ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/header_item.html)

Triclosan

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004188/00001

Material Information

Title: Triclosan Reviews of its Environmental and Health Effects and a Detection and Quantification Analytical Method
Physical Description: Book
Language: English
Creator: Valentine, Michael
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2009
Publication Date: 2009

Subjects

Subjects / Keywords: Triclosan
SPME
Health
Environment
Method
Analytical
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Triclosan is a ubiquitous chemical with possibly profound impacts on the environment and in medicine. It adversely affects aquatic and terrestrial organisms and is linked to antibiotic resistance. This chemical can be found in a wide variety of consumer products, such as soap, toothpaste, toys, athletic equipment, and kitchen cutting boards. Triclosan is a polychloro-phenoxy phenol, IUPAC name: 5-chloro-2-(2,4- dichlorophenoxy)phenol. Triclosan has several known metabolites, including a dioxin (Aranami & Readman 2006) and two (2,4-DCP and 2,4,6-TCP) (Canosa et al. 2005) that are acutely toxic and are listed as probable carcinogens (CDC 1997, CDC 1998). Triclosan has profound effects on the environment, both aquatic and terrestrial. It may interfere with agriculture (Liu et al. 2009) and kill or mutate aquatic life (Harada et al. 2008). It flows down the drain when hands are washed, toothpaste is discarded (EWG 2009), and other products have been utilized. These sources are diverted to waste treatment plants here some quantity of triclosan is removed from waste water. Triclosan kills bacteria by inhibiting enoyl (acyl carrier protein) reductase and may have the same or similar effects on enoyl (ACP) reductase in plants. Aquatic organisms such as microalgae and invertebrates can be killed by doses of ng/L concentration. Studies in vertebrates have found that it inhibits pregnancy functional hormones (James et al. 2009) and testosterone (Chen et al. 2007), and decrease testicular weight in rats (Kumar et al. 2009). Triclosan also may adversely impact human health. It can do this by acting as a contributor to the hygiene hypothesis (Mullooly et al. 2007). In non-lethal doses, bacteria can become resistant to it, and some bacteria show a triclosan-linked cross resistance to several common antibiotics (Schweizer et al. 2001). The method of Canosa et al. (2005) was reviewed and duplicated in the laboratory. Their methods used solid phase microextraction in combination with gas chromatography � mass spectrometry (GC-MS) to detect and quantify triclosan. My findings moderately supported Canosa et al. (2005); however procedural and statistical discrepancies arose. Variability increased to nearly three times the published values, possibly due to due to operator error, published descriptions, GC-MS column type, or experimental conditions. Linearity of the standard curve was excellent (R2 = .9834). Since raw data were not presented in the published work, direct comparisons of this data pool could not be used to determine method discrepancies. Overall, extraction, detection, and quantitation of triclosan using SPME were successful.
Statement of Responsibility: by Michael Valentine
Thesis: Thesis (B.A.) -- New College of Florida, 2009
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: McCord, Elzie

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2009 V15
System ID: NCFE004188:00001

Permanent Link: http://ncf.sobek.ufl.edu/NCFE004188/00001

Material Information

Title: Triclosan Reviews of its Environmental and Health Effects and a Detection and Quantification Analytical Method
Physical Description: Book
Language: English
Creator: Valentine, Michael
Publisher: New College of Florida
Place of Publication: Sarasota, Fla.
Creation Date: 2009
Publication Date: 2009

Subjects

Subjects / Keywords: Triclosan
SPME
Health
Environment
Method
Analytical
Genre: bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Triclosan is a ubiquitous chemical with possibly profound impacts on the environment and in medicine. It adversely affects aquatic and terrestrial organisms and is linked to antibiotic resistance. This chemical can be found in a wide variety of consumer products, such as soap, toothpaste, toys, athletic equipment, and kitchen cutting boards. Triclosan is a polychloro-phenoxy phenol, IUPAC name: 5-chloro-2-(2,4- dichlorophenoxy)phenol. Triclosan has several known metabolites, including a dioxin (Aranami & Readman 2006) and two (2,4-DCP and 2,4,6-TCP) (Canosa et al. 2005) that are acutely toxic and are listed as probable carcinogens (CDC 1997, CDC 1998). Triclosan has profound effects on the environment, both aquatic and terrestrial. It may interfere with agriculture (Liu et al. 2009) and kill or mutate aquatic life (Harada et al. 2008). It flows down the drain when hands are washed, toothpaste is discarded (EWG 2009), and other products have been utilized. These sources are diverted to waste treatment plants here some quantity of triclosan is removed from waste water. Triclosan kills bacteria by inhibiting enoyl (acyl carrier protein) reductase and may have the same or similar effects on enoyl (ACP) reductase in plants. Aquatic organisms such as microalgae and invertebrates can be killed by doses of ng/L concentration. Studies in vertebrates have found that it inhibits pregnancy functional hormones (James et al. 2009) and testosterone (Chen et al. 2007), and decrease testicular weight in rats (Kumar et al. 2009). Triclosan also may adversely impact human health. It can do this by acting as a contributor to the hygiene hypothesis (Mullooly et al. 2007). In non-lethal doses, bacteria can become resistant to it, and some bacteria show a triclosan-linked cross resistance to several common antibiotics (Schweizer et al. 2001). The method of Canosa et al. (2005) was reviewed and duplicated in the laboratory. Their methods used solid phase microextraction in combination with gas chromatography � mass spectrometry (GC-MS) to detect and quantify triclosan. My findings moderately supported Canosa et al. (2005); however procedural and statistical discrepancies arose. Variability increased to nearly three times the published values, possibly due to due to operator error, published descriptions, GC-MS column type, or experimental conditions. Linearity of the standard curve was excellent (R2 = .9834). Since raw data were not presented in the published work, direct comparisons of this data pool could not be used to determine method discrepancies. Overall, extraction, detection, and quantitation of triclosan using SPME were successful.
Statement of Responsibility: by Michael Valentine
Thesis: Thesis (B.A.) -- New College of Florida, 2009
Electronic Access: RESTRICTED TO NCF STUDENTS, STAFF, FACULTY, AND ON-CAMPUS USE
Bibliography: Includes bibliographical references.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The New College of Florida, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Local: Faculty Sponsor: McCord, Elzie

Record Information

Source Institution: New College of Florida
Holding Location: New College of Florida
Rights Management: Applicable rights reserved.
Classification: local - S.T. 2009 V15
System ID: NCFE004188:00001


This item is only available as the following downloads:


Full Text

PAGE 1

TRICLOSAN: REVIEWS OF ITS ENVIRONMENTAL AND HEALTH EFFECTS AND A DETECTION AND QUANTIFICATION ANALYTICAL METHOD BY MICHAEL VALENTINE 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. Elzie McCord, Jr. Sarasota, Florida May, 2009

PAGE 2

ii Dedication This is dedicated to all the amazing people in my l ife that helped keep me sane. My family, especially Mom & Dad, my SCA family, and my insomniac friends that all gave me someone to rant to when I needed it, then told m e to get back to work. Also for Jenna Nobody else was crazy enough to st ay in the lab with me until ridiculous hours of the morning. This is as much their work as it is mine.

PAGE 3

Table of Contents Dedication .................................................. ................................................... ..................... ii List of Tables .................................................. ................................................... ..................v List of Figures .................................................. ................................................... ................v Abstract .................................................. ................................................... ........................ vi Chapter 1: Introduction .................................................. ..................................................1 Triclosan and its Metabolites ..................... ................................................... ...................1 Uses, Products, and Efficacy ...................... ................................................... ...................3 Mode of Action ................................ ................................................... .............................6 Triclosan, its Metabolites, and mammals ....... ................................................... .............10 Triclosan in Sewage ........................... ................................................... ......................... 11 Triclosan in the Aquatic Environment .......... ................................................... ..............14 Triclosan in the Terrestrial Environment ...... ................................................... ..............16 Triclosan and its interactions with the human i mmune system .....................................2 0 Triclosan and antibiotic resistance ........... ................................................... ...................21 Chapter 2: Materials and Methods .................................................. ..............................23 Hypothesis ........................................ ................................................... ...........................23 Equipment ......................................... ................................................... ..........................23 Samples, reagents, and solvents ............... ................................................... ...................24 Sample Preparation ............................ ................................................... .........................25 Analysis ...................................... ................................................... .................................27 Quantification ................................ ................................................... .............................29 Pitfalls in recreating the method ............. ................................................... ....................31 Chapter 3: Results .................................................. ................................................... .......36 Standard Deviation (SD) vs. Standard Error (SE) ................................................. .......36 Derivatization ................................ ................................................... ..............................40 Elution Time and Quantification Limits ........ ................................................... .............42 Linearity of the Method ....................... ................................................... .......................46 Lack of Data in Canosa et al. (2005) ......... ................................................... ................48 Chapter 4: Discussion .................................................. ................................................... .50

PAGE 4

iv Experimental Results .......................... ................................................... ........................50 Statistical Ruminations ....................... ................................................... ........................51 Possible Methods of Improvement ............... ................................................... ..............53 Triclosan Safety .............................. ................................................... ............................54 Conclusion .................................... ................................................... ..............................56 Works Cited .................................................. ................................................... .................57

PAGE 5

v List of Tables Table 1: Area under the curve, including standa rd deviation, dried and non-dried SPME fiber ........................... ................................................... ....................................27 Table 2: Peak Areas for all Concentrations, sho wing mean, standard deviation, and standard error .................. ................................................... ..............................49 Table 3: Peak areas for all concentrations usin g population standard deviation ...........52 Table 4: Peak areas for all concentrations, out liers removed from calculation .............53

PAGE 6

vi List of Figures Figure 1: Triclosan ........................... ................................................... .............................1 Figure 2: 2,3,7,8-TCDD ........................ ................................................... ........................2 Figure 3: X,Y-DCDD ............................ ................................................... ........................2 Figure 4: 2,4-DCP ............................. ................................................... ............................3 Figure 5: 2,4,6-TCP ........................... ................................................... ...........................3 Figure 6: Methyl Triclosan..................... ................................................... .......................3 Figure 7: Products containing Triclosan ....... ................................................... ................5 Figure 8: Triclosan-FabI interaction .......... ................................................... ...................9 Figure 9: 2,4-DCP ............................. ................................................... ..........................18 Figure 10: 2,4-D .............................. ................................................... ............................18 Figure 11: SPME Fiber adsorption holder ....... ................................................... ...........24 Figure 12: Headspace Derivatization ........... ................................................... ...............27 Figure 13: Triclosan MS Chromatogram with secon dary peak .....................................30 Figure 14: ECD Chromatogram of blank SPME desor ption .........................................32 Figure 15: Zoomed ECD Chromatogram of blank SPM E desorption ...........................34 Figure 16: Triclosan MS Chromatogram & Spectrum from Canosa et al. (2005) .........35 Figure 17: Triclosan standard curve showing sta ndard deviation ..................................3 8 Figure 18: Triclosan standard curve showing sta ndard error ....................................... ..39 Figure 19: Tetradecanoic acid chromatogram & sp ectrum ............................................ 41 Figure 20: Triclosan 2ng/L chromatogram and spe ctrum ............................................. .43 Figure 21: Extraction time and peak area relati onship ............................................ ......45 Figure 22: Linear regression analysis of triclo san standard curve, with standard deviation .................... ................................................... .................................47

