics will be evaluated in depth over the course of this section to eluc idate the nature of the long controversial status of teleostean forebrain homologies. Actinopterygii The Actinopterygii are a diverse group of fish united by their radial fin s, as is the literal translation of the class from ancient Greek, which are composed of true bone extending from their skeleton. The earliest fossil evidence of a true Actinopterygian was found approximately 425 million years ago in fresh water, where the fishes remained until around 340 million years ago when they began to venture into brackish and eventually sea water. There are two su bclasses present among the Actinopterygii, the Chondrostei and Neopterygii, the former consis t-
151 ing primarily of sturgeons a nd the latter representing two further infraclasses, the Holostei and Teleostei. The Holostei are further composed of two orders, Lepisosteiformes (gars) and Ami iformes (bowfins), and the Teleostei may be reduced indeterminately. (Cavin, 2008; Braford, 200 9) Teleostei is composed of at least 30,000 known species as diverse as eels and anglerfish to salmon and sea horses, completely dominating not only the class Actinopterygii but in fact the vast majority (>99% of total species) of extant bony fishes belon g to this infraclass. The first fo ssil evidence for the teleosts dates them to 284 MYA, though mitochondiral DNA analysis predicts a divergence within the range of 268 326 MYA. The origin of Teleostei has long been considered to be derived from the crown g roup Neopterygii, which is comprised of the lepisosteids, amiids, and teleosts as sister groups. This has been debate, especially in light of recent mitochondrial model fails to statistically reject the opposing Neopterygian model; likewise, an analysis of var ious nuclear genes produces a model consistent with the Neopterygian hypothesis, but fails to reject the Ancient Fish Clade. While this issue remains to be resolved recent Bayesian analyses favor the Neopterygian hypothesis as well as strong evidence from synapomorphic analysis
152 (Hu r ley et al, 2006). Regardless of which hypothesis one accepts, if one is to consider establis hing homology across these groups it is nece ssary to identify the homologous organ in the ance stor to the teleosts, the polypterids, of which the only extant member is the bichir. Considering the huge scope of the Actinopterygii, their behaviors vary widely according to their ecological niche and t heir cognitive capacities follow suit. Many different lifestyles have been adopted by different members of this class, such as the trigger fish which has a hugely d eveloped optic tectum to measure the exact trajectory and force necessary to hit a flying ar thr opod, or the intense sexual competition that has led thousands of species of teleosts to wrestle for territory, build nests, negotiate sexual roles, and compete for hierarchal dominance (Barlow, 1961; Munakata et Kobayashi, 2010; Oliveira, 2009; Groseni ck et al, 2007). The specific abilities of the Actinopterygii, typically Teleostei, will be discussed in terms of laboratory experiment ation of cognition in order to elucidate the behavioral implications for many of the proposed h omologies to Tetrapodia. C hondichthyes To fully evaluate the distinctive forebrain of the Actinopterygii, consideration of earlier gnathostome outgroups may shed light on the evolutionary basis of this oddity. There is su bstantial disagreement as to whether the Chondrichthyes, the cartilaginous fishes, belong as a sister group to tetrapods or Actinopterygii, which genetic evidence has not resolved (Mine, 2006). Regardless, this class is considered the nearest out group to Actinopterygii (Nie uwenhuys, 1963; Wourms, 1997; Lisney et C ollin, 2006). Elasmobranchs, an abundant subclass of Chondrichthyes, have a considerably larger brain to body ratio than most ray finned fishes and exhibit a diverse set of skills: a set of acute sensory apparati for short and long range electror eception (i.e., bioelectric vs. migratory magnetic sense), vision adapted to predation, strong olfa ctory sense, ovoviviparous brooding, and social hierarchy in many species (Linsey et Collin, 2006;
153 Tricas, 2001; McComb, 2010; Pratt et al, 2001; Economakis, 1998; G uttridge et al, 2009). While sharks tend to have larger telencephalon to whole brain mass ratios than Actinopterygii, this is predominantly due to drastically larger olfactory bulbs rather than non sensory regions (Linsey et Collin, 2006). As noted previou sly, the telencephalons of all Chondrichthyes undergo embryonic eva gination of the telencephalon (Wourms, 1997). While this class parallels the Actinopterygii in many neuroanatomical features beyond the telencephalon, this ontogenetic paradigm shift makes it difficult to ascertain how much of the forebrain morphology is retained from this co mmon ancestor of ray finned fishes and tetrapods. Recent work on this front by Rodriguez Moldes (2009) in the Lesser Spotted Dogfish, a common model elasmobranch, has re vealed that this and topological patterns previously believed to be of later descent: Sonic hedgehog (Shh) defines its alar basal boundary with higher concentrations in the basal region
