This item is only available as the following downloads:
DEVELOPMENT OF PROCEDURES TO DETECT PATHOGENS IN Rhizophora mangle L. BY ALYCIA SHATTERS 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 Dr. Amy Clore Sarasota, Florida March, 2010
ii Acknowledgments I would like to thank my Dad for instilling in me m y love of learning. He has always been the driving force in my head pushing me to do no less than I know I am capable of. Thank you to my Mom, whose strength has been my inspiration and whose support I would be lost without. My parents believe d in me when I would not. Warm wishes for a bright future go out to my belov ed friends who went through this with me, step by step. We struggled together, laughed together, and now we will graduate together. My dear friends Marisol, Miranda Lindsey, Elena, Elizabeth, and Ahmad, thank you for being my support. I would also like to thank Dr. Donna Devlin for al lowing me to work with her on this project and for answering all of my questions with patience. Thank you to Dr. Amy Clore, my sponsor, who has helped me shape this the sis. Thank you also to Dr. Mariana Sendova and Dr. Margaret Lowman for being on my com mittee.
iii Table of Contents Contents Page # Acknowledgments ii List of Figures v List of Tables vi Abstract vii Chapter I: Introduction to Mangroves and Associated Pathogens 1 1.1 Distribution and zonation of Rhizophora mangle L. 1 1.2 Adaptations of Rhizophora mangle L. 3 1.3 Reproduction and dispersal 6 1.4 Rhizophora mangle L. as a keystone species 8 1.5 Traditional uses of Rhizophora mangle L. 12 1.6 Rhizophora mangle L., a species at risk 13 1.7 Known Rhizophora mangle L. colonizers, including pathogens 15 1.7.1 Fungi associated with R. mangle 16 1.7.2 Bacteria associated with R. mangle 18 1.8 Global climate change affecting mangrove healt h 20 1.9 Dwarf mangrove forests 21 1.10 Rhizophora mangle L. in Biscayne Bay, Including those with 23 abnromal morphology Chapter II: Procedures 29 2.1 Tissue collecing 29 2.2 DNA extraction 30 2.3 Selection of a positive control 31
iv 2.4 PCR analysis to test for the presence of funga l and bacterial 33 sequences 2.5 Adaptations of a cotton genomic DNA isolation protocol for 34 R. mangle tissue 2.6 Application of modified cotton genomic DNA iso lation protocol 36 for abnormal R. mangle tissue 2.7 Isolation and identification of amplified sequ ences 36 Chapter III: Results 39 Chapter IV: Discussion 56 4.1 Development of new methods was necessary for i solation of high-quality PCR-amplifiable DNA from R. manlge tissue 56 4.2 Identification of differences, using DNA profi ling, between R. mangle samples taken from abnormal and control trees growing in Biscayne National Park 57 4.3 Future studies 60 4.4 Conclusions 63 Appendix 64 Works Cited 70
v List of Figures Figure Page # Figure 1.1 Rhizophora mangle L. 2 Figure 1.2 Map of mangroves along Florida's coastli ne 3 Figure 1.3 Photograph of cork warts 5 Figure 1.4 Red mangrove propagules 7 Figure 1.5 Rhizophora mangle L. dwarf forest 23 Figure 1.6 Example of observed abnormal morphology in the height 24 Figure 1.7 Example of observed abnormal morphology in the leaves 25 Figure 1.8 Example of observed abnormal morphology in the branches 26 Figure 1.9 Location where samples were collected 28 Figure 2.1 Isolation variations 38 Figure 3.1 Results of PCR using Rhizophora microsat ellite primer 40 Figure 3.2 Results of temperature gradient PCR usin g microsatellite primer 46 42 Figure 3.3 Results of PCR using Rhizophora PAL prim ers 44 Figure 3.4 DNA quality test 45 Figure 3.5 Results of PCR using the rigorous isolat ion protocol with 50 ng of DNA 47 Figure 3.6 Results of PCR using the rigorous isolat ion protocol with 100 ng of DNA 48 Figure 3.7 Results of PCR using rigorous isolation protocol with 200 ng of DNA 49 Figure 3.8 Universal primer run with bacterial and phytoplasma primers 57 Figure 3.9 Universal primer run with fungal primers 53
vi Figure 3.10 BLAST results for band A 55 Figure 3.11 BLAST results for band B 55 List of Tables Table 1 Information about primers used for PCR 67 Table 2 Information from the NanoDrop Spectrophotom eter 68 Table 3 Rating of DNA template quality (as assessed by A260/A280 ) and band intensity after PCR 69
vii DEVELOPMENT OF PROCEDURES TO DETECT PATHOGENS IN Rhizophora mangle L. Alycia Shatters New College of Florida, 2010 ABSTRACT Rhizophora mangle L. trees form a forest on the edges of Biscayne Ba y National Park, where a unique abnormal morphology, including club-shaped branch termini with extremely short internodes and clumped leaves, was observed in several sections of this R. mangle forest. These abnormal trees also displayed more s evere herbivory than neighboring healthy looking trees, and some localiz ed symptoms were so severe that dead patches of the forest were observed. Visual sy mptoms seemed to be consistent with abnormalities in other trees caused by microbial pa thogens. Therefore, universal polymerase chain reaction (PCR) primers for fungi, phytoplasmas, and other bacteria were used to test for the possible presence of path ogens in both abnormal and normal R. mangle tissues collected from this site. After some initi al difficulties with the PCR using conventionally extracted DNA templates, a new DNA e xtraction procedure, meant to get rid of contaminants that inhibit PCR, was developed Using this procedure, templates were produced and used with the universal primers a nd PCR in an attempt to identify a possible pathogen from the abnormal tissue. Candida te amplification products were
viii sequenced and possible explanations were formulated as to the cause of the abnormalities. Although definitive results were not obtained, pursuing the cause of the abnormalities is important due to the integral role R. mangle plays in its environment. Dr. Amy Clore Division of Natural Sciences
1 Chapter I: Introduction to Mangroves and Associated Pathogens 1.1 Distribution and zonation of Rhizophora mangle L. Mangroves are an identifying feature of Florida's coastline. Mangroves are a group of facultative halophytes, or plants that hav e adapted to life in saline conditions, that are divided into eight families: Avicenniaceae Chenopodiacea, Combretaceae, Meliaceae, Myrsinaceae, Plumbaginaceae, Rhizophorac eae, and Sonneratiaceae (Lugo, 1974). These families, while all called mangroves, are related only distantly, having been grouped together based solely on their ecological f unctions within their environments and the habitats they colonize (Lugo, 1974). Rhizophora mangle L., in the family Rhizophoraceae is more commonly known as the red mangrove (figure 1.1). These adaptive tr ees fringe saltwater and brackish shores along the southern half of Florida's East an d West coasts (figure 1.2). Globally, there are two genetically isolated regions of R. ma ngle that are separated by the Eurasian landmass and the Eastern Pacific Ocean (Duke et al. 2002). Their range is temperaturedependent in that they are limited to subtropical a nd tropical regions. There are approximately 117 countries and territories that ha ve mangroves, the largest concentration being in Indonesia (Barbier and Cox, 2003).
2 Figure 1.1: Rhizophora mangle L. The red mangrove can be identified by its unique pr op roots which are visible above ground and become submerged as they take hold in th e sediment. They have elliptical leaves and long, cylindrical propagules (not shown here) that hang from the ends of branches. (http://www.nt.gov.au/nreta/wildlife/nature/pdf/man groves/2_mangrove_ecosystem.pdf)
3 Figure 1.2: Map of mangroves along Florida's coastl ines In the state of Florida, mangroves are found on bot h the east and west coasts. Their range starts from the very tip of the peninsula and exten ds north to St. Augustine on the Atlantic Coast and Cedar Key on the Gulf Coast. (http://www.sfrc.ufl.edu/extension/pubtxt/for43.htm ) 1.2 Adaptations of Rhizophora mangle L. R. mangle grows right along the water's edge, often below th e high-tide mark, and has made several unique adaptations that allow it t o do so. It has taken advantage of an ecosystem where it does not have many competitors f or space. The roots and methods of reproduction that R. mangle has developed allow this mangrove to survive in waterlogged, anaerobic soils where available water is high in salinity (Thibodeau and Nickerson, 1986). The characteristic roots of R. mangle are mainly long extensions from the stem and branches that grow down toward the soi l and end underground. There are
4 two types: prop roots which grow down from the base of the stem, and drop roots which grow down from the upper part of the stem and the b ranches (McKee et al. 1988). These roots do not go deep underground, as the ground in the estuarine habitats favored by mangroves is too soft and would not support trees w ith deep roots. Instead R. mangle has shallow root systems that spread out widely, lendin g support to the upward growth of the stem and branches and bearing the weight of the lea ves and propagules (McKee et al. 1988; Nagelkerken et al. 2000). Red mangroves have had to adapt a unique mechanism to provide their submerged roots with oxygen from the atmosphere. A study done by Evans et al. (2005), showed the path of air flow through R. mangle and how oxygen is provided to the roots by submerging different sections of a mangrove plant, injecting air into one area, and observing where the air bubbles left the submerged plant. The results of this study suggested that air normally enters the plant throug h cork warts, or areas of tightly packed sclerified cells that are apparent as dark brown to black spots on the abaxial surface of aerial leaves (figure 1.3). Air enters the leaves t hrough the cork warts and becomes pressurized within the layer of aerenchyma tissue t hat lies below. This layer of aerenchyma tissue in the leaf is connected to the a erenchyma tissue in the petiole, and air diffuses into this tissue, then continues to diffus e into the inner aerenchyma of the mangrove's stem (Evans et al 2005). Air travels down the stem and diffuses int o the inner root aerenchyma tissue where oxygen is provid ed throughout the submerged roots all the way to the termini. The air then ascends th e outer aerenchyma of the roots and is released into the surrounding environment through l enticels, or pores on the surface of the submerged roots that allow gas exchange (Evans et al 2005).
Figure 1.3: Photograph of cork warts Abaxial surface of a mangrove leaf submerged to sho w air bubbles that came from the cork warts (visible in this photograph as dark spot s on direction of air flow in non R. mangle has a filter system in its roots that prevents most of the salt from ever entering the plant's tissues, but there is some unc ertainty about how this is accomplished. One of the main theories is that that is high enough to prevent the uptake of salt w hile allowing the intake of freshwater (Smith and Snedaker, 1995). Figure 1.3: Photograph of cork warts Abaxial surface of a mangrove leaf submerged to sho w air bubbles that came from the cork warts (visible in this photograph as dark spot s on the leaf), which is the opposite direction of air flow in non -submerged mangrove leaves (Evans et al. 2005). has a filter system in its roots that prevents most of the salt from ever entering the plant's tissues, but there is some unc ertainty about how this is accomplished. One of the main theories is that R. mangle maintains a negative internal osmotic pressure that is high enough to prevent the uptake of salt w hile allowing the intake of freshwater (Smith and Snedaker, 1995). 5 Abaxial surface of a mangrove leaf submerged to sho w air bubbles that came from the the leaf), which is the opposite 2005). has a filter system in its roots that prevents most of the salt from ever entering the plant's tissues, but there is some unc ertainty about how this is accomplished. maintains a negative internal osmotic pressure that is high enough to prevent the uptake of salt w hile allowing the intake of freshwater
6 1.3 Reproduction and dispersal R. mangle is viviparous, meaning it releases live young. Its leathery, yellowish flowers are mainly pollinated by wind, and perhaps also by insects such as bees (Kathiresan and Bingham, 2001). Because it tends to form dense forests along coastlines, with little species variation, self-pollination is heavily relied upon for reproduction (Proffitt et al 2006). Relatively little is known about the repro duction and dispersal of R. mangle; recent studies have more of a focus on habitat los s and degradation. Vegetative reproduction accounts for very little of the disper sal of R. mangle, thus they depend heavily on seedling recruitment, as explained below (McKee, 1995a). The seedlings, called propagules (figure 1.4), germinate while sti ll on the plant, without a seed dormancy period as is found in many species. The lo ng propagules form elongated, buoyant hypocotyls that hang vertically on the ends of branches. When the propagules are sufficiently mature, a collar is formed from the fu sed cotyledons. This marks where the seedling detaches from the fruit, while the cotyled ons are left behind on the tree (Lowenfeld and Klekowski, 1992). This happens mainl y during September and October, but individuals may also disperse off-season, durin g the rest of the year (Rabinowitz, 1978).
