MIAMI UNIVERSITY The Graduate School

Certificate for Approving the Dissertation

We hereby approve the Dissertation

of

Sandra J. Connelly

Candidate for the Degree:

Doctor of Philosophy

______Director Dr. Craig E. Williamson

______Reader Dr. Maria González

______Reader Dr. David L. Mitchell

______Graduate School Representative Dr. A. John Bailer ABSTRACT

EFFECTS OF ULTRAVIOLET RADIATION (UVR) INDUCED DNA DAMAGE AND OTHER ECOLOGICAL DETERMINANTS ON CRYPTOSPORIDIUM PARVUM, GIARDIA LAMBLIA, AND DAPHNIA SPP. IN FRESHWATER ECOSYSTEMS

Sandra J. Connelly

Freshwater ecosystems are especially susceptible to climatic change, including anthropogenic-induced changes, as they are directly influenced by the atmosphere and terrestrial ecosystems. A major environmental factor that potentially affects every element of an ecosystem, directly or indirectly, is ultraviolet radiation (UVR). UVR has been shown to negatively affect the DNA of aquatic organisms by the same mechanism, formation of photoproducts (cyclobutane pyrimidine dimers; CPDs), as in humans. First, the induction of CPDs by solar UVR was quantified in four aquatic and terrestrial temperate ecosystems. Data show significant variation in CPD formation not only between aquatic and terrestrial ecosystems but also within a single ecosystem and between seasons. Second, there is little quantitative data on UV-induced DNA damage and the effectiveness of DNA repair mechanisms on the damage induced in freshwater invertebrates in the literature. The rate of photoproduct induction (CPDs) and DNA repair (photoenzymatic and nucleotide excision repair) in Daphnia following UVR exposures in artificial as well as two natural temperate lake systems was tested. The effect of temperature on the DNA repair rates, and ultimately the organisms’ survival, was tested under controlled laboratory conditions following artificial UVB exposure. The results of these studies suggest a significant interaction of UVR and temperature on individual survival and ultimately population dynamics in freshwater systems. Lastly, freshwater human pathogens have negative effects ranging from gastrointestinal distress in otherwise healthy individuals to death in the immunocompromised and elderly. The control of infectious pathogens in water treatment is imperative. The abiotic and biotic environmental stressors of human pathogens are not well understood. Herein, solar radiation and artificial UVB are shown to significantly decrease the infectivity of Cryptosporidium parvum in vitro. The generalist filter feeder, Daphnia pulicaria, was shown to have significant effects on the viability, excystation, and infectivity of both Cryptosporidium parvum and Giardia lamblia under laboratory- controlled conditions. Both of these studies have significant implications for the natural control and potable water pretreatment approaches to human pathogen control.

EFFECTS OF ULTRAVIOLET RADIATION (UVR) INDUCED DNA DAMAGE AND OTHER ECOLOGICAL DETERMINANTS ON CRYPTOSPORIDIUM PARVUM, GIARDIA LAMBLIA, AND POTENTIAL ZOOPLANKTON GRAZERS IN FRESHWATER ECOSYSTEMS

A DISSERTATION

Submitted to the Faculty of Miami University in partial fulfillment of the requirements for the degree of Doctor of Philosophy Department of Zoology

by

Sandra J. Connelly Miami University Oxford, Ohio 2007

Dr. Craig E. Williamson, Miami University, Chair, Major Advisor Dr. Maria González, Miami University Dr. David L. Mitchell, MD Andersen Cancer Center, University of Texas Dr. James Oris, Miami University Dr. Michael Vanni, Miami University Dr. A. John Bailer, Miami University, Graduate School Representative

©

Sandra J. Connelly 2007

ii Table of Contents

LIST OF TABLES IV

LIST OF FIGURES IV

DEDICATION VI

ACKNOWLEDGEMENTS VII

CHAPTER 1: INTRODUCTION AND OVERVIEW 1

CHAPTER 2: ANNUAL AND SEASONAL VARIABILITY OF UV-INDUCED DNA DAMAGE IN A TEMPERATE ECOSYSTEM ASSESSED BY DOSIMETRY 15

CHAPTER 3: EXAMINATION OF UV-INDUCED DNA DAMAGE IN DAPHNIIDS: DNA DAMAGE VS. REPAIR AT DEPTH IN TWO LAKE SYSTEMS 37

CHAPTER 4: TEMPERATURE EFFECTS ON UV-INDUCED DNA DAMAGE AND ITS REPAIR: BIOLOGICAL RESPONSES OF FOUR SPECIES OF THE FRESHWATER CLADOCERAN DAPHNIA 61

CHAPTER 5: ARTIFICIAL UV-B AND SOLAR RADIATION REDUCE IN VITRO INFECTIVITY OF THE HUMAN PATHOGEN CRYPTOSPORIDIUM PARVUM 88

CHAPTER 6: IMPACT OF ZOOPLANKTON GRAZING ON THE EXCYSTATION, VIABILITY, OR INFECTIVITY OF THE PROTOZOAN PATHOGENS CRYPTOSPORIDIUM PARVUM AND GIARDIA LAMBLIA 108

CONCLUDING REMARKS 129

iii List of Tables

TABLE 2.1. AQUATIC IRRADIANCE MEASURES IN LAKES GILES AND LACAWAC (2005-2006)...... 24

TABLE 2.2. INCIDENT SOLAR IRRADIANCE (APRIL & JULY 2006)...... 24

TABLE 3.1. ABIOTIC MEASUREMENTS IN LAKES GILES AND LACAWAC

(APRIL – AUGUST 2006) ...... 45

TABLE 4.1. RATE OF REPAIR ± PHOTOREPAIR RADIATION IN DAPHNIA...... 86

TABLE 4.2. UVB RESPONSE PARAMETER COMPARISON OF DAPHNIA ...... 87

TABLE 5.1. SOLAR EXPOSURES OF CRYPTOSPORIDIUM PARVUM

(JULY & SEPTEMBER 2006) ...... 96

TABLE 5.2. IN VITRO INFECTIVITY OF CRYPTOSPORIDIUM PARVUM FOLLOWING UVB

AND SOLAR EXPOSURE...... 99

TABLE 6.1. CLEARANCE RATES OF CRYPTOSPORIDIUM PARVUM , GIARDIA LAMBLIA, AND

SELENASTRUM BY DAPHNIA PULICARIA ...... 119

TABLE 6.2. VIABILITY, EXCYSTATION, OR INFECTIVITY OF CRYPTOSPORIDIUM PARVUM

AND GIARDIA LAMBLIA POST GRAZING BY DAPHNIA PULICARIA ...... 121

iv List of Figures

FIGURE 1.1: SCHEMATIC OF DISSERTATION HYPOTHESES...... 5

FIGURE 2.1. TERRESTRIAL UV-INDUCED DNA DAMAGE (2005-2006) ...... 26

FIGURE 2.2. AQUATIC UV-INDUCED DNA DAMAGE (2005-2006) ...... 27

FIGURE 2.3. TERRESTRIAL NITRATE AND NITRITE ACTINOMETRY (APRIL & JULY 2006)... 29

FIGURE 2.4. AQUATIC NITRATE AND NITRITE ACTINOMETRY (APRIL & JULY 2006)...... 30

FIGURE 3.1. UVB LAMP EMISSION SPECTRUM ...... 43

FIGURE 3.2. SEASONAL AQUATIC UV-INDUCED DNA DAMAGE (APRIL – AUGUST 2006). 54

FIGURE 4.1. UVB LAMP AND PHOTOREACTIVE RADIATION (PRR) EMISSION SPECTRA...... 67 2 FIGURE 4.2.SURVIVAL OF DAPHNIA AT TWO TEMPERATURES (20 KJ/M UVR EXPOSURE).72

FIGURE 4.3. DNA DAMAGE INDUCTION IN DAPHNIA (20 KJ/M2 UVR EXPOSURE)...... 73

FIGURE 4.4. PHOTOPROTECTION IN DAPHNIA AT 10° AND 20°C ...... 74

FIGURE 4.5. DNA REPAIR RATES IN DAPHNIA AT 10° AND 20°C...... 75

FIGURE 4.6. NON-REPAIRABLE DAMAGE IN DAPHNIA ± PHOTOREPAIR RADIATION ...... 76

FIGURE 6.1. DAPHNIA PULICARIA GUT CONTENT (CRYPTOSPORIDIUM PARVUM AND GIARDIA

LAMBLIA) ...... 113

FIGURE 6.2. NAMARSKI CONTRAST IMAGE OF GIARDIA LAMBLIA, CRYPTOSPORIDIUM

PARVUM, AND SELENASTRUM USED IN DAPHNIA PULICARIA GRAZING EXPERIMENTS .. 114

FIGURE 6.3. PERCENT CHANGE IN EXCYSTATION, VIABILITY, OR INFECTIVITY OF GIARDIA

LAMBLIA OR CRYPTOSPORIDIUM PARVUM POST GRAZING BY DAPHNIA PULICARIA ...... 122

v Dedication

To all those who have shared in this journey – we made it!

This final compilation of many years of work, which was full of both great achievements and heart-wrenching failures, was only possible through the support, compassion, and unending love of my family, my boyfriend, and the amazing network of friends that I have amassed over the years. My mother and father have taught me that only through hard work and persistence will we ever prove to ourselves how great we can be. My siblings have taught me that no matter the problems that life throws at you, it is nothing that a little ice cream can’t fix! My boyfriend has taught me that anything can be overcome if you are passionate. My friends have taught me that life is not life without the love of friends and that when all else fails, friends will never fail you. This dissertation is dedicated to these people, and so many more. Seemingly insurmountable mountains were no more than pebbles in the road because of you – thank you.

vi Acknowledgements

The work herein would never have been possible without the guidance, support, and undying commitment of my major advisor, Dr. Craig Williamson, my DNA damage mentor, Dr. David Mitchell, and my pathogen mentor, Dr. Kristen Jellison. They have taught me so much and helped me grow not only as a scientist and teacher, but as a person. Thank you.

Thanks to my doctoral committee and the Department of Zoology at Miami for being so amenable during my transfer between Universities.

A special thanks to the members of the Williamson Ecology of UV Lab, past and present, especially Dr. Robert Moeller, Erin Overholt, and Dr. Gaby Dee for both the positive and negative critiques through the years! Without them both, I would have never made it this far!

Thanks to the students and technicians that have been a driving force in my research, both in the lab and the field, through the years.

Thanks to the undergraduates who have molded me in to the mentor I am today, especially Katie Dieter and Stephen Gerhard. Their enthusiasm and drive could motivate any one to be better!

vii

Chapter 1

Introduction and overview

1 Ultraviolet radiation (UVR) plays a critical role in the formation of cancer in humans, especially skin cancer (Agar et al., 2004; Douki and Cadet, 2001; Wallin, 2000). UVR can directly damage the DNA of aquatic organisms by the same mechanism as in humans. This damage in aquatic organisms presumably results in protective physiology and behavior that minimizes genomic damage and death (De Lange and Van Reeuwijk, 2003; Grad et al., 2001; Ravanat et al., 2001; Witkin, 1995; Zellmer, 1995). Protection from UVR may be accomplished by several means: behavioral avoidance of UVR, increased levels of UVR-absorbing molecules [melanin, carotenoids, or mycosporine-like amino acids (MAA’s)], or repair of the damaged DNA. All of these options are physiologically costly to the organism (Mitchell and Karentz, 1993). All three basic mechanisms of UVR protection have been shown in aquatic organisms. Behavioral avoidance and increased levels of UVR filtering compounds are some of the more thoroughly studied mechanisms. Freshwater zooplankton alter their vertical migration patterns to avoid intense UVR exposure (Leech and Williamson, 2001). Rhode et al (2001) showed a direct relationship between the exposure of Daphnia sp. to UVR and their depth of vertical migration. In addition, they reported an inverse relationship between this response to UVR exposure and the level of pigmentation that the species possess. Increased melanization levels in Daphnia sp. with increased UVR exposure has been shown repeatedly in the literature (e.g., Rautio and Korhola, 2002; Hurtubise et al., 1998). These two protection mechanisms can be very costly to the organism and make it more susceptible to predation by altered migration patterns and increased coloration, concomitantly increasing the visibility to predators and affecting the survival rate of the zooplankton species (Morgan and Christy, 1996). More recent studies have shown the potential of zooplankton to repair UV induced damage, such as cyclobutane pyrimidine dimers (CPDs). Grad et al (2001) reported variations in the abilities of different freshwater species to utilize photolyase mechanisms have been shown. Photoenzymatic repair (PER) is driven by the presence of the photolyase enzyme in an organism and is activated by exposure of the organism to longer wavelength light. PER is ubiquitous in the plant and animal kingdoms with only placental mammals lacking the required photolyase enzyme (Sancar and Sancar, 1988). PER is a simple repair mechanism involving the photolyase and a small number of