154 Paired box gene 6 (PAX6) is found to be expressed in heterogeneous densities in all prosomeric regions, but stands starkly in contrast to the mesencephalon where it is e ntirely absent Prosomere 2 (P2) has its cauda l border defined by the fasciculus retroflexus and its ro stral limens encircled by the zona limatans intrathalamica, the latter of which is defined by a gradient of Shh The first developmental period is defined by the absence of GABAergic cells in the alar component (pallium) and the absence of PAX6 expressing cells in the basal region (su bpallium); by the second period, the former cells have migrated tangentially dorsal and the latter radially and tangentially ventral The identification of the three prosom eres in the shark separated by the same semaphores found in Actinopterygii and even Tetrapodia indicates that this developmental process is almost ogenetically earli est example of tangential migration observed in any species, a complex process that requires many chemokines to be selectively expressed by radially migrated cells of specific fates and, foremost, a prerequisite for advanced axonal targeting (Handel, 2009; Kershaw, 2009; Li, 2009). These findings comprise reasonable evidence to conclude that many of the structures found in Chondrichthyan telencephalons may in fact be homologous to those of their descen dants. Coelacanths The closest possible descendent to the Actinopterygii lies in its sister group of lobe finned fishes, consisting of tetrapods, lungfishes, and coelacanths. In terms of telencephalic o rganization, one of two extant members of crossopterygii, Latimeria chalumnae is the closest
155 relative. Long only considered a part of the fossil record, this co e gnited scientific curiosity into this a r mor plated relic of the Mesozoic age and represented the pote n bclass of Sarcopte rgyii including lungfish). An analysis by R u dolf Nieuwenhuys concluded that this animal does i ndeed satisfy these criteria in some respects: its forebrain reveals a complex kind of evagination that is reminiscent of the eversion found in Actinopterygii and closely resembles the forebrain of the mon o pneumonian Ceratodus, a primitive lungfish (1965). The subpall i um has undergone a true evagination whereas the dorsal pallium has everted in a manner akin to the Actinopterygii; altogether, thi l lial structures are also noted to be exceptionally primitive, di s pla y ing a strong amalgamation of cells near the se ptal areas with almost no neural migration, which Nieuwenhuys asserts as being t he most rud imentary subpallial organization of all extant Gnathostomes. A more recent evaluation by Nort hcutt and Gonzalez has supplemented these findings in light of recent advances in the Actino pterygian forebrain and the fortuitous procur e ment of two ne w specimens (2011): they conclude with a more complex model of the pallium which demonstrates at least four distinct cellular groups which are preliminarily considered h o mologous to the typical pallial divisions in tetr apods based on connectivity to the la teral olfact o ry tract and the eversion of the telencephalon (a discussion of these principles in teleosts fo l lows below). The comparatively large medial pall ium is thought to explain the advanced spatial capacities of most lungfish and amphibians, which
156 ar e likely derived their enlarged hippocampal formations from an originating species common to all lobe finned fishes or a prece ding species. This new study also rebukes the supposed superior superoptic nucleus, instead believing it to be the medial amygd al ar nucleus. This implies that the all of the major functional co mponents of the amygdala were retained from a species predating the lobe finned fishes, all extant members of which have been shown to possess these comp o nents. While the se features certainly make this fish interesting in its evolutionary context, it is not considered to be a direct ance stor of Tetrapodia but rather an offshoot that did not proceed to evolve any known species (Miles, 1965). While little is known of this sp ecies behavior due to difficulties in husbandry and incred ible rarity, it is widely considered to be very primitive and unlikely to engage a more complex behavioral pattern than is afforded by its large size, armored scales, and residence in the mesopelagi c volcanic caves, though it is has been observed to form peaceful