7 Figure 1.4: Red mangrove propagules Red mangroves are viviparous, and their seeds germi nate into propagules while still attached to the parent plant, as shown here. The pr opagules grow on the ends of branches and when they are mature, the seedlings separate fr om the cotyledons and disperse. ( http://ecolibrary.org/page/DP4110 ) After they drop from the parent plant, the propagu les are able to float on the water for extended periods of time before they reach a su itable environment to grow, taking into account light, salinity, pH, water levels, and soil content, as stated in Proffitt et al. (2006). The propagules will begin to grow whether they are placed vertically or horizontally in the soil, and there are two different possibilities as to how they grow in the wild. The first is that they strand in a suitable environment horiz ontally, then the roots begin to grow into the soil which anchors them and the propagule will erect itself. The second is that as the propagules float vertically, the tips embed the mselves into the soil by abrasion and Place of seedling detachment from cotyledons
8 then the propagules begin to grow. Rhizophora propagules often sink some time after dispersal and start to grow underwater. If the area proves to be unsuitable, because the propagules are large, it is thought that they have the ability to become buoyant once again and can disperse further (Rabinowitz, 1978; Tomlins on 1986). 1.4 Rhizophora mangle L. as a keystone species R. mangle is a keystone species, or a species that plays a f undamental role in maintaining an ecosystem, in many estuarine environ ments. R. mangle plays critical roles in the development and shelter of many other specie s of plants and animals as well as contributing to the preservation of their ecosystem s. One of the main benefits of this species is by the network of prop roots that act a s a nursery for many species of juvenile fish, creating an environment rich in available foo d that is protected from the threat of predation by larger species (Shervette et al. 2007). The roots have a large surface area on which epiphytic algae proliferate and provide food for many invertebrate species, mainly crabs and shrimp, which in turn provide food for no t only the juvenile fish sheltering there, but also many species of birds (Laegdsgaard and Johnson, 2001). Also, the complex structure of the roots of R. mangle and the vegetation that grows on them combine to create a refuge for young fish. During h igh tides, they travel far into the structured mangrove forest, thus reducing the incid ence of predation from larger fish, which stay on the edges of the forest because they lose their ability to quickly maneuver when swimming within the prop roots (Ronnback, 1999 ). Mangrove areas are also less frequented by large predators, due in part because they are separate habitats from coral reefs and deeper offshore waters (Nagelkerken et al 2001). Finally, the mud in which the mangroves grow is very soft, which is ideal for pro tecting many burrowing animals such
9 as crabs (Ronnback, 1999). A study by Laegdsgaard and Johnson (2001) showed t hat mangrove prop roots do not provide the only coastal habitat that shelters and feeds juvenile fish, but mangrove roots are the most beneficial as compared to seagra ss beds and mudflats, which are habitats that are often found adjacent to mangrove forests. These researchers found that mudflats alone offer little protection in the way o f physical structure and that the survival rate of juveniles in this habitat is extremely low. Seagrass beds offer the same amount of protection as mangrove prop roots, but there is les s food available and therefore seagrass beds are not as opportune a habitat as the prop roo ts (Laegdsgaard and Johnson, 2001). As the water flows through mangrove forests, the p rop roots of R. mangle filter and trap sediment that adds to the soil surrounding the area. This helps to stabilize the soft ground and also plays a role in the upkeep of seagrass beds, which are also important to the ecosystem. Soil erosion is a major threat to many coastlines throughout the world, resulting from land clearing for agricultural and a quacultural use, housing development, and the building of roads (Victor et al 2004). When the mangrove forests are cleared, the land erosion increases, caused by tidal changes or ocean waves. The mangroves no longer act as a buffer to trap much of the sediment, and i t gets dumped elsewhere, effecting other ecosystems down the line (Victor et al. 2004). Airai Bay, Palau, Micronesia, is a small bay that is a specific example of an area aff ected by such clearing of the mangrove forests and is experiencing an increase in soil ero sion in the estuarine area. It has been shown by Golbuu et al (2003) that, on average, about 30% of the riverin e sediment that flows through Airai Bay settles in the remaining ma ngrove fringe forest, and about 1% of that is resuspended by currents and wind. As is des cribed below, it is this resuspended
10 sediment that gets dumped onto other fragile ecosys tems, and the percent trapped by the mangroves is shrinking along with the shrinking R. mangle population. This finding shows the need to preserve what mangroves are left. Golbuu et al (2003) also showed that the coral reefs of Airai Bay are indirectly affected by the clearing of mangroves. The mangrove s forests slow water flow, so with their loss there has been a significant increase in water turbidity in the bay area, which stresses coral reefs that are located near the frin ging mangrove forests. The increase in water turbidity causes less light to penetrate the water and effects the growth patterns of coral in that area (Fabricius, 2005). The silt that is not trapped by the prop roots is often deposited on top of reef structures and is held the re by the algal mat that forms on the surface, together creating a muddy layer that smoth ers reefs (Golbuu et al 2003). Since the coral is already stressed from the increase in water turbidity, this accumulation of sediment may prevent the recovery of already establ ished corals, as well as the recruitment of new coral larvae to rebuild what was damaged (Victor et al 2004). The soil in estuarine ecosystems is made of mainly dead and decaying plant and some animal material, or detritus. Mangroves in gen eral are responsible for producing an estimated 8.86-14.16 t DW ha-1year-1 of litter fall, the main component being leaves (Ng a et al 2005). This is a large contribution to the amount of detritus accumulated in estuarine ecosystems. The mangroves take up nitroge n and phosphorus from their environment, then when their leaves fall to the gro und and decompose, those nutrients help form the base of the food chain (Nga et al 2005). The main process by which plant matter is turned into detritus in mangroves involve s physical and chemical agents as well as many microorganisms. Almost as soon as the plant matter falls from the tree, microbial
11 colonization begins and fungal hyphae grow into the plant matter and release exoenzymes that break down cellulose in the cells, weakening t he plant tissue (Bowen, 1987). From there, mechanical agents such as water movement and the abrasion by particulates in the water cause further fragmentation. When the plant m atter is still coarse, fungi are the major colonizers, while bacteria are most abundant on fine particles. As this fragmentation pathway takes place, some protein fro m the original plant matter is lost in solution, but more importantly, some is degraded in to small amino acid polymers, individual amino acids, and organic nitrogen (Bowen 1987). Over time, these nutrients are bound into organic complexes that make them dif ficult to digest, so organisms called detritivores have evolved mechanisms that allow the m to digest and use the nutrients trapped in detritus, and the detritivores are in tu rn eaten by animals further up on the estuarine food chain. The amino acids form a necess ary part of the diets of many animals such as fish, which require a high percent of their diet to be amino acids in order to attain maximum growth (El-Dahhar and El-Shazly, 2008). The re is some debate over whether detritivores get these necessary amino acids direct ly from the detritus or through the microorganisms that are attached to the detritus pa rticles and are ingested by detritivores along with the detritus, but scientists have favore d the later explanation (Bowen, 1987). Because of these functions, the health of the mang roves is directly connected to the health and productivity of their estuarine ecos ystems. It has been shown that mangrove forests in some parts of Australia's coast s have high rates of litter production, however they are not producing below maximum capaci ty, as compared to mangrove forests in Thailand (Boto and Wellington, 1983). Le ss than optimal growing conditions, mainly limited amounts of nitrogen and phosphorous in the Australian soils, were shown
12 to be the culprit of this decrease in productivity (Boto and Wellington, 1983). The large amounts of biomass that comes from mangroves is an important transfer of energy from terrestrial to marine ecosystems. With the disappea rance of the mangroves, food chains are broken down from the bottom up. 1.5 Uses of Rhizophora mangle L. Historically, mangroves have been used by humans ar ound the world for a variety of uses. Some of the products provided by mangroves are timber, fuel for cooking, charcoal, food, and construction materials for boat s, housing, furniture, and posts for pastures and rural housing; Rhizophora specifically is known for its abundant tannins and medicinal uses (Kovacs, 1999). Tannins are very ast ringent, meaning they are able to precipitate proteins from a solution, and for this reason they are used to tan animal hides and cure leather. Tannins have other abundant uses in cooking and medicine, as described by Bandaranayake (1998). They are water soluble pol yphenols, which aid in disease prevention, and have other medicinal qualities like being anti-inflammatory and are used to stop infection of internal wounds. Large concent rations of tannin in the wood cause it to be highly resistant to rot, an asset when choosi ng building material (Bandaranayake, 1998; Berenguer et al 2006). Even though they are useful in so many different w ays, in several areas today there are extremely strict restrictions on the harv esting and use of R. mangle According to the Florida Statutes, alteration, which includes defoliation, destruction, removal, and trimming of mangroves, is prohibited unless a permi t is applied for. Any legal trimming is usually restricted to a height of 6 feet. These res trictions extend to cover the alteration of mangroves by homeowners ( 200-2010 State of Florid a; http://www.flsenate.gov).
13 1.6 Rhizophora mangle L., a species at risk There has long been a failure to recognize the int egral role that mangroves, including R. mangle play in estuarine ecosystems. As evidencied by th e restrictions mentioned above, recently this ignorance has starte d to be reversed, although there is still a shortage of conservation efforts. Since 1900, abo ut 50% of the global mangrove area has been lost, and what is left remaining is declin ing at a global average annual rate of about 2.1% (Gilman and Ellison, 2007). A change mus t occur soon in order to save the mangroves since it has been estimated they might di sappear within the next 100 years (Duke et al. 2007). Many places are trying to counter the loss of mangroves by simply planting new trees among older ones where dense pop ulations used to occur. Trying to replant or start up new populations of mangroves is a complicated process that involves many factors. One problem that often arises with th is solution is that the stressor, the condition that first affected the older generations of trees in the area, is neither identified nor removed. Therefore the young, newly planted tre es may not replenish the mangrove forest, since they may die just as their predecesso rs (Gilman and Ellison, 2007). There are several major threats to mangroves; popul ations are suffering severe devastation in areas like Asia where much of the la nd is being cleared for aquaculture farms (Nga et al 2005). Waterfront development is also a major thr eat to mangrove ecosystems. Improper agricultural land use and cle aring has caused an increase in soil erosion with degenerative effects only starting wit h the mangrove forests and reaching further into other ecosystems, like offshore seagra ss beds and coral reefs. From the beginning, shrimp aquaculture has been an unsustainable industry. According to Graaf and Xuan (1998), it has grown dr amatically since the early 1990's,
14 causing destruction of many mangrove forests as wel l as self-pollution of the farms and viral disease outbreaks. It has been recorded that in the Mekong Delta of Vietnam, an area where shrimp farming is one of the major forms of a griculture, one hectare of mangrove forest can support a marine catch of 450 kg per yea r, a number that is similar in mangrove stands throughout the world. (Graaf and Xu an, 1998). However, this total marine catch that mangrove forests can support is r apidly shrinking and will continue to do so as long as mangroves are still disappearing. The growing shrimp culture in the Asian region has been governed solely by market for ces, and has been absent in national development plans and lacking in industry managemen t (Graaf and Xuan, 1998). Like many coastal aquaculture development projects, the free benefits provided by the ecosystem, in this case the mangrove stands, were t aken for granted while the possibility of short-term profits caused the exploitation and d estruction of the natural ecosystem. This is just one example of the lack of regulation that has occurred in regards to the exploitation of mangrove populations worldwide. R. mangle is also sensitive to environmental factors. There is no seed bank because it is viviparous, so if a mangrove forest s uffers severe damage, it may take great effort to reestablish a population (Proffitt et al 2006). Hurricanes and tsunamis are especially destructive, and have wreaked havoc on m any populations in the recent years (Baldwin et al. 2001). Fringe mangrove forests often receive the f ull force of the storm as it first comes ashore, and these storms can both de stroy the existing trees and flowers, and also strand propagules at higher elevations tha n they can grow (Proffitt et al 2006). However, if there is a dense population of seedling s growing below the canopy, then an R. mangle forest would stand a good chance of complete regen eration according to
15 Baldwin et al. (2001). However, as is typical of populations of sp ecies on islands (where mangroves are often found), rates of regrowth are s o slow that a decimated population will never reach the size or density it was before the devastation (Simberloff and Abele, 1976). One of the major hurricanes that devastated Florida's coastlines was Hurricane Andrew, in 1992. In the mangrove forests on the sou thwest coast, about 60% of the trees were uprooted or broken, and about 25% of the uprig ht, unbroken trees died (McCoy et al. 1996). The Asian Tsunami in December of 2004 proved just how valuable mangrove forests can be. While suffering extreme damage, the mangrove fringe forests of Thailand severely reduced the tidal wave energy form the for ce of the tsunami, protecting the inland population from the worst of the storm (Barb ier, 2006). 1.7 Known Rhizophora mangle L. colonizers, including pathogens There are many microorganisms, or colonizers, that live in and on trees. Colonizers occur in all natural ecosystems and are sometimes known for their deleterious effects on organisms, however in forests (stands of Rhizophora included), as well as other ecosystems, they can have advantageous and disadvan tageous effects. In general, most species of trees support a community of organisms t hat depend on the tree for survival. R. mangle grow in a restricted habitat, and strands of these mangroves are high in plant density and low in species diversity. In areas with a high density of mangroves, especially monocultures, plant populations are more susceptibl e to diseases, which increases the threat posed by pathogenic colonizers to R. mangle communities (Gilbert et al. 2002). The most common of these colonizers are fungi and b acteria, both of which are found in great abundance both on and within R. mangle trees.