PAGE 7

vii Triclosan: Reviews of its Environmental and Health Effects and a Detection and Quantification Analytical Method Abstract Triclosan is a ubiquitous chemical with possibly p rofound impacts on the environment and in medicine. It adversely affects a quatic and terrestrial organisms and is linked to antibiotic resistance. This chemical can be found in a wide variety of consumer products, such as soap, toothpaste, toys, athletic equipment, and kitchen cutting boards. Triclosan is a polychloro-phenoxy phenol, IUPAC nam e: 5-chloro-2-(2,4dichlorophenoxy)phenol. Triclosan has several known metabolites, including a dioxin (Aranami & Readman 2006) and two (2,4-DCP and 2,4,6 -TCP) (Canosa et al. 2005) that are acutely toxic and are listed as probable carcin ogens (CDC 1997, CDC 1998). Triclosan has profound effects on the environment, both aquatic and terrestrial. It may interfere with agriculture (Liu et al. 2009) and kill or mutate aquatic life (Harada et al. 2008). It flows down the drain when hands are wash ed, toothpaste is discarded (EWG 2009), and other products have been utilized. These sources are diverted to waste treatment plants where some quantity of triclosan i s removed from waste water. Triclosan kills bacteria by inhibiting enoyl (acyl carrier protein) reductase and may have the same or similar effects on enoyl (ACP) reductase in plants. Aquatic organisms such as microalgae and invertebrates can be killed by doses of ng/L concentration. Studies in vertebrates have found th at it inhibits pregnancy functional hormones (James et al. 2009) and testosterone (Chen et al. 2007), and decrease testicular

PAGE 8

viii weight in rats (Kumar et al. 2009). Triclosan also may adversely impact human health. It can do this by acting as a contributor to the hygiene hypothesis (Mullooly et al. 2007). In non-lethal doses, bacteria can become resistant to it, and some bacteria show a triclosan-linked cross resistance to several common antibiotics (Schweizer et al. 2001). The method of Canosa et al. (2005) was reviewed and duplicated in the laboratory. Their methods used solid phase microext raction in combination with gas chromatography – mass spectrometry (GC-MS) to detec t and quantify triclosan. My findings moderately supported Canosa et al. (2005); however procedural and statistical discrepancies arose. Variability increased to nearl y three times the published values, possibly due to due to operator error, published de scriptions, GC-MS column type, or experimental conditions. Linearity of the standard curve was excellent (R2 = .9834). Since raw data were not presented in the published work, direct comparisons of this data pool could not be used to determine method discrepancies Overall, extraction, detection, and quantitation of triclosan using SPME were successfu l. X nrr nn

PAGE 9

1 Chapter 1: Introduction Triclosan and its Metabolites Triclosan (IUPAC – 5-chloro-2-(2,4-dichlorophenoxy) phenol) is a polychlorophenoxy phenol (Figure 1). Figure 1: Structure of Triclosan (from www.epa.gov) Triclosan is a white powder with a molecular weigh t of 289.5 and a melting range of 55-57C. It is slightly soluble in water, but is soluble in alcohols, ethers, and moderately strong basic solutions (i.e. 1M sodium h ydroxide) through the removal of the hydroxyl hydrogen, increasing its polarity (Bhargav a & Leonard 1996). In the environment, triclosan may not remain in th is form for extended periods. Aranami & Readman (2006) found that triclosan has a half life of four days in marine environments, and eight in freshwater, possibly due to ionic influences. Triclosan has been shown to adsorb to suspended molecules in the water, settle into silt deposits, and degrade very slowly. Singer et al. (2002) found triclosan in Greifensee sediment over thirty years old. Triclosan can be rapidly transformed into other co mpounds in the presence of sunlight and some aerobic microorganisms. Aerobic m icroorganism degradation, combined with activated sludge sorption, is the mos t commonly used method of

PAGE 10

removing triclosan in wastewater treatment plants (Thompson It has been shown dichlorodibenzo-pdioxin (2,8 Readman 2006). This demonstrates the poten dioxin compounds in water. containing triclosan exposed to free chlorine dioxins (TCDDs) (figure 2) contaminant in Agent Orange, the substance used to poison Ukrainian Yushchenko (BBC 2006) Figure 2 : 2,3,7,8 (from www.p esticideinfo.org) It has also been shown that triclosan can form othe r compounds. (2005) found two, 2,4dichlorophenol (2,4 (2,4,6-TCP), (figure 5) to be DCP and 2,4,6-TCP are formed (2006) showed that 2,4DCP, monochlorophenol, tentatively identified as dichl in wastewater treatment plants (Thompson et al. 2005). shown that photolytic degradation of triclosan produces 2 ,8 dioxin (2,8 -DCDD) (figure 3 ) in both fresh and seawater (Aranami & demonstrates the poten tial for triclosan to spontaneously form dioxin compounds in water. Kanetoshi et al. (1998) found that incinerating products exposed to free chlorine creates di-, tri, and tetrachloro (figure 2) – TCDDs are the class of dioxins that contain the contaminant in Agent Orange, the substance used to poison Ukrainian p resident Viktor ) : 2,3,7,8 -TCDD Figure 3: X,YDCDD esticideinfo.org) (From www.quantexlabs.com) It has also been shown that triclosan can form othe r compounds. Canosa et al dichlorophenol (2,4 -DCP), (figure 4) and 2,4,6trichlorophenol to be triclosan metabolites. Canosa et al (2005) theorize that formed through photolytic degradation. Sanchez DCP, monochlorophenol, 2,8-DCDD, and another compound tentatively identified as dichl orohydroxydibenzofuran formed from triclosan 2 2005). that photolytic degradation of triclosan produces 2 ,8 ) in both fresh and seawater (Aranami & triclosan to spontaneously form (1998) found that incinerating products and tetrachloro -p-dibenzo that contain the resident Viktor DCDD (From www.quantexlabs.com) Canosa et al trichlorophenol Canosa et al (2005) theorize that 2,4Prado et al. another compound from triclosan degradation.

PAGE 11

3 Figure 4: 2,4-DCP Figure 5: 2,4,6-TCP (From www.chemexper.com) (From www.chemexper.com ) Methyl triclosan is another triclosan derivative. T his metabolite is much more stable in the environment, and due to its lower pol arity, it bioaccumulates to a much greater degree than triclosan. Ingested methyl tric losan is deposited into fat bodies. It bioaccumulates in top predators feed on contaminate d organisms. This is similar to DDT, a pesticide that caused increased eggshell fragilit y in raptor populations, especially highprofile creatures such as bald eagles (Stokstad 200 7). Figure 6: Methyl Triclosan (from www.galchimia.com) Uses, Products, and Efficacy Triclosan was invented over forty years ago by Ciba Inc., a pharmaceutical company that sells its product under the name Irgas an and Microban (Ciba 2008). Originally patented as an herbicide (Heath et al. 2001), triclosan has been in use since 1972 as an ingredient in a surgical scrubs. It was restricted to clinical use until the mid 1990's, when its popularity in consumer products be gan to rise (Glaser 2004).

PAGE 12

4 Triclosan is a biocide and bacteriostat that is al so an effective fungicide in some species. Its popularity has increased exponentially in an increasingly germophobic society. It functions by inhibiting a fatty acid bi osynthesis pathway that bacteria and some fungi use to create cell walls. Without this fatty acid, the cell eventually fails (Russell 2004). Russell’s (2004) findings contradict previou s claims that triclosan is a multi-site biocide that physically lyses the cell wall. Today, triclosan can be found in numerous products with hygiene products being the most common. Skin Deep, the cosmetic safety da tabase for the Environmental Working Group (2008), found 932 different products containing triclosan. These included deodorant, soap, shaving cream, facial cleanser, ac ne treatments, and toothpaste. It can also be found in other products, both commercial an d home use. Due to its high thermal stability, up to 200C for two hours (Bhargava & Le onard 1996), triclosan is compatible with plastics, i.e. kitchen cutting boards, childre n’s’ toys, and dishwashing machines. It is also formulated into several other products, such a s clothing, mulch, tabletops, dish racks, paint, fabric softener, credit cards, and paper cur rency (EWG 2008). Figure 7 shows a small sample of triclosan-containing products.

PAGE 13

5 Figure 7: A small sample of products containing tri closan (Taken by Valentine 2009) Triclosan is a broad spectrum biocide that has pro ven effective against most bacteria (both Gram-positive and Gram-Negative) and some mycobacteria. Due to the wide range of bacterial strains in the world, it is difficult to assign a broad descriptor of efficacy. However, several studies have examined th e effect of triclosan exposure on a variety of microbes, the results of which were comp iled by Jones et al. (2000). Triclosan is present from .2% to 2% concentrations in most consumer products. These levels are often enough to exhibit bactericid al or at least bacteriostatic effects. It has been shown to be effective against common germs such as Escherichia coli and Salmonella, with minimum inhibitory concentrations of .1-.3ppm for both (Bhargava & Leonard 1996), and kill rates of >99.99% in a 15-s econds at 1% concentration (Jones et al. 2000). Kill rates were much lower against other or ganisms, such as the mycobacteria Candida tropicalis (a species in normal human flora that can cause can didiasis) or Trichophyton tonsurans (a species that causes ringworm) with 73.9% and 52% mortality,

PAGE 14

6 respectively (Jones et al. 2000). The problem with this test is that most consumer pr oducts do not contain 1% triclosan. Most “antibacterial” hand and dish soaps contain concentrations well below the 1% concentration proved effective by Jones et al. (2000), ranging from .1% to .15%, or 1 – 1.5 mg/L (EWG 2008). This amount may be furt her diluted during hand washing or dishwashing, given the amount of water involved. An assumed 100:1 water : soap ratio (a conservative estimate for hand washing) reduces the amount of triclosan to 10 to 15 g/L (.001% .0015%). This concentration may be ef fective against highly susceptible strains such as numerous Staphylococci, including methylcillin-resistant Staphylococcus aureus (MRSA) (Bhargava & Leonard 1996) but against more resistant species, such as some strains of E. coli and Pseudomonas aeruginosa this concentration is ineffective (Russell 2004). Many of the triclosan-impregnated plastics and fabr ics available to consumers compare similarly. Heath et al. (2001) found no difference between bacterial viabi lity levels on triclosan-impregnated and normal plastic wrap used to cover supermarket beef. This lack of inhibition may be a sign of increased bacterial resistance to triclosan. Mode of Action Triclosan is not a multi-site biocide, as many prev iously thought. It was believed to lyse the bacterial wall and cause death. Studies show that triclosan blocks the active site of enoyl-acyl-carrier protein reductase, an en zyme in the type II fatty acid synthesis system (Heath et al. 2001, Kapoor et al. 2001, Heath & Rock 2000). These fatty acids are required to produce and maintain the bacterial cell membrane.