2 cofactors, in addition to the energy associated with UVA/blue light absorption, to specifically remove UV photoproducts. While PER is specific to UV damage and is thought to be metabolically benign (use of light energy as opposed to cellular ATP), it is likely not the most important repair mechanism for organisms due to the light energy requirement. Nucleotide excision repair (NER) is likely the most important DNA repair mechanism associated with the removal of UV-induced DNA damage. There are two major NER pathways, transcription coupled repair (TCR) and global genome repair (GGR), that remove pyrimidine dimers and replace the damaged site with a “patch” that is approximately 29 nucleotide bases in length. TCR removes DNA damage from transcribed genes, whereas GGR repairs damage in nontranscribed regions. While these pathways repair DNA damage at different rates, in general CPD photoproducts are removed much more slowly than other UV photoproducts, believed to occur because other photoproducts highly distort the DNA. GGR is regulated indirectly by p53 in the cells, but both are considered to be metabolically costly to the organism, using ATP for strand correction (Cleaver and Mitchell, 2006). Variability in organism survival following UV exposure seems to be a function of both repair mechanisms and temperature (Williamson et al., 2001). Temperature has also been shown to influence the induction of the DNA damage by UVR (MacFadyen et al., 2004). Further, the functional success of these damage avoidance and repair mechanisms has been shown to vary directly with changes in their environment, such as increased dissolved organic carbon (DOC) and the presence or absence of predators. Therefore, it is critical to obtain a better understanding of how variations in UVR across environments may impact the species survival and population dynamics. Seasonal cycles, natural or anthropogenically driven, of temperature and precipitation affect not only the terrestrial ecosystems, but also aquatic ecosystems (ecosystem interactions). Terrestrial and aquatic systems are linked through carbon transport. With increased temperatures and precipitation, there is increased decomposition of leaf litter and other carbon sources in the terrestrial ecosystem. With increased precipitation, this decomposed organic matter is transported to aquatic ecosystems and the carbon is leached into the water as chromophoric dissolved organic

3 matter (CDOM). This CDOM acts to color the water, alters the UVR attenuation in the water column, and indirectly affects (potentially decreasing) the DNA damage induced by UVR exposure (e.g., Williamson et al., 1996; Wetzel, 2001; Hargreaves, 2003). There are three major components to this research: UVR induced DNA damage in temperate ecosystems, the role of DNA repair mechanisms in zooplankton following exposure to UVR, and the effects of UVR and zooplankton grazers on human pathogens. Figure 1.1 illustrates the hypotheses proposed herein which are focused on Cryptosporidium parvum, Giardia lamblia, Daphnia spp., and the effects of abiotic and/or biotic stressors on their survival. DNA damage is considered to be an experimentally uncontrollable factor in all organisms; it is induced through direct effects of environmental conditions (in this case, UVR) and cannot be manipulated independently. However, this study will address the direct effect of UVR on DNA damage of individual organisms under otherwise controlled conditions. These studies consider direct effects of UVR on temperate ecosystems (DNA dosimetry), Daphnia spp., and the human pathogen Cryptosporidium parvum. In addition, the direct effect of Daphnia pulicaria, a freshwater zooplankton filter-feeder, on Cryptosporidium parvum and Giardia lamblia has also been tested.

4

Human Impacts & Climate Change

- +/-

Ozone Depletion DOC Quantity & Quality + - +/- Solar UV or Equivalent Zooplankton Grazers II III / IV - (-)

UV-induced Damage VI ? V

- ?

Cryptosporidium parvum & Giardia lamblia

Figure 1.1: The hypotheses are labeled as they are presented as chapters herein (roman numerals). Known direct and indirect effects are labeled as being positive or negative on the resulting component. Effects that are unknown but hypothesized to be negative are indicated by “-?”. Effects that are unknown, and too little is known to imply a positive or negative effect are indicated by”?”.

5 Cryptosporidium spp. and Giardia lamblia, two freshwater protozoan pathogens, cause acute gastrointestinal disease (cryptosporidiosis and giardiasis, respectively) in humans which can be prolonged and potentially fatal for people with compromised immune systems. Both pathogens reproduce in the intestinal (or gastrointestinal) tract of multiple organisms, including humans, dogs, cats, cattle, deer, geese, reptiles, and many others. The reproduction of the pathogen causes intestinal distress (diarrhea) even in people with healthy immune systems. Although very healthy people may experience no distinct symptoms of cryptosporidiosis or giardiasis, they can still be infected and propagate the pathogens until they are able to ward off the infection (ca. 2 weeks for cryptosporidiosis and potentially several months, or longer, for giardiasis). Following reproduction in the host, the Cryptosporidium spp. oocysts and Giardia spp. cysts are shed in human waste and may potentially contaminate the food and/or water supply. The typical disinfectant steps (chlorination and filtration) in water treatment systems do not completely inactivate either pathogen in the potable water supply, leading to occurrences of disease outbreaks due to contaminated drinking water even in developed countries. Perhaps the most familiar is the Milwaukee outbreak in 1993 when 400,000 people were infected with cryptosporidiosis (Centers for Disease Control, Atlanta, GA, USA). While filtration can be effective against Giardia spp., outbreaks of giardiasis still occur. Giardia is reported by the World Health Organization as being the chief cause of gastrointestinal distress worldwide. Cattle, deer, and other wildlife shed these pathogens in the same manner directly to the environment. Rainwater washes the pathogen from fields and forests into neighboring streams and rivers. From this, it is easy to assume that in such cases the water the cattle and wildlife drink from the stream is contaminated with the pathogens. Cryptosporidiosis, in particular, is a serious issue for the health of livestock herds and may increase the susceptibility of the animals to other diseases that are transmittable to humans. It is critical that we fully understand the effects of the environment on these pathogens; such factors may ultimately enhance our ability to control the propagation of these pathogens even before water treatment. There have been several papers published on the use of UVR as a disinfectant of Cryptosporidium spp. and Giardia spp. in water treatment (Zimmer et al., 2003; Linden

6 et al., 2001; Lorenzo-Lorenzo et al., 1993). This research has focused primarily on the removal/inactivation of viable (oo)cysts in the potable water supply. The maximum disinfection rate at specific intensities can be determined from dose response curves generated for the specific UVR lamps. Low pressure (LP), high output mercury based UVR lamps with a monochromatic output (254 nm) provide the greatest disinfection rates with the lowest temperature variation (bulb temperature 100°C). Medium pressure (MP) polychromatic UVR lamps, produce lower disinfection rates because they are not wavelength specific (less efficient disinfection) and require a tenfold increase in power (increasing the temperature variation; bulb temperature 600 – 900°C) (Clancy et al., 2004). MP lamps also encompass a greater portion of the spectrum (185 – 1367 nm) but do not provide the specificity needed to address the most damaging UVR at the Earth’s surface (UVB: 280 – 315 nm) (Wetzel, 2001). Low doses of UVR light using LP or MP mercury lamps potentially inactivate Cryptosporidium spp. in the lab (Craik et al., 2001). These data are very specific to the water treatment industry and give little insight into the broader picture of natural UVR damage to Cryptosporidium spp. when exposed to natural solar radiation. Almost all of these data have been determined using UVC (200-280nm) disinfection lamps, and do not consider any of the longer wavelengths present in nature. If it can be shown that naturally occurring wavelengths of UVR render the oocysts non-infectious at intensities far less than those currently used in disinfection, there could be innumerable applications in both watershed management and the water treatment industry. Presence does not equate to infectiousness of Cryptosporidium spp. or Giardia spp. (Rochelle and De Leon, 2001; Slifko et al., 1997; Fayer et al., 1996; Fayer, 1994). Numerous studies report that not all viable (living) pathogens are infectious to humans or other hosts. It is critical to test these pathogens for excystation (i.e., the ability to release sporozoites), viability [outer wall of (oo)cyst is intact], and infectivity (i.e., the ability to reproduce and cause infection sites in the host or in vitro) following any experimental treatment. The only exception is that is has been shown repeatedly in the literature that in vitro Cryptosporidium spp. oocyst infectivity levels following exposure to UVR disinfection levels are habitually lower than any other measure (e.g., excystation or viability) of the same oocysts (Oguma et al., 2001; Slifko et al., 1999). The effect of

7 UVR on the Cryptosporidium spp. oocysts was tested by the infectivity assay only because it provides the most accurate information following treatment. The hypotheses presented herein test the effects of abiotic or biotic environmental stressors on Cryptosporidium parvum, Giardia lamblia, Daphnia spp. In addition, direct quantification of UV-induced DNA damage and repair in Daphnia spp. was assessed in the field and lab. The hypotheses tested are: 1. A. There is a significant increase in photoproduct concentrations (CPDs) in systems with increased UVR exposure (surface of aquatic systems, low DOC aquatic systems, open terrestrial systems, etc.). B. There is a significant change in photoproduct concentration in systems with increased seasonal UVR exposure. 2. Zooplankton exposed to natural radiation are capable of repairing DNA damage induced by an artificial UVB radiation device. Repair is defined as the number of CPDs removed by photoenzymatic (light) or dark repair mechanisms in a given period of time. 3. Zooplankton exposed to acute levels of UVR in the laboratory will repair CPDs by photoenzymatic (light) or dark repair mechanisms. 4. Artificial UVB and naturally occurring levels of solar radiation will have a negative effect on the infectivity of Cryptosporidium parvum 5. Grazing by Daphnia pulicaria will have a direct negative effect on the excystation, viability, and/or infectivity of Cryptosporidium parvum oocysts and Giardia lamblia cysts.

The chapters herein are presented as independent experiments in a self-contained manuscript-style format. The induction of DNA damage by solar and artificial radiation is discussed in Chapters 2 (ecosystems), 3 (Daphnia field experiment), 4 (Daphnia laboratory experiment), and 5 (Cryptosporidium field and laboratory experiment). The effect of a biological grazer, Daphnia pulicaria, on the human pathogens Cryptosporidium parvum and Giardia lamblia is presented in Chapter 6.

8 References

Agar, N.S., G.M. Halliday, R.StC. Barnetson, H.N. Ananthaswamy, M. Wheeler, and A.M. Jones. 2004. The basal layer in human squamous tumors harbors more UVA than UVB fingerprint mutations: A role for UVA in human skin carcinogenesis. Proceedings of the National Academy of Sciences. 101(14): 4954-4959.

Balseiro, E.G., B.E. Modenutti, and C.P. Queimalinos. 2001. Feeding of Boeckella gracilipes (Copepopda, Calanoida) on ciliates and phytoflagellates in an ultraoligotrophic Andean lake. Journal of Plankton Research. 23(8):849-857.

Brookes, J.D., J. Antenucci, M. Hipsey, M.D. Burch, N.J. Ashbolt, and C. Ferguson. 2004. Fate and transport of pathogens in lakes and reservoirs. Environmental International. 30:741-759.

Burns, C.W. and J.J. Gilbert. 1993. Predation on ciliates by freshwater calanoid copepods: rates of predation and relative vulnerability of prey. Freshwater Biology. 30: 377-393.

Campbell A.T., W. Robertson, and H.V. Smith. 1992. Viability of Cryptosporidium parvum oocysts: correlation of in vitro excystation with inclusion or exclusion of flurogenic vital dyes. Applied and Environmental Microbiology. 58(11):3488-3493.

Clancy, J.L., M.M. Marshall, T.M. Hargy, and D.G. Korich. 2004. Susceptibility of five strains of Cryptosporidium parvum oocysts to UV light. Journal of the American Water Works Association. 96(3): 84-93.

Cleaver, J.E. and D.L. Mitchell. 2006. Ultraviolet radiation carcinogenesis. In: Cancer Medicine: 7th Edition. BC Decker, London, UK. 303-311.

Craik, S. A., D. Weldon; G.R. Finch, J.R. Bolton, and M. Belosevic. 2001. Inactivation of Cryptosporidium parvum oocysts using medium- and low-pressure ultraviolet radiation. Water Research. 35: 1387-1398.

9 De Lange, H.J. and P.L. Van Reeuwijk. 2003. Negative effects of UVB-irradiates phytoplankton on life history traits and fitness of Daphnia magna. Freshwater Biology. 48: 678-686.

Douki, T. and J. Cadet. 2001. Individual determination of the yield of the main UV- induced dimeric pyrimidine photoproducts in DNA suggests a high mutagenicity of CC photolesions. Biochemistry. 40(8): 2495-2501.

Fayer, R. 1994. Effect of high temperature on infectivity of Cryptosporidium parvum oocysts in water. Applied and Environmental Microbiology. 60(8): 2732-2735.

Fayer, R. and T. Nerad. 1996. Effects of low temperatures on viability of Cryptosporidium parvum oocysts. Applied and Environmental Microbiology. 62(4): 1431-1433.

Fayer, R., J.M. Trout, E. Walsh, and R. Cole. 2000. Rotifers ingest oocysts of Cryptosporidium parvum. Journal of Eukaryotic Microbiology. 47:161-163.

Grad, G., C.E. Williamson, and D. Karapelou. 2001. Zooplankton survival and reproduction responses to damaging UV radiation: A test of reciprocity and photoenzymatic repair. Limnology and Oceanography. 46(3): 584-591.