16 1.7.1 Fungi associated with R. mangle There are two main types of fungi that can be found within the R. mangle habitat. Marine fungi colonize the submerged and often decay ing detached roots, stems, and leaves, while the terrestrial fungi are found on th ose parts of the mangroves above the water line (Lee and Baker, 1973). The types of fun gi associated with mangroves are diverse, some of them are beneficial to the mangrov es while others are pathogenic (Suryanarayanan et al. 1998). Some species of fungi associated with the m angroves serve crucial roles to their surrounding environments, li ke the manglicolous fungi which are vital in nutrient cycling. They colonize and break down the dead, submerged mangrove tissues, and release the stored nutrients into the environment for use by other species (Kathiresan and Bingham, 2001). Other fungi are det rimental to the health of the mangroves, causing high levels of foliar disease. D iseased trees display symptoms such as necrosis, or the premature death of living cells and tissues, and chlorosis, in which the leaves appear yellow or white due to the breakdown and/or insufficient production of chlorophyll (Gilbert et al. 2002). R. mangle has been found to suffer more folier damage due to fungal pathogens than A. germinas or L. racemosa (Gilbert et al. 2002) This could be due to the different methods of salt excretion used by the different man groves. Since A. germinas excretes excess salt through glands on the leaves and L. racemosa accumulates salt in its leaves, large concentrations of excess salt would be allowe d to build up on the surface of such leaves, which would not happen with the root filtra tion system of R. mangle (Drennan, 1982). It has been speculated by Gilbert et al. (2002) that the salt could act as an accessory mechanism for resistance to pathogenic fu ngi. In their observations they found
17 similar amounts of insect herbivory on leaves of bo th R. mangle and L. racemosa however there were significantly lower levels of fu ngal disease on L. racemosa Due to the high internal salt content of the leaves, the w ounds would be quite saline and thus potentially protected against many fungal infection s (Gilbert et al. 2002). Stands in the Carribean and Gulf of Mexico have be en the sites of many studies on resident fungi of R. mangle because this mangrove dominates coastal forests of this area (Wier et al. 2000). There have been several species identified and associated with the afflictions they cause in the trees: species of Anthostomella and Cercospora cause leaf spot in Puerto Rico and Florida, respectively; a sp ecies of Cytospora caused a dieback of branches and roots in R. mangle seedlings in Hawaii; and Cylindrocarpon didymum has been associated with a stem gall disease in South F lorida and Gambia, Africa (Wier, et at. 2000). One of the more recent findings in this fi eld linked Cytospora rhizophorae, an imperfect fungus (meaning that they lack a sexual s tage), to stands of R. mangle in the coastal areas of Puerto Rico. This fungus was consi stently associated with canker tissue, characterized by leaf spotting, and stem dieback an d mortality, and was isolated from the symptomatic tissues (Wier, et at. 2000). C. rhizophorae is a facultative pathogen so it is not restricted to marine habitats, and it requires a wound to infect a host (Wier, et at. 2000). Such wounds occur naturally by herbivory and breakage from strong winds. As breifly mentioned above, some fungi are benefic ial, such as the manglicolous fungi that help recycle nutrients from parts of the trees that have died or fallen off (Kathiresan and Bingham, 2001). There are many diff erent families of such fungi that live in the mangrove habitat including Ascomycetes, Deuteromycetes, and Basidiomycetes (Kathiresan and Bingham, 2001). Many produce lignocellulose-
18 modifying exoenzymes, like laccase, which is associ ated with white-rot fungi, known for their ability to decompose polysaccharides of cellu lose, hemicellulose, and lignin (Luo, 2005). In this case, this destructive activity is w hat drives the estuarine food chain. 1.7.2 Bacteria associated with R. mangle Bacteria, like fungi, are abundant in mangrove for ests. Some are exogenous, colonizing the apoplast, and others are endogenous, infecting vascular tissues, specifically the vessels in xylem and sieve tubes i n phloem (Bove and Garnier, 2002). Pathogenic bacteria affect their host cells in diff erent ways. For example, many produce extracellular products, like certain proteins, that alter the cellular pathways of the host plant in order to facilitate the growth of the path ogens within the plant (Salmond, 1994). These extracellular proteins are often structurally similar between different bacteria, and thus serve similar functions. Some examples of secr eted proteins include cellulases, proteases, and lipases (Buttner and Bonas, 2003). M any bacteria are highly specialized and only infect specific hosts, while others can sp read to a wide variety of hosts (Buttner and Bonas, 2003). A specific group of bacteria that are restricted t o the phloem and that, until recently, were thought to have been viruses, are ca lled phytoplasmas. Because they are similar to mycoplasmas (class Mollicutes), in that they are both prokaryotic and lack cell walls, they were originally labeled mycoplasma-like organisms (MLOs), but are now called phytoplasmas (Hogenhout et al. 2008). Phytoplasmas are carried by insect vectors from plant to plant, so their life cycle requires t wo separate hosts. They are mainly vectored by leafhoppers and have been identified in over 100 different plant species (Christensen et al 2005) in which they interfere with normal develop ment. Infected
19 plants produce extra leaves, shoots, and branches, and also yellow, leading to witch's broom, a condition characterized by dense clusters of branches or leaves (Strauss, 2009). When a leafhopper feeds on an infected plant, the p hytoplasmas enter the insect through the stylet, travel through the intestine, a nd circulate in the hemolymph (Christensen et al 2005). Eventually, they multiply within the saliv ary glands so that when the insect feeds on a subsequent plant, an inf ective dose of the phytoplasma is administered (Christensen et al 2005). Phytoplasmas have not been found to have a ny major adverse effects on their insect hosts, most l ikely because they need the insects as a means of traveling between plant hosts so induction of ill effects would not be advantageous (Christensen, 2005). In plants, phytop lasmas are mainly concentrated in the sieve tubes of the phloem and spread through the si eve plates (Bove and Garnier, 2002). Phytoplasmas, like mycoplasmas, have reduced genom es, lacking the genes needed for many basic metabolic pathways (Hogenhout et al 2008). In fact, their genomes are even more reduced than those of mycopla smas. Most bacteria have a phosphotransferase system (PTS), which imports and phosophorylates sugars. Most phytoplasma genomes, however, do not code for any P TS system components. In their place, these genomes have many transporter-related genes not present in bacteria with the PTS pathway (Hogenhout et al 2008). Because of this, the phytoplasma are extre mely dependent upon their host cells since these genes c onfer the ability to import host nutrients. Phloem sap is deficient in some nucleoti des, which the phytoplasmas cannot synthesize (Christensen et al. 2005). Therefore it is highly likely that the comp anion cells of the phloem that provide necessary compounds to t he sieve tube elements also supply the phytoplasmas with such resources (Christensen et al 2005). Phytoplasma genomes
20 also lack genes coding for ATP synthase, but they d o have a large membrane potential, a sign of an extremely efficient active transport sys tem. They do have five genes encoding P-type ATPases, in some ways analogous to the Na+/K+ and H+/K+ pumps found in animals, which have the ability to create the elect rochemical gradient across the membrane (Hogenhout et al 2008). 1.8 Global climate change affecting mangrove health With global climate changes affecting ecosystems w orldwide, the ability to detect and identify possible pathogens early on is of weig hty importance. Sustained climate change, such as global warming, imposes pressures t hat cause individuals within an environment to adapt over time. The increase in glo bal temperatures may accelerate the rate at which insect vectors develop, thus creating a favorable environment for an increase in the reproduction rate of pathogens, whi le not necessarily affecting the rates of the host reproduction, in this case R. mangle (Ayers and Lombardero, 2000). Currently, the populations of some insects are kept in check b y annual low temperatures that cause die-offs. But with increases in temperature, insect vectors will be able to survive in a wider area than they previously could, introducing the pathogens they carry to an increasing number of hosts (Strauss, 2009). Another direct threat to mangroves due to global c limate change is the rise in sea levels worldwide (Alongi, 2002). Suitable coastline s for replanting and reestablishing mangrove populations are already rapidly shrinking because of the large amount of coastal housing development (Nicholls et al. 1999). Coastlines that once could have supported the growth of mangrove trees are now cove red with waterfront houses and hotels. A significant rise in sea level will wipe a way what little suitable coastline is left
21 for the mangroves. Due to the large amounts of biomass produced by ma ngroves, there is some vertical rise in the habitat that will counter a gr adual rise in sea level (Alongi, 2002). However, if the rate of sea level rise is greater t han the rate of vertical biomass build-up, then the ecosystem will become watterlogged and die Inland wetland migration will then be vital for the upkeep of the estuarine ecosystem (Nicholls et al. 1999), but if that land is highly developed, it will never provide a suitable wetland environment. 1.9 Dwarf mangrove forests Dwarf mangroves have a low stature, usually growin g only several feet high, and have tough, scleromorphic leaves, meaning that they are leathery and resistant to desiccation (Feller, 1995). This toughness is thoug ht to be a defense mechanism of the dwarf trees against herbivory, as is an increased c oncentration of defense compounds found in the leaves (McKee, 1995b). In addition, such tough leaves tend to be long-l ived and have chemical and physical properties that redu ce rates of decomposition, so they stay on the tree for a long time and may help retai n nutrients in plants that live in such conditions as phosphorus-limited soils (Feller, 199 6). A study by Feller (1996), done in a phosphorus-limited environment in Belize, confirmed that in dwarf R. mangle, leaf sclerophylly (or measured leaf toughness), is relat ed to nutrient deficiency. When these trees were fertilized with phosphorus, they showed a reduction in leaf lamina thickness due to the reduction of the hypodermis. The charact eristic expanded hypodermis found in nutrient-deficient R. mangle leaves was once thought to function for water stor age, osmoregulation, or salt accumulation as proposed by Camilleri and Ribi (1983). However, there is evidence against this notion as s imilar water content and sodium
22 concentrations were found among R. mangle plants growing in differing nutrient levels and having significant differences in leaf thicknes s between them, suggesting that leaf thickness is not related to salt accumulation or wa ter storage (Feller, 1995). Dwarf forests are large areas, typically in which only one species of mangrove is growing, and that are covered with trees that do no t reach the average height for that species (figure 1.5). R. mangle is the dominate species in dwarf forests in the so uthern half of Florida that cover almost 6,000 ha, some of which fringe the coastlines of southern Biscayne Bay and eastern Florida Bay (Davi s, et. al 2001). Dwarf forests are characterized by extremely low productivity, in ter ms of biomass, and several factors that have been associated with such growth in mangroves include salinity, saturation levels of the soil, and soil compactness (Davis, et.all, 2001 ). Nutrient availability is also believed to be an important factor in the formation of dwarf forests. It has been shown that slower growth rates, higher concentrations of carbon-based defense compounds, and lower rates of herbivory are characteristic of plant species ad apted to growing in resource-limited habitats (McKee, 1995b). Defense compounds are an indicator of the produc tivity of a species; plants with higher productivity tend to ha ve less defensive compounds in their tissue because they are energetically expensive to produce. There are various compounds that can protect the foliage from being eaten and m ost fall into one of three groups: terpenes, alkaloids, and phenolics (McKee, 1995b). Phenolic compounds vary in concentration within trees according to light and n itrogen availability. They protect mangroves through their ability to bind proteins; f or example they will bind insect salivary proteins to prevent the digestion of leave s (McKee, 1995b). Many wet, lowland tropical forest habitats are either nitrogen or pho sphorus-limited, and plants living in both
23 types of nutrient-limited environments have adapted similarly (Feller, 1996). Figure 1.5: Rhizophora mangle L. dwarf forest Located at Biscayne National Park, these trees are characterized by below-average height. (Photograph taken by D. Devlin and A. Shatt ers) 1.10 Rhizophora mangle L. in Biscayne Bay, including those with abnormal morphology The present study was conducted on abnormal and no rmal trees from within a dwarf forest. The R. mangle trees in which the abnormal morphology was manifes t were all tall relative to their surrounding dwarf mangro ve forest (figure 1.6). They occurred in almost round patches that stood out among the dwarf R. mangle trees that comprised the rest of the mangrove-covered area in this particula r part of Biscayne Bay National Park. The two types of trees grew right next to each othe r, branches almost touching, and yet
24 only the tall trees displayed the abnormal qualitie s. There was no visible difference in the dwarf mangroves growing on the proximity of the abn ormal patches verses those growing further away (data not shown). Figure 1.6 Example of observed abnormal morphology in height The R. mangle trees manifesting abnormal morphology at Biscayne National Park were all several feet taller than the surrounding dwarf trees, as shown in the tree on the right of the above photograph. (Photograph taken by D. Devli n and A. Shatters) Several patches of Rhizophora mangle L. have been found in the Biscayne Bay area that exhibit abnormal morphology including clu mped leaves (figure 1.7) and short, almost nonexistent internodes at the ends of branch es, and club-shaped branch termini (figure 1.8 and D. Devlin, unpublished observations ). Associated with these patches were large dead areas and heavy herbivory on the areas t hat were still alive. Often the
25 abnormal trees appeared mostly dead, with only one or two branches that had leaf clumps on the ends. The goal of this thesis was identifica tion of the potential cause of this morphology, suspected to be a pathogen because of t he nature of the symptoms. However, another possible explanation that was not a focus of the current study was a heterogeneous distribution of nutrients, such as ni trogen and phosphorus, throughout the site, which could have contributed to the abnormal morphology. Figure 1.7: Example of observed abnormal morphology in the leaves Abnormal morphology seen at Biscayne National Park. The leaves are growing such that they are tightly packed together on the ends of the branches. (Photograph taken by D. Devlin and A. Shatters)
26 Figure 1.8 Example of observed abnormal morphology in the branches Abnormal morphology seen in R. mangle at Biscayne Bay. Leaf scars are seen along with extremely short internodes and club-shaped branch t erminals. (Photograph taken by D. Devlin and A. Shatters) Mangroves are an irreplaceable part of many ecosys tems and many species depend on their survival, perhaps most of all Rhizophora mangle L. Therefore, it is a precautionary measure, and not premature, to begin pathogen detection studies before symptoms are widespread. The R. mangle at Biscayne Bay National Park might act as a
27 sentinel, cluing us in on what is to come in the fu ture. It is prudent to begin looking into a potential problem while it is still contained, sinc e once it is widespread any efforts to stop further spreading of a possible pathogen throughout the entire mangrove population will be many times harder, if not impossible. As will be described in detail in the next chapter in preparation for this study, samples were collected at the Biscayne National Par k site (figure 1.9), from both trees displaying the abnormal morphology mentioned above and control trees not displaying the abnormal morphology. DNA was isolated and teste d for the presence of pathogens of fungal, phytoplasmic, or other bacterial origin usi ng polymerase chain reaction (PCR) with universal primers (i.e., primers that are desi gned to amplify the most conserved DNA sequences from the genomes of specific microbia l groups of bacteria and fungi). A specific type of PCR (gradient PCR) was used since the optimal temperature for each reaction was not known, a priori. In gradient PCR, replicates of the same reaction are set up using identical DNA templates and primers and ea ch is run on a different temperature from within a specific range. When the PCR products are run on a gel and stained, the most intense band will correlate to the temperature at which that reaction works best, or the optimal temperature. In addition, primers for a positive control were s ought that could reliably amplify sequences in both abnormal and normal tissue. Both microsatellite primers and phenylalanine ammonia-lyase (PAL) gene primers were tested. Microsatellites are small sequence repeats (typically 2 base to ~6 base repeat sequences) of DNA that occur in the genomes of many organisms (Gupta and Varshney, 2000 ). They can be amplified with PCR by creating primers to the sequences flanking t he microsatellite region of the DNA