PAGE 15

7 Two types of fatty acid synthase systems are encou ntered. The first is type I, which is a large, multifunction enzyme used by mamm als and some fungi. Type II is used by bacteria and employs multiple discrete enzymes ( Ling et al. 2004). Enoyl-ACP reductase reduces the carbon-carbon double bond nea r the carbonyl to elongate the chain and form the building blocks of the bacterial membr ane (Kapoor et al. 2001). Without this enzyme, the bacterium produces fatty acids whi ch cannot maintain the integrity of the cell membrane. This prevents further growth and repair of the cell, eventually leading to cell failure. Triclosan's perceived biolytic action was based on a misinterpretation of experimental results. Vertebrate cells lack rigid peptidoglycan-based cel l walls like plants and bacteria, Triclosan was originally patented as an herbicide ( Regs et al. 1979), thus the cell wallattack hypothesis seemed valid in light of the orig inal experimental results. Heath & Rock (2000) performed experiments examining the form-dependent interactions between phenotypically different fab sites and triclosan, the results of which are summarized here Triclosan binds tightly to the active site of typ e II ENR, and prevents NADH from binding. It binds more tightly i n the presence of NAD+, forming an extremely stable complex. Triclosan binds with vary ing degrees of tightness to the f abI site. The degree of binding depends on variations i n the specific ENR structure. This structure is determined by the fabI and inhI genes. Binding affinities are one source of eventual resistance in bacteria. Several mutant iso forms of the enoyl-ACP reductase enzyme have been found in bacteria. These isoforms, termed fabK (originally found in S. pneumoniae), and fabL (first found in B. subtilis) display varying levels of resistance to

PAGE 16

8 Triclosan. When the fabK isoform of enoyl-ACP reductase was expressed by E. coli its resistance became great enough to grow in triclosan levels of 2 mg/mL. This also gives a possible insight into why triclo san is only effective against some mycobacteria – as mentioned before, some fungi use type I FAS systems (Ling et al. 2004), which triclosan does not affect. The actual site interactions, compiled by Heath et al. (2001) for fabI are quite extensive, also shown in figure 8. The hydroxychlorophenyl ring of triclosan is coplan ar with the nicotinamide ring of the NAD+ at a distance of about 3.4 Ar, facilita ting pi–stacking interactions, and also interacts with the side chains of Tyr156 a ndTyr146. The 20-hydroxyl of the NAD+ ribose and the hydroxyl of Tyr156 hydrogen bond to the hydroxyl of triclosan. The chlorine at the 4 position of the ph enyl ring accepts a hydrogen bond from amide of Ala95 and forms hydrophobic inte ractions with the side chain of Met159. The substrate binding loop becomes struc tured on drug binding, strengthening the interaction with additional Van d er Waals interactions, as seen with diazaborine binding.

PAGE 17

9 Figure 8: Site interaction of triclosan with FabI s howing the ternary complex formed between NAD+, tri closan, and the active site of enoyl-ACP reductase (From He ath et al. 2001)

PAGE 18

10 Triclosan, its Metabolites, and Mammals Triclosan is marketed as harmless to humans and ma mmals. Research shows this to be true in acute cases (Bhargava & Leonard 1996) In higher concentrations it seems to be a hormonal disruptor (Kumar et al 2009, James et al. 2009). Its metabolites are documented toxicants. The formation of TCDD through combustion (Kanetoshi et al. 1988) is of concern due to 2,3,7,8-TCDD's extremely high bioaccumulation rate (EWG 2008). 2,4-DCP and 2,4,6-TCP, proven triclosan meta bolites (Canosa 2005), are acutely toxic in high concentrations (CDC 1997, CDC 1998 ) and at low concentrations act as endocrine disruptors. 2,4-DCP and 2,4,6-TCP are als o listed as probable human carcinogens (EPA 2007, OSHA 2007). Triclosan possesses a low octanol/water coefficien t (4.8). When introduced to an organism, it will sequester in fat bodies. Comparat ive data from the Danish Environmental Protection Agency shows bioaccumulati on rates of up to 8,400 times environmental levels (Samsoe-Petersen 2003). Freque nt use of triclosan may cause high levels in the fat bodies of the exposed organism. M ethyl triclosan is an even more hydrophobic compound. It would logically bioaccumul ate to an even greater degree. Triclosan's polyphenol structure also suggests tha t it may mimic and disrupt endocrine pathways. Research supports this idea, th ough in higher dose concentrations than generally will be encountered by most organism s (Kumar et al 2009, James et al. 2009). Chronic or subchronic effects may occur, lin ked to the high bioaccumulation rate. I found no published research at the time of this w riting to support this hypothesis. When orally administered to male Norway rats, a dos e of 10 mg/kg/day for 60 days decreased testis size from 2.334g to 1.709g, a difference of 26.78% (Kumar et al.

PAGE 19

11 2009). Triclosan's antiandrogenic effects have also been shown in vitro with human testosterone. These studies showed that a 1 M tric losan decreased testosterone transcriptional activity by 38.8%, and 10 M decrea sed it by 92% (Chen et al. 2007). Triclosan's hormonal disruption is not limited to androgens. James et al. (2009) found that triclosan is a potent inhibitor of estra diol and estrone sulfonation in sheep placentas. These compounds are critical in fetal de velopment in both sheep and humans. In vitro experiments showed that 0.31 M triclosan was effec tive in inhibiting 50% of placental sulfotransferase activity (James et al 2009). These findings show that complications in the preg nancy, spontaneous abortion, and fetal development issues are possible. While tr iclosan is partially glucoronidated and sulfonated in the human liver, one pass typically r emoved only 53-78% of free triclosan (James et al. 2009), suggesting that triclosan may not be as saf e for humans and mammals as previously thought. These new findings s uggest that triclosan may be better used in a prescribed setting, as opposed to the ubi quitous use of today. Triclosan in Sewage All chemicals not completely degraded by the body eventually enter the sewage system. Triclosan enters most often through sink dr ains after washing hands or dishes, or washing out of triclosan-impregnated materials (Bes ter 2003). Once in the sewer, it goes through one or more processes in sewage treatment t o remove or degrade all waste materials. After treatment, water is often reused f or various purposes, ejected into a surface body of water, or injected into underground aquifers (FDEP 2009). The problem occurs when we realize that waste treatment is not 100% effective. Chemicals, especially

PAGE 20

12 common pharmaceuticals, are often released back int o the environment unchanged in quantities that have a real potential for impact on the environment and public health. Sewage treatment occurs a number of stages. The fi rst stage is mechanical, removal of unprocessable materials such as rocks, s ticks, plastics, sand, and gravel. After this initial processing, the waste goes on to a sec ondary, largely biological treatment (FDEP 2009). Sarasota uses a modified Bardenpho process to trea t wastewater (FDEP 2009). This process is similar to most other activated slu dge treatments. The influent is held in a tank and is mixed with an organic sludge containing a wide variety of microorganisms, which break down organic matter and create more slu dge, which is quite lipophilic in nature (Eimco 2007). A large number of pharmaceutic als adsorb into this sludge at high percentages, and are thus are prevented from exitin g with the effluent. A review of several papers shows that triclosan adsorbs to acti vated sludge basins at efficiencies ranging between 95% and 97.5% (Thompson et al. 2005). This apparently high removal rate may be misleadin g. Bester (2003) investigated the fate of triclosan in a wastewater treatment pla nt in Dortmund, Germany. This plant had a similar triclosan removal rate (95%), however the researchers made an interesting discovery. Only 50 grams of triclosan were released into the environment, of the approximately 1000 grams that entered in a five-day span. The anomaly was that only 300 g were absorbed into the sludge. This means tha t 65% of the triclosan that entered the plant over five days seemingly disappeared. The law of conservation of matter dictates that matter cannot be created nor destroyed, thus o ne must assume that the remaining 65% of the triclosan was metabolized (either throug h photolysis or biological

PAGE 21

13 degradation) into other chemicals that were not ide ntified in Bester’s (2005) work. It is possible that the remaining 65% was minerali zed into elemental carbon, or used as a carbon source for the microorganisms; how ever, this is likely not the case in purely activated-sludge sewage treatments, as most of the process is carried out in anoxic conditions (Heidler & Halden 2007). In the Bardenph o process, oxygen is introduced in two of the five stages to promote aerobic biotransf ormation of chemicals and organic waste (Eimco 2007). Much of the remaining triclosan may be mineralized to CO2 or used to increase biomass since triclosan removal is link ed to microbial oxygen demand (Federle et al. 2002). Triclosan does not appear to have deleterious effe cts on wastewater treatment systems. It may be counterintuitive, as most system s use bacteria to degrade sewage. Stasinakis et al. (2008) found that 10 mg/L of triclosan inhibited m icrobial respiration rates by ~23%. Federle et al. (2002) found triclosan at concentrations generally present in wastewater (<16,000 ng/L) showed little evidence of bacterial inhibition. Triclosan has been in use for over thirty years, an d sewage treatment flora has a marked resistance to triclosan (Stasinakis et al. 2008). As demonstrated by Federle et al. (2002), when unacclimated sludge was treated with a 200 g/L concentration of triclosan and was then re-dosed with a higher concentration ( 1 mg/L), the radio labeled total CO2 recovery rates at fifteen days increased from 33% t o 58% for one group and 65% for the second batch. In reviewed studies, triclosan in wastewater treat ment plant influent was not found in concentrations greater than 1 mg/L (Heidle r & Halden 2007, Canosa et al. 2005, McAvoy et al. 2002, Bester 2003, Thompson et al. 2005), thus it can be safely assumed