Haney J.F. and M.A. Trout. 1985. Size selective grazing by zooplankton in lake Titicaca. Archives Hydrobiologia. 21:147-160.

Hurtubise, R.D., J.E. Havel, and E.E. Little. 1998. The effects of ultraviolet-B radiation on freshwater invertebrates: Experiments in a solar stimulator. Limnology and Oceanography. 43(6): 1082-1088.

Leech, D.M. and C.E. Williamson. 2001. In situ exposure to ultraviolet radiation alters the depth distribution of Daphnia. Limnology and Oceanography. 46(2): 416-420.

10 Linden, K. G., G.A. Shin, and M.D. Sobsey. 2001. Relative efficacy of UV wavelengths for the inactivation of Cryptosporidium parvum. Water Science and Technology. .43: 171-174.

Lorenzo-Lorenzo, M. J., M.E. Aresmazas, I.V.M. Dematurana, and D. Duranoreiro. 1993. Effect of Ultraviolet Disinfection of Drinking-Water on the Viability of Cryptosporidium-Parvum Oocysts. Journal of Parasitology. 79(1): 67-70.

MacFadyen, E.J., C.E. Williamson, G. Grad, M. Lowery, W.H. Jeffrey, and D.L. Mitchell. 2004. Molecular response to climate change: temperature dependence of UV- induced DNA damage and repair in the freshwater crustacean Daphnia pulicaria. Global Change Biology. 10: 408-416.

Mitchell D.L. and D. Karentz. 1993. “The induction and repair of DNA photodamage in the environment.” In: Environmental UV Photobiology. Plenam Press, New York. 345- 377 pp.

Mitchell, D.L., J. Meador, L. Paniker, D. Gasparutto, W.H. Jeffrey, and J. Cadet. 2002. Development and application of a novel immunoassay for measuring oxidative DNA damage in the environment. Photochemistry and Photobiology. 75(3): 257-265.

Morgan S.G. and J.H. Christy 1996. Survival of marine larvae under the countervailing selective pressures of photodamage and predation. Limnology and Oceanography. 41(3): 498-504.

Oguma, K., H. Katayama, H. Mitani, S. Morita, T. Hirata, and S. Ohgaki. 2001. Determination of pyrimidine dimers in Escherichia coli and Cryptosporidium parvum during UV light inactivation, photoreactivation, and dark repair. Applied and Environmental Microbiology. 67: 4630-4637.

Rautio, M. and A. Korhola. 2002. UV-induced pigmentation in subarctic Daphnia. Limnology and Oceanography. 47(1): 295-299.

11 Ravanat, J-L., T. Douki and J. Cadet. 2001. Direct and indirect effects of UV radiation on DNA and its components. Journal of Photochemistry and Photobiology B: Biology. 63: 88-102.

Rhode, S.C., M. Pawlowski, and R. Tollrian. 2001. The impact of ultraviolet radiation on the vertical distribution of zooplankton of the genus Daphnia. Nature. 412: 69-72.

Rice, E.W. and F.W. Schaefer III. 1981. Improved in vitro excystation procedure for Giardia lamblia cysts. Journal of Clinical Microbiology. 14:709-710.

Robertson, L.J., A.T. Campbell, and H.V. Smith. 1993. In vitro excystation of Cryptosporidium parvum. Parasitology. 106(Pt 1):13-19.

Rochelle, P.A. and R. De Leon. “A review of methods for assessing the infectivity of Cryptosporidium parvum using in-vitro cell culture.” In: Cryptosporidium: The analytical challenge. pp. 88-95. Royal Society of Chemistry. Cambridge. 2001.

Rochelle, P.A., M.M. Marshall, J.R. Mead, A.M. Johnson, D.G. Korich, J.S. Rosen, and R. De Leon. 2002. Comparison of in vitro cell culture and mouse assay for measuring infectivity of Cryptosporidium parvum. Applied and Environmental Microbiology. 68(8): 3809-3817.

Sancar, A. and G.B. Sancar. 1988. DNA repair enzymes. Ann. Rev. Biochem. 57: 29-68.

Slifko, T.R., D. Friedman, J.B. Rose, and W. Jakubowski. 1997. An in vitro method for detecting infectious Cryptosporidium oocysts with cell culture. Applied and Environmental Microbiology 63(9):3669-3675.

Slifko, T.R., D.E. Huffman, and J.B. Rose. 1999. A most-probable number assay for enumeration of infectious Cryptosporidium parvum oocysts. Applied and Environmental Microbiology. 65(9):3936-3941.

12 Stott R., E. May, E. Ramirez, and A. Warren. 2003. Predation of Cryptosporidium oocysts by protozoa and rotifers: implications for water quality and public health. Water Science and Technology. 47(3): 77-83.

Thys, I., Leporcq, B. and J. Descy. 2003. Seasonal shifts in phytoplankton ingestion by Daphnia galeata, assessed by analysis of marker pigments. Journal of Plankton Research. 25(12): 1471-1484.

Upton, S., M. Tilley, M.V. Nesterenko, and D.B. Brillhart. 1994. A simple and reliable method of producing in vitro infections of Cryptosporidium parvum (Apicomplexa). FEMS Microbiology Letters. 118: 45-50.

Upton, S., M. Tilley, and D.B. Brillhart. 1995. Effects of select medium supplements on in vitro development of Cryptosporidium parvum in HCT-8 cells. Journal of Clinical Microbiology. 33(2): 371-375.

Wallin, H. 2000. A molecular pathway for UV-induced CC to TT mutations. Mutation Research. 447: 317-318.

Wetzel, R.G. 2001. Limnology: Lake and River Ecosystems, 3rd Edition. Academic Press.

Williamson, C.E., G. Grad, H. De Lange, and S. Gilroy. 2002. Temperature-dependent ultraviolet responses in zooplankton: Implications of climate change. Limnology and Oceanography. 47(6): 1844-1848.

Witkin, E. 1969. Ultraviolet-induced mutation and DNA repair. Annual Review of Genetics. 3: 525-552.

Zellmer, I.D. 1995. UV-B-tolerance of alpine and arctic Daphnia. Hydrobiologia. 307: 153-159.

13 Zimmer, J. L.; R.M. Slawson, and P.M. Huck. 2003. Inactivation and potential repair of Cryptosporidium parvum following low- and medium-pressure ultraviolet irradiation. Water Research. 37: 3517-3523.

14

Chapter 2

Annual and seasonal variability of UV-induced DNA damage in a temperate ecosystem assessed by dosimetry

To be submitted: Sandra J. Connelly, Craig E. Williamson, Bruce R. Hargreaves, Jesse Phillips-Kress, Patrick J. Neale, and David L. Mitchell. Annual and Seasonal Variability of UV Induced DNA Damage in a Temperate Ecosystem Assessed by Dosimetry. To Global Change Biology.

15 Abstract

Ecosystems are dynamic entities, influenced significantly by time of day, season, anthropogenic forcing, and climate change. Light is an environmental stress, either in excess or deficit. Solar ultraviolet (UV)B radiation (280-315 nm) is the most biologically relevant solar irradiance because of its direct absorption by the DNA molecule, causing mutations, cell death, and organism distress. Assessing the variability in solar radiation to better understand these potentially harmful effects of UVB across ecosystems is difficult. Point measurements using radiometers are expensive, highly subjective (placement of radiometer, duration of logging, etc.) and to infer DNA damage, data must be modeled, correlating measured irradiance to potential biological effects. Herein we test two inexpensive alternatives to standard radiometry in an attempt to better understand this system irradiance variability and its biological effects. These studies were conducted in Northeastern Pennsylvania using DNA dosimetry and nitrate/nitrite actinometers. We quantified the variability in ultraviolet (UV)-A and -B solar radiation using actinometry in six temperate ecosystems: four terrestrial (hemlock and deciduous forests, open and tall grass fields) and two aquatic (high and low UVR transparency lakes). Further, we used DNA dosimetery to quantify the potential DNA damage as cyclobutane pyrimidine dimers (CPDs) per megabase of DNA by radioimmunoassay. In these temperate ecosystems, DNA damage was consistently greater in deciduous forests than hemlocks. In July, we observed higher levels of DNA damage at 0.5 m in the more ultraviolet (UVR) transparent lake (333 CPDs) than in the hemlock (155 CPDs) or deciduous forests (268 CPDs). From April to July, the open field was the only ecosystem that showed a significant increase in UVB:UVA (0.192 to 0.297), while the others remained nearly constant. This spatial and seasonal variability has significant implications for UVR stress in light of anthropogenic forcing and climate change.

16 Introduction

Solar radiation is variable in nature, both within and among ecosystems (e.g., Flint and Caldwell, 1998; Brown et al., 1994; Williamson et al., 2001). These systems are in a constant state of flux, with anthropogenic forcing directly influencing all ecosystems (e.g., Brovkin et al., 2004). Successive changes in both aquatic and terrestrial systems alter habitat structure and function, directly impacting all its inhabitants, both plant and animal species. Within terrestrial systems, the ratio of damaging to beneficial radiation is highly variable, with the ultraviolet UVB to photosynthetically active radiation ratio (UVB:PAR) fluctuating daily and seasonally within any given system. Shaded areas, including forested canopies with gaps, are relatively rich in UVB (305nm), although this pattern is much more predominant in shaded clearings as compared to shaded forest gaps (Flint and Caldwell, 1998). Shaded gaps (40m2; <10% open sky) have very little diffuse radiation, UVB or PAR; however, these areas are exposed to columns of light, typically during peak solar hours, but with lower UVB:PAR than open spaces. Shaded portions of these gaps experience UVB:PAR on the order of 2-3 times that of open spaces, similar to that of deep forested spaces (Flint and Caldwell, 1998). This suggests that fragmented forests, resulting from clear cutting or natural stand loss (e.g., disease), have extreme variability in UV exposure in relatively small habitat areas. The ratio of UVB:PAR determines the effect of solar exposure on plant species. While PAR, by definition, is critical to plant function, high UVB exposure can induce physiological damage to the plants, as with all species. Plant species in areas of high variability, such as the gaps described by Flint and Caldwell (1998) were thought to be disrupted by the UVB:PAR ratio only if damage to the photosynthetic processes was induced. Barnes et al (1996) report significant morphological changes to plant species with UVB exposure than that which induces damage to the primary photosynthetic pathways. The plants remaining in these gaps typically are most similar to understory plants, they are not adapted to intense light, or the increased UVB:PAR. While damage to metabolic pathways would be imminently detrimental to the plant, damage that alters the plant morphology suggests longer-term effects, those of which might not be immediately

17 apparent (e.g., biomass reduction, susceptibility to pests and disease, ecosystem composition shifts, etc.). Seasonal cycles, natural or anthropogenically driven, of temperature and precipitation affect not only the terrestrial ecosystems, but also aquatic ecosystems (ecosystem interactions). Terrestrial and aquatic systems are linked through carbon transport. With increased temperatures and precipitation, there is increased decomposition of leaf litter and other carbon sources in the terrestrial ecosystem. With increased precipitation, this decomposed organic matter is transported to aquatic ecosystems and the carbon is leached into the water as chromophoric dissolved organic matter (CDOM). This CDOM acts to color the water, alters the UVR attenuation in the water column, and indirectly affects (potentially decreasing) the DNA damage induced by UVR exposure (e.g, Williamson et al., 1996; Wetzel, 2001; Hargreaves, 2003). With development in and around forested and aquatic systems come increased anthropogenic inputs, such as agricultural and domestic runoff, containing high levels of nutrients and possibly disease-causing agents, such as pathogens. Changes in both terrestrial and aquatic systems alter the persistence of infectious pathogens in nature. Cryptosporidium and Giardia are two common human pathogens that are propagated by wildlife, domesticated animals, and humans. These pathogens are intestinally passed from the host and may enter both terrestrial and aquatic ecosystems. Fecal matter from deer and geese are common sources for these pathogens. The feces deposited in forests and open fields by these animals are transported to lakes and reservoirs by surface runoff. Others are deposited directly into the waters. These pathogens are disinfected directly by UVC in water treatment, but Cryptosporidium has shown significant reduced infectivity with exposure to natural solar radiation (Connelly, et al., 2007). Shaded gaps with increased UVB:PAR may reduce infectivity in the terrestrial systems, however, natural disinfection is most likely in the surfaces of aquatic systems after the fecal matter has been washed away, exposing the pathogenic cysts. Variable UVB:PAR ratios could have a significant impact on infectious pathogens in water supplies and the surrounding watershed. DNA directly absorbs UV, making it the most basic metric for comparing UV- induced damage in all living organisms across ecosystems. DNA is damaged directly by

18 UVR (Mitchell and Nairn, 1989); specifically UVB is the most biologically relevant solar radiation (MacFadyen et al., 2004). Solar UVR induces two major photoproducts in DNA, cyclobutane pyrimidine dimers (CPDs) and (6-4) pyrimidine-pyrimidone dimers [(6-4)PDs]. CPDs are the predominant photoproduct, targeting adjacent thymine (T) or cytosine (C) bases (T<>T, C<>T, T<>C, and C<>C) and causing base change mutations. The (6-4)PD occurs at 5 - 50% of the CPD frequency level; however, the greater distortion of the DNA strand by the (6-4)PD often results in greater cell mortality compared to the CPD (Mitchell, 1997). Both of these photoproducts are quantifiable using a radioimmunoassay specifically targeting the UVR damage sites. Through this technique, we are able to quantify the potential UV-induced damage in any organism, in any ecosystem. DNA damage has been shown to have negative effects on individual organisms, ranging from decreased reproduction rates to death. Microhabitats, which are protected in some way from UVR damage or provide some means by which the individual may avoid UVR exposure (e.g., depth distribution in lakes and streams), should be most beneficial for organisms. Increased UV may drive organisms to suboptimal habitats (i.e. colder temperatures), may alter their diel periodicity, and/or force a crepuscular lifestyle. Other organisms may increase melanization in order to reduce UV effects; however, this will also increase their visibility and susceptibility to predation. Without any such protection or lifestyle changes, these organisms would incur UVR damage and must repair the damage in order to survive. Longer wavelength solar radiation, including UVA and photosynthetically active radiation (PAR) (> 320 nm), is the driving force behind DNA repair in natural systems (Essen and Klar, 2006). Organisms which possess the photolyase enzyme are dependent on these wavelengths, also known as photoreactive radiation (PRR) to drive the photoenzymatic repair (PER) process, and specifically repair the UV-induced DNA damage at little or no cost to the organism (e.g., Malloy et al., 1997; Mitchell, 1997; Rocco et al., 2002). In the absence of photolyase or PRR, DNA is repaired through a nucleotide excision process (NER, or dark repair), which is not specific to UV-DNA damage and is fueled by ATP, making the repair metabolically costly to the organism.