28 and are often used in population genetics studies b ecause they mutate rapidly, creating diversity even within populations of the same speci es (Gupta and Varshney, 2000). Previous work by Dr. D. Devlin resulted in the iden tification of microsatellite primers specifically for R. mangle but these primers had yet to be tested. Phenylal anine ammonia-lyase (PAL) is an enzyme that catalyzes the first step in phenylpropanoid metabolism, which is a series of biochemical reacti ons that provides the plant with essential phenolic compounds, like lignin (Weisshaa r and Jenkins, 1998). Primers made from the sequences of the PAL genes specific to R. mangle were made and also tested for use as a positive control, along with the universal primers used to test for the presence of a pathogen. Figure 1.9: Location where samples were collected The red box indicates the general area where the ab normal and normal samples were collected. The dwarf forest where the samples for t his study were collected is located just east, (to the right), of the cooling ducts, seen in this photo as long vertical lines, at the Turkey Point Nuclear Generating Station. ( 2010 Eu ropa Technologies; 2010 Google)
29 Chapter II: Procedures Note that since much of the methods involved troubl e shooting and modifications of procedures based on initial results, brief allus ions to certain results, which are described more thoroughly in the Results and Discus sion chapters, will be provided in this chapter to explain the logic behind the method ologies chosen. 2.1 Tissue collecting Samples were collected from R. mangle trees at Biscayne National Park located about 25 miles south of Miami near the Turkey Point Nuclear Generating Station (figure 1.9). For all samples collected, the branches were cut five centimeters above the terminal end and the cutting was stored in a plastic bag con taining a label, numbered 1 through 25. Twenty-five samples were taken from a patch, approx imately 20 feet in length, of Rhizophora mangle L. trees that showed characteristic abnormal morph ology (as shown in figures 1.6, 1.7 and 1.8). In these patches were trees several feet taller than the surrounding mangrove forest. Samples were taken fro m branches that had club-shaped ends, severe herbivory, clumped leaves, and/or dead or bare patches. Sampling within the patch was not completely random; trees were inspect ed for the above mentioned abnormal morphologies and when one or more was clea rly displayed on a tree, a sample was taken that included the abnormal part of that t ree. Twenty-five control samples were also taken from dwarf trees that did not have abnor mal morphology and which surrounded that patch and representative of the maj ority of the mangrove forest in the area. Again, sampling was not random throughout the entire dwarf forest, rather branches were cut from dwarf trees showing no abnormalities and surrounding the abnormal patch. A photograph of the branch that showed the abnormal ities and the GPS coordinates of the
30 branch were taken for each abnormal sample (data no t included). Samples were placed on wet ice and transported to the laboratory of Dr. D. Devlin at Harbor Branch affiliated with Florida Atlantic University in Ft. Pierce, Florida. The midribs were excised from the leaves in order to isolate the phloem, the tissue i n which many plant pathogens, including phytoplasmas, are located (see literature review in Introduction). 2.2 DNA extraction The midribs and stem tissue were ground, separatel y, in liquid nitrogen with a chilled mortar and pestle. The tissue was never al lowed to thaw completely, even after it was ground, to prevent oxidation and minimize degra dation. DNA was extracted using two methods, the DNeasy Plant Mini Kit from QIAGEN and the FastDNA kit from MP Biotechnology, as described in the protocols of the manufacturers. In brief, the former uses a QIAshredder spin column to remove cel l debris and homogenize the samples, then binds the DNA to a silica-gel membran e from which the DNA is finally eluted with water (DNeasy Plant Mini Kit, QIAGEN, manual). The second method requires the use of the FastPrep 24 instrument (M P Biomedicals) to homogenize the tissues after lysis and then the DNA is pelleted us ing a binding matrix and eluted in water (FastDNA kit, MP Biotechnology, manual). The DNA e xtracts were put on ice and DNA concentration was determined by spectrophotomet ric analysis using the NanoDrop 1000 Spectrophotometer (Thermo Scientific). Specifi cally, it analyzes 1.5 l of the eluted DNA, calculates the concentration of DNA (in ng/l), and indicates the quality of the DNA, (A260/A280), and the level of contaminates (A260/A230). These numbers were used to compare the two methods of DNA extraction, as will be discussed in the results chapter. The FastDNA kit consistently yielded the highest co ncentration (ng/l) and quality of
31 DNA and was therefore used in all subsequent extrac tions. All PCR reactions below, unless otherwise stated, were run using approximate ly 60-80 ng of this DNA per reaction. 2.3 Selection of a positive control Optimization of PCR conditions was initially accom plished through PCR amplification of R. mangle genomic DNA fragments. In early attempts to establ ish a positive control (i.e. a sequence that should relia bly amplify DNA from both abnormal and normal plants), temperature gradient PCR reacti ons were run with R. mangle microsatellite primer pairs previously identified b y Dr. D. Devlin (unpublished data) and labeled as Rm 11, 21, 36, 38, 46, 47. Each reaction mixture contained 17L Platinum Super Mix, including Taq polymerase (Invitrogen); 0 .5 l each of the forward and reverse primers diluted to a concentration of 10 pmoles/l with sterile water; and 2 l of the DNA template, originally extracted from an abnormal leaf midrib sample. The gradient thermocycler protocol used an activation at 94 C f or 2 minutes, then 40 cycles of the following: denaturation at 94 C for 30 seconds, an nealing at 50.0 55.0 C for 30 seconds, and extension at 72 C for 1 minute. The P CR products were electrophoresed on a 1.2% agarose gel and stained with ethidium bromid e. The results showed mixed success (shown and discussed in subsequent chapters). Several tests were run to work out the inconsisten cies with the microsatellite primers, for example the fact that there was amplif ication of microsatellite primer 46 during the first PCR reaction, but not during the s econd (see next chapter). The first run using R. mangle primer pair 46, which showed the darkest band out of all the microsatellite primers on the first gel, was run on a wider temperature gradient PCR, range 45-60 C, with an abnormal leaf midrib sample Microsatellite primer pair 46 was
32 also used with two different abnormal leaf midrib s amples and an abnormal stem sample at the annealing temperature where the most intense band was found in the successful reaction mentioned above, that was at 49.2 C. The gradient thermocycler protocol was the same as above, using the temperature gradient f or the annealing period of 45 60 C. The PCR products were run on a 1.5% agarose gel and stained with ethidium bromide. Unfortunately, primer pair 46 was deemed unreliable at this point. An additional test used an abnormal leaf midrib sample with microsatellite primer pairs 11, 21, 36, and 38, but inconsistent results with these R. mangle microsatellite primers led us to discontinue their use as positive controls in future reactions. In the continued search for a positive control, se quences for Rhizophora phenylalanine ammonia-lyase (PAL) genes 5, 6, 7, 8, and 10 (Cheesemen and Dassanayake, unpublished sequences) were R. mangle genomic sequences present in the National Center for Biotechnology Information (NCBI ) Genbank (accessed July, 2009). Because they belonged to the same gene family, the sequences were aligned so that a conserved region could be used for identification o f PCR primers. Using a conserved region allowed the greatest chance of the primers m atching exactly to all populations of R. mangle that may be studied in the future. The PAL sequenc es were aligned using a web-based molecular biology program suite, Biology Workbench 3.2 (San Diego Supercomputer Center, http://workbench.sdsc.edu/). Within this suite of programs, CLUSTAL W, a multiple sequence alignment tool, was used to create the PAL alignment. The alignment showed that the PAL gene 7 was the le ast conserved, so that particular gene was taken out of the alignment. Then the text shade function, which color codes the conserved regions of pre-aligned sequences, was use d to create a consensus sequence of
33 Rhizophora PAL genes 5, 6, 8, and 10. This conserved region w as entered into a webbased primer design program, Primer 3 (version 4.0; Rozen and Skaletsky, 2000; http://frodo.wi.mit.edu/primer3/), and primers were selected that amplified a 217 bp fragment of the PAL gene. The primers were then ord ered from Eurofins MWG Operon (see table 1 in Appendix for sequences) and ultimat ely, these primers were used as positive controls. To test whether or not some inconsistencies previo usly encountered arose from the amount (ng) of DNA used per reaction, another P CR assay was run with the PAL gene primers, this time using varying concentration s of the previously run DNA templates. The samples used were abnormal leaf samp les 1 (80.7 ng/reaction), 10 (80.9 ng/reaction), and 13 (78.9 and 34.6 ng/reaction); a nd control samples 10 (76.5 and 95.7 ng/reaction), 11 (76.1 ng/reaction) and 13 (76.5 ng /reaction). All were run at 46.3 C, chosen because it was the optimal temperature for t he PAL primers in the previous PCR reaction. Results will be given and discussed in th e next chapter. 2.4 PCR analysis to test for the presence of fungal and bacterial sequences Various plant DNA samples were used as templates i n PCR reactions containing several universal primer sets designed specifically for phytoplasma, bacterial, or fungal pathogens. The universal primers for phytoplasmas w ere P1/P7 and fU5/rU3 (Chen et al. 2009; see table 1 in Appendix for sequences). The u niversal primers for fungi used were NSA3/NLC2 (Martin and Rygiewicz, 2005), LSD/LS1 (Ha usner et al. 1993), and NS1/NS8 (White et al. 1990). The universal bacteria primers used were 10 F/480R (Sandstrom et al. 2001; Hansen et al. 2007). For all reactions, the PCR products were separated on a 1.2% agarose gel and stained with et hidium bromide.
34 2.5 Adaptation of a cotton genomic DNA isolation pr otocol for R. mangle tissue After encountering numerous inconsistencies with o ur early amplifications, and in order to extract DNA of a higher quality for use in further studies, two procedures originally designed for tissues with many oxidizabl e contaminants like tannins, which are abundant in R. mangle were modified specifically for R. mangle genomic DNA. The first was a protocol originally used for isolating cotton genomic DNA (Permingeat, 1998). The modifications made were to add polyethylene glycol (PEG) 4000 to the lysis buffer, to put the samples through the FastPrep 24 (MP Biome dicals) shaker, which lyses and homogenizes cells, the DNA spooling step was replac ed by a precipitation, a second precipitation with chloroform-isoamyl alcohol was a dded, and EB buffer (Qiagen) was used instead of the original buffer (for specific p rocedures see protocol 1 in the Appendix). The flowchart (figure 2.1) shows all of the modifications that were compared using normal R. mangle DNA (as is explained in depth in next paragraph). The variation that produced the cleanest DNA extracts is written out in protocol 1 in the Appendix. A procedure published by Fu et al (2004) for isolating RNA from mangroves in the same family (Rhizophoraceae) as Rhizophora mangle L. was also modified in multiple ways. The initial buffer in the original p rocedure was changed such that the Cell Lysis Solution (CLS-VF) plus Protein Precipitation Solution (PPS) buffers for plants was used from the FastDNA kit (MP Biotechnology), and the volumes used were according to manufacturer's directions. The second buffer use d was the lysis buffer from the the cotton DNA isolation protocol, and the FastPrep 2 4 (MP Biomedicals) shaker was used to homogenize the samples. This procedure (modified Fu et al. 2004) was followed, with the above modifications, for the first several step s of the protocol, then the second half of
35 the cotton DNA isolation procedure was followed (fo r specific steps see protocol 2 in the Appendix) since DNA, not RNA, was the desired produ ct. These procedures were tested first on normal Rhizophora mangle L. leaf midrib and stem tissue collected from Oslo Road, Ft. Pierc e, located near the lab. Samples from four branches were used. Each of five variations (f igure 2.1) were used with one sample from each branch, four samples total for each proce dure. Variation 1 used protocol 1 (Appendix), the tissue was ground in liquid nitroge n, and 1% PEG 4000 was added to the lysis buffer. Variation 2 used a modified protocol 1, in which the tissue was chopped with a razor blade and 1% PEG 4000 was added to the lysi s buffer. Variation 3 used a different modification of protocol 1 in which the tissue was ground in liquid nitrogen and PEG was not added to the lysis buffer. Variation 4 used yet another modification of protocol 1 such that the tissue was chopped with a razor blade and PEG was not added to the lysis buffer. Variation 5 began with protocol 2 and ended with steps from protocol 1. Two PCR reactions were run with each of these extr actions; one with 50 ng of DNA per reaction and one with 100 ng of DNA per rea ction. The second was used to test if the new DNA isolation procedures cleaned up some of the contaminants from the DNA that were suspected to interfere with the reaction when present at high concentrations (which should occur when high amounts of DNA were u sed) in earlier PCR runs. The PAL gene primers were used in all of the initial te st reactions. The PCR products were electrophoresed on a 1.5% agarose gel and stained w ith ethidium bromide. This gel contained more intense bands than the gel containin g the samples extracted with the FastDNA kit, MP Biotechnology. Because the bands f or the reactions that had 100 ng of DNA per reaction were of greater intensity than the ones for 50 ng of DNA, another
36 PCR reaction was run with 200 total ng of DNA and t he products were again run out on a 1.5% agarose gel. Because strong bands appeared, th e modified procedures for DNA isolation and 200 ng of DNA per reaction were deeme d successful and used in subsequent DNA extractions. 2.6 Application of modified cotton genomic DNA isol ation protocol for abnormal R. mangle tissue The modified isolation protocol 1 (Appendix), vari ation 2 (refer to figure 2.1), was used on the leaf midrib samples collected from Biscayne National Park, and both abnormal and control samples were used. A temperatu re gradient (45-60) PCR reaction was run using universal primers NSA3/NLC2 (Martin a nd Rygiewicz, 2005), LSD/LS1 (Hausner et al. 1993), and NS1/NS8 (Cheng et al 2004) for fungi; 10F/480R (Hansen et al 2007; Sandstrom et al 2001) for bacteria; P1/P7, and fU5/rU3 for phytop lasma (Chen et al 2009); and the PAL gene primers (Cheeseman and Da ssanayake, unpublished data). The products were run on a 1.5% agarose gel and sta ined with ethidium bromide. 2.7 Isolation and identification of amplified seque nces The differential bands, meaning the bands in which there was a difference between the abnormal and control tissues in either intensity or presence of band, from the above gel were excised. For two bands, labeled 1 an d 2, and both amplified using the phytoplasma primers fU5/rU3 (Chen et al 2009) and found only in the abnormal samples, the DNA was eluted from the agarose using the NucleoSpin Extract kit (Clontech, Mountain View, CA). The DNA was sent to the genomics lab, USDA Horticultural Research Laboratory in Ft. Pierce, Fl orida, to be sequenced using the original PCR-primers as sequencing primers. Using t he TOPO TA Cloning Kit
37 (Invitrogen), a portion of the purified DNA was als o inserted into a plasmid vector (pCR2.1-TOPO) and then transformed into One Shot Top 10 Chemically Competent E. coli following manufacturers recommended protocol (Invit rogen, Carlsbad, CA). The sequence of each purified product was put through the Basic Local Alignment Search Tool, BLAST (Altschul, 1990) search, which c ompares a given sequence to those in a publically available database and statisticall y calculates the similarity between sequences. Results will be presented and discussed in chapters 3 and 4, respectively.