PAGE 22

14 that sewage flora are acclimated to the levels in t heir environment. This resistance is further evidence that bacteria are capable of becom ing resistant to triclosan through a number of pathways. Triclosan in the Aquatic Environment Triclosan affects aquatic organisms in a variety of harmful ways. Studies involving mussels, algae, anurans, crustaceans, pro tozoa, and bacteria have all demonstrated various effects of triclosan on the en vironment. Triclosan, and especially its methylated derivative methyl triclosan, also bioacc umulate in numerous organisms. Triclosan is toxic to Medaka in early life stages, and metabolites may act as weak androgens (Ishibashi et al. 2004). South African clawed frogs experience simil ar effects. Mussels suffer more immediate effects of exposure ( Harada et al. 2008). In a study involving Zebra mussels, low levels of triclosan in duced hemocyte apoptosis and caused rapid, irreversible genetic damage (Binelli et al. 2009). In microalgae, photosynthesis is inhibited and growth is slowed (Harada et al. 2008, Yang et al. 2008). Harada et al. (2008) performed a broad study involving several p harmaceuticals and personal care products, one of which was triclo san, on South African clawed frogs, algae, crustaceans, protozoa, and bacteria. These o rganisms all demonstrated triclosan's toxicity in aquatic settings. None of the other tes ted chemicals, including several common antibiotics and analgesics, affected all test organ isms. Algae were the most sensitive to triclosan in this study. In two studies, the EC50 of triclosan on algal (Specifically Pseudokirchneriella Subcapitata) growth inhibition ranged from 0.53 g/L (72 hr IC50) to 12 g/L (96 hr EC50) (Yang et al. 2008, Harada et

PAGE 23

15 al. 2008). In the study performed by Yang et al (2008) triclosan was the most toxic to the microalga P. Subcapitata of the twelve compounds tested. Other compounds te sted showed IC50 concentrations of 1 g/L to 210 g/L. Concentratio ns of 0.53 g/L are low enough that waters directly downstream of wastewate r effluent sites will frequently experience levels at or above this, and is a cause for concern. Triclosan also affects crustaceans and protozoa. I n the study conducted by Harada et al. (2008), the freshwater organisms Daphnia magna and Tetrahymena pyriformis were both exposed to triclosan. The results showed an un favorable environmental safety profile. The EC50 (48 hour exposure for D. magna 96 for T. pyriformis ), measured as mobility impediment for D. magna and cell growth inhibition for T. pyriformis, were 260 g/L and 210 g/L, respectively (Harada et al. 2008). Xenopus laevus the South African clawed frog, was the most resis tant to triclosan in the studies reviewed. Embryos were exposed to tr iclosan over a 96 hour period. It was found from these experiments that the embryonic LC50 for X. laevus is 0.82 mg/L (Harada et al. 2008). This level is unlikely to be found in water s receiving wastewater effluent. Plants sprayed with triclosan-containing pesticides or runoff ditches containing triclosan from pesticide applications may become in imical to amphibian embryos. The effects on Zebra mussels ( Dreissena polymorpha ) may be the most severe. Hemocytes were extracted from healthy Zebra mussels and exposed to 0.1 0.3 M (28.95-86.85 g/L) triclosan. The cells were then e xamined after sixty minutes of exposure. Significant cellular and genetic damage w as seen at the lowest concentration (0.1 M) (Binelli et al. 2009). Control and DMSO-tr eated cells showed no significant difference between them. When exposed to 0.1 M tri closan, cellular damage in the mid-

PAGE 24

16 to-extreme range increased from 0% in DMSO to 51.9% 24.4% of the cells exposed to 0.1 M triclosan had “minimal” damage, as opposed t o 94.4% from the DMSO group. Cellular apoptosis also increased with application of triclosan. Apoptotic cells increased from 3.2% in the control group to 6% after exposure to 0.1 M triclosan. At 0.3 M triclosan, the highest level, apoptosis was 14.3% ( Binelli et al. 2009). These data show that triclosan's safety profile mu st be revised to include the notable hazards to aquatic life. Heavy use of this antibacterial, which is becoming very common, could eventually cause tremendous environme ntal harm to an ecosystem that is vital to human survival. Triclosan in the Terrestrial Environment As noted previously, triclosan is a lipophilic com pound. When it is treated in the sewage plant, a great deal of it is absorbed into t he sludge. While this means that much less is released into the aquatic environment, it m ay actually be concentrated and unwittingly incorporated into soil. This may have a dverse impacts on plant growth and alter the soil's microfauna species distribution. B eneficial fungi may also be affected, depending on their FabI system. Excess sludge biomass is recycled into fertilizer i n many cities and municipalities (Liu et al. 2009). The sludge is first de-watered using polyme rs and then heat, which also kills the sewage microfauna (FDEP 2009). Due to its high thermal stability and fairly low polarity (Bhargava & Leonard 1996), triclosan is li kely to be unaffected by heating or sorption on to highly polar water-targeting polymer s. Fertilizers are then disbursed to surrounding farm s, gardens, parks, etc (FDEP

PAGE 25

17 2009). Triclosan is thus introduced to the soil in levels that may reach 55 g/g (55 mg/kg) (Heidler & Halden 2007) or higher, depending on the extent of water extraction. A number of studies have examined the effects of t riclosan on the soil, and Liu et al (2008) examined how it affects rice and cucumbers growth in triclosan-treated soil. Triclosan’s high anaerobic soil half-life may cause a number of problems. Since it is hydrophobic, it is unlikely that rainfall will dilu te concentrations. Triclosan is expected to have a half-life of 120 days in soil using EPA fate models (Ying et al. 2007). Its half life is not as long as the EPA fate model predicts. Aero bic bacteria, proven to degrade triclosan, are present in soil. In aerated soil, tr iclosan’s half-life in a 1mg/kg concentration of is 18 days (Ying et al. 2007). Triclosan in anaerobic soil persists much lo nger, with little reduction in concentration over the seventy days it was studied (Ying et al. 2007). This is much closer to the predictions of the EPA f ate model. Triclosan does seem not harm soil bacteria, despit e its fairly long half life and initial depressant effect on soil respiration (Liu et al. 2009). In the first two days, triclosan (in concentrations ranging from 1-50 mg/k g) decreased respiration rates by 4.88.8%. This decrease did not persist. After four day s, respiration rates in all treated samples were equal to or greater than the control g roup (Liu et al. 2009). Bacterial diversity increased in the samples. After 72 hours, the Shannon Diversity index at 30 and 50 mg/kg actually from the control of ~2.95 to 3.25 and 3.2, respectively (Liu et al. 2009). This may be due to slower-dividing, more res istant bacteria with increased fitness in an adverse environment. Neither cucumbers nor rice showed similar resistan ce to triclosan. Liu et al. (2008) examined the effects of triclosan in levels of 1 – 300 mg/kg on rice and cucumber

PAGE 26

18 plants. The results were interesting and unsurprisi ng knowing that triclosan was originally developed as an herbicide. They looked a t two factors, root length and shoot length. In cucumbers, root length was significantly inhibited (27.3%) at concentrations of 30mg/kg. Shoot length was also inhibited at this co ncentration, with a mean inhibition of 20.6%. Rice was found to be much more sensitive. Si gnificant root length inhibition (17.6%) was encountered when grown in soil only con taining 10 mg/kg triclosan. Shoot length, interestingly, increased 12.5% in rice pla nts exposed to that concentration. When 30 mg/kg triclosan was used, the shoot length stand ard deviation and shoot length increased. 30 mg/kg triclosan-exposed shoots were m uch more variable in size, with a mean height of 16.2 cm, compared to the control gro up's 10.6 cm mean height. The standard deviation of 30mg/kg treated plants was 10 .2 cm, a large amount compared to the control group's .8 cm. This may be due to small amounts of triclosan breaking down into 2,4-DCP (Fig. 9), which is similar to the know n auxin mimic 2,4dichlorophenoxyacetic acid (2,4-D, Fig. 10). Triclo san may also act as a weak auxin mimic, causing experimental error. In higher concen trations, much less variation was seen. 50 mg/kg concentrations inhibited plant growt h 13.4% (Liu et al. 2009). Figure 9: 2,4-DCP Figure 10: 2,4-D (From www.chemexper.com) (From www.phyt otechlab.com)

PAGE 27

19 A likely explanation of triclosan’s inhibitory ef fects on plants is that triclosan inhibits the ENR enzyme of plants as it does in bac teria. A study by Dayan et al. (2008) involving the effects of triclosan and cyperin, a f ungal phytotoxin (structure is similar to triclosan) on Arabidopsis ( Arabidopsis thaliana ) showed that triclosan is highly toxic to this particular plant. Arabidopsis is a phytologica l model species and numerous plants possess similarly structured ENR enzymes, because E NR is highly conserved in plant biology. Dayan et al (2008) determined that plants respond similarly t o bacteria when exposed to triclosan. Cell membrane stability was g reatly reduced due to the necessity of fatty acid synthesis in producing the lipid bilayer s. A concentration of only .01 M inhibited purified Arabidopsis ENR activity by 20% and 0.05 M decreased it by 40%. At the 1 M (2.895 mg/L) concentration, ENR activit y was completely inhibited (Dayan et al. 2008). In vivo experiments on Arabidopsis may yield somewhat diffe rent results, as plants have detoxification and toxin sequestration mechanisms like most other organisms. In the study involving cucumbers, a 3.45 M (1mg/kg concentration, density calculated at 1.0 g/mL) inhibited cucumber shoot growth by 8.6 %, and root growth 9.6%. This concentration increased rice root and shoot height by 1.9% and 12.3%, respectively (Liu et al. 2008). It is easy to see why triclosan contaminated water and fertilizer could be an enormous problem in the agricultural world, when on e recalls triclosan’s herbicidal origin. The monetary savings of free or reduced pri ce supplies of water and fertilizer