19 The UVR:PAR is highly susceptible to changes in cloud cover, sun angle (particularly across seasons), atmospheric ozone and aerosols, and orientation of radiometers within the ecosystem due to differential scattering of the wave bands (e.g., Flint and Caldwell, 1998). Quantification of this natural variability of UVB:PAR between natural systems requires an integrative metric for any comparisons and estimations of the potential for UV-induced damage in this ecosystems. Spot measurements using radiometers and the like are not sufficient when considering cumulative and long-term effects of UVR exposure in systems with such high levels of variability. DNA dosimeters provide a biologically integrative metric of UV-induced DNA damage that is relevant to all living organisms (e.g., Regan et al., 1992; Mitchell and Karentz, 1993; Olson and Mitchell, 2006) and are minimally susceptible to the intricacies normally associated with environmental UVR assessment over an extended period (e.g., expensive radiometers, temperature, cloud cover, weather conditions, etc.). DNA dosimeters have been shown to provide one of the best estimates of potential DNA damage across marine systems in numerous studies (Malloy et al., 1997; Mitchell, 1997; Mitchell and Karentz, 1993; Mitchell and Nairn, 1993; Regan et al., 1992; Karentz and Lutze, 1990). These data are more reflective of the total potential damage to which an organism is exposed than data collected at instantaneous intervals with radiometers or other instrumentation, making them highly advantageous in dynamic systems. Chemical actinometry is used extensively to measure irradiance flux in photochemical studies. The liquid-based actinometers are particularly appropriate for aquatic studies where there are concerns of UVR reflection and/or refraction; however, with the appropriate temperature correction measures, they can also be very useful in the determination of terrestrial irradiance variability. Nitrate actinometers are used to quantify UVB (290-320 nm) irradiance, while nitrite actinometers quantify UVA (320 – 400 nm) though photolysis (Jankowski et al., 1999). Again, like dosimeters, actinometers can be used in integrate irradiance over time. Unlike dosimeters, which can be used potentially for weeks without replacement in the absence of any significant fouling, actinometers are most valuable over shorter time periods (hours). Due to the constraints of exposure length in the actinometers in this study we have conducted seasonal experiments that compare actinometers and dosimeters exposed for a 24 h period. To

20 better estimate potential DNA damage, we conducted annual experiments using dosimeters exposed over a 7 d period. We quantified potential DNA damage using dosimetry in four terrestrial (2004- 2006) and two aquatic (2005-2006) temperate ecosystems in Northeastern Pennsylvania (Pocono Plateau). This region has been highly susceptible to deforestation, acid rain, and land use impacts in recent years, making it an ideal location to consider solar stresses on varying habitats. In addition to DNA damage measurements, we use actinometers and dosimeters to quantify seasonal variation in UVA and UVB radiation in conjunction with potential DNA damage from early spring before leaf-out (April 2006) and mid summer during peak foliage (July 2006).

Materials and methods

DNA Dosimeters Potential DNA damage can be measured in any environment using DNA dosimeters (“raw” DNA; no repair mechanisms) that are assessed for CPD photoproducts following exposure using a competitive binding radioimmunoassay (RIA) (Mitchell and Karentz, 1993). DNA dosimeters are hollow quartz tubes, approximately 5 cm in length (< 1cm diameter), filled with herring testes DNA (Sigma-Aldrich Co., USA) dissolved in sterile 1x tris-EDTA (TE) buffer solution (Jeffrey et al., 1996) and hermetically sealed with rubber stoppers.

Actinometers Chemical actinometers respond to sunlight, allowing for characterization of photon exposure over a defined wavelength. Due to their relative low cost, ability to integrate exposures over time in areas with variable irradiance, and be distributed in many ecosystems providing spatial data, actinometers are a valuable tool in this work. The methods of Jankowski et al. (1999) were followed except that nitrate and nitrite concentrations were 10% less than they report to reduce potential for saturation of the

21 reaction with relatively long exposures (24 h). Temperature was recorded every 15 min and used to correct the quantum yield of photoproducts for the formation of salicyclic acid in solution. Calculation of overall photoproduct yield used the known exposure weighted temperature based on independently measured incident UVB or UVA. In some cases for April exposures, total exposure was high enough that salicyclic acid formation approached saturation. This non-linearity was corrected based on calibration curves constructed from concurrent exposures of actinometers in shallow surface tanks vs. measurements of spectral irradiance with a SR19 UV scanning spectroradiometer (Biospherical Instruments, Inc.). Neutral density screening and Mylar®-D (DuPont, Inc., USA) were used in July to reduce total exposure to within a region of fixed yield (maximum exposure).

Experimental design A survey of seasonal trends in a natural light gradient was conducted to assess DNA damage as a proxy for variable solar UV exposure. The effect of season on UV induced DNA damage was assessed using both DNA dosimetry and nitrate/nitrite actinometry in April (high UVR: low temperature) and July (high UVR: high temperature). DNA dosimeters and UVA and UVB actinometers were deployed in 400ml Whirl-Pak® bags for 24 h along with temperature loggers in the same locations as the annual dosimeter experiment (two aquatic and four terrestrial). DNA dosimeters were placed in two aquatic (high or low DOC lakes) and four terrestrial ecosystems (hemlock and deciduous forests, open and tall grass fields). The dosimeters were deployed during the same time period (i.e. 7 days of exposure with all dosimeters set out on Day 0 and collected on Day 7) at all locations. All four terrestrial ecosystems and the high DOC lake (DOC = 5 mg/L; Lake Lacawac) are located at Lacawac Sanctuary (Mt. Ariel, PA, USA; N 41°22’57” W 75°17’35”, elevation 439m). The low DOC lake (DOC = 1.1 mg/L; Lake Giles) is on the property of the Blooming Grove Hunting and Fishing Club (Blooming Grove, PA, USA; N 41°22’34” W 75°05’33”, elevation 428m). The survey was conducted during three consecutive years (2004-2006) and seasonally in 2006 (April and July).

22 Environmental data Temperature and solar UVR data were collected for the duration of both the annual and seasonal experiments. Temperature loggers (iButtons, Maxim Integrated Products, Inc., USA) were placed with the actinometers to allow for the necessary temperature corrections at each site. Aquatic irradiance data were collected at both Lake Giles and Lake Lacawac using a submersible BIC radiometer (Biospherical Instruments, Inc.) (Table 2.1). The BIC radiometer quantifies down-welling solar irradiance at 305, 320, and 380 nm and photosynthetically active radiation (PAR; 400-700 nm). Underwater profiles were not available for all dates of this experiment due to instrument unavailability. However, profiles that were recorded as close as possible to the experimental dates were acquired and the data of interest for this experiment (depths equivalent to 1% of the surface irradiance of 320, 380 nm, and PAR for each lake and date) are presented (Table 2.1). Incident solar UVR was recorded using a SR19 UV scanning spectroradiometer (Smithsonian Environmental Research Center, Edgewater, Maryland, USA) located near Lake Giles and/or a GUV 521 radiometer located at Lacawac Sanctuary (Table 2.2). The SR19 and GUV radiometers were inter-calibrated and the irradiance data presented herein can be compared directly regardless of source radiometer. A full description of the SR19 instrument is available (Neale, et al. 2005). Briefly, the SR19 is comprised of 18 channels in the 290 – 324 nm wavelength range spaced in approximately 2 nm widths and one 10 nm channel centered at 330 nm. Spectra are acquired in 4 sec intervals and averaged over 12 min. These spectra were extended over the UVR and PAR range (290 – 700 nm) using a radiative transfer model (System for Transfer of Atmospheric Radiation or STAR) using the procedure described by Neale et al. (2005). The GUV 521 records irradiance as 15 min averages with channels centered on 305, 320, 340, 380 nm and PAR. Irradiance data were summed and total exposures are presented as kJ/m2 for each experimental period.

23 Table 2.1. Aquatic irradiance data from the BIC submersible radiometer. Data are presented as depth at which 1% of the subsurface irradiance is measured. Measurements were collected on one date during the survey period.

Lake Date 1% 320 nm 1% 380 nm 1% PAR (m) (m) (m) Giles July 2005 2.44 3.40 15.80 April 2006 2.50 6.25 18.50 July 2006 1.80 4.50 15.00 Lacawac July 2005 April 2006 0.25 0.50 4.00 July 2006 0.25 0.50 5.00

Table 2.2. Incident solar radiometer data compiled from the GUV 521 radiometers. Data are presented as cumulative exposures for the duration of the experiment (kJ/m2). The mean 320:PAR ratio is provided for July 12-19, 2006 for direct comparison to the seasonal dates of 2006.

Experiment Date 305nm 320nm PAR 320:PAR (kJ/m2) (kJ/m2) Seasonal April 6, 2006 36.1 430 2.74 121.75 July19, 2006 98.3 790 5.63 135.19

Annual 2006 July 12 – 19 585.8 4856 34.2 x = 139.59

DNA Damage analysis Cyclobutane dimers (CPDs) were quantified using radioimmunoassay (RIA). The RIA is a competitive binding assay between very small amounts of radiolabeled DNA and sample DNA for antisera raised against UV-irradiated DNA. For the RIA, 50 ng of heat- denatured sample DNA was incubated with 5-10 pg of poly(dA):poly(dT) (labeled to >5 x 108 cpm/µg by nick translation with 32P-dTTP) in a total volume of 1 mL 10 mM Tris, pH 7.8, 150 mM NaCl, 1 mM EDTA, and 0.15% gelatin (Sigma). Antiserum was added

24 at a dilution that yielded ~50% binding to labeled ligand and after incubation overnight at 4ºC the immune complex was precipitated with goat anti-rabbit immunoglobulin (Calbiochem) and carrier serum from non-immunized rabbits (UTMDACC, Science Park/Veterinary Division, Bastrop, TX). After centrifugation, the pellet was dissolved in tissue solubilizer (NCS, Amersham), mixed with ScintiSafe (Fisher) containing 0.1% glacial acetic acid by volume, and the 32P quantified by liquid scintillation spectrometry. Under these conditions, antibody binding to an unlabeled competitor inhibits antibody binding to the radiolabeled ligand. These details, as well as those concerning the specificities of the RIA and standards used for quantification, are described by Mitchell (1996 and 2006).

Data analyses DNA dosimetry data were corrected for background controls by subtracting the dark control dosimeter from each experimental dosimeter prior to statistical analyses. These data were tested statistically for annual and habitat differences using two-way ANOVA (location x date). Actinometer data were corrected as by Jankowski et al (1999) and were tested statistically for seasonal differences within ecosystems using one-way ANOVA. UVB:UVA were compared using two-way ANOVA (location x date) (SPSS v. 15, SPSS, Inc.; α = 0.05).