social groups and possibly permanent homes (Fricke et al, 1990). The ancient coelacanth has thus presented a dead end for r esearchers of brain evolution, though the apparent transitional qua lities of its half everted, half evaginated telencephalon may represent a divergence from a common ancestor evolving from the everted condition of Actinopterygii; unfortunately, this is a difficult hypothesis to further given the uniqueness of this ancient species and the extreme pa ucity of scientific specimens. F IGURE 26 : T HE FOREBRAIN OF L ATIMERIA CHALUMNAE A) A N ISSL STAINED LEFT HEMISPHERE OF THE TE LE NCEPHALON B) N ORTHCUTT AND G ONZALEZ (2011) I NTERPRETATION OF THE FUN CTIONAL REGIONS OF TH E PA LLIUM C) N IEWENHUYS (1965) INTERPRETATION OF THE PALLIUM
157 Anatomical and Hodological Models of Pallial Arrangemen While the simpler representatives of this class are analyzed less often and generally with the explicit intention of comparing to the most pre valent and advanced infraclass, Teleostei, there is much to be gained from observing the less developed telencephalons present in the Polypteriformes (bichirs and Reedfish), Acipenseriformes (sturgeons and paddlefishes), Semio notiforms (g ars), and Amiiform es (bowfins). T here are many striking differences amongst the three species shown; namely, the single common ventricle decreases in size as the complexity of the brain increases from birchir to sturgeon to gar and as well as total pallial size and density increasing, as is particularly evident from Polypterus to more modern Actinopterygians. I mportantly, it should be noted that the nissl stain density becomes increasingly more visible at regions distal from the paraventricular zone (i.e., the ependyma), ind icating increased mit ochondrial metabolism in non proliferative regions, thus neural greater migration; additionally it is particularly evident in the gar that there are distinct regions of various morphologies from the nissl stain alone. The teleost telen cephalon is consistent with these trends and advances them significantly. Unfortunately, the pallial zones demarcated on all of these images remain contr oversial, especially in Teleostei. From Polypteriforms to Acipenseriformes to Semionotiforms to Amiifor mes to Teleosts there is an increasing trend toward the parcellation of the migrated rostrolateral preglomerular nuclei, or posterior tubercle (Demski, 1984; Braford, 2009). The posterior tubercle projects pr imarily to the pallium and the development of th is diencephalic structure has been correlated to the development of more complex pallial morphology, standing in constrast to the dorsal preglomerular nuclei which are highly conserved from the basal species of Actinopterygii (Northcutt, 2007; Mueller, 201 2). The basal fish also tend to have a large migrated nucleus m edianus of the posterior tubercle, a nucleus with afferents from the optic tectum and efferents toward the areas of subpallium pars posteriori (Vp), Dl, and Dm; this nucleus shows a strong
158 tren d toward increased volume as species advance, but is mysteriously absent or parcellated in the most advanced taxon, Teleostei (Northcutt, 2009). While the nucleus medianus dyes pos itively for calretinin, the only diencephalic nucleus with similar connectio ns to the nucleus med ianus, the lateral preglomerular nucleus, is monocellular and is calretinin negative. While the nucleus medianus may be split up into several nuclei or entirely replaced in teleosts, either ou tcome represents a paradigm shift in the ho doloy of the Actinopterygian prosencephalon, which suggests the presence of other critical forebrain changes in this most proliferant and diverse infraclass. In the pallium of the basal species there are similar divisions to the better studied tel eostean made: it is apparent that Dm and Dl are in approximately the same location in each taxon, that Dc is localized in vastly different regions, and that Dd is conspicuously absen t in all but the tel eembryological pallial units, though the lesser degree of eversion allows for the simple localiz ation of these units in Polypterus : ventra l P1 is considered topological equivalent of VP, dorsal P1 is the LP, P2 is the DP, and P3 is the MP (Braford, 2009; Northcutt, 2007); this analysis is well supported by hodological and genetic evidence (Nieuwenhuys, 2009). In the sturgeon it is thought by Northcutt and Braford that the situation is highly similar to teleosts, expressing all of the same regions except Dd and that these regions correspond to all four pallial divisions, poss ibly lacking the DP (Northcutt, 2007; Braford, 2009). Nieuwenhuys tak es a different stance and asserts that the sturgeon possesses only two divisions, the LP and MP, corresponding to Dm and Dl combined with Dp, respectively (Nieuwenhuys, 2009).