38 Figure 2.1: Isolation variations Five variations on the isolation protocols used to extract DNA from R. mangle were used in an attempt to produce DNA with less contaminants than previously tested protocols. Variation 1 Variation 5 Variation 4 Variation 3 Variation 2 Excise 100 mg leaf midrib tissue Follow protocol 2 Chop finely with razor blade Grind in N2 Grind in N2 Chop finely with razor blade Go to step g) in protocol 1 and follow through to end Add 200 l lysis buffer with PEG Centrifuge at 16.1 xg 20 min, at 4 C. Discard supernatant Go to step c) in protocol 1 and follow through to end Add 200 l lysis buffer with PEG Add 200 l lysis buffer (no PEG) Add 200 l lysis buffer (no PEG)
39 Chapter III: Results The high concentrations of tannin in Rhizophora mangle L. makes extracting high quality DNA from the tissue difficult. For this rea son, two widely used extraction methods were initially compared; the DNeasy Plant Mini kit, QIAGEN, and the FastDNA kit, MP Biotechnology. The FastDNA kit gen erally produced extractions that contained more DNA, based on the readings made usin g the NanoDrop 1000 Spectrphotometer (see table 2 in Appendix). Thus, t hose extractions were used for both the leaf midrib and stem tissues in the PCR reactio ns. Unfortunately, the quality of the DNA, based on the A260/280 (see table 2 in Appendix) was poor for the extra ctions from both procedures and varied considerably between pre parations. Rhizophora mangle L. microsatellite primers were tested on a temperat ure gradient PCR to see if they would amplify one of th e previously extracted abnormal DNA templates, and also to determine the optimal temper ature. When the microsatellite primer pairs numbered 11, 21, 36, 38, 46, and 47 were used in the temperature gradient PCR, only 11, 21, 38, and 46 produced bands visible on a garose gels. The bands of greatest intensity (indicating the greatest amplification) w ere produced using microsatellite primer pairs 38 and 46, both at 50.0 C (figure 3.1, lanes 2, 8). Use of primer pairs 11 and 21 also resulted in bands at 50.0 C, but these were much l ess efficient (data not shown).
40 Figure 3.1: Results of PCR using Rhizophora microsa tellite primers Top Lanes: Lane 1, 20: D7058-Wide Range DNA markers; lanes 27: Rm 38; lanes 8-13: Rm 46; lanes 14-19: Rm 47. All primers were run wit h an abnormal leaf midrib DNA template. The temperature gradient went from 50.0 t o 55.0 C for each set of primers. Results: Bands were visible for Rm 38 and 46, both most int ense at 50.0 C, and less intense bands for the rest of the temperatures for both primers. There were no distinct bands for Rm 47, only smudges that were all too low in molecular weight and showed no variation between the temperatures. Therefore they most likely resulted from primer artifacts, which would explain why they are so far down on the gel (corresponding to smaller DNA fragments). Due to the initial, somewhat positive results obta ined using the microsatellite primers, which proved that the previously extracted DNA could be used for amplification, universal phytoplasma primers were then used in a t emperature gradient PCR reaction with the extracted DNA from abnormal samples. Since the specific pathogen is unknown, the annealing temperature gradient was used to help find the optimal temperature at which phytoplasma sequences, if any, would amplify. R. mangle microsatellite primer pair 46 was used as the positive control at 50.0 C because it produced the most intense
41 band from the first gel. Unfortunately no bands wer e produced on this gel (data not shown), not even in the positive control. Due to its failure to produce bands the second ti me it was used, primer pair 46 was chosen to run again, this time using a wider te mperature gradient PCR and with different types of tissue (figure 3.2). Bands were slightly visible for two different DNA samples extracted from abnormal leaf midribs, one w ith microsatellite primer pair 38 (lane 9) and one with primer pair 46 (lane 11). The re was no band for the DNA sample extracted from stem tissue, so the stem extractions were no longer used in the reactions. The phytoplasma primers were also run again in the wider temperature gradient PCR along with R. mangle microsatellite primer pairs Rm 11, 21, 36, and 38. The only band was produced by primer pair 38 (resul ts not shown), whereas previously use of the same microsatellite primers showed bands at approximately 50.0 C (figure 3.2, lanes 9,11). The R. mangle microsatellite primers were originally going to be used for the positive control in all of the PCR reaction s with the collected samples, but they could not be used because of the inconsistencies th ey showed. To recap, primer pair 46 resulted in amplification at 50.0 C (figure 3.1), was not amplified at all the second time it was run, then was sporadically amplified at 49.2 C (figure 3.2). The bands produced up to this point were all assigned a number, on a scal e 0-4 based on the intensity of the band, which was compared to the quality of the DNA, using the A260/280 reading from the NanoDrop spectrophotometer (see table 2 in Appendix ). There was no correlation found between the visual quality of the amplified band an d the quality of the DNA template used for the amplification as assessed by the NanoD rop spectrophotometer (see table 1 in Appendix). Because reliable amplification of a R. mangle genomic DNA fragment is
42 necessary to show reaction conditions are conducive to PCR prior to the search for potential pathogen DNA, it was decided that the R. mangle microsatellite primers were not going to be used as controls and new positive c ontrol primers needed to be found. Figure 3.2: Results of temperature gradient PCR usi ng microsatellite primer 46 lm = leaf midrib, s = stem, blue boxes indicate the presence of a band Lanes: Lane 1, 14: D7058-Wide Range DNA markers; lanes 27: Rm 46; lane 8: empty; lane 9: Rm 38 used as positive control; lane 10: Rm 46 negative control; lane 11-13: Rm 46 positive controls; lane 15-20: empty. The temper ature gradient for reactions in lanes 27 was 45-60 C. All of the positive controls were r un at 49.2 C. The reactions in lanes 27 and 9 were run with an abnormal leaf midrib DNA t emplate. The negative control was run with sterile water instead of a DNA template. T he product in lane 11 was run with a different abnormal leaf midrib DNA template, and in lane 13 with a third abnormal leaf midrib template. The product in lane 12 was run wit h an abnormal stem sample. Results: The was a band associated with Rm 38, used as a po sitive control, and for (LM) Rm 46 at 49.2 there was only one reaction that prod uced a band. There was no band associated with the stem DNA template, temperature gradient, or the negative control.
43 The only other R. mangle genomic DNA sequences present in public databases were a partial genomic sequence of five phenylalani ne amonia lyase (PAL) genes. When primers designed to the PAL genes were run along wi th the universal fungal and bacterial primers on two different DNA templates both extract ed from abnormal leaf midribs, there were mixed results. The PAL primers produced modest to faint bands at all of the temperatures, but only for one of the DNA templates (figure 3.3). The more visible bands were at 45.0 C (lane 2, top), 46.3 C (lane 3, top ), and 59.7 C (lane 7, top). What was unexpected was the lack of bands produced using the bacterial primers (lanes 8-11, top and 2-4, bottom) because bacteria are so abundant i n mangroves that the detection of at least a few in the extracted DNA was expected, alth ough they were not necessarily expected to be the causal pathogen. This result may also have been an indication that the quality of the DNA extracted was interfering with t he PCR reactions. The PAL primers were then used in PCR reactions co ntaining different DNA templates and varying amounts of DNA per reaction ( figure 3.4). There were only three DNA templates with associated bands. Two were for D NA template numbered 10, one with approximately 77 ng of DNA per reaction (lane 6), and the other with around 81 ng of DNA per reaction (lane 3). The third visible ban d was for DNA template numbered 13 with around 77 ng of DNA per reaction (lane 9). As in DNA template 10 (c.f. lanes 6 and 3, figure 3.4), the highest concentration of DNA te mplate did not yield any products. This seemed to further indicate that there were too many contaminants in the extracts and that when higher amounts of the DNA template (and thus l ikely also associated contaminants) were added to the PCR reactions, complete inhibitio n of PCR often occurred. The conclusion was made that cleaner extractions were n eeded.
44 Figure 3.3: Results of PCR using R. mangle PAL primers red dots = expected size for amplicon using bacteri al primers, green dot = expected size for amplicon using fungal primers, NC = negative co ntrol, blue box indicates the presence of bands Top Lanes: Lane1: D7058-Wide Range DNA markers; lanes 2-7 PAL primers; lanes 810: 10F/480R bacterial primers, lane 11: PAL negati ve control. All primers were run with an abnormal leaf midrib DNA template on the tempera ture gradient from 45 to 60 C. Sterile water was used instead of a DNA template fo r the negative control. Bottom Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-4: 1 0F/480R bacterial primers; lanes 5-10: NSA3/NLC2 fungal pri mers; lane 11: 10F/480R negative control. All primers were run with an abnormal leaf midrib DNA template on the temperature gradient from 45 to 60 C. Sterile wate r was used instead of a DNA template for the negative control. Results: There were bands for the PAL primers at all the te mperatures, the more visible were at 46.3 C, 45.0 C, and 59.7 C.
45 Figure 3.4: DNA quality test red dot = expected size for PAL primers, NC = negat ive control, blue boxes indicate the presence of a band Lanes: Lane 1: D7058-Wide Range DNA markers; lane 2: abnor mal leaf midrib DNA template 1 (81 ng DNA per reaction); lane 3: abnorm al leaf midrib DNA template 10 (81 ng DNA); lane 4: abnormal leaf midrib DNA template 13 (79 ng DNA); lane 5: abnormal leaf midrib DNA template 13 (96 ng DNA); lane 6: ab normal leaf midrib DNA template 10 (77 ng DNA); lane 7: abnormal leaf midrib DNA te mplate 10 (96 ng DNA); lane 8: abnormal leaf midrib DNA template 11 (77 ng DNA); l ane 9: abnormal leaf midrib DNA template 13 (77 ng DNA); lane 10: sterile water (77 ng DNA). PAL gene primers were used in all of the reactions, which were all run at 46.3 C. Results: There were only three bands visible, associated wi th template 10 (~77 and ~81 ng DNA) and template 13 (77 ng DNA). A new DNA preparation procedure (referred to as the 'rigorous isolation procedure' in the figure legends) was tested. This one was originally designed to work with cotton plant samples, which are known to have high levels of PCR inhibiting
46 compounds. Using the modified cotton DNA isolation protocol (see protocol 1 in Appendix) as the basic procedures, several modifica tions were compared and tested on normal R. mangle tissue collected at a site near the lab. It was fo und that grinding of the tissue in liquid nitrogen before the extraction pro cedure, a routine method used because it is thought to provide the best homogenization of ti ssue, was not necessary since the bands corresponding to the sequences amplified usin g samples that were ground in liquid nitrogen were equally as intense as bands correspon ding to samples that were finely chopped with a razor blade (e.g., c.f. lanes 2,3, t op and 2,3, bottom in figure 3.6). The addition of PEG 4000 was also tested because it is known to sequester PCR inhibiting compounds. When PEG 4000 was present in the lysis b uffer (at 1% weight/volume) it did result in more intensely stained PCR products in ag arose gels (e.g., c.f. lanes 4 and 9, top in figure 3.6). Using the cotton protocol with the addition of PEG 4000, amplification was more efficient (as determined by agarose gel DN A band intensity), than with any of the same PCR reactions using DNA isolated using the FAST prep kit. The amount of DNA template that could be used was greater as well further supporting the thought that the new DNA isolation procedure is more efficient a t removing PCR inhibitors. Agarose gel band intensity for leaf midrib tissue continued to increase with increasing amounts of DNA template added to the rea ctions with up to 200 ng of DNA per reaction (figures 3.5, 3.6, and 3.7). The amoun t of DNA from the pathogen that must be detected is quite small in comparison to the tot al amount of plant DNA, so the more DNA that can be used per reaction the better the ch ances of the primers amplifying any pathogen DNA. The results indicate that the new iso lation protocol separated most of the contaminants from the DNA so that the PCR reaction was able to run unhindered.
47 Figure 3.5: Results of PCR using PAL primers and th e rigorous isolation protocol with 50 ng of DNA LM = leaf midrib, S = stem, v(1-5) = number of vari ation of isolation protocol (figure 2.1) Top Lanes: Lane 1: D7058-Wide Range DNA markers lanes 2-4: le af midrib templates extracted with variation 2; lane 5: stem template e xtracted with variation 2; lane 6: empty; lanes 7-9: leaf midrib templates extracted w ith variation 4; lane 10: stem template extracted with variation 4; lane 11: empty. All rea ctions were run at 46.3 C with 50 ng/l of DNA and using the PAL gene primers. Bottom Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-3 le af midrib templates extracted with variation 1; lane 4: stem template extracted with variation 1; lane 5: empty; lane 6-7: leaf midrib templates extr acted with variation 3; lane 8: stem template extracted with variation 3; lane 9: empty; lane 10-11 leaf midrib templates extracted with variation 5; lane 12: stem template extracted with variation 5. All reactions were run at 46.3 C with 50 ng/ l of DNA and using the PAL gene primers. Results: There were bands associated with all the leaf midr ib templates that were not ground in liquid nitrogen, and most of them that we re.
48 Figure 3.6: Results of PCR using PAL primers and th e rigorous isolation protocol with 100 ng of DNA LM = leaf midrib, S = stem, v(1-5) = number of vari ation of isolation protocol (figure 2.1) Top Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-4: m idrib templates extracted with variation 2; lane 5: stem template e xtracted with variation 2; lane 6: empty; lanes 7-9: leaf midrib templates extracted w ith variation 4; lane 10: stem template extracted with variation 4; lane 11: empty. All rea ctions were run at 46.3 C with 100 ng/l of DNA and using the PAL gene primers. Bottom Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-3: l eaf midrib templates extracted with variation 1; lane 4: stem template 1; lane 5: empty; lane 6-7: leaf midrib templates extracted with variation 3; lane 8 : stem template extracted with variation 3; lane 9: empty; lane 10-11 leaf midrib templates extracted with variation 5; lane 12: stem template extracted with variation 5. All reactions were run at 46.3 C with 100 ng/l of DNA and using the PAL gene primers. Results: There were bands associated with all the leaf midr ib templates that were not ground in liquid nitrogen, and most of them that we re.