PAGE 28

20 could be offset by the reduced yield or death of cr ops supplemented by these supplies. Triclosan and its interactions with the human immun e system Triclosan may contribute to dysfunctional immune s ystems when used around toddlers, infants, and small children. It also may contribute to bacterial resistance to antibiotics. This combination means that not only c ould widespread triclosan use be weakening the body's natural defenses, it could als o be decreasing the effectiveness of the available weapons when the body’s defenses are insu fficient. The hygiene hypothesis, originally put forth by St rachan (1989) postulates that increasing environmental sanitization could lead to autoimmune disorders caused by a lack of early-age stimulation of the immune system. This means that people with more resources (developed countries) to spend on improve d sanitation such as developed sewage systems, antibacterial soaps, low population density in the home, and products such as triclosan and Lysol, will encounter fewer p athogens in their lifetime, especially in early childhood. Triclosan may therefore contribute to an increasingly dysfunctional immune system. Studies have found that a lack of exposure to path ogens is associated with an increase in autoimmune malfunctions such as allergi es, asthma, and atopic dermatitis (atopy) (da Costa Lima et al 2003, Strachan 1989, Mullooly et al. 2007). Other autoimmune diseases include inflammatory bowel dise ase, multiple sclerosis, and type I diabetes. A cohort study by da Costa Lima et al. (2003) found that the prevalence rates of asthma positively correlated with mean household in come (Prevalence Rates: <$50,000, 16.4%; >$300,000, 23.8%) and were inversely correla ted with increased number of

PAGE 29

21 people per bedroom and number of other children liv ing in the home. Prevalence rates were 23.6% in houses with no other children in the home, and 17.2% with other children present. The increase was even larger when correlat ed with individuals per bedroom. In cases where two or fewer people inhabited a bedroom the asthma rate was 26.7%. If more than two individuals shared a bedroom, the rat e dropped to 16.8%. Mullooly et al. (2007) found that larger numbers of antibiotic pre scriptions in the first two years of an infant’s life were inversely associated with atopic dermatitis. The number of prescriptions was used as a marker in the study for the amount of pathogens the child was exposed to. In the first two years of life, the control (non-atopic) group was prescribed an average of 4.78 antibiotics, versus t he 4.15 in the atopic group. Triclosan and antibiotic resistance Triclosan is also increasingly connected to antibi otic resistance in bacteria (Schweizer 2001). Since its mode of action was disc overed, several experiments have demonstrated that bacteria can become resistant to triclosan. Some of these mechanisms of resistance are effective against antibiotics. Ba cteria are commonly exposed to sublethal doses of the chemical and thus have the opportunity to develop resistance. The most common methods of triclosan resistance ar e gene mutations of fabI and inhA and efflux pumps (Schweizer 2001). Other methods o f resistance include marA, soxS, and acrAB over expression (Russell 2004). InhA mutations and efflux pumps correspond to cross-resistance with antibiotics (Sc hweizer 2001). FabI mutations occur in two forms – over expression and triclosan-insensitivity. These mutations do not appear to decrease the effic acy of antibiotics (Schweizer 2001).

PAGE 30

22 This is likely because the examined antibiotics do not target enoyl-ACP reductase. If new antibiotics were introduced that targeted this syst em, they would likely prove ineffective against triclosan-resistant bacteria. Triclosan resistance due to InhA mutations and increased drug efflux pump activity, however, engenders cross-resistance to an tibiotics (DiPetrillo et al. 2004). Drug efflux pumps are broad-spectrum bacterial responses to harmful substances in the cell. Efflux pump activity was documented in E. coli, P. aeruginosa, S. aureus, and L. lactis (Schweizer 2001) Triclosan-induced antibiotic cross resistance has b een documented in E. coli and P. aeruginosa (Schweizer 2001) InhA is a gene similar (48% homology) to fabI that also controls the structure of the enoyl-ACP enzyme ((DiPetrillo et al. 2004). InhA triclosan/antibiotic cross-resistance is documented in several bacteria including Mycobacterim smegmatis and M. tuberculosis, the mycobacterium that causes tuberculosis (Schweizer 2001). Though the hygiene hypothesis is still a matter of some debate and controlled trials are difficult to perform, both logistically and ethically, it seems reasonable that precautions should be taken. If taken into consider ation with high probability of creating wild-type mutants, ubiquitous triclosan use could c ause a looming race between antibiotic development and bacterial evolution.

PAGE 31

23 Chapter 2: Materials and Methods Hypothesis The hypothesis for this experiment was that the res ults published by Canosa et al. (2005) would be reproducible. The null hypothesis for this experiment was that t he results published by Canosa et al. (2005) would not be reproducible. Equipment Manual, polyacrylate-coated (PA) solid-phase microe xtraction (SPME) fibers, syringe, and injection port holder were purchased f rom Varian (Palo Alto, California). The SPME fibers were conditioned according to manuf acturer's specifications before use. The mass spectrometer (GC-MS) was a Varian Saturn 2200 (Ion trap) attached to a CP3800 gas chromatograph. The column was a Varian VF -5ms, a low bleed, 5% phenylmethyl 95% dimethylpolysiloxane, 30 meter column with an 0.25 mm inner diameter. Glassware, stirrers, scoops and vials were obtained from laboratory supplies, and were washed with acetone and Milli-Q water before and be tween uses. A clamp (Figure 11) to stabilize and hold the SPME syringe during adsorpti on was manufactured by the author.

PAGE 32

24 Figure 11: SPME fiber with adsorption holder Samples, Reagents, and Solvents Triclosan (>97% purity) was extant in the laboratory and kept at -16.2 C, isolated from light before and between uses, and was previou sly obtained from Fluka Analytical (Subsidiary of Sigma-Aldrich, St. Louis, MO). N -MethylN -[tert-butyldimethylsilyl]trifluoroacetimide (MTBSTFA) was purchased fr om Regis Technologies (Morton Grove, IL). HPLC-grade methanol was extant in the l aboratory, previously purchased from Thermo Fisher Scientific (Waltham, MA). Ultrapure water was generated via Milli-Q filtration. Acetone and 6 M HCL were obtai ned from laboratory supplies.

PAGE 33

25 Sample Preparation Triclosan stock solution was prepared by adding 1 mg triclosan to 10 mL HPLC grade methanol to create a 100 mg/L stock solution. This solution was kept in a glass vial with an opaque covering at -16.2 C in the dark between uses to prevent possible sample degradation. Time outside the freezer was limited t o <30 seconds per sample in order to further prevent degradation. The opaque covering wa s added to further prevent photolytic degradation when outside the freezer. Sample solutions were prepared by adding 60 L 6 M hydrochloric acid to 99.94 mL Milli-Q water to acidify to pH 4.5 (tested with Hydrion pH test paper). Acidified water was then removed in amounts corresponding to the amount of stock 100 mg/L triclosan/MeOH stock solution to be added. Stock so lution was then added to create the sample solution (while stirring, 500 RPM) to create 100 mL of sample solution. This procedure was used for all concentrations except 2 ng/L. Solutions of 2 ng/L were prepared by adding 10 mL of 20 ng/L solution to 90 mL acidified Milli-Q water. Sample solution concentrations of 2, 20, 50, 100, 3 00, 500, 750, 1000, and 2000 ng/L were made. Sample solution was kept at room temperature under dark conditions to prevent photolytic degradation while maintaining a constant testing temperature. This solution was stirred 30 seconds before transfer to the 22 mL vial to ensure sample homogeneity. Solution from the 100 mL beaker was added to a 22 m L vial containing a stir bar for a total solution volume of 21 mL. Each 100 mL batch o f solution yielded enough to create four 21 mL samples. Sample vials were stirred at 350 RPM for 30 second s before SPME fiber

PAGE 34

26 exposure. Stirring was maintained at 350 RPM during SPME exposure. Attempts at the 500 RPM stirring rate given by Canosa et al. (2005) were unsuccessful. At this speed, the magnetic stir bar caused the vial to vibrate and sp in. This motion was likely to increase variation and the possibility of spilling the sampl e, so the stirring was maintained at 350 RPM. Fiber adsorption was accomplished by exposing it t o the sample for 30 minutes at room temperature (23.7 C) (figure 11). Fiber expos ure depth was maintained at 1.1 cm, to prevent capillary action from depositing sample into the SPME needle. Sample depth was maintained with a stable syringe holding appara tus (Figure 11) custom-built for the experiment. Fiber exposure deviated from Canosa et al (2005), as the vapor pressure of triclosan (7x104 Pa @ 25 C) did not warrant headspace exposure fro m a septum capped vial. Any loss of triclosan through evaporation sho uld have been negligible, and variation in loss rates should have been small due to similar repetition conditions. After fiber exposure, it was removed from solution and retracted into the needle. The tip of the needle was dabbed with a Kimwipe to remove droplets. Some droplets remained attached to the fiber after retraction. Ho wever, removing droplets directly from the fiber with a Kimwipe caused an increase in sta ndard deviation on the sample tested (table 1). It also reduced sample delivery and peak area under the curve, resulting in smaller peaks. A second method of drying the SPME f iber was also attempted. Neither air drying at room temperature for 20 minutes nor dryin g for 20 minutes in an increased temperature environment appeared to decrease the va riability of adsorption. All samples in the standard curve were performed as described.

PAGE 35

27 Table 1: Area under the curve, including standard d eviation, dried and non-dried SPME fiber Sample Run 1 (Area) Run 2 (Area) Run 3 (Area) Run 4 (Area) Mean (Area) Std. Dev. (RSD) 50 ng/L (dried) 2347000 1797000 1492000 1601000 1806750 382866 (21. 19%) 50 ng/L (non-dried) 3343000 2872000 1927000 2111000 2585000 696188 (26. 93%) After exposure, the fiber was exposed to the headsp ace of a 2 mL autosampler vial with septa cap containing 30 L MTBSTFA at room tem perature for 10 minutes (Figure 12). The fiber was then retracted and taken directl y to the GC-MS. Figure 12: Headspace Derivatization in 2 mL septumcapped vial Analysis The fiber was desorbed into the injection port for 3 minutes under splitless mode at a temperature of 280 C. An additional secondary desorption was carried out for three minutes in the ECD injection port, which was not us ed for the final analysis. This was to prevent carryover of any undesorbed triclosan into the next analysis. Gas chromatography (GC) paired with an electron ca pture detector (ECD) and ion

PAGE 36

28 trap mass spectrometer (MS) were used to investigat e the most efficient detectors for determining triclosan concentrations in a solution. MS is superior to an ECD, especially when dealing with low concentrations. Initially, a calculation error occurred and the sol ution concentration was much smaller than believed (2 ng/L vs. 2000 ng/L). No pe ak was detectable using the electron capture detector, whereas mass spectrometry showed peak identification through spectral analysis. All concentrations in the standard curve were analyzed by mass spectrometry. The analytical method was identical to the method used by Canosa et al. (2005) with the exception of ion trap temperature. Injecto r temperature was set at 280 C in splitless mode. Column pressure was 8.7 psi (60 kPa ) using analysis-grade helium. The GC oven was programmed as follows: 3 minutes at 50 C, 10 C per minute to 260 C, 260 C held for 10 minutes (total run time: 34 minu tes). The GC-MS transfer line was set at 260 C. The ion trap was set at 200 C, a deviat ion from the published method of 220 C. Consultation with a Varian representative dete rmined that prolonged temperatures of 220 C may be harmful to the ion trap. Additiona l time at 200 C after analysis was determined sufficient to volatilize any substances that may have accumulated on the filament after analysis. The Varian representative believed that Canosa et al. (2005) ion trap temperature of 220 C removed analyte accumula tion.