159 Nieuwenhuys, Braford, and Northcutt have attempted to resolve the problem of teleo stean telencephalic anatomy through attempting phylogenetic rationalizations and predicting the topological outcomes of a simple eversion of the basic telencephalic design of their nearest rel atives, the so called basal groups of the Actinopterygii, and the closest noneverted out group, the Chondrichthyes. If a simple eversion had occurred in the tel eosteon pallium, there are several possibilities for homologies to the typical four fold divison. One could start with the MP as the most dorsal division the primordial state, thus making it the most lateral in adulthood and corr esponding to either Dc or D p, the former the potential result of a dorsomedial migration or the latter representing the most lateral proliferative region. The clearly migratory (due to its di stance from the common ventricle) region Dc is thus problematic and indicates that simple ev e rsion cannot solely explain the origin of this zone nor is it probable that the Actinopterygii have five histogenetic pallial divisions since the four divisions are inherited from more primitive taxa (Nieuwenhuys, 2009). From either Dc or Dp the regions would proceed medially in the order defined embryologically in a true eversion: MP, DP, LP, VP. F IGURE 27 : THE REGIONS DESCRIBED BY NIEUWENHUYS AT THE LEVEL OF THE PREOPTIC AREA. (ADAPTED FROM ITO ET YAMAMOTO, 2008)
160 Various conclusions have been reached, with Northcutt and Braford adhering to the n otion that Dm, Dp, Dl, and Dd correspond to the VP, LP, MP, and DP, respectively (Braford, 2009). This hypothes is excludes Dc as a pallial division and obviously defies simple eversion excluding Dc, which would be characterized by the order Dm, Dd, Dp, and Dl being equated to the VP, LP MP and DP The reasoning for this is based on the connections present in thes e regions, esp ecially the termination of the medial olfactory tract into Dp, and the topological awkwardness of this non linear progression is justified by the comparatively larger Dm and Dl (or VP and MP) crowding the puny Dp into its caudolateral positio n, pressed against the ependymal wall. F IGURE 28 COMPARISON OF THREE BASAL ACTINOPTERYGIAN FOREBRAIN. LEFT SIDES ARE NISSL STAINED, RIGHT SIDE ILLUSTRATES ANATOMICAL REG IONS PER BRAFORD AND NORTHCUTT. (ADAPTED FROM BRAFORD, 2009)
161 Nieuwenhuys provides the rationale for this argument, though he ultimately arrives at a diffe rent conclusion (Nieuwenhuys, 2009): In each of the taxa within Actinopterygii, there are distinct, homologous lateral olfactory tracts which present tightly against the ventricular space This lateral olfactory tract is not in a homologous region in evaginated brains, but rather another lateral olfactory tract (named based on the correct anatomical position in i nverted bra ins) is present and homologous within all members of Gnathostoma with evaginated telencephalons Polypterus contains both the lateral olfactory tracts of everted and evaginated brains, referred to as the lateral and medial olfactory tracts respectively, in which the medial o lfactory tract targets the dorsal part of P1 and the lateral olfactory tract terminates in the P3 The evaginated lateral olfactory tract always sends efferents to the LP in all other Gn athostomes Dp receives strong secondary olfactory in put from both the medial and lateral olfactory tracts Since Dm, Dd, and Dl are topologically equivalent to P1, P2, and P3, they must be h omologous to the LP, DP, and MP, respectively; Dl and Dp both receive olfactory input and are thus probably extensions of the same division Nieuwenhuys presumes a simple eversion beginning at Dm and terminating the at Dl, since he finds Dp to be an extension of Dl and omits the VP. This is certainly supported by his analysis of the trajectory of these two unique olfactory tracts, but hodology cannot define homology, i n
162 auxiliary criterion, [sic] because a cell mass may change one or more of its connections, [sic] without losing summary analysis of the whole state of affairs: lacking for teleosts and for actinopterygians in ge neral. Because of uncertai nties regarding the embryonic origin of several pallial cell masses and therefore their primary topological position (see table 1), it will be of great interest for comparative anatomists to establish whether these four basic pall ial zones are also present in representatives of each of the five actinopterygian clades (fig. 1), and if so, to trace the migratory paths of their derivatives. Given the consider able variation in forebrain structure in these clades, it will not be suffici ent to confine these studies to a single species (and to proclaim this species subs eNieuwenhuys also admits to being unable to trace the apparent histolo gsuspected by Northcutt and Braford (1980). Altogether, he stands alone in his a s serti on that Dm is homologous to the LP and he is the first to claim Dp as an extension of Dl. (Nie uwenhuys, 2009) F IGURE 29 : C ONTRASTING THE STATE OF AFFIARS OF THE MOST POPULAR MODEL O F THE TELEOSTEAN DOR SAL TE LE NCEPHALON WITH THE MO USE (A DAPTED FROM M UELLER ET AL 2010)