Results

Dosimetry In terrestrial ecosystems, there was high inter-annual variation in UV-induced DNA damage, particularly in July. Within a given year, significantly less DNA damage was accrued in the hemlock forest within the exposure period than in the deciduous forest (ecosystem p<0.002) (Figure 2.1). DNA damage in the aquatic systems was less variable, with no significant effect of depth on UV-induced CPDs in either system within an experimental year (p>0.1622) (Figure 2.2). Damage was negligible deeper than 0.5 m in Lake Lacawac (high DOC). In Lake Giles (low DOC), significant DNA damage was possible to depths up to at least 2.0 m in 2005 (not significant). There was no effect of

25 experimental year (July 2005 vs. 2006) in Lake Giles (p=0.0807) or Lake Lacawac (p=0.4133). Further, there was no significant year effect within a terrestrial ecosystem (p=0.1740). There was no UV-induced DNA damage seasonal effect (April vs. July 2006) in the terrestrial ecosystems (p=0.6583). DNA damage induction was independent of temperature, therefore, in this study, we observed no differences in UVR exposure (quantified as the photoproducts induced, CPDs) between Spring and Summer.

100 2004 2005 2006 of control)

-1 80

60

40

20

% DNA Damage (CPDs% mb DNA 0 Open Tall Deciduous Hemlock

Figure 2.1. UV induced DNA damage in four temperate ecosystems, Open field, tall grass, deciduous forest, and hemlock forest quantified by RIA to determine CPD photoproduct concentration in DNA dosimeters. Data reported as CPDs as percent of light control in July of 2005 and 2006. Error bars = standard error of the mean (N = 3).

26 % DNA Damage (CPDs mb DNA-1 of control)

0 20406080100

0.0

0.5

1.0

1.5

2.0

Lake Lacawac Depth (m) Lacawac Depth Lake 2.5

3.0 A

0.0

0.5

1.0

1.5

2.0 Lake Giles Depth (m) Depth Giles Lake

2.5

3.0 B

Figure 2.2. UV induced DNA damage in high DOC Lake Lacawac (A) and low DOC Lake Giles (B) quantified by RIA to determine CPD photoproduct concentration in DNA dosimeters. Data reported as CPDs as percent of lake surface light control in July of 2005 and 2006. Error bars = standard error of the mean (N = 3).

27 Actinometry The pattern of UVA and UVB exposure quantified using actinometry mimics very closely the potential UV-induced DNA damage, independent of wavelength, observed in the dosimeters during the same seasonal exposure periods. As predicted, UVA was greater than UVB in all systems, terrestrial (Figure 2.3) and aquatic (Figure 2.4), in April and July, due to preferential scattering of the shorter wavelength UVB. There was no effect of season on UVA or UVB, however (p=0.5649 and p=0.9284, respectively). The UVA:UVB ratios varied by season and habitat. open, tall, deciduous, and hemlock UVA:UVB ratios were 0.192, 0.166, 0.162 and 0.128, respectively, in April, and 0.297, 0.166, 0.169, and 0.000, respectively, in July 2006. There was no significant effect of season on the UVA:UVB ratio (p=0.6583), however, there was a strong ecosystem effect on the UVA:UVB ratio (p<0.0001).

28 400 3500 Open A Tall Deciduous 3000 Hemlock 300

) 2500 CPDs mb DNA -2

2000 200 1500 Einstein cm μ -1

Ep ( 1000 100

500

0 0 400 3500 B 3000

300

) 2500 CPDs mb DNA -2

2000 200 1500 Einstein cm μ -1

Ep ( 1000 100

500

0 0 UVA UVB CPDs

Figure 2.3. Analysis of potential UVR damage by DNA dosimetry and actinometry in four temperate terrestrial ecosystems. DNA damage was quantified by RIA to determine CPD photoproduct concentrations (CPDs mb DNA-1) and nitrite and nitrate actinometers (Ep) was used as a quantification of UVR dose corrected for temperature. Panel A: April 2006, Panel B: July 2006. Error bars = standard error of the mean (dosimeter N = 3; actinometers N=2).

29

Figure 2.4. Analysis of potential UVR damage by DNA dosimetry and actinometry in two temperate aquatic ecosystems. UV induced DNA damage potential quantified by RIA to determine CPD photoproduct concentration ( ) and nitrite (UVA; gray bars) and nitrate (UVB; black bars) actinometers as a quantification of UVR dose corrected for temperature. Panel A: April 2006, Panel B: July 2006. UVB:UVA at 0.5m in Lake Giles (July) = 0.144; all others UVB:UVA = 0. Error bars = standard error of the mean (dosimeter N = 3; actinometers N=2).

Discussion

Anthropogenic forcing has caused significant shifts in land use and structure. This invokes immediate shifts in habitat structure and concomitantly ecosystem community changes. The ultimate result of changes such as these is unknown, from any perspective. Further, stressors on environments (e.g., disease, temperature, precipitation, UVR, etc.) are known to significantly impact habitat structure and ecosystem dynamics. For instance, the invasive Hemlock Woolly Adelgid (Adelges tsugae) is causing massive die- offs of Eastern and Carolina hemlocks on the East Coast of the US. The hemlocks are 30 being replaced mainly by ground cover ferns and eventually by deciduous trees of varying species. This change in botanical structure and, perhaps most importantly based on the results of this study, canopy cover in forested areas (Orwig and Foster, 1998), are likely to have significant impacts at the population and community levels. In addition, changes in vegetation can cause a significant change in the CDOM/DOC additions to lakes and streams within the watershed (e.g., Dittman et al., 2007). Studies have shown significant effects of solar UVR as an abiotic environmental control of insects, bacteria, and protozoan pathogens (e.g., Hernandez et al., 2004; Connelly et al., 2007). Water treatment facilities rely on UVC as a disinfectant of common pathogens such as Cryptosporidium and Giardia (Clancy et al., 2004; Linden, et al., 2001). A recent study shows significant effects of UVR exposure in both July and September on Cryptosporidium parvum oocysts (Chapter 5; Connelly et al., 2007), an obligate protozoan pathogen that causes cryptosporidiosis, a gastrointestinal disease which can be debilitating and potentially deadly to the immunocompromised and elderly. Peak concentrations of Cryptosporidium sp. have been reported during the same periods (July and September) in natural water supplies (Tsushima et al., 2003). These findings are recognized as being regional and potentially driven by agricultural inputs. However, it is during this period annually (July – September) that photobleaching of CDOM is maximal in the northern hemisphere, concomitantly increasing the UVR transparency of natural waters (Morris and Hargreaves, 1997) and providing the greatest potential for natural solar disinfection. Seasonal fluctuations in solar UVR and ecosystem exposures could play a significant role in the natural disinfection of some pests and pathogens. The comparison of actinometers and dosimeters used in this study to assess seasonal fluctuations in solar irradiance and subsequent inducible DNA damage has provided important information for ecosystem assessment. Actinometers provide more specific information by isolating UVA vs. UVB exposure. Further, the technique is likely more feasible for general research laboratories to establish than the RIA used for CPD quantification, which requires isotope usage and specialized equipment. The DNA damage dosimeters, however, are most useful for directly assessing the biological effect of exposure over time (integrated). To apply the information obtained from the

31 actinometers, one must still model the relationship and make higher-level assumptions between the exposures and the biological significance of the exposures. The application of low-cost, integrative tools for assessing UVR within and among ecosystems provides important information for population and community studies. The long-term effects of variable UVR on these systems are unknown. The ability of organisms, plants and animals, to respond to this variability is also unknown. However, quantification of UV:PAR ratios could play a significant role in assessing such things as the potential for waterborne human pathogen outbreaks and potential species invasions of native areas. The dynamic nature of ecosystems does not allow for far- reaching correlations between exposure and DNA damage, but dosimeters in conjunction with actinometers provide a means by which biologically relevant UVR variability can be assessed in a greater number of areas and under a broader range of conditions than radiometers.

Acknowledgements

We thank the Zoology Department (Miami University) for graduate research support to SJC. We thank Erin Overholt for laboratory and field assistance during this project. This work was partially supported by the National Science Foundation grant DEB-IRCEB- 0552283 to CEW.

32 References

Barnes PW, Ballare CL, Caldwell, MM (1996) Photomorphogenic effects of UV-B radiation on plants: consequences for light competition. Journal of Plant Physiology. 148, 15-20.

Brovkin V, Sitch S, von Bloh W, Claussesn M, Bauer E, Cramer W (2004) Role of land

cover changes for atmospheric CO2 increase and climate change during the last 150 years. Global Change Biology. 10, 1253-1266.

Brown MJ, Parker GG, Posner NE (1994) A survey of ultraviolet-B radiation in forests. Journal of Ecology. 82, 843-854.

Clancy JL, Marshall MM, Hargy TM, Korich DG (2004). Susceptibility of five strains of Cryptosporidium parvum oocysts to UV light. Journal of the American Water Works Association. 96, 84-93.

Connelly SJ, Wolyniak EA, Williamson CE, Jellison KL (2007) Artificial UV-B and solar radiation reduce in vitro infectivity of the human pathogen Cryptosporidium parvum. Environmental Science and Technology. In press.

Dittman, JA, Driscoll CT, Groffman PM, Fahey TJ (2007) Dynamics of nitrogen and dissolved organic carbon at the Hubbard Brook Experimental Forest. Ecology. 88, 1153- 1166.

Essen LO, Klar T (2006) Light-driven DNA repair by photolyases. Journal of Cellular and Molecular Life Sciences. 63, 1266-77.

Flint SD, Caldwell MM (1998) Solar UV-B and visible radiation in tropical forest gaps: measurements partitioning direct and diffuse radiation. Global Change Biology. 4, 863– 870.

33 Hargreaves BR (2003) Water Column Optics and Penetration of UVR. In: UV Effects in Aquatic Organisms and Ecosystems. E.W. Helbling & H. E. Zagarese (eds). pp.59-105. Comprehensive Series in Photochemical and Photobiological Sciences, Royal Society of Chemistry, Cambridge, UK.

Hernandez, EA, Ferreyra, GA, MacCormack, WP. 2004. Effect of solar radiation and the subsequent dark periods on two newly isolated and characterized Antarctic marine bacteria. Polar Research. 23, 67-77.

Jankowski JJ, Kieber DJ, Mooper K (1999) Nitrate and nitrite ultraviolet actinometers. Photochemistry and Photobiology. 70, 319-328.

Jeffrey WH, Pledger RJ, Aas P, Hager S, Coffin RB, VonHaven R, Mitchell DL (1996) Diel and depth profiles of DNA photo damage in bacterioplankton exposed to ambient solar ultraviolet radiation. Marine Ecology-Progress Series. 137, 283-291.

Karentz D, Lutze LH (1990) Evaluation of biologically harmful ultraviolet-radiation in Antarctica with a biological dosimeter designed for aquatic environments. Limnology and Oceanography. 35, 549-561.

Linden KG, Shin GA, Sobsey MD. (2001) Relative efficacy of UV wavelengths for the inactivation of Cryptosporidium parvum. Water Science and Technology. 43, 171-174.

MacFadyen EJ, Williamson CE, Grad G, Lowery M, Jeffrey WH, Mitchell DL (2004) Molecular response to climate change: temperature dependence of UV-induced DNA damage and repair in the freshwater crustacean Daphnia pulicaria. Global Change Biology. 10, 408-416.

Malloy KD, Holman MA, Mitchell D, et al. (1997) Solar UVB-induced DNA damage and photoenzymatic DNA repair in Antarctic zooplankton. Proceedings of the National Academy of Science USA. 94, 1258-1263

34 Mitchell, DL (2006) “Quantification of DNA photoproducts in mammalian cell DNA using radioimmunoassay.” In: Henderson, D.S. (ed) Methods in Molecular Biology, DNA Repair Protocols: 2nd Edition.Humana Press Inc, New Jersey. 239-249.

Mitchell DL (1997) “Ultraviolet radiation damage to DNA” In: Encyclopedia of Molecular Biology and Molecular Medicine. Volume 6. VCH, Weinheim.

Mitchell, DL (1996) “Radioimmunoassay of DNA damage by ultraviolet light.” In: Pfeifer, G (ed) Technologies for Detection of DNA Damage and Mutations. Plenum Publishing, New York. 73-85.

Mitchell D, Karentz D (1993) “The induction and repair of DNA photodamage in the environment.” In: Environmental UV Photobiology. Plenam Press, New York. 345-377 pp.

Mitchell D, Nairn RS (1989) The Biology of the (6-4) Photoproduct. Photochemistry and Photobiology. 49, 805-819.

Morris DP, Hargreaves BR (1997) The role of photochemical degradation of dissolved organic carbon in regulating the UV transparency of three lakes on the Pocono Plateau. Limnology and Oceanography. 42, 239-249.

Neale PJ, Goodrich VR, Brinley WR (2005) The Smithsonian-NIST-USDA-FPL network for monitoring solar ultraviolet irradiance: comparison of radiometer measurements and radiative transfer model calculations. In: Martin JW, Ryntz R, Dickie RA (eds) Service life prediction: challenging the status quo. Federation of Societies for Coatings Technology, Blue Bell, PA 159–170 pp.

Olson MH, Mitchell DL (2006) Interspecific variation in UV defense mechanisms among temperate freshwater fishes. Photochemistry and Photobiology 82, 606-610.