PAGE 37

29 Quantification Triclosan was quantified using peak area integrati on. Peaks were identified by their ionization spectrum and m/z ratio. TCS Concentrations of >500 ng/L were characterized by the presence of a smaller secondar y peak (Fig. 13, peak B). This peak was always below 0.4% of the total volume of the ma in peak (Fig. 13, peak A). This area was not added to the main peak for any calculations as any effect on the results would be negligible. The cause of this secondary peak is unknown. The s pectrum is an exact version of all other triclosan peaks. It may be a form of peak tailing caused by the large amount of triclosan in the column.

PAGE 38

30 Figure 13: MS Chromatogram showing secondary peak a t 1000 ng/L concentration of triclosan. Peak A is t he main triclosan peak. Peak B is a secondary triclosan peak.

PAGE 39

31 Pitfalls in re-creating the method A number of problems occurred while trying to re-cr eate the method. The largest issue was due to my sample calculation error. The o riginal concentration tested using the ECD was only 2 ng/L, which is the published limit o f detection using the method developed by Canosa et al. (2005). Calculation errors led me to believe the s ample was 2000 ng/L. This resulted in an inordinate amount of time spent detecting triclosan in a noisy chromatogram. This miscalculation yielded val uable information, despite the excessive time used. At low concentrations, this method must be used wit h mass spectrometry. The SPME fiber’s polyacrylate coating released molecule s that created numerous peaks on the chromatogram, and produce higher baseline noise (Fi gure 14). This overshadows triclosan at levels near the detection limit.

PAGE 40

32 Figure 14: ECD chromatogram for blank desorption (3 minutes) of pre-conditioned polyacrylate SPME fibe r. All peaks are produced solely by the polyacrylate SPME fiber.

PAGE 41

33 Mass spectrometry is preferred because it enables peaks to be identified by their ionization spectrum. At much higher levels, it woul d be feasible to use an electron capture detector. At lower levels, identification o f triclosan becomes difficult or impossible due to the noise shown in figure 14. Another interesting difference was encountered in the elution time. In the published method, triclosan eluted at 24.25 minutes Though absorption time, fiber type, solution properties, column makeup, and GC-MS progr am were identical in this experiment, triclosan eluted at approximately 26.3 minutes (26.24 – 26.38 minutes). This initially made analysis difficult, as the SPME fibe r created a great deal of noise in the 24 minute region (figure 15). Canosa et al. (2005) did not report this difficulty. As shown in their chromatogram (figure 16), no other peaks eluted near the triclos an peak.

PAGE 42

34 Figure 15: ECD chromatogram for blank desorption (3 minutes) of pre-conditioned SPME fiber, zoomed to show noise near published elution time

PAGE 43

35 Figure 16: Elution Time and spectrum of derivatized triclosan, concentration unknown (from Canosa et al. 2005)

PAGE 44

36 Chapter 3: Results Canosa et al. (2005) published a moderately reproducible method for the detection and quantification of triclosan. Large differences in standard deviation were encountered, and were the main issue in reproduction. Other area s of the method were good to excellent. MTBSTFA was confirmed to be highly effec tive in derivatizing triclosan. An interesting spontaneous reaction in the column invo lving MTBSTFA may have been found. The linearity of the results was also excell ent. Quantification comparison based on peak area was impossible to analyze due to the lite rature's lack raw of data in that respect. Standard Deviation (SD) vs. Standard Error (SE) The most salient issue in reproducing this method is the standard deviation. Canosa et al. (2005) reported “Relative standard deviations of t he extraction– derivatization process, using samples spiked at 200 ng/l, ranged from 7 to 10% using a PA fibre” and “The variability of the extraction-de rivatization procedure remained below 12% for all compounds”. Triclosan's relative standa rd deviation is reported at 8.7%, assuming a 200 ng/L concentration (N = 3). This con centration is not explicitly stated in the table, but is referred to earlier in the text. Experimental values from my findings show a standard deviation at 100 ng/L and 300 ng/L to be 24.59% and 24.29%, respectively. This deviation is a significant departure from the data published by Canosa et al (2005). The conflict between the results published by Cano sa et al. (2005) and my analysis may be the result of a poor statistical kn owledge. If “relative standard deviation” is replaced with “standard error”, the results of t his paper correspond much more closely to those published by Canosa et al. (2005). At 100 ng/L, the standard error (N = 4) wa s

PAGE 45

37 12.30%, at 300 ng/L, 12.14%. These figures are much closer to the published data. The mean relative standard deviation (RSD) for all samp les (9 concentrations, N = 4) in my data was 18.66%. The mean relative standard error ( RSE) was 9.34%. Removing the two low concentrations (2 ng/L and 20 ng/L, RSD = 6.45% and 6.71%), the mean RSD increased to 22.11%. The mean RSE increased to 11.0 5%. Canosa et al. (2005) calculated the standard deviation only a si ngle concentration (200 ng/L, N=4). This is an enormous problem given the nature of the experiment. As shown in table 2, and figure 17 standard deviation and standard error (table 2, figure 18) varied with concentration.

PAGE 46

38 Figure 17: Triclosan standard curve showing standar d deviation nr nnrnn !"#

PAGE 47

39 Figure 18: Triclosan standard curve showing standar d error nr nnrnn !"#

PAGE 48

40 If the published paper mistakenly gave values for standard error and labeled them as relative standard deviation, this method is very repeatable. If the published values are, in fact, relative standard deviation, then this met hod needs to be re-confirmed. A mean relative standard deviation of 18.66% (22.11% at co ncentrations >20 ng/L) is approximately 215% greater than the 8.7% published variation (approx. 254% greater at concentrations >20 ng/L). This increased variabilit y does not invalidate the method, as it is still more convenient and rapid than the standar d solid phase extraction using cartridges. Lengthy extraction, washing, and evapor ation steps are eliminated with the Canosa et al. (2005) method. Derivatization All triclosan was fully derivatized in all concentr ations examined. MTBSTFA is an extremely aggressive and effective silylation re agent. Many derivatizers require lengthy exposures or heating, or activation. MTBSTF A does not require these. Complete derivatization was carried out in 10 minutes at roo m temperature. A possible spectrum corresponding to silylated tetradecanoic acid (Prim ary match in the NIST library), figure 19, was encountered in all samples at ~23.05 minutes It is unknown whether this is an inherent part of the fiber that is being silylated, or if reactions involving triclosan are taking place inside the column.

PAGE 49

41 Figure 19: Peak possibly corresponding to t-dimethy lsilyl ester of tetradecanoic acid

PAGE 50

42 Elution Time and Quantification Limits Elution time varied from Canosa et al. (2005), which placed elution time at ~24.25 minutes. Experimental data consistently place d triclosan elution time between 26.24 and 26.38 minutes. This created a number of d ifficulties, as mentioned in the “Pitfalls” section that were later resolved by usin g mass spectrometry. Detection limits corresponded to those published by Canosa et al. (2005), 2 ng/L (figure 20). In highly pure samples, the possibilit y of lower quantification limits may be realized with an increased exposure time. That poss ibility was not examined in this analysis, and could be the subject of further inqui ries.

PAGE 51

43 Figure 20: Triclosan 2 ng/L concentration chromat ogram and spectrum – See figure 8 for comparison

PAGE 52

44 To increase sensitivity at or below 2 ng/L, extrac tion time should be increased corresponding inversely to the amount of analyte in the solution. Figure 21 shows the results of increased extraction time with two diffe rent fibers. Both trials were carried out for two hours. Examining the graph shows that the s lope is >0 at two hours, which implies the feasibility of further extraction. The polyacrylate fiber at 2 hour shows peak area increases to 71.42%. It may be prudent to use a polydimethylsiloxanedivinylbenzene (PDMS-DVB) fiber for extraction time s greater than 60 minutes, as the sensitivity of that fiber is higher than that of th e PA at higher adsorption times.

PAGE 53

45 Figure 21: Relationship between extraction time and peak area (unknown concentration). Triclosan is re presented by the dotted line (From Canosa et al. 2005).

PAGE 54

46 Linearity of the Method Published linearity (R2) of the triclosan standard curve is 0.9999. Experi mental data yielded an R2 value of 0.9834 (figure 22). This linearity is exc ellent and is sufficient to quantify results in real-life data. The extreme difference in slope; however, should be noted. Experiments yielded a slope of 29x103 (figure 21), compared to the 84x105 given by Canosa et al (2005). If Canosa et al. (2005) had given a standard curve and raw data, more analysis of these differences could be underta ken. As it stands, only general hypotheses can be made. One possible hypothesis for this cause is that the mass spectrometer used by Canosa et al. (2005) is more sensitive than the Varian Saturn 2200.

PAGE 55

47 Figure 22: Linear regression analysis of triclosan standard curve, with standard deviation nr nnrnn n !"# "n $$

PAGE 56

48 Lack of Data in Canosa et al. (2005) The paper published by Canosa et al. (2005) lacked raw data. All data was presented in percentages; when presented. This decr eased the ability to compare experimental data and samples to the published find ings. No standard curve for any of the investigated compounds was presented Canosa et al. (2005). Peak areas were not included by Canosa et al. (2005) as they are in table 2. The concentrations to create the standard curve were also excluded from the publishe d article. In method creation work, data such as these (which in this paper is shown in table 2) are an invaluable addition to comparatively re-create published work. Quantitative comparison is rendered impossible with out these data. Even comparability of standard deviation is reduced. It would be a matter of simple arithmetic to determine their method of calculating their rela tive deviation, if a table presenting the peak areas of the analytes were available. Without peak areas, calculation methods and possible translation errors cannot be examined. A p resented standard curve and data would show differences in the calculated slope of t he regression analysis.