49 Figure 3.7: Results of PCR using PAL primers and th e rigorous isolation protocol with 200 ng of DNA LM = leaf midrib, S = stem, v(1-5) = number of vari ation of isolation protocol (figure 2.1) Lanes: Lane 1: D7058-Wide Range DNA markers; lane 2: leaf midrib template; lanes 34: PEG leaf midrib templates; lanes 5-6: RNA leaf m idrib templates; lane 7: negative control; lane 8: positive control with 100 ng of PE G leaf midrib sample. All reactions were run at 46.3 C with 200 ng/l of DNA and using the PAL gene primers. Results: Strong bands were produced for all of the template s, except the negative control. When universal primers for fungi, bacteria, and phy toplasma were used in conjunction with the R. mangle DNA isolated from the Biscayne samples, both abnor mal and normal samples produced PCR products that were visible on agarose gels (figures 3.8 and 3.9). There were many products for all of the u niversal primers, and the PAL genes produced intense bands. There were many differentia l bands when comparing the abnormal samples to the control samples. Some of th e bands were present in both the abnormal and control samples, but more intense in o ne or the other representing quantitative differences. Some bands were present i n either the abnormal or control samples, and absent in the other, representing qual itative differences. Some of the PCR products visible on the agarose ge ls migrated at or close to the
50 proper distance as would be expected given the spec ific primer pair used (indicated in figure 3.8 and 3.9 by red dots). All PCR products w ere compared to the D7058-Wide Range DNA markers (Sigma Aldrich, St. Louis, MO). I t is possible that some of the sequences we obtained, because they are of unknown origin, were not the expected size due to deletions or insertions that may have occurr ed during the pathogen's evolution. The bacterial primers 10F/480R produced many PCR p roducts which differed both qualitatively (c.f. lanes 4 and 10, top) and q uantitatively (c.f. lanes 2 and 8, top) depending on the annealing temperature for both the abnormal and control samples. As is indicated by the red dot (figure 3.8), the expected size of the bacterial fragment is 2502 bp (table 1; Sandstrom et al. 2001; Hansen et al. 2007). There are PCR products of this size for both the normal and abnormal samples for a ll temperatures. For annealing temperatures of 53.9 C and above, this is the only band visible. There were two sets of phytoplasma primers used (f igure 3.8). For the phytoplasma primer pair P1/P7, the only PCR product s of the expected size (1800 bp; Chen et al 2009)) were at 47.5 C (lane 5, middle) for a nor mal control DNA template, and at 49.2 C (lane 6, middle) for an abnormal DNA template. None of the products obtained using the P1/P7 primers were specific to e ither the abnormal or control samples. The second primer pair, fU5/rU3, only produced PCR products at low annealing temperatures, and none were the expected size (882 bp; Chen et al. 2009). However, there were two bands of interest (lanes 2, 4, and 6 bottom, boxes A and B) produced by the fU5/rU3 primer set exclusively in abnormal samp les and not in the control.
51 Figure 3.8: Universal primer run with bacterial and phytoplasma primers red dot = expected size for the respective primer, A = abnormal DNA template, C = control DNA template, blue boxes indicate the prese nce of a band Top Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-13: 10F/480R bacterial primers; lane 14: empty; lane 15: abnormal PAL posi tive control; lane 16: control PAL positive control. An abnormal leaf template was use d with the 10F/480R primers in the even lanes, and in the odd lanes a control leaf mid rib DNA template was used. The temperature gradient used was from 45-60 C. Middle Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-13: P1/P7 phytoplasma primers; lanes 14-16: empty. An abnorma l leaf template was used with the
52 P1/P7 primers in the even lanes, and in the odd lan es a control leaf midrib DNA template was used. The temperature gradient used was from 45 -60 C. Bottom Lanes: Lane 1: D7058-Wide Range DNA markers lanes 2-13: f U5/rU3 phytoplasma primers; lanes 14-17: empty. An abnorma l leaf template was used with the fU5/rU3 primers in the even lanes, and in the odd l anes a control leaf midrib DNA template was used. The temperature gradient used wa s from 4560 C. Results: There were many bands for all of the primers, seve ral of expected size. Of most interest were the differential bands displaying bot h qualitative and quantitative differences between the abnormal and control templa tes. Three primer pairs were used to amplify fungal pat hogens (figure 3.9). Primer pair NS1/NS8 produced a PCR product the correct size, bu t it was only in the control samples at low annealing temperatures (lanes 3 and 5 in fig ure 3.9, middle). NSA3/NCL2 and LSD/LS1 primer pairs both produced PCR products of the correct size, 1379 (Martin and Rygiewicz, 2005) and 1475 (Hausner et al 1993) base pairs respectively, at all annealing temperatures. However, the 1379 bp band produced f rom the NSA3/NCL2 primer pair appears to be a doublet in the abnormal samples (e. g. lane 4, top) while only a single band in the control samples (e.g. lane 3, top) at a nnealing temperatures below 50 o C.
53 Figure 3.9: Universal primer run with fungal primer s red dot = expected size for the respective primer, A = abnormal DNA template, C = control DNA template, blue boxes indicate the prese nce of a band Top Lanes: Lane 1: D7058-Wide Range DNA markers; lanes 2-13: NSA3/NCL2 fungal primers. An abnormal leaf template was used with th e NSA3/NLC2 primers in the odd lanes, and in the even lanes a control leaf midrib DNA template was used. The temperature gradient used was from 45-60 C. Middle Lanes: Lane 2: D7058-Wide Range DNA markers, lanes 2-13: LSD/LS1 fungal primers. An abnormal leaf template was used with th e LSD/LS1 primers in the odd lanes, and in the even lanes a control leaf midrib DNA tem plate was used. The temperature gradient used was from 45-60 C. Bottom Lanes: Lane 1:D7058-Wide Range DNA markers; lanes 2-13: N S1/NS8 fungal
54 primers. An abnormal leaf template was used with th e NS1/NS8 primers in the odd lanes, and in the even lanes a control leaf midrib DNA tem plate was used. The temperature gradient used was from 45-60 C. Results: There were many bands for all of the primers. Of m ost interest were the differential bands displaying both qualitative and quantitative differences between the abnormal and control templates. Of all the bands that were excised from the univers al primer runs, two of the bands, labeled A and B (lanes 2, 4, and 6, bottom i n figure 3.8) that showed the most potential of being a causal agent were associated w ith the fU5/rU3 phytoplasma primers. Since they were well defined bands that were presen t only in the abnormal tissues, they were sequenced and put through BLAST (Altschul, 199 0). The resulting sequences are shown in figures 3.10 and 3.11. Upon BLAST searching, there were several matches f or the sequenced bands. The closest match to band B (figure 3.11) wa s an putative ubiquitin-protein (gene ID: 8265613 RCOM 0985450) from P opulus trichocarpa or California poplar, a deciduous tree native to North America. There were two close matches for band A. The first was to a gene from Medicago truncatula, a legume. The second was to a Glycine max (soybean) retrotransposon. These results will be d iscussed in the next chapter.
55 GAGGAAACTAATAAAAGAAGGAAGAATTTTGATACTTTGAACAAATGACAA G TCTATGGTATTTATAGATGTGATAATGATTATTTGTCAATAGATACTTTTG GTTTG GTGGCTAACAACTTGACCTTTTAACTTTACTTTATTTTTTGTCATTTATTA TGAAA TGATATATAAAATAGTGTTAATCCTTTTGTTAAAGGATCAGCCATGTTTTG ACTT GATCTTACAAAATTAATTATGATCACTCCATTTGTGATCAAGTCTCTCACA TAGT TATATCTAAGTCCAATGTATTTGGACTTACCATTATATACTTAACTATACA ATCTT GCTAGAGTTGCAGCACTATCACAATATATATAAATAGGTGCTAGTGGCTTT TGCC ACAAAGGAATGTTATAAAGCAAATTCCTTAACCATTTTACTTCTTTACTTG TTG AAGCTAATGTTAC Figure 3.10: BLAST results for band A DNA sequence of differential band A produced with a bnormal leaf midrib DNA and phytoplasma primers fU5/rU3. GGTTTAATGTCGCCTCGAGCCCCCTCTGACCTTGACAGAATAGCAAGTGCA CC TAGAGCTTCAGATATGGCATTAGATCCATCATTTGCAGATTCTAAAAGAGA AAC TAATGCAAGGACTGTTCCAGCACGATTCACACAATCTGTCACAGCATTGTC AA TTCTGCGGGAATGAAGCAGACGACCAATTGCAGCTGCTGCATGTGTCTTTC CA GAAACTGTTCCTTCACGAAGAACTCTCGTTGCAGGCAGAATAATTTCCTCA GG TATTGCTTTCTCTGAAGCTTCACTGTCCAAAATAAGATTTGCCAGAGCACA TGT TGCCTGTTCAGCGACTTCTAGATTTAGAGAGTTAGCGAGTGTAACTAAAGG AG GATATGCATCTCGAGCAATAGCGGCCACATCTCTATTATCCTTAATTGAGA GAA ATAGTGCAGCAAGACAATGAGATGTCTCTAACAAAATGTTTTCAGATTCGA CA TTTAGCAATTTCATCACTGACCAGAAAGTTTTTAAAGCAATCCCACTTTCA CGC AAATCCTTCCTGACTTCAAAAATTCCAGCCAAAGCCGAGGCAGATTTTGCT TG AGTCTCCACTTTGGTAGAGCTCAATAGTTTAATCATGGTCTCCATTGCATC ATTT GCTGCACTGCCTTCCCGTAATATATCACTCAACGGAACCACAGAGAGCATG CT TCTTAAAGCATCTAGAACGTATATTTTTGATTCAGGCAAGTCACTGGTTAA TAAT GCTGTCAGCTGGCTGATAGTTGCCGTATCAGATTTATGGATCAAATGGTTT AAA GTCTTCGCTGCAATTTCTTTTCCATTAGGGCTCCCATTCTTCAATAGCCAT AACA ATGCTGGAACAGCATCAGCACTTTCAACACATGCACGTATGTCTTCACTGT GAT TGCATAAGTTCCGAAGGATTGTTGCAGAATCCTCCTTGGCTTTTGCAGATC CTG TCTCTAGTATCTGAACCAGTGGGGGTATGCCACCAGCAGCAGTAATGGCCC AC TTGCTTTCATCATTTTCATTAGACAGAAGGCAAAGCAAAGCAACTGCACAT TC TTGCTGTTGTTCTGATGACAGTCCAAGAAGAGATATCAACAACTGAACACC CT CACGGCCTGAAG Figure 3.11 BLAST results for band B DNA sequence of differential band B produced with a bnormal leaf midrib DNA and phytoplasma primers fU5/rU3.
56 Chapter IV: Discussion 4.1 Development of new methods was necessary for is olation of high-quality, PCRamplifiable DNA from R. mangle Standard kits for convenient and rapid DNA isolati on are available from numerous biotechnology companies, with two of the m ost common ones being DNeasy Plant Mini Kit (QIAGEN) and the FastDNA kit (MP Bi otechnology). Although these kits are designed to be suitable for use on tissues from many biological sources, our data show that they are not suitable for isolating PCR-a mplifiable DNA from Rhizophora mangle L. This is not an uncommon problem with many plant tissues because of secondary metabolites (i.e., tannins, phenolics, ex cessive carbohydrates and other PCR inhibiting molecules). We have shown that by modify ing a procedure originally developed for cotton leaves (which is another type of recalcitrant tissue due to high levels of phenolic terpenoids and tannins present in the c ells; Maliyakal, 1992), with the addition of PEG 4000, we can obtain exceptionally h igh quality R. mangle DNA that efficiently serves as a PCR template. We also neede d to develop a suitable set of primers for amplification of R. mangle DNA that could be used as a positive control in te sting the PCR efficiency of different preparations of R. mangle DNA. Our first attempt was to use R. mangle microsatellite primers as the positive control. Un fortunately, testing of these primers for use as PCR positive controls showed tha t they were inefficiently and inconsistently amplified. This led to the use of th e R. mangle genome sequences present in Genbank. These were partial gene sequences for f ive different phenylalanine ammonia lyase (PAL) genes. Alignment of these sequences led to the identification of a primer set that functioned reliably as a positive control due to the development of the modified
57 cotton DNA isolation protocol. 4.2 Differences were found, using DNA profiling, be tween R. mangle samples taken from abnormal and control trees growing in Biscayne National Park Universal primers, which are designed to amplify t he most conserved DNA sequences from the genomes of specific microbial gr oups of bacteria and fungi, were used in the attempt to identify a pathogen within t he abnormal R. mangle tissue taken from trees growing in Biscayne National Park that d isplaed disease-like symptoms. The use of universal primers was necessary in this stud y because, if a pathogen is present, its particular type is unknown so specific primers cann ot be used. Six sets of universal primers were tested in this study; three for fungi, one for bacteria, and two for phytoplasmas (mollecute-type bacteria). These were chosen because the abnormal morphologies in some ways similar to those observed at Biscayne National Park have been associated with these types of pathogens in kn own diseases, as discussed in the Introduction. The first PCR runs with any of the universal prime rs produced no bands. We showed that early DNA preparations using either DNe asy or FAST prep commercial kits likely had too many contaminants that inhibited the PCR reaction. When the more rigorous isolation protocol, specifically using met hods that remove excess phenolic compounds (protocol 1), was developed and the DNA w as used along with the PAL primers, PCR products were produced that were more intense than any produced with other protocols. Also, successful PCR reactions occ urred within a larger temperature range, indicating the superiority of the new isolat ion protocol over the commercial kits
58 for extracting R. mangle DNA. When protocol 1 was used for DNA isolation, amplif ication of R. mangle DNA with the universal primers resulted in differences in the PCR products produced between the trees with abnormal morphology and the trees wi thout (figures 3.8 and 3.9). Our hypothesis was that if a pathogen is the cause for the abnormal morphology, unique microbial PCR bands should be produced in DNA isola ted from the abnormal trees. Bands A and B in figure 3.8 produced using phytopla sma primers fU5/rU3 were both present only in the abnormal tissues and thus were considered the most promising possibility of being from the putative causal agent for the observed abnormal morphology at Biscayne National Park. However, they were not t he expected size and they were only amplified at low annealing temperatures which are f actors that indicate they may not be from a pathogen. They, in fact, are most likely non -specific plant DNA amplification products, as explained below. For band B in figure 3.8, sequence analysis showed that its sequence was closely matched to an uncharacterized protein from Populus trichocarpa a plant protein. It is unlikely that the DNA has anything to do with a pat hogen. For band A in figure 3.9, there were two close matches for the sequence but the fir st match, to a gene from Medicago truncatula, a legume, was less intriguing than the second. The second match, which is of more interest, was to a Glycine max (soybean) retrotransposon. Retrotransposons are sequences of DNA that act as jumping genes, in that they move from one place to another within the genome (Pray, 2008). Retrotransposons ca n become activated when under stressful conditions, such as severe changes in car bon levels, temperature, and UV light (Ilves, 2001). It is possible this could be happeni ng in the abnormal plants and not in the
59 control plants, because the bands coresponding to t his gene were only present in the abnormal plants (figure 3.8). This could possibly b e the result of a transposition event that fortuitously created primer binding sites clos e enough to allow PCR amplification as a result of the DNA rearrangement. The sites are m ost likely not an exact match to the primers since they only amplify at the lowest annea ling temperatures. This result could indicate a stressor in the environment present in o nly the abnormal plants, which could be correlated with the observed abnormal morphology. T o pursue this possibility, DNA would have to be isolated from more replicates of a bnormal and control plants from Biscayne National Park with similar results as abov e. Transpositional elements were first found in maize and have since been associated with unstable mutant phenotypes in many plants, such as soybean (Feschotte, et al. 2002). Two retrotransposons have been identified i n tobacco, the Tnt1A and Tto1 elements. Tnt1A element is activated by wounding and viral, bacter ial, and fungal attacks; Tto1 element is similarly induced by viral attacks, wou nding, and salicylic acid (Grandbastien, 1998). These retrotransposons are ac tivated by both biotic and abiotic factors that normally elicit the activation of defe nse responses in plants. Transposable elements provide much genetic variation and, becaus e they are induced by stress, it is speculated they play a role in an organisms ability to adapt to environmental changes, however no proof has been provided thus far (Grandb astien, 1998). Taking this into account, however, presents a case that a retrotrans poson was amplified from the abnormal R. mangle tissue and not the control tissue because it was a ctivated in response to an environmental stressor that may be the cause of the abnormal morphologies observed in the Biscayne area. However, it is also possible tha t the unknown plant protein (the other