163 Embryonic Origins of Pallial Zones Recent genetic evidence by a scientist thoroughly engrossed in this field of research, Dr. Thomas Mueller, and his German collaborators have recently identified a new proliferative zone that represents a significant advancement to this heavily conjectured question. By dually stai ning the pallium of the most common model teleost, the zebrafish, with nicotine adenine din ucleotide phosphate diphorase (NADPHd) and parvalbumin, this new publication has exposed the embryonic origins of three essential regions in question, Dp, Dd, and Dc. NADPHd binds nitric oxide synthase(an essential enzyme for the formation of nitric ox ide gas and a basic component of many kinds of GABAergic, acetylcholinergic, and glutamatergic cells in the CNS), thus provi dprogressively more caudal tranver se sections to form the dorsal border of Dc. This provides Dc with an embryonic origin and identifying it as a histogenetic zone. By employing a marker for DNA uncoupling to test for neurogenesis, Bromodeoxyuridine (Brd u), the scientists also o bserved the migration of cells originating from the proliferative zone at Dl to the paraventricular area at Dp, thus eliminating Dp as a candidate for an independent pallial zone. Combined with previous evidence, Mueller and co lleagues present their findings definin g the pallial zones thus: the VP is Dm (the homologue of the amygdala of tetrapods), the LP is the combination of Dp and Dd (the homologue of the piriform cortex), the DP is Dc (homologous to the isoform cortex), and the MP is Dl (the homologue to the hipp ocampus). This coincides with the behavioral evidence presented as well as with the models of eversion put forth by Northcutt and Braford, who prev iously recognized Dp as a migratory region. (2011) Northcutt, the long standing authority on the teleostean d oral pallium, has since pu bnote that that teleostean dorsal pallium is unlikely to be resolved through research in a single
164 species of cyprinid; he continues to correctly state the the question of tetrapodian homology relies on the demonstration of these four distinct pallia in the preceding clades, the chondro steans (polypteriforms and acipenseroids) being the most essential as the closest extant taxa to the orig inal ancestor of both actinopterygii and tetrapods. Northcutt expounds a series of doubts regarding the conclusion that Dd is simply the narrow zone from which Dc derives its neural progenitors: the Dd has not been conclusively identified in non teleosts; likewise, Dc/DP has not been found in polypteriforms despite previous attempts to find a pallial connection to the optic tectum; finally, the sulcus ypsiloniformi, which Mueller and others suggest is the entry with the hypertrophied MP and VP, is only pr esent in teleosts. While Northcutt acknowledges Mueller et al's work as valuable, he finds the claims of the discovery of the dorsal pallia in teleosts to be extravagant. Mueller rebuts that is is less disputed i f the distal outgroup chondrichthyes possess a dorsal pallium or even agnathans, but concedes that the identification of which remains to be proven in studies of chondrosteans (2011).Despite uncertainties regarding the exact layout of the pallia, he states that it the exis tence of regions homologous to the amygdala, hippocampi, and piriform cortex of tetrapods has never been in doubt, merely anatomically implacable. In conclusion, it is apparent that while much debate surrounds the exact locations of the hi ppocampal and amygdalar formations of the many diverse species in Actinopterygii remains ambiguous, it is often taken for granted that the behavioral and genetic evidence, while inco mplete, suggests that the commonly used model teleost species have regions in the brain that are brain. While the questions of the unique teleostean telencephalon excite further inquiry, the answers are unlikely to emerge from theoretic al derivations of embryological processes, which
165 have already been thoroughly evaluated, but rather from further genetic and hodological ev idence to confirm or deny preexisting theories.