PAGE 57

49 Table 2: Peak Areas for all Concentrations, showing mean, standard deviation, and standard error Conc. (ng/L) Run 1 (Peak Area) Run 2 (Peak Area) Run 3 (Peak Area) Run 4 (Peak Area) Mean (Peak Area) Std. Dev. (%SD) Std. Err. (RSE) 2 327912 288830 298291 327340 310953.25 20045 (6.4 5%) 10022 (3.23%) 20 805339 881636 805222 919836 853008.25 57275 (6.7 1%) 28638 (3.36%) 50 2347000 1797000 1492000 1601000 1809250 380062 ( 21.01%) 190031 (10.5%) 100 2775000 2263000 3486000 4012000 3134000 770781 (24.59%) 770781 (12.3%) 300 11230000 6394000 8386000 10700000 9177500 22289 84 (24.29%) 1114492 (12.14%) 500 12080000 12880000 8391000 11180000 11137250 195 5300 (17.56%) 977649 (8.78%) 750 19550000 14160000 20370000 11730000 16452500 41 83255 (25.43%) 2091627 (12.71%) 1000 29620000 24980000 20960000 34360000 27480000 5 792938 (21.08%) 2896469 (10.54%) 2000 76020000 63250000 45020000 60650000 61235000 1 2726939(20.78%) 6363469 (10.39%)

PAGE 58

50 Chapter 4: Discussion Experimental Results Overall, the results were not consistent enough wi th the published findings for this to be considered a true confirmation of the method. At best, a translation error accounts for most of the differences in results. More likely the differences in stirring or septum use between the published method and the method used in this experiment caused significant variation increases. Slight variations in fiber pla cement between runs may also play a factor in this. At worst, the published method rel ied on insufficient sample number or selective data use. Despite this, the method remain s useful in certain applications. The causes of variation at concentrations >20 ng/L should be examined further. The lack of raw data in Canosa et al. (2005)’s paper reduces the ability to make meaningful comparisons or clear up uncertainties re garding calculations. An appendix showing peak areas and concentrations used to creat e the standard curve would increase the quality of this work. Further repetition of this method should be perfor med to validate the method or show conclusively that the observed variations are truly greater than those published. An appendix should also be made available by the autho rs showing the data collected from the experiments. Despite these issues, this method remains useful f or several reasons. First, the method is very convenient. Samples can be taken in remote locations where transporting several liters of sample would be inconvenient. The needles can be removed from the syringe and stored for analysis upon return to the laboratory. Secondly, it is quite rapid.

PAGE 59

51 No lengthy extraction, washing, and evaporation tim es are necessary as they are in solid phase extraction. Third, this method is very sensit ive. A concentration of only 2 ng/L can be readily detected with a fiber exposure time of 3 0 minutes. Sensitivity may well increase into the range of pg/L if adsorption time is increased. Statistical Ruminations A great deal of thought went into examining the dif ference between the experimental and published results (Canosa et al. 2005). It was decided to examine the possibility of alternative methods of calculating t he deviation. Two methods of calculation, aside from standard error, arose which gave results closer to those found in the paper. The first method used the population standard devi ation (Table 3) instead of straight standard deviation or sample standard devi ation. Population standard deviation is used in conjunction with a small sample size. Popul ation standard gave results slightly closer to, but still significantly higher than the published results. Using this method, the average relative standard deviation (RSD) is 16.16% vs. the 18.66% given by the standard deviation.

PAGE 60

52 Table 3: Peak areas for all concentrations using po pulation standard deviation Concentration (ng/L) Run 1 Run 2 Run 3 Run 4 Mean Pop. StDev. Pop. RSD 2 327912 288830 298291 327340 310593 17359 5.59 20 805339 881636 805222 919836 853008 49602 5.81 50 2347000 1797000 1492000 1601000 1809250 329143 1 8.19 100 2775000 2263000 3486000 4012000 3134000 667516 21.30 300 11230000 6394000 8386000 10700000 9177500 1930357 21.03 500 12080000 12880000 8391000 11180000 11132750 1693339 15.21 750 19550000 14160000 20370000 11730000 16452500 3622805 22.02 1000 29620000 24980000 20960000 34360000 27480000 5016832 18.26 2000 76020000 63250000 45020000 60650000 61235000 11021852 18.00 Another method explored involved finding outliers. Using Grubbs’ Test, which is a statistical equation used to determine outliers, yielded no outliers. Outliers were instead calculated as those values lying more than one stan dard deviation from the mean. These outliers were removed from the calculations, and a new mean and standard deviation was calculated. Results from this method were very clos e to those published in by Canosa et al. (2005) (Table 4). The mean relative standard devia tion was 11.38%. The RSD at 100 ng/L was 16.06%, and 14.97% at 300 ng/L. These valu es are much closer to the published error interval. If this calculation metho d was used in tandem with standard error, it would give a mean RSE of 5.69%. Percent e rror for 100 ng/L would be 8.03%, and for 300 ng/L 7.49%. These values are actually s lightly below the Canosa et al. (2005) variation of 8.7%

PAGE 61

53 Table 4: Peak areas for all concentrations, outlier s removed from calculation Concentration (ng/L) Run 1 Run 2 Run 3 Run 4 Mean Std. Dev % Dev 2 327912 Outlier 298291 327340 317847.667 16938.98475 5.33 20 805339 881636 805222 Outlier 830732.333 44083.90729 5.31 50 Outlier 1797000 1492000 1601000 1630000 154554.1 976 9.48 100 2775000 Outlier 3486000 Outlier 3130500 502752. 9214 16.06 300 11230000 Outlier 8386000 10700000 10105333.3 1512383.99 14.97 500 12080000 12880000 8391000 11180000 11132750 1955299.7 17.56 750 19550000 14160000 20370000 Outlier 18026666.7 3373637.997 18.71 1000 29620000 24980000 Outlier Outlier 27300000 3280975.465 12.02 2000 Outlier 63250000 Outlier 60650000 61950000 1838477.631 2.97 The validity of these calculations is questionable This is only to demonstrate that finished data can be misleading if no raw data is g iven. Possible Methods of Improvement There are a number of ways this method may be impr oved to increase sensitivity and decrease variability. First, the extraction tim e increases have shown increases in the peak area, due to the high affinity of triclosan to the SPME fiber. If extraction time is increased beyond 60 minutes, a PDMS-DVB may be used due to its increased rate of triclosan adsorption. Care must be taken to thoroug hly clean the PDMS-DVB fiber frequently using the manufacturer’s guidelines, as fiber sensitivity decreased with the number of injections, possibly due to pore blockage At a point past 120 minutes, it should be possible to extract all triclosan from a given sample or fully load the SPME fiber. Extracti ng all triclosan from the sample should eliminate the large amount of variation betw een peak areas at the same concentration. Changing the method of stirring may also decrease variability. Using a vibratory

PAGE 62

54 stirring apparatus may reduce the possibility of an effect in which the solution is not effectively flowing past the fiber, similar to that experienced in the eye of a hurricane. Vortex stirring caused by magnetic stir bars may al so cause some analyte to be removed from the fiber through shear forces generated by th e water. Further study should examine the effects of fiber placement and peak area variab ility. Triclosan Safety Triclosan continues to be one of the most popular household chemicals in industrialized countries. Some governments and inde pendent watch groups have reevaluated the safety profile triclosan and found it to be less benign (Samsoe-Petersen et al. 2003, EWG 2008) than its previous supporting resea rch indicated. Environmental problems, autoimmune diseases, agricultural issues, and even human health complications may all be impacted by prevalent tric losan use. As the research shows, aquatic environments are th e most impacted (Yang et al. 2008, Harada et al. 2008, Binelli et al. 2009). These environments often go overlooked by the layman, although scientists have found a gre at deal of information using them as biomarkers, which are increasingly used as early-wa rning systems in environmental toxicological studies. Continuing widespread and ca reless use of triclosan could significantly deplete microalgae, the basis of the aquatic food chain. This would obviously have a cascade effect that, in a worst-ca se global disaster scenario, could cause the breakdown of the entire aquatic ecosystem in ar eas where triclosan is released. This scenario is unlikely, but given the fragile ledge u pon which the environment hangs in today’s industrialized, overcrowded world, any figu rative straw could break the camel’s

PAGE 63

55 back. Triclosan also has serious effects on aquatic invertebrates and amphibians (Harada et al. 2008). Fish are seemingly little affected by tricl osan, but it may be that triclosan affects systems that have not been examined. Triclosan’s environmentally deleterious effects ar e not only limited to the water as it was originally patented as an herbicide. Similar fatty acid enzyme pathways exist in both plants and bacteria. Triclosan has been shown to reduce the growth of rice and cucumbers (Liu et al. 2008), and was demonstrated to be an extremely pot ent inhibitor of Arabidopsis Enoyl (acyl carrier protein) reductase (Dayan et al. 2008). While the most profound impacts are on the aquatic environment, perhaps the most concerning to humans are the effects triclosan has on mammals. Triclosan has been found to be a potent disruptor of ovine pregnancy h ormones, which are the same as human hormones (James et al. 2009). It affects rat testicular size (Kumar et al. 2009), and the function of human testosterone (Chen et al. 2007). It may also be linked to increases in autoimmune diseases such as asthma and atopic de rmatitis via the hygiene hypothesis. These are issues that should be of great import as they could directly and immediately impact human health. Another impact on human health is triclosan’s inte raction with bacteria. Triclosan has been found to act on a pathway which bacterial mutations can render ineffective (Dayan 2004). Developed resistance has been shown t o carry over to other common antibiotics in several species (Schweizer 2001). If the dose of triclosan is insufficient to kill the bacterium, the potential for increased dru g efflux pumps or altered enzymes rises. This is a common occurrence due to the recent trend of impregnating surfaces that have a high bacterial encounter rate with triclosan. This includes many kitchen cutting boards,

PAGE 64

56 athletic equipment, and children’s toys (EWG 2008). Triclosan slowly leeches from the material in doses high enough to be bacteriostatic, but too low to actually kill the organism. Conclusion Overall, this method is a convenient and rapid mean s of detecting triclosan in aqueous solutions. The variability of the experimen tal results contrasts significantly with the findings given by Canosa et al. (2005). Differences in methods and results should be further investigated. The lack of raw data makes re sult comparisons difficult. Several possibilities remain for further improvement of the precision and sensitivity of the method. The development of precise, sensitive, and rapid m ethods of detecting emerging environmental contaminants must be a high priority. The environmental, agricultural, and health impacts of triclosan are shown to be of grea t importance. Numerous chemicals are being introduced to consumers with little data on t heir environmental or chronic effects. These chemicals, like triclosan, are quite often on ly partly removed from waste water before being introduced to the environment. This me thod allows for the detection of environmentally pertinent chemicals. If the precisi on of this method can be confirmed or improved, it will be a valuable tool in the detecti on of possible environmental contaminants.