60 match) is an uncharacteristic defense protein activ ated by stress conditions within the abnormal plants. There was also a notable difference in PCR fragmen ts between reactions using the abnormal versus the normal tree DNA with the fungal primer pair NSA3/NCL2. This difference is visible at the expected size for this primer pair at lower annealing temperatures. The bands for the abnormal DNA templa tes are a doublet while there is only a single band for the control DNA templates at the lower temperatures. Further characterization of these bands through cloning and sequencing the PCR products would reveal more about the identity of these bands. Many of the bands produced by the bacterial and fu ngal primers were similar between the abnormal and normal control DNA templat es. Since the primers will produce a PCR product of the same size for a broad range of microbes that may be present, it is quite possible that the presence of pathogen-specif ic PCR products is masked by the presence of non-pathogenic microbial PCR products o f the same size. These can only be distinguished by characterization of multiple clone s isolated from the purified band seen on the agarose gels. This was beyond the scope of my project, but is being persued by Dr. D. Devlins laboratory. 4.3 Future studies If this study were to be continued, now that a str ong positive control has been established and an isolation protocol has been foun d that produces extracts with little contaminants, more samples would have to be collect ed from different areas of the mangrove forest located at Biscayne National Park i n order to compare abnormal trees from different areas throughout the dwarf forest. T here were also many other differential
61 bands that showed differences between the abnormal and control tissues for all of the bacterial, fungal, and phytoplasma primers (figures 3.8 and 3.9) that were excised that can be sequenced and put through BLAST to find any possible relations to any known pathogens. There are several tests that could be conducted to further the knowledge about the abnormal R. mangle observed at the park. Sediment cores could be take n from the areas where the normal dwarf trees were growing and the a reas where the abnormal tress were growing. Comparing the levels of nutrients availabl e to both types of trees might provide a reason for the differences in growth patterns and degree of herbivory. Setting up insect traps on and around the trees would provide informa tion about what is causing the herbivory seen on the leaves of the R. mangle plants, which could be compared between the abnormal and normal trees. Insects are known ve ctors of pathogens (Hogenhout et al. 2008), and different insects carry different pathog ens, so the type of insects found could give a clue as to what type of pathogen, if any, ex ists within the plants. A study done on Sparina alterniflora cord grass that grows in salt marshes, by Sullivan and Daiber (1974) showed that plants recei ving nitrogen fertilizer yielded three times as much (measured in terms of fresh weight) a s plants receiving no fertilizer. The cord grass that was receiving the nitrogen fertiliz er grew more densely and was a darker green than the control plants. This situation of di fferent morphologies within the same environment is similar to that observed with the R. mangle trees at Biscayne National Park. The trees with the abnormal morphology were a ll several feet taller and those that weren't dead had a greater amount of foliage than t he control plants surrounding them (unpublished observations, D. Devlin). This could s uggest a possible inequality in the
62 distribution of nutrients, like nitrogen or phospho rous, throughout the mangrove forest that could be causing the difference between the ab normal and control plants. The increase in foliage and nutrient uptake of the abno rmal trees indicates higher levels of productivity, which may be correlated to a decrease in the amount of defense compounds stored in the leaves (Feller, 1996). Such a decreas e in defense compounds makes the plant more susceptible to attacks by bacteria and f ungi. These trees also tend to suffer from heavier herbivory, perhaps because a decrease in defense compounds may be making the leaves more palatable, and many of the i nsects that feed on plants serve as vectors for pathogens. In short, an uneven distribu tion of nutrients throughout the mangrove forest could make some trees more suscepti ble to pathogens than others, which would explain why only some of the trees at Biscayn e National Park displayed abnormal morphology while others appeared normal. As described in the introduction, much of today's mangrove forests are surrounded by both urban sprawl and agricultural la nd and thus are subject to runoff and wastes, often including pollutants such as heavy me tals and nitrogen fertilizers (MacFarlane and Burchett, 2002). The mangrove fores t at Biscayne National Park grows right along the miles of long, extremely deep cooli ng channels for the water that comes out of Turkey Point nuclear generating system. The water comes out of the power plant extremely hot and much of it evaporates quickly, wh ich concentrates solutes in the water. If there were toxins, such as metals, in the water within the cooling channels, it is possible they could be seeping over to the dwarf ma ngrove forest. Mangroves do possess a tolerance to increased levels of heavy metal poll ution, but the accumulation of metals such as copper, zinc, and lead in the tissues can c ause growth inhibition and increased
63 mortality (MacFarlane and Burchett, 2002). Pesticid es affect mangroves similarly, often causing species-specific diebacks when present in h igh concentrations within an ecosystem (Duke et al. 2001). It is possible that a toxin or pesticide co uld be the cause of the abnormal morphology at Biscayne Bay, but most o ften these are associated with large regions of dieback, not differing morphology. 4.4 Conclusions Mangroves serve a fundamental role in estuarine eco systems throughout the world, providing many ecoservices, like a nursery f or juvenile fish, that are not available from any other species. Because they are so importa nt to their environment, mangroves need to be protected. When populations are threaten ed by an unknown agent, as is the case in Biscayne National Park, finding the cause i s essential to protecting that population. A procedure was developed in this thesi s that allows clean DNA to be extracted from the midrib of the leaves of R. mangle Once clean DNA was extracted that did not contain PCR-inhibiting contaminants, univer sal primers were used to test for the presence of a pathogen. Definitive results were not obtained, but, through the development of a DNA isolation protocol for R. mang le that consistently produced high quality extracts, the way was paved for future stud ies not only related to potential pathogens, but also to other numerous aspects of ma ngrove genetics.
64 Appendix Protocol 1: Modification of Permingeat et al (1998) Cotton Genomic DNA Isolation Procedure RT = room temperature a) Excise 100 mg leaf midrib tissue, chop finely wi th a razor blade, add to 2 ml tube with 500 l cold extraction buffer. b) Centrifuge at 16.1g 20 min at 4 C, discard supe rnatant. c) Add 200 l lysis buffer (to which 1% PEG 4000 has already be en added) and beads. Put in FastPrep 24 homogenizer for 40 seconds at g rade 6. d) Incubate in 65 C water bath for 30 min. Agitate tubes every 10 min. e) Add 240 l chloroform:isoamyl alcohol (24:1) and mix gently, inverting tube, until an emulsion forms. Centrifuge at 16.1g for 20 min at 4 C. f) Carefully transfer aqueous phase (top phase) to a clear 1.5 ml microfuge tube. Add 240 l chloroform:isoamyl alcohol (24:1) and shake sampl e vigorously. Centrifuge at 12,000g for 8 min. g) Carefully transfer aqueous phase (top phase) to a clear 1.5 ml microfuge tube. Add equal volume of cold isopropanol; mix gently and let stand at room temperature for at least 1 hour. h) Centrifuge at 16.1g for 10 min (RT). Decant supe rnatant and let tubes air dry by placing them on the side on a paper towel (alternat ively, tubes can be put in fast vac for 5 min). i) Add 200 l cold 76% ethanol/0.2 M Na Acetate. Place at 4 C for 1 hour to overnight (good stopping point).
65 j) Centrifuge tubes at 16.1g for 10 min. Carefully discard supernatant, rinse the pellet with ~100 l cold 70% ethanol. Let tubes air dry by placing th em on the side on a paper towel (can be put in fast vac for 5 min). k) Resuspend pellets in 100 l EB buffer using the force of a 1000 l pipette. Incubate at 65 C for 15 min. l) Centrifuge at 16.1g for 10 min (15C). Transfer supernatant to new tube. Add RNase to a concentration of 20 l/ml, mix gently and incubate at 37 C for 15 min. m) Add 0.1 volume of 3 M Na Acetate and 2 volumes o f 95% ethanol for DNA precipitation. Place at -20 C for 1 hour (can let set overnight if needed). n) Centrifuge at 16.1g for 10 min. to form DNA into pellet. Once pellet is dry, add 50 l EB buffer, put in 65 C water bath for 30 min, mixi ng after 15 min. o) Analyze the samples via Nanodrop spectrophotomet er. Extraction Buffer (EB, pH 6.0) .35 M glucose 0.1 M Tris Hcl 5.0 mM Na-EDTA (pH 8.0) 2% Polyvinylpyrrolidone (PVP), 40,000 MW 0.1% DIECA, diethyldithiocarbamic acid sodium salt 0.2% beta-mercaptoethanol (add just before use) Lysis Buffer 0.1 M Tris HCl 1.4 M NaCl 20 mM Na-EDTA (ph 8.0) 2% CTAB, cetyl trimethylammonium bromide 0.1% DIECA, diethyldithiocarbamic acid sodium salt 0.2% beta-mercaptoethanol (add just before use) optional: 1% PEG 4000, polyethylene glycol
66 Protocol 2: Modification of Fu et al (2004) Mangrove RNA Isolation Procedure a) Put 400 l CLS-VF and 100 l PPS from FastDNA kit into a tube with beads. b) Put 1 cm2 or 0.065 g excised leaf midrib tissue into the tub e. Put in FastPrep 24 homogenizer for 20 sec at grade 5.5. Centrifuge bri efly. c) Add 500 l CTAB extraction buffer (see lysis buffer from pro tocol 1, including 1% PEG 4000). d) Centrifuge at 16.1g for 8 min. transfer supernat ant to new 1.5 ml tube. e) Add 750 l chloroform:isoamyl alcohol (24:1) and shake sampl e vigorously. Centrifuge at 10,000g for 8 min at 4 C. f) Carefully transfer aqueous phase (upper phase) t o new tube. Add 750 l chloroform:isoamyl alcohol (24:1) and shake sample vigorously. Centrifuge at 12,000 g for 8 min at 4 C. g) Go to step g) in protocol 1.
67 Table 1: Information about primers used for PCR Primer Pair Sequence (5'-3') Amplico n Size (bp) Specificity Reference RmPAL-L /RmPAL-R cgcaatttttgcagaagtca/ccactgtggggaagtcctta 217 Rhizophora mangle L. PAL gene Cheeseman & Dassanayake, unpublished sequences P1/P7 aagagtttgatcctggctcagg att/cgtccttcatcggctctt 1800 universal phytoplasmas Chen et al. 2009 fU5/rU3 cggcaatggaggaaact/ttca gctactctttgtaaca 882 universal phytoplasmas Chen et al. 2009 10F/480R agtttgatcatggctcagattg/ cacggtactggttcactatcggtc 2502 universal bacteria Sandstrom et al. 2001; Hansen et al. 2007 NS1/NS8 gtagtcatatgcttgtctc/tccg caggttcacctacgga ~1769 universal fungi Cheng et al. 2004 LSD/LS1 ggaacctttccccacttc/agta cccgctgaacttaag 1475 universal fungi Hausner et al. 1993 NSA3/NLC2 aaactctgtcgtgctggggata/gagctgcattcccaaacaactc 1379 universal fungi Martin and Rygiewicz, 2005
68 Table 2: Information from the NanoDrop Spectrophoto meter A = abnormal tissue, C = control tissue, 260/280 va lues around 1.8 were considered reasonably pure, results either higher or lower tha n 1.8 were considered impure reasonably pure, results either higher or lower tha n 1.8 were considered impure Sample IDFast PrepDNeasy with beadsDNeasytypetissue260/280260/280260/280260/280260/280260/28 0 Aphloem31.861.5336.11.6910026.871.8610026.721.76253 .031.161003.871.2125 Aphloem36.441.536.11.64100Aleaf26.911.66300.81.9310012.131.34257.411.3425Aleaf34.991.911008.591.24258.161.1625Aleaf 4.71.09253.160.6625 Aleaf 4.210.88250.50.1925 Aleaf 6.261.19253.670.8625 Aleaf 16.121.64251.790.6125 Aleaf24.411.827.91.941009.111.7610022.451.92251.241 .271003.741.1925 Aleaf26.971.8129.91.91100Aleaf34.961.7736.21.851000.560.541002.942.44250.711 .941002.221.0325 Aleaf26.561.7830.51.75100Aleaf31.171.7131.61.9510010.511.641001.150.92250.82 0.491000.02-0.0125 Aleaf31.541.6235.61.73100Cleaf20.671.9826.42.13100Cleaf19.132.7125.52.28100Cleaf19.272.0325.71.86100Cleaf25.491.4123.72.25100Astem42.941.3348.11.47100 15.681.82504.682.1250 Astem39.551.4240.51.41100 8.31.67501.937.8250 Astem28.981.3933.31.45100 8.221.69503.0717.4350 Astem27.91.9335.61.9100 5.853.34506.652.0450 Cstem25.742.0232.21.87100Cstem15.932.4920.42.46100Cstem20.512.2323.42100Cstem30.991.5336.71.51100 1st analysis2nd analysis1st elution2nd elution1st elution2nd elution ng/lng/le(l)ng/le(l)ng/le(l)ng/le(l)ng/le(l)
69 Table 3: Rating of DNA template quality (as assesse d by A260/A280 ) and band intensity after PCR Bands and their corresponding DNA templates were co mpared to try and establish a correlation between the quality of the DNA used in a reaction and the band produced, but no correlation was found. Figure # Procedures DNA Template Quality + (0-4) 260/280 3.1 Smaller temperature gradient PCR with microsatellite primers abnormal leaf midrib 4 1.91 Results not shown Smaller temperature gradient PCR with microsatellite primers abnormal leaf midrib 4 1.91 Results not shown Temperature gradient PCR with universal phytoplasma primers abnormal leaf midrib 0 1.77 Results not shown Temperature gradient PCR with universal phytoplasma primers abnormal stem 0 1.33 3.2 Temperature gradient PCR with microsatellite primers abnormal leaf midrib 2 1.66 3.2 Temperature gradient PCR with microsatellite primers abnormal leaf midrib 3 1.91 3.2 Temperature gradient PCR with microsatellite primers abnormal leaf midrib 2 1.77 3.2 Temperature gradient PCR with microsatellite primers abnormal leaf midrib 0 1.33 Results not shown Temperature gradient PCR with universal phytoplasma primers abnormal leaf midrib 1 1.71 + Individual bands in several gels were given a num ber (0-4) based on the intensity of the band (assessed visually) The A260/A280 is a measure of the quality of the DNA extracts
70 Works Cited Alongsi, D. (2002). Present state and future of the world's mangrove forests. Environmental Conservation, 29 (3), 331-349. Altshcul, S., Gish, W., Miller, W., Myers, E., & Li pman, D. (1990). Basic local alignment search tool. Journal of Molecular Biology, 215 (3), 403-410. Ayres, M., & Lombardero, M. (2000). Assesing the co nsequences of global change for forest disturbance from herbivores and pathogens. The Science of the Total Environment, 262 (3), 263-286. Baldwin, A., Egnotovich, M., Ford, M., & Platt, W. (2001). Regeneration in fringe mangrove forests damaged by hurricane andrew. Plant Ecology, 157 (2), 151-164. Bandaranayake, W. (1998). Traditional and medicinal uses of mangroves. Mangroves and Salt Marshes, 2 133-148. Barbier, E. (2006). Natural barriers to natural dis asters: Replanting mangroves after the tsunami. Frontiers in Ecology and the Environment, 4 (3), 124-131. Barbier, E., & Cox, M. (2003). Does economic develo pment lead to mangrove loss? A cross-country analysis. Contemporary Economic Policy, 21 418-432. Berenguer, B., Sanchez, L., Quilez, A., LopezBarr eiro, M., Haro, O., Galvez, J., et al. (2006). Protective and antioxidant effects of rhizophora mangle L. against NSAIDinduced gastric ulcers. Journal of Ethnopharmacology, 103 (2), 194-200.