35 Orwig DA, Foster DR (1998) Forest response to the introduced hemlock woolly adelgid in southern New England, USA. Journal of the Torrey Botanical Society. 125, 60-73.

Regan JD, Carrier WL, Gucinski H, Olla BL, Yoshida H, Fujimura RK, Wicklund RI (1992) DNA as a solar dosimeter in the ocean. Photochemistry and Photobiology. 56, 35- 42.

Rocco, V.E., O. Oppezzo, R. Pizarro, R. Sommaruga, M. Ferraro, and H.E. Zagarese. (2002) Ultraviolet damage and counteracting mechanisms in the freshwater copepod Coeckella poppei from the Antarctic Peninsula. Limnol Oceanogr. 47, 829-836.

Tsushima Y, Karanis P, Kamada T, Makala L, Xuan X, Tohya Y, Akashi H, Nagasawa H (2003) Seasonal change in the number of Cryptosporidium parvum oocysts in water samples from the rivers in Hokkaido, Japan, detected by the ferric sulfate flocculation method. Journal of Veterinary and Medical Science. 65, 121-123.

Wetzel RG (2001) Limnology: Lake and River Ecosystems, 3rd Edition. Academic Press.

Williamson CE, Sanders RW, Moeller RE, Stutzman PL (1996) Utilization of subsurface food resources for zooplankton reproduction: Implications for diel vertical migration theory. Limnology and Oceanography. 41, 224-233.

Williamson CE, Neale PJ, Grad G, De Lange HJ, Hargreaves BR (2001) Beneficial and detrimental effects of UV on aquatic organisms: Implications of spectral variation. Ecological Applications. 11, 1843-1857.

36

Chapter 3

Examination of UV-induced DNA damage in daphniids: DNA damage vs. repair at depth in two lake systems

37 Abstract

Seasonal fluctuations in solar ultraviolet radiation (UVR) are well documented. The direct impacts of these fluctuations on community structure in freshwater systems are not. Short wavelength solar radiation (UVB) is known to directly induce DNA damage, while the longer wavelengths (UVA and photosynthetically active radiation (PAR)) drive the photoenzymatic (PER) DNA repair process specifically targeting the UVR damage. With scattering of solar wavelengths in the water column, the shortest, most damaging, wavelengths are most important in the surface waters and the longer wavelengths penetrate to much greater depths. Because of this, organisms that display diel vertical migrations (DVM) may be exposed to higher levels of damaging UVR in the early morning before migration, will migrate down through the longer wavelength radiation, and will likely reach a depth of total darkness for the remainder of the solar day. The availability of longer wavelengths to drive PER during migration may be critical for the survival of the organism before it reaches a depth where no light is present, and it must rely on dark DNA repair mechanisms, which are not UV-damage specific. Here, we test the response of Daphnia catawba to a fixed acute exposure of artificial UVB irradiation and a fixed duration of solar radiation at fixed depths in both a high UVR and a low UVR lake. We quantified the induction of cyclobutane pyrimidine dimers (CPDs) using radioimmunoassay immediately following the UVB exposure and again immediately following the repair period. Further, we quantified the potential for additional DNA damage induction at each depth using DNA dosimetery. Dosimetry results indicated that high levels of additional UV-induced DNA damage are likely to occur in the surface waters. DNA damage (CPDs) in the Daphnia was increased in the surface waters above the initial induction levels, whereas CPDs generally decreased with depth in both lake systems. Results of this study suggest the potential for significant DNA damage induction by UVR in this environment.

38 Introduction

The natural seasonal fluctuations of both ultraviolet radiation (UVR) and temperature are known to drive community composition, particularly in small, highly transparent freshwater systems (e.g., Rae and Vincent 1998; Williamson et al., 2002; Strecker et al., 2004; Boeing et al., 2004). The within system variability (e.g., dissolved organics and temperature) will also significantly effect species distributions (e.g., see Boeing et al., 2004) and community composition. In concert these natural variations can have significant effects on individuals, populations, and communities within these systems, perhaps the most significant of which may be solar UVR. Acute UVR exposure (high intensity, short duration) induces erythema in humans (Clydesdale et al., 2001) and is sufficient to initiate DNA repair mechanisms (Matsumura and Ananthaswamy, 2004). DNA damage has been quantified in natural marine populations (Karentz, 1991) and is believed to be a significant environmental stressor in most ecosystems. Acute induction of DNA damage, and the subsequent repair, in freshwater species has been demonstrated in the lab (e.g., MacFadyen et al., 2004; Olson and Mitchell, 2006). In natural systems, UV-induced DNA damage has been detected in freshwater species (Zellmer et al., 2004). Cyclobutane pyrimidine dimers (CPDs) are the dominant UVB photoproducts in DNA (Cadet and Vigny, 1990). The distribution of organisms in natural systems will ultimately affect the level of DNA damage they sustain and their potential to counteract this damage. UV-induced DNA damage is specifically repaired by photoenzymatic repair (PER) in organisms with the photolyase enzyme. This repair is driven by the presence of the enzyme and long wavelength radiation (> 320 nm), and therefore is not considered a stress on the organism. However, these repair wavelengths are in greatest concentration in the surface waters where visual predation risks are high for organisms. Organisms which depend on PER for survival risk high predation if the system lacks a refuge for both predation and UVR. In the absence of photolyase, or long wavelength radiation, organisms rely on the nucleotide excision (dark) repair (NER) process which is not UV-DNA damage specific and is fueled by ATP in the organism, making it metabolically costly. Both repair

39 processes are believed to be temperature dependent based on the enzyme kinetic hypothesis (reaction rate is directly proportional to temperature) (e.g., Li, et al., 2002; Mitchell, 1997). An individual’s ability to maintain optimal environmental conditions for (1) avoidance of UV-induced DNA damage and (2) repair of inevitable UVR damage is imperative for its long-term survival. Many zooplankton species exhibit diel vertical migration (DVM) and thus are believed to avoid high levels of UV induced DNA damage (Alonso et al., 2004). It has been suggested that Daphnia spp. in particular will exhibit DVM under high UVR and/or high fish predation conditions (e.g., Alonso et al., 2004; Ringleberg and van Gool, 2003). Further, Daphnia have been shown to actively avoid UVR controlled conditions in natural systems (e.g., see Leech et al., 2005). Although this behavioral avoidance of surface waters will decrease UVR exposure and reduce visual predation risks, this DVM is believed to come at a significant energetic cost to the individuals from increased physical stresses associated with the act of migration, in addition to colder temperatures and, potentially, reduced food quantity and quality at depth (Winder et al., 2004). Migration to colder temperatures and low light conditions may have negative effects on DNA repair potential in these organisms. In addition, Daphnia can still be exposed to levels of UVR significant enough to induce damage even at depth in the more transparent lakes, particularly high alpine and arctic systems (e.g., Winder et al., 2003; Zellmer, 1995). DNA repair in these relatively cold water systems at high latitude and altitude is likely to be slowed and the constant high levels of UVR exposure will be cumulative. The ability of aquatic species to tolerate potentially significant UVR exposure in natural systems is not well understood and the effect of DVM on the induction and repair of UV- induced DNA damage in Daphnia is unknown. The basis for the current study is the hypothesis that UV-induced damage sustained in the early morning hours in freshwater systems will not be completely repaired if the Daphnia migrate to lower depths. These depths lack the necessary photoenzymatic repair radiation (PRR) to repair the damage without metabolic cost to the organism using NER. NER is a global genome repair mechanism responsible for the repair of most bulky adducts in DNA regardless of source, and is driven by the organisms’ main energy source, ATP. Concomitantly, Daphnia, which exhibit DVM, will

40 be more dependent on dark DNA repair processes that are not specific to UV-induced damage. Further, based on the enzyme kinetics hypothesis stating that temperature will have a direct effect on any enzyme-based reaction, all repair processes that do occur will be slowed significantly by decreased temperature. Here we examined DNA damage induced by artificial acute (short term) UVB radiation exposure and its removal in Daphnia catawba in the presence of natural solar radiation at fixed depths in two freshwater lakes from April to August 2006. The repair of UV-induced CPDs by the Daphnia was assessed at experimental deployment (UVB exposure only) and following six h of natural solar exposure at fixed lake depths twice per month during the experimental period. Temperature and solar spectral composition were quantified for each experimental deployment. DNA dosimeters were used to assess the potential DNA damage induced by artificial UVB and any additional damage potentially incurred during the course of the solar repair period.

Materials and Methods

Source lakes These experiments were conducted in two temperate lakes located in northeast Pennsylvania in the Pocono Mountain range. Lake Giles is a 48 ha lake located within the Blooming Grove Hunting and Fishing Club in Blooming Grove, PA (41°23’ N, 75°06’ W; 428m elevation). It is a transparent, oligotrophic lake with a mean DOC concentration of 1.1mg/L, a pH of 5.6 (Cooke et al., 2006), and a mean depth of 10.1m. Historically Lake Giles has had a summer UVR-320 attenuation up to 15m, however significant decreases in that attenuation have been observed in recent years (Williamson et al, 1999; Fischer et al., 2006). Lake Giles has a mixed zooplankton community, dominated by Daphnia catawba, Leptodiaptomus minutus, Aglaodiaptomus spatulocrenatus, and Cyclops scutifer. Lake Lacawac is located within Lacawac Sanctuary in Mt. Ariel, PA (41°23’ N, 75°18’ W; 439m elevation). It is a 21 ha lake with a mean depth of 5.2m. It is a low transparency lake with an average DOC concentration of 5 mg/L and summer UVR-320 attenuation only up to 0.5m. This mesotrophic lake is less acidic than Lake Giles (pH=6.2

41 – 6.5) and has a mixed zooplankton community dominated by Daphnia ambigua, Daphnia catawba (low abundance), Holopedium gibberum, Leptodiaptomus minutus, and Cyclops scutifer.

Daphnia Daphnia catawba from two lakes (Lake Lacawac and Lake Giles) in Northeastern Pennsylvania were used in this experiment. Daphnia were acquired from the source lakes less than 24 h prior to each experiment from their respective source and experiment lakes. Animals were collected by horizontal towing at ~5m using a 363 μm 0.5 m diameter net. Gravid stage females from each lake were sorted into 48 μm filtered source water and placed in the dark at least 9 h prior to UVB exposure in an environmental control chamber at their respective lake temperature (temperature at 0.5 m). Background damage levels were assessed from three aliquots of the sorted animals prior to UVB exposure.

UVB damage irradiation Approximately 2,000 Daphnia from Lake Giles and 1,200 Daphnia from Lake Lacawac were exposed to UVB at the current lake temperatures in the environmental chamber. UVB irradiation exposures were conducted in an environmentally controlled chamber (Environmental Growth Chambers, Chagrin Falls, OH) at source lake temperature (ranging from 6°C in April to 24°C in August). Six open Petri dishes (600 mL) containing the daphnia were exposed to the UVB irradiation from the top at a distance of 24cm. The UVB source was from one unfiltered Spectronics Spectroline BLE-1T158 15 W lamp (Westbury, NY) yielding a spectral output of 281-405 nm (for damage radiation emission spectrum see Figure 3.1). Samples were exposed for 60 min to UVB yielding a total exposure of 160 kJ/m2. All exposures were conducted for the same length of time and under the same culture conditions on each experimental date with temperature as the only variable throughout the season. Following exposure, the animals from the six exposed dishes were combined, mixed, and equal aliquots (100 mL) were placed in each of the repair bags for deployment in the lakes (~100 Daphnia per bag, see Repair design.). This was done to ensure that a mixed population of the Daphnia was present at each depth and to limit any possible shading (i.e., reduced UVB exposure/DNA damage

42 induction) effects present in the Petri dishes due to the higher concentrations of Daphnia. Daphnia were returned to their source lakes in the repair bags as described below (no translocation).

1000 -1

nm 100 -2

10

1 Damage Irradiance m mW

0.1 300 400 500 600 Wavelength (nm)

Figure 3.1. Emission spectra. The emission spectra for the UVB lamp used to induce DNA damage. (Data courtesy of R. Moeller).

Repair design Aliquots of the Daphnia as described above were placed in 18x20 cm Bitran Series PE bags (VWR, USA). The bags were filled with 48 μm-filtered water at lake temperature and placed in the dark until deployment in their respective lakes. Daphnia were transported to the lakes in the dark and deployed into a rack system (travel + deployment time < 60 min). PVC® racks (12”x24”) covered in 4 mil Aclar (UVR transmitting plastic; Honeywell International) were suspended at fixed depths by surface buoys in each lake. Elastic rope was used to divide the racks into quadrants to keep the bags in place. The racks were fixed at 0.5, 1, 3, 5, and 7 m in Lake Giles and 0.5, 1, and 3 m in Lake Lacawac, due to the variable depth and transparency of the lakes. Three random bags from each lake were sampled for initial UV induced DNA damage at the time of deployment. The bags were left in the lakes for six h.