PAGE 65

57 Works Cited 2,4-Dichlorophenol. (8/ 4/1997). In Center for Disease Control & Prevention Retrieved March 1, 2009, from http://www.cdc.gov/niosh/ipcsneng/neng0438.html 2,4,6-Trichlorophenol. (11/25/1998). In Center for Disease Control & Prevention. Retrieved March 1, 2009, from http://www.cdc.gov/niosh/ipcsneng/neng1122.html Tests Confirm Yushchenko Poison. (2/6/2006). In BBC News. Retrieved November 21, 2008, from http://news.bbc.co.uk/2/hi/europe/5040378.stm Bester, K. (2003). “Triclosan in a sewage treatment process – balances and monitoring data”. Water Research, 37, 3891-3896. Bhargava, H.N., Leonard, P. (2006)“Triclosan: Appli cations and Safety”. American Journal of Infection Control, 24, 209-218 Binelli, A., Cogni, D., Parolini, M., Riva, C., Pro vini, A. (2009). “Cytotoxic and genotoxic effects of in vitro exposure to Triclosa n and Trimethoprim on zebra mussel (Dreissena polymorpha) hemocytes”. Comparative Biochemistry and Physiology Part C: Toxicology and Pharmacology, in press, corrected proof. Canosa, P., Morales, S., Rodrguez, I., Rub, E., Cela, R., Gmez, M. (2005) “Aquatic degradation of triclosan and formation of toxic ch lorophenols in presence of low concentrations of free chlorine”. Analytical & Bioanalytical Chemistry, 383 (7), 1119-1126. Canosa, P., Rodriguez, I., Rubi, E., Cela, R. (2005 ) “Optimization of solid-phase microextraction conditions for the determination o f triclosan and possible related compounds in water samples ”. Journal of Chromatography A, 1072, 107-115. Chen, J., Ahn, K.C., Gee, N.A., Gee, S.J., Hammock, B.D., Lasley, B.L. (2007) “Antiandrogenic properties of parabens and other p henolic containing small molecules in personal care products”. Toxicology and Applied Pharmacology, 221(3), 278-284. Da Costa Lima, R., Victora, C.G., Menezes, A.M.B., Barros, F.C.. (2003). “Do Risk Factors for Childhood Infections and Malnutrition P rotect Against Asthma? A Study of Brazilian Male Adolescents”. American Journal of Public Health, 93 (11), 1858-1864. Dayan, F.E., Ferreira, D., Wang, Y.H., Khan, I.A., McInroy, J.A., Pan, Z.. (2008). “A Pathogenic Fungi Diphenyl Ether Phytotoxin Targets Plant Enoyl (Acyl Carrier Protein) Reductase”. Plant Physiology, 147, 1062-1072.

PAGE 66

58 DiPetrillo, C., Philburn, R.T., Rovinsky, M.A., Mey er, B., Schneider, T., Peteroy-Kelly, M.A.. (2004). “Effects of the Antimicrobial Agent T riclosan on the Emergence of Antibiotic Resistance in Staphylococcus aureus”. BIOS, 75 (3), 109-117. Domestic Wastewater Facilities. In Florida Department of Environmental Protection Retrieved February 2, 2009, from ftp://ftp.dep.state.fl.us/pub/reports/wafr/wafrdw.x ls Wastewater Program. In Florida Department of Environmental Protection Retrieved May 5, 2009, from http://www.dep.state.fl.us/water/wastewater/index.h tm Eimco Water Technologies. (2007). “Bardenpho Proces s: Biological Nutrient Removal System”. Triclosan in Consumer Products. In Environmental Working Group. Retrieved on November 12, 2008, from http://www.ewg.org/node/26861 Federle, T.W., Kaiser S.K., Nuck, B.A. (2002) “Fate and effects of triclosan in activated sludge”. Environmental Toxicology and Chemistry, 21 (7), 1330–1337. Glaser, A. (2004) “The Ubiquitous Triclosan: A comm on antibacterial agent exposed”. Pesticides and You, 24 (3). Beyond Pesticides / National Coalition Against the Misuse of Pesticides. Heath, R.J., Rock, C.O.. (2000) “Microbiology: A tr iclosan-resistant antibacterial enzyme”. Nature, 406, 145-146 Heath, R.J., White, S.W., Rock, C.O. (2001) “Lipid biosynthesis as a target for antibacterial agents”. Progress in Lipid Research, 20, 467-497. Heidler, J., & Halden, R.U. (2007) “Mass balance as sessment of triclosan removal during conventional sewage treatment”. Chemosphere, 66, 362-360. History. In Ciba Retrieved on January 13, 2009, from http://www.ciba.com/ind-pc-triclosaninfo-101-histor y.htm Ishibashi, H., Matsumura, N., Hirano, M., Matsuoka, M., Shiratsuchi, H., Ishibashi, Y., Takao Y., Arizono, K.. (2004). “Effects of triclosa n on the early life stages and reproduction of medaka Oryzias latipes and inductio n of hepatic vitellogenin”. Aquatic Toxicology, 67, 167-179 James, M.O., Li, W., Summerlot, D.P., Rowland-Faux, L., Wood, C.E. (2009). “Triclosan is a potent inhibitor of estradiol and estrone sul fonation in sheep placenta”. Environment International, in press, corrected proof.

PAGE 67

59 Jones, R., Jampani, H.B., Newman, J.L., Lee, A.S.. (2000). “Triclosan: A Review of Effectiveness and Safety in Health Care Settings”. Journal of American Infection Control, 28, 184-196. Kapoor, M., Jamal Dar, M., Surolia, A., Surolia, N. (2001) “Kinetic Determinants of the Interaction of Enoyl-ACP Reductase from Plasmodium falciparum with Its Substrates and Inhibitors”. Biochemical and Biophysical Research Communications, 289 (4) 832-837. Kanetoshi A., Ogawa, H., Katsura, E., Kaneshima, H ., Miura, T. (1988). “Formation of polychlorinated dibenzo-p-dioxin from 2,4,4’-trichl oro-2’hyroxydiphenyl ether (Irgasan DP300) and its chlorinated derivati ves by exposure to sunlight”. Journal of Chromatography A, 442, 289-299. Kumar, V., Chakraborty, A., Kural, M.R., Roy, P.. ( 2009). “Alteration of testicular steroidogenesis and histopathology of reproductive system in male rats treated with triclosan”. Reproductive Toxicology, 27 (2), 177-185. Ling, L.L., Xian, J., Ali, S., Geng, B., et al. (2004). “Identification and characterization of inhibitors of bacterial enoyl-acyl carrier protein reductase”. Antimicrobial Agents and Chemotherapy, 48 (5), 1541-1547. Liu, F., Ying, G.G., Yang, L.H., Zhou, Q.X. (2009). “Terrestrial ecological effects of the antimicrobial agent triclosan”. Ecotoxicology and Environmental Safety, 72, 86-92. McAvoy, D.C., Schatowitz, B., Jacob, M., Hauk, A., Eckhoff, W.S.. (2002). “Measurement of Triclosan in Wastewater Treatment S ystems”. Environmental Chemistry 21 (7), 1323-1329. Mullooly, J.P., Schuler, R., Barrett, M., Maher, J. E.. (2007) “Vaccines, antibiotics, and atopy”. Pharmacoepidemiology and Drug Safety, 16 (3), 275-288. Russell, A.D.. (2004). “Whither Triclosan?”. Journal of Antimicrobial Chemotherapy, 53 (5), 693-695. Regs, J., Zak, O., Solf, R., Vischer, W.A., Weiric h, E.G.. (1979). “Antimicrobial spectrum of triclosan, a broad-spectrum antimicrobi al agent for topical application. II. Comparison with some other antimic robial agents”. Dermatologica 158( 1), 72-79. Samsoe-Petersen, L., Winther-Nielsen, M., Madsen, T .. (2003). “Fate and Effects of Triclosan” Danish EPA September 2003. Sanchez-Prado, L., Llompart, M., Lores, M., GarciaJares, C., Bayona, J.M., Cela, R..

PAGE 68

60 (2006). “Monitoring the photochemical degradation o f triclosan in wastewater by UV light and sunlight using solid-phase microextrac tion”. Chemosphere (65), 1338-1347. Schweizer, H.P.. (2001). “Triclosan: a widely used biocide and its link to antibiotics”. FEMS Microbiology Letters, 202, 1-7. Singer H., Muller S., Trixier C., Pillonel L. (2002 ) "Triclosan: occurrence and fate of a widely used biocide in the aquatic environment: fi eld measurements in wastewater treatment plants, surface waters, and lake sedimen ts". Environmental Science & Technology, 36 (23), 4998–5004 Stasinakis, A.S., Mamais, D., Thomaidis, N.S., Dani ka, E., Gatidou, G., Lekkas, T.D. (2008). “Inhibitory effect of triclosan and nonylp henol on respiration rates and ammonia removal in activated sludge systems”. Ecotoxicology & Environmental Safety, 70, 199-206. Stokstad, E.. (2007) "Species conservation. Can the bald eagle still soar after it is delisted?", Science, 316(5832), 1689-1690 Strachan, D.P.. (1989). “Hay fever, hygiene and hou sehold size”. British Medical Journal, 299 1259–1260. Thompson, A., Griffin, P., Stuetz, R., Cartmell, E. (2005) “The Fate and Removal of Triclosan during Wastewater Treatment”. Water Environment Research, 77 (1) Yang L.H.,Ying G.G., Su H.C., Stauber J.L., Adams M .S., et al. (2008) “GrowthInhibiting Effects of 12 Antibacterial Agents and t heir mixtures on the Freshwater Microalga Pseudokirchneriella Subcapitata”. Environmental Toxicology and Chemistry, 27 (5), 1201–1208 Ying, G.G., Yu, X.Y., Kookana, R.S.. (2007). “Biolo gical degradation of triclocarban and triclosan in a soil under aerobic and anaerobic co nditions and comparison with environmental fate modeling”. Environmental Pollution, 150, 300-305.


ERROR LOADING HTML FROM SOURCE (http://ncf.sobek.ufl.edu//design/skins/UFDC/html/footer_item.html)