71 Boto, K., & Wellington, J. (1983). Phosphorus and n itrogen nutritional status of a northern australian mangrove forest. Marine Ecology, 11 63-69. Bowen, S. (1987). Composition and nutritional value of detritus. In D. Moriarty, & R. Pullin (Eds.), Detritus and microbial ecology in aquaculture, JCLA RM conference proceedings (pp. 192-216) Buttner, D., & Bonas, U. (2003). Common infection s trategies of plant and animal pathogenic bacteria. Current Opinion in Plant Biology, 6 312-319. Camilleri, J., & Ribi, G. (1983). Leaf thickness of mangroves ( rhizophora mangle ) growing in different salinities. Biotropica, 15 (2), 139-141. Cheng, T., Jia, X., Ma, X., Lin, H., & Zhao, Y. (20 04). Phylogenetic study on shiraia bambusicola by rDNA sequence analysis. Journal of Basic Microbiology, 44 (5), 339-350. Christensen, N., Axelsen, K., Nicolaisen, M., & Sch ulz, A. (2005). Phytoplsmas and their interactions with hosts. Trends in Plant Science, 10 (11), 526-535. Drennan, P., & Pammenter, N. (1982). Physiology of salt excretion in the mangrove avicennia marina (forsk.) vierh. New Phytologist, 91 (4), 597-606. Duke, N., Roelfsema, C., Tracey, D., & Godson, L. ( 2001). Preliminary investigations into dieback of mangroves in the mackay region. Report to Queensland Fisheries Service, Northern Region (DPI) and the Community of Mackay Region. Duke, N., Yuk Ying Lo, E., & Sun, M. (2002). Global distribution and genetic
72 discontinuities of mangrovesemerging patterns in the evolution of rhizophora Trees, 16 65-79. El-Dahhar, A., & El-Shazly, K. (2008). Effect of es sential amino acids (methionine and lysine) and treated oil in fish diet on growth perf ormance and feed utilization if nile tilapia, tilapia nilotica (L.). Aquaculture Research, 24 (6), 731-739. Evans, L., Okawa, Y., & Searcy, D. (2005). Anatomy and morphology of red mangrove (rhizophora mangle) plants in relation to internal airflow. Journal of the Torrey Botanical Society, 132 (4), 537-550. Fabricius, K. (2005). Effects of terrestrial runoff on the ecology of corals and coral reefs: Review and synthesize. Marine Pollution Bulliten, 50 125-146. Feller, I. (1995). Effects of nutrient enrichment o n growth and herbivory of dwarf red mangrove ( rhizophora mangle ). Ecological Monographs, 65 (4), 477-505. Feller, I. (1996). Effects of nutrient enrichmet on leaf anatomy of dwarf rhizophora mangle L. (red mangrove). Biotropica, 28 (1), 13-22. Feschotte, C., Jiang, N., & Wessler, S. (2002). Pla nt transposable elements: Where genetics meets genomics. Nature Reviews, 3 329-341. Gilbert, G., Mejia-Chang, M., & Rojas, E. (2002). F ungal diversity and plant disease in mangrove forests: Salt excretion as a possible defe nse mechanism. Oecologia, 132 278-285. Gilman, E., & Ellison, J. (2007). Efficacy of alter native low-cost approaches to mangrove
73 restoration, american samoa. Estuaries and Coasts, 30 (4), 641-651. Golbuu, Y., Victor, S., Wolanski, E., & Richmond, R (2003). Trapping of fine sediment in a semi-enclosed bay, palau, micronesia. Estuarin, Coastal, and Shelf Science, 57 941-949. Graaf, G., & Xuan, T. (1998). Extensive shrimp farm ing, mangrove clearance and marine fisheries in the southern province of vietnam. Mangroves and Salt Marshes, 2 159166. Grandbastien, M. (1998). Activation of plant retrot ransposons under stress conditions. Trends in Plant Science, 3 (5), 181-187. Gupta, P., Varshney, R. (2000). The development and use of microsatellite markers for genetic analysis and plant breeding with emphasis o n bread wheat. Euphytica, 113 163-185. Hansen, A., Jeong, G., Paine, T., & Stouthamer, R. (2007). Frequency of secondary symbiont infection in an invasive psyllid relates t o parasitism pressure on a geographic scale in california. Applied and Environmental Microbiology, 73 (23), 7531-7535. Hogenhout, S., Oshima. Kenro, Ammar, E., Shigeyuki, K., Kingdom, H., & Namba, S. (2008). Phytoplasmas: Bacteria that manipulate plan ts and insects. Molecular Plant Pathology, 9 (4), 403-423. Ilves, H., Horak, R., & Kivisaar, M. (2001). I nvolvement of S in starvation-induced
74 transposition of pseudomonas putida transposon tn 4652 Journal of Bacteriology, 183 (8), 5445-5448. Kathiresan, K., & Bingham, B. (2001). Biology of ma ngroves and mangrove ecosystems. Advances in Marine Biology, 40 84-254. Kovacs, J. (1999). Assessing mangrove use at the lo cal scale. Landscape and Urban Planning, 43 (4), 201-208. Laegdsgaard, P., & Johnson, C. (2001). Why do juven ile fish utilize mangrove habitats? Journal of Experimental Marine Biology and Ecology, 257 (2), 229-253. Lee, B., & Baker, G. (1973). Fungi associated with the roots of red mangrove, Rhizophora mangle Mycologia, 65 (4), 894-906. Lowenfeld, R., & Klekowski, E. (1992). Mangrove gen etics. I. mating systems and mutation rates in rhizophora mangle in florida and san salvador island, bahamas. International Journal of Plant Science, 153 (3), 349-399. Lugo, A., & Snedaker, S. (1974). The ecology of man groves. Annual Review of Ecological Systems, 5 39-64. MacFarlane, G., & Burrchet, M. (2002). Toxicity, gr owth and accumulation relationships of copper, lead, and zinc in the grey mangrove avic ennia marina (forsk.) vierh. Marine Environmental Research, 54 (1), 65-84. Maliyakal, J. (1992). An efficient methos for isola tion of RNA and DNA from plants containing polyphenolics. Nucleic Acids Research, 20 (9), 2381.
75 Martin, K., & Rygiewicz, P. (2005). Fungal-specific PCR primers developed for analysis of the ITS region of environmental DNA extracts. BMC Microbiology, 5 (28) McCoy, E., Mushinsky, H., Johnson, D., & Meshaka, W (1996). Mangrove damage caused by hurricane andrew on the southwestern coas t of florida. Bulliten of Marine Science, 59 (1), 1-8. McKee, K. (1995). Interspecific variation in growth biomass partitioning, and defense characteristics of neotropical mangrove seedlings: Response to light and nutrient availability. American Journal of Botany, 82 (3), 299-307. McKee, K. (1995). Seedling recruitment patterns in a belizean mangrove forest: Effects of establishment ability and physico-chemical facto rs. Oecologia, 101 448-460. McKee, K., Mendelssohn, I., & Hester, M. (1988). Re examination of pore water sulfide concentrations and redox potentials near the aerial roots of rhizophora mangle and avicennia germinans American Journal of Botany, 75 (9), 1352-1359. Nagelkerken, I., van der Velde, G., Gorissen, M., M eijer, G., van't Hof, T., & den Hartog, C. (2000). Importance of mangroves, seagrass beds a nd the shallow coral reef as a nursery for important coral reef fishes, using a vi sual census technique. Estuarine, Coastal, and Shelf Science, 51 31-44. Nagelkerken, I., Kleijnen, S., Klop, T., van der Br and, R., Cocheret de la Morinire, E, & van der Velde, G. (2001). Dependance of caribbean r eef fishes on mangrove and seagrass beds as nursery habitats: A comparison of fish faunas between bays with and without mangroves/seagrass beds. Marine Ecology Progress Series, 214 225-
76 235. Nga, B., Tinh, H., Tam, D., Scheffer, M., & Roijack ers, R. (2005). Young mangrove stands produce a large and high quality litter inpu t to aquatic systems. Wetlands Ecology and Management, 13 569-576. Nicholls, R., Hoozemans, F., & Marchand, M. (1999). Increasing flood risk and wetland losses due to global sea level rise: Regional and g lobal analyses. Global Environmental Change, 9 (1), 69-87. Nickerson, N., & Thibodeau, F. (1985). Association between pore water sulfide concentrations and the distribution of mangroves. Biogeochemistry, 1 183-192. Permingeat, H., Romagnoli, M., & Vallejos, R. (1998 ). A simple method for isolating high yield and quality DNA from cotton ( gossypium hirsutum L.) leaves. Plant Molecular Biology Reporter, 16 1-6. Proffitt, E., Milbrandt, E., & Travis, S. (2006). R ed mangrove ( rhizophora mangle ) reproduction and seedling colonization after hurric ane Charlie: Comparisons of Charlotte Harbor and Tampa Bay. Estuaries and Coasts, 29 (6), 972-978. Rabinowitz, D. (1978). Dispersal properties of mang rove propagules. Biotropica, 10 (1), 47-57. Ronnback, P. (1999). The ecological basis for econo mic value of seafood production supported by mangrove ecosystems. Ecological Economics, 29 (2), 235-252. Salmond, G. (1994). Secretion of extracellular viru lence factors by plant pathogenic
77 bacteria. Annual Review of Phytopathology, 32 181-200. Sandstrom, J., Russell, J., White, J., & Moran, N. (2001). Independent origins and horizontal transfer of bacterial symbionts of aphid s. Molecular Ecology, 10 217228. Shervette, V., Aguirre, W., Blacio, E., Cevallos, R ., Gonzalez, M., Pozo, F., et al. (2007). Fish communities of a disturbed mangrove wetland an d an adjacent tidal river in palmar, ecuador. Estuarine, Coastal, and Shelf Science, 72 115-128. Simberloff, D., & Abele, L. (1976). Island biogeogr aphy theory and conservation practice. Science, 191 (4224), 285-286. Smith, S., & Snedaker, S. (1995). Salinity response s in two populations of viviparous rhizophora mangle L. seedlings. Biotropica, 27 (4), 435-440. Strauss, E. (2009). Phytoplasma research begins to bloom. Microbiology, 325 388-390. Suryanarayanan, T., Kumaresan, V., & Johnson, J. (1 998). Foliar fungal endophytes from two species of the mangrove rhizophora Canadian Journal of Microbiology, 44 1003-1006. Thibodeau, F., & Nickerson, N. (1986). Differential oxidation of mangrove substrate by avicennia germinas and rhizophora mangle. American Journal of Botany, 73 (4), 512-516. Victor, S., Golbuu, Y., Wolanski, E., & Richmond, R (2004). Fine sediment trapping in two mangrove-fringed estuaries exposed to contrasti ng land-use intensity, palau,
78 micronesia. Wetlands Ecology and Management, 12 277-283. Weir, A., Tattar, T., & Klekowski, E. (2000). Disea se of red mangrove ( rhizophora mangle ) in southwest puerto rico caused by cytospora rhizophorae Biotropica, 32 (2), 299-306. Weisshaar, B., & Jenkins, G. (1998). Phenylpropanoi d biosynthesis and its regulation. Current Opinion in Plant Biology, 1 251-257. Wilson, I. (1997). Inhibition and facilitation of n ucleic acid amplification. Applied and Environmental Microbiology, 63 (10), 3741-3751. Wuo, W., Vrijmoed, L., & Jones, E. B. (2005). Scree ning of marine fungi for lignocellulose-degrading enzyme activities. Botanica Marina, 48 379-386.