43

Environmental abiotic assessment An additional Bitran PE bag was filled with 48 μm lake water and placed in the racks at each depth alongside the Daphnia bags. These abiotic measurement bags contained two DNA dosimeters [purified herring testes DNA in a solution of 1x TE buffer at a concentration 90 ng/mL in UVR transparent tubing ~4 cm in length; DNA source Sigma Aldrich; (see Olson et al., 2006; Mitchell and Karentz, 1993)] to quantify potential UV- induced DNA damage and an iButton temperature logger (Maxim/Dallas integrated Products, Inc., USA) which recorded the temperature ± 0.5°C every 15 min for the duration of the experiment. Underwater UVR was measured with a BIC radiometer (Biospherical Instruments, USA), which quantified down-welling solar irradiance at 305, 320, and 380 nm and photosynthetically active radiation (PAR; 400-700 nm). Underwater profiles were not available for all dates of this experiment due to instrument unavailability. However, profiles that were recorded as close as possible to the experimental dates were acquired and the data of interest for this experiment (depths equivalent to 1% of the surface irradiance of 320, 380 nm, and PAR for each lake and date) are presented (Table 3.1). Incident solar UVR was recorded using a SR19 UV scanning radiometer located near Lake Giles (~ 41°23’ N, 75°06’ W). Full description of the instrument is available (Neale, et al. 2005). Briefly, the SR19 is comprised of 18 channels in the 290 – 324 nm wavelength range spaced in approximately 2 nm widths and one 10 nm channel centered at 330 nm. Spectra are acquired in 4 sec intervals and averaged over 12 min. These spectra were extended over the UVR and PAR range (290 – 700 nm) using a radiative transfer model (System for Transfer of Atmospheric Radiation or STAR; Neale et al. 2005).

44 Table 3.1. Abiotic factors for each experimental deployment date. Temperature daily average per depth recorded every 15 min (°C ± 0.5°C). Water spectral composition was obtained using a BIC radiometer. One percent depth (m) for 320 and 280 nm and PAR are given for the closest available profile date for each lake.

Experiment Lake Depth Temp BIC Date 1% 1% 1% PAR Date (m) (°C) Reference UVR- UVR-380 (m) 320 (m) (m) April 6 Giles 0.5 6.0 April 10 2.5 6.25 18.5 1.0 5.6 3.0 5.7 5.0 5.5 7.0 5.5 April 6 Lacawac 0.5 6.9 April 18 0.25 0.5 4.0 1.0 7.5 3.0 7.3 April 7 Giles 0.5 6.0 April 10 2.5 6.25 18.5 1.0 5.5 3.0 5.7 5.0 5.4 7.0 5.5 April 7 Lacawac 0.5 7.0 April 18 0.25 0.5 4.0 1.0 7.0 3.0 7.0 May 18 Giles 0.5 15.7 May 17 3.0 6.5 16.0 1.0 15.7 3.0 15.2 5.0 14.5 7.0 11.8

45 Experiment Lake Depth Temp BIC Date 1% 1% 1% PAR Date (m) (°C) Reference UVR- UVR-380 (m) 320 (m) (m) May 18 Lacawac 0.5 16.9 May 17 0.25 0.5 4.25 1.0 16.5 3.0 14.9 May 19 Giles 0.5 15.2 May 17 3.0 6.5 16.0 1.0 15.0 3.0 15.2 5.0 14.9 7.0 12.5 May 19 Lacawac 0.5 15.9 May 17 0.25 0.5 4.25 1.0 15.9 3.0 14.6 June 1 Giles 0.5 23.2 June 15 3.0 7.0 16.0 1.0 22.3 3.0 17.9 5.0 15.4 7.0 13.7 June 1 Lacawac 0.5 26.6 June 15 0.5 0.75 3.8 1.0 24.2 3.0 14.7 June 21 Giles 0.5 22.4 June 22 3.25 7.5 18.0 1.0 22.0 3.0 20.5 5.0 18.5 7.0 15.3 June 21 Lacawac 0.5 24.1 June 23 0.25 0.75 4.2 1.0 23.9 3.0 16.7

46 Experiment Lake Depth Temp BIC Date 1% 1% 1% PAR Date (m) (°C) Reference UVR- UVR-380 (m) 320 (m) (m) July 18 Giles 0.5 27.4 July 17 1.8 4.5 15.0 1.0 26.8 3.0 25.9 5.0 22.1 7.0 16.5 July 18 Lacawac 0.5 28.3 July 18 0.25 0.5 5.0 1.0 27.8 3.0 18.2 July 19 Giles 0.5 27.2 July 17 1.8 4.5 15.0 1.0 26.6 3.0 26.1 5.0 21.9 7.0 16.4 July 19 Lacawac 0.5 28.6 July 18 0.25 0.5 5.0 1.0 27.9 3.0 18.7 August 8 Giles 0.5 26.3 August 9 2.25 5.5 13.75 1.0 26.4 3.0 26.4 5.0 26.2 7.0 18.3 August 8 Lacawac 0.5 27.1 August 9 0.5 0.75 5.5 1.0 26.7 3.0 23.1 August 9 Giles 0.5 26.5 August 9 2.25 5.5 13.75 1.0 26.2 3.0 25.9 5.0 25.5 7.0 18.4 47 Experiment Lake Depth Temp BIC Date 1% 1% 1% PAR Date (m) (°C) Reference UVR- UVR-380 (m) 320 (m) (m) August 9 Lacawac 0.5 26.3 August 9 0.5 0.75 5.5 1.0 26.1 3.0 23.7

48 DNA damage analysis. All samples were immediately frozen and stored at -20°C. Samples were then thawed in a 2.0 mL microcentrifuge tube (Eppendorf) containing 750 µL of lysis buffer solution (10 mM Tris-Cl, pH 8.0 (Sigma), 1 mM EDTA pH 8.0 (Sigma), and 0.25% sodium dodecyl sulfate (Boehringer-Mannheim)). After incubation with 10 µg/mL heat inactivated RNAse (Sigma) at 37°C for 1 h the lysates were treated with 30 µL Proteinase K (10mg/mL) (Boehringer-Mannheim) in a variable speed shaker (Eppendorf) overnight at 37°C with vigorous mixing. Following deproteination, samples were extracted sequentially with equal volumes of buffer-saturated phenol (Boehringer-Mannheim), phenol:Sevag (1:1), and Sevag (24:1 chloroform:isoamyl alcohol). Finally, the aqueous phase was transferred to a centrifuge tube containing 1/10 volume 3 M sodium acetate (pH 5.2). One volume ice-cold isopropanol was added and the mixture was inverted several times before overnight precipitation at -20°C. Following precipitation samples were centrifuged at 10,000 rpm (Eppendorf μicro) for 10 min at 4°C, washed with 70% ethanol, air-dried and resuspended in 1 mL 10 mM Tris, 1 mM EDTA (pH 8.0). The DNA concentration and purity were determined by absorbance at 260/280nm using a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). Cyclobutane pyrimidine dimers (CPDs) were quantified by radioimmunoassay (RIA). The RIA is a competitive binding assay between very small amounts of radiolabeled DNA and sample DNA for antisera raised against UV-irradiated DNA. For the RIA, 50 ng of heat-denatured sample DNA was incubated with 5-10 pg of poly(dA):poly(dT) (labeled to >5 x 108 cpm/µg by nick translation with 32P-dTTP) in a total volume of 1 mL 10 mM Tris, pH 7.8, 150 mM NaCl, 1 mM EDTA, and 0.15% gelatin (Sigma). Antiserum was added at a dilution that yielded ~50% binding to labeled ligand and after incubation overnight at 4ºC the immune complex was precipitated with goat anti-rabbit immunoglobulin (Calbiochem) and carrier serum from non-immunized rabbits (UTMDACC, Science Park/Veterinary Division, Bastrop, TX). After centrifugation, the pellet was dissolved in tissue solubilizer (NCS, Amersham), mixed with ScintiSafe (Fisher Scientific) containing 0.1% glacial acetic acid, and the 32P quantified by liquid scintillation spectrometry. Under these conditions, antibody binding to an unlabeled competitor inhibits antibody binding to the radiolabeled ligand. These

49 details, as well as those concerning the specificities of the RIA and standards used for quantification, are described by Mitchell (1996 and 2006).

Data analyses CPD data were corrected for population background damage levels (taken just prior to UVB exposure) by subtracting the control CPD level from the experimental CPD level for each sample. The effects of temperature and CPD repair were evaluated using two- way ANOVA for the two lakes independently. Significant effects of temperature and depth on DNA repair were evaluated using two-way ANOVA (SPSS v. 15, SPSS, Inc.; α = 0.05) (Quinn and Keough, 2002).

Results

Abiotic environmental conditions Temperatures in both Lake Lacawac and Lake Giles range from 5.4 to 7.5 in April, and 18.4 to 27.1 in August with the coldest temperatures at 7 m in Lake Giles and the warmest at the surface (0.5 m) of Lake Lacawac (Table 3.1). The transparency of both lakes was highly variable from April through August. The greatest 1% depth for UVR- 320 in Lake Giles was 3.25 m on June 21, whereas the greatest UVR-320 1% depth in Lake Lacawac was 0.5 m in mid June and again in early August. The greatest 1% PAR depth in Lake Giles was 18.5 m in early April and the shallowest 1% depth was 13.75 m in early August (more shallow beginning in late June and persisting through August). The 1% PAR depth in Lake Lacawac was 4.0 m in early April and slowly increased throughout the experimental period to 5.5 m in early August.

DNA damage induction The frequency of CPDs measured immediately following exposure of D. catawba to 160 2 kJ/m UVB (T0) for each experimental date is shown in Figure 3.2 (vertical line). Damage induction differences were observed, particularly on May 18 and June 1, but were not significantly different between populations (p>0.25) (Figure 3.2).

50 DNA repair There were significant CPD differences between experimental dates in both Lake Giles and Lake Lacawac D. catawba populations (p<0.0001) following the 6 h repair period (Figure 3.2). There were also significant differences in CPDs following the repair period by depth in Lake Lacawac D. catawba (p=0.0013) with the highest damage levels following repair observed at 3 m. Damage differences were not significantly different at the different depths in Lake Giles as a function of time (p=0.1839).

51 -1 LAKE GILES: CPDs mb DNA-1 LAKE LACAWAC: CPDs mb DNA

0 100 200 300 0 100 200 300

0.5m

1.0m

3.0m Depth (m)

April 6, 2006 April 5.0m

7.0m

0.5m

1.0m

3.0m Depth (m) 5.0m April 7, 2006 7.0m

0.5m

1.0m

3.0m Depth (m) 5.0m May 18, 2006 May 7.0m

0.5m

1.0m

3.0m Depth (m) 5.0m May 19, 2006 19, May

7.0m

52 LAKE GILES: CPDs mb DNA-1 LAKE LACAWAC: CPDs mb DNA-1

0 100 200 300 0 100 200 300

0.5m

1.0m

3.0m Depth (m) Depth

June 1, 2006 5.0m

7.0m

0.5m

1.0m

3.0m Depth (m) Depth 5.0m June 21, 2006

7.0m

0.5m

1.0m

3.0m Depth (m) Depth

July 18, 2006 5.0m

7.0m

0.5m

1.0m

3.0m Depth (m) Depth

July 19, 2006 5.0m

7.0m

53 LAKE GILES: CPDs mb DNA-1 LAKE LACAWAC: CPDs mb DNA-1

0 100 200 300 0 100 200 300

0.5m

1.0m

3.0m Depth (m) 5.0m August 8, 2006

7.0m

0.5m

1.0m

3.0m Depth (m) 5.0m August 9, 2006 August 9,

7.0m

Figure 3.2. DNA damage assessed by RIA in Daphnia and DNA dosimeters for the experimental period April – August 2006 in two temperate lakes at given depths. Initial DNA damage by 160 kJ/m2 UVB (vertical line) for each population was quantified immediately preceding deployment in the lakes. DNA damage in the dosimeters (points) (N=2) and Daphnia (bars) (N=3) was assessed following a six h solar incubation (simultaneous damage and repair). 1% UV-320 depth is given (dashed horizontal lines) as reference (Table 3.1).

Discussion

The quantification of environmental stress by molecular techniques in a natural system is problematic. Changing conditions add constant variability and individuals within populations may or may not respond similarly to the same stress. Further, the quantity of organisms required, particularly of Daphnia (100+ per sample), make it impossible to measure individual effects. By this, identifying the presence of a stressor and attempting to assess its direct (or indirect) effect on an individual, population, and/or community can in and of itself be daunting.

54 The variability in the induction of DNA damage observed in this experiment was not surprising. Individual variability, even in clonal species such as Daphnia, contributes to the range of damage susceptibility present in these populations. Furthermore, we believe the abiotic and biotic variability of the two source lakes may have contributed significantly to the susceptibility of Daphnia catawba to UV-induced DNA damage. Previous UVR exposure was likely to reduce susceptibility of individuals to UVR damage, and when damage was previously induced, repair rates are expected to be very efficient (high repair and survival). Lake Lacawac is a high DOC, low UVR system, suggesting low adaptation of D. catawba to UVR. Lake Giles is a low DOC, high UVR system, thereby suggesting the potential for high adaptation of D. catawba to UVR. However, the DNA damage induction patterns (Figure 1) do not reflect this hypothesis, and, actually, induction was almost always lower in the high DOC acclimated Lake Lacawac population (lower UVR tolerance?) as compared to the low DOC acclimated Lake Giles population (higher UVR tolerance?) on any given experiment date. The dosimeter data reported in Chapter 2 showed no quantifiable CPDs in Lake Lacawac and very low levels in Lake Giles between 0 and 3m in April. By this we would predict high levels of light repair in the surface waters with additional photoproduct induction. However, at depth highs level of dark repair would be occurring in the absence of additional photoproduct induction resulting in a significant decrease in final CPDs as compared to the surface samples. Furthermore, significant quantifiable DNA damage was observed in only Lake Giles in July and only down to 2m. Again, we would predict high levels of light repair in the surface waters and significant NER, thus reducing CPDs in the absence of additional UV-induced DNA damage. In this study we actually saw an increase in CPDs with depth in both lakes on at least two occasions, suggesting an accumulation of DNA damage in the absence of UV, which is impossible. We hypothesize that this increase in CPDs is an artifact resulting from the population sub-sampling technique associated with this type of experiment. Because we were not able to sub-sample a single individual during the experimental period, we relied on the clonal attributes of the Daphnia populations and biological uniformity within the population. While this is a logical design, it is feasible that different individuals will respond differently to the accumulation and repair of UV-induced DNA damage. In response to these results, we suggest that an increase in CPDs is a function of population variability and not the replication of DNA damage sites as

55 the data suggest. We believe future studies on a controlled population subset in the laboratory will support this hypothesis. The results of this study are not conclusive, particularly since we see an increase in CPDs beyond initial induction at the lowest depths which we cannot explain within the context of this experiment. In addition, heavy rains in June resulted in reduced transparency in both systems during July and August, causing both lower, atypical, UVR and PRR conditions during that period. Based on UVR transmittance data Lake Lacawac was less affected by these rain events compared to Lake Giles. However, we still observed increased CPDs at the greatest depths even late in the season when 1% PAR was nearly twice the depth of the 3 m experimental rack suggesting that sufficient light for PER was present. Due to the confounding, and uncontrollable, factors of this experiment, a study of UV-induced DNA damage and repair was conducted in the laboratory using four Daphnia spp. in an attempt to address some of the questions not sufficiently answered in the in situ experiments (see Chapter 4).

Acknowledgements

We thank Blooming Grove Hunting and Fishing Club (Lake Giles) and the Lacawac Sanctuary (Lake Lacawac) for limitless lake access that was crucial to the success of this experiment. We thank the Zoology Department (Miami University) for graduate research support to SJC. This work was partially supported by the National Science Foundation grant DEB-IRCEB-0552283 to CEW.

56 References

Alonso, C., V. Rocco, J. P. Barriga, M. A. Battini and H. E. Zagarese (2004) Surface avoidance by freshwater zooplankton: Field evidence on the role of ultraviolet radiation. Limnology and Oceanography, 49, 225-232.

Boeing, W. J., D. M. Leech, C. E. Williamson, S. Cooke and L. Torres (2004) Damaging UV radiation and invertebrate predation: conflicting selective pressures for zooplankton vertical distribution in the water column of low DOC lakes. Oecologia, 138, 603-612.

Cadet, J. and P. Vigny. 1990. “The photochemistry of nucleic acids.” In: Morrison, H. (ed). Bioorganic Photochemistry: Vol 1. Wiley, New York. 1-272.

Cadet, J., F. Odin, J. F. Mouret, M. Polverelli, A. Audic, P. Giacomoni, A. Favier and M. J. Richard (1992) Chemical and biochemical postlabeling methods for singling out specific oxidative DNA lesions. Mutation Research, 275, 343-54.

Clydesdale, G. J., G. W. Dandie and H. K. Muller (2001) Ultraviolet light induced injury: immunological and inflammatory effects. Immunology and Cellular Biology, 79, 547-68.

Cooke, S. L., C. E. Williamson, B. R. Hargreaves and D. P. Morris (2006) Beneficial and detrimental interactive effects of dissolved organic matter and ultraviolet radiation on zooplankton in a transparent lake. Hydrobiologia. 568: 15-28.

Fischer, J. M., P. A. Fields, P. G. Pryzbylkowski, J. L. Nicolai and P. J. Neale (2006) Sublethal Exposure to UV Radiation Affects Respiration Rates of the Freshwater Cladoceran Daphnia catawba. Photochemistry and Photobiology, 82, 547–550.

Karentz, D., J. E. Cleaver and D. L. Mitchell (1991) Cell survival characteristics and molecular responses of Antarctic phytoplankton to ultraviolet-B radiation. Journal of Phycology, 27, 326- 341.

57

Leech, D. M., A. Padeletti, C. E. Williamson (2005) Zooplankton behavioral responses to solar UV radiation vary within and among lakes. Journal of Plankton Research, 27, 461-471

Li, S., M. Paulsson, L.O. Bjorn (2002) Temperature-dependent formation and photorepair of DNA damage induced by UV-B radiation in suspension-cultured tobacco cells. Photochemistry and Photobiology B. 66, 67-72.

MacFadyen, E. J., C. E. Williamson, G. Grad, M. Lowery, W. H. Jeffrey, and D. L. Mitchell (2004) Molecular response to climate change: temperature dependence of UV-induced DNA damage and repair in the freshwater crustacean Daphnia pulicaria. Global Change Biology. 10, 408-416.

Matsumura, Y. and H. N. Ananthaswamy (2004) Toxic effects of ultraviolet radiation on the skin. Toxicology and Applied Pharmacology, 195, 298-308.

Mitchell, DL (2006) “Quantification of DNA photoproducts in mammalian cell DNA using radioimmunoassay.” In: Henderson, D.S. (ed) Methods in Molecular Biology, DNA Repair Protocols: 2nd Edition.Humana Press Inc, New Jersey. 239-249.

Mitchell, D. (1997) Ultraviolet radiation damage to DNA. In Encyclopedia of Molecular Cell Biology and Molecular Medicine: Volume 6. (Edited by R.A. Meyers), pp. 149-164. Wiley Press, New Jersey.

Mitchell, DL (1996) “Radioimmunoassay of DNA damage by ultraviolet light.” In: Pfeifer, G (ed) Technologies for Detection of DNA Damage and Mutations. Plenum Publishing, New York. 73-85.

Neale, P., R. Goodrich, and W. Brinley. (2005) The Smithsonian-NIST-USDA-FPL Network for Monitoring Solar Ultraviolet Irradiance: Comparison of Radiometer Measurements and Radiative Transfer Model Calculations. In Service Life Prediction: Challenging the Status Quo.

58 J.W. Martin, R.A. Ryntz, and R.A. Dickie, editors. Federation of Societies for Coatings Technology, Blue Bell, PA. 159-170.

Olson, M.H. and D. L. Mitchell. 2006. Interspecific variation in UV defense mechanisms among temperate freshwater fishes. Photochemistry and Photobiology 82: 606-610.

Quinn, G.P and M.J. Keough. Experimental Design and Data Analysis for Biologists. Cambridge University Press. New York. 2002.

Rae, R. and W. F. Vincent (1998) Effects of temperature and ultraviolet radiation on microbial foodweb structure: potential responses to global change. Freshwater Biology, 40, 747-758.

Ringelberg, J. & E. van Gool (2003) On the combined analysis of proximate and ultimate aspects in diel vertical migration (DVM) research. Hydrobiologia 491: 85–90.

Strecker, A. L., T. P. Cobb and R. D. Vindebrooke (2004) Effects of experimental greenhouse warming on phytoplankton and zooplankton communities in fishless alpine ponds. Limnology and Oceanography, 49, 1182-1190.

Williamson, C.E., D.P. Morris, M.L. Pace, and O.G. Olson (1999) Dissolved organic carbon and nutrients as regulators of lake ecosystems: Resurrection of a more integrated paradigm. Limnology and Oceanography. 44(3, part 2): 795-803.

Williamson, C. E. and G. Grad (2002) Temperature-dependent ultraviolet responses in zooplankton: Implications of climate change. Limnology and Oceanography, 47, 1844-1848.

Winder, M., P. Spaak and W. M. Mooij (2004) Trade-offs in Daphnia habitat selection. Ecology, 85, 2027-2036.

Zellmer, I. D. (1995) UV-B tolerance of alpine and arctic Daphnia. Hydrobiologia, 307, 153- 159.

59

Zellmer, I. D., M. T. Arts, D. Abele and K. Humbeck (2004) Evidence of sublethal damage in Daphnia during exposure to solar UV radiation in subarctic ponds. Arctic, Antarctic, and Alpine Research, 36, 370-377.

60

Chapter 4

Temperature effects on UV-induced DNA damage and its repair: Biological responses of four species of the freshwater cladoceran Daphnia

To be submitted: Sandra J. Connelly, Craig E. Williamson, Robert E. Moeller, and David L. Mitchell. Temperature Effects on UV-Induced DNA Damage and Repair: Biological Responses of Four Species of the Freshwater Cladoceran Daphnia. To Photochemistry and Photobiology.

61 Abstract

The responses of four freshwater daphniid species (Daphnia middendorffiana, D. pulicaria, D. pulex, and D. parvula) to a single acute dose of ultraviolet B (UVB) radiation were compared. In addition to survival, the induction and repair of cyclobutane pyrimidine dimers were quantified using radioimmunoassay immediately after UVB exposure and at increasing times post- irradiation in the presence and absence of photoreactivating light. A broad spectrum of survival and an equally broad diversity in photoprotection and light- and dark-dependent DNA repair capacity were observed. All four daphniid species showed high levels of photoprotection against DNA damage induced by UVB irradiation when compared to damage induced in purified DNA solutions (DNA dosimeters), with significant variation among species. In addition, considerable variation was observed between species in their ability to repair ultraviolet radiation (UVR) photoproducts using photoenzymatic repair (PER) and nucleotide excision repair (NER). Contrary to our predictions, both PER and NER occurred at significantly higher rates at 10ºC compared to 20ºC. Damage reduction by NER was only significant at 10°C. Analysis of melanin concentration in the whole animal, carapace, and ephippia of D. middendorffiana versus the non- melanic D. pulicaria was conducted at 10, 15, and 20°C. Ephippia and carapaces of D. middendorffiana showed no significant difference in melanin concentration at the three temperatures. These data show a high reliance of these Daphnia species on DNA repair mechanisms as opposed to photoprotection (melanin) at colder temperatures.

62 Introduction

Decreased stratospheric ozone has historically altered the ultraviolet radiation (UVR) penetration of the atmosphere (Madronich et al, 1998) and current elevated UVR conditions are likely to persist through mid century (McKenzie et al., 2007). DNA is thought to be the primary target of UVR damage and induction of damage in DNA is linearly related to UVB exposure (dose) in isolated DNA (Mitchell and Karentz, 1993). Less is known about either the extent of variability in net DNA damage among closely related organisms (e.g. in the same genus), the relative role of UVR tolerance versus molecular repair processes, or the temperature dependence of these repair processes. UVR induces DNA damage by the formation of photoproducts which interfere with or terminate DNA replication and transcription (Mitchell and Nairn, 1989). Cyclobutane pyrimidine dimer (CPDs) account for 80-90% of the photoproducts induced by UVB at the earth’s surface (290-315nm). To maintain DNA replication and cell function, the CPDs are repaired by the organism using nucleotide excision repair (NER) and photoenzymatic repair (PER). NER is present in all organisms and uses adenosine triphosphate (ATP) to remove large sections of nucleotide bases containing any damaged DNA. In contrast, PER is not present in all organisms (e.g., Li et al., 1993; Mitchell and Karentz, 1991) and is driven by a UVA/visible light enzyme reaction that is believed to be physiologically benign (Essen and Klar, 2006). Further, PER is specific to UV-induced DNA damage, potentially increasing the efficiency of the repair (e.g., Rocco et al., 2002; Malloy et al., 1997; Mitchell, 1997). Based on the enzyme kinetic hypothesis, which suggests that an optimal temperature exists for any enzymatic reaction, the influence of temperature on both NER and PER is believed to directly impact the rate of DNA repair (Sancar and Sancar, 1988). In natural systems, UV induced DNA damage has been detected in freshwater species (Zellmer et al 2004). The distribution of organisms in natural systems will affect the level of DNA damage sustained and their ability to counteract this damage. Species such as Daphnia exhibit diurnal migration and thus are thought to avoid high levels of UV induced DNA damage (Alonso et al, 2004). This behavior is costly due to exposure to decreased temperatures and potentially lowers food levels at lower depths (Winder et al, 2004). Furthermore, although this behavioral avoidance of surface waters decreases UVR exposure, Daphnia are still be exposed to

63 levels of UVR with damage induction potential even at increased depths in the more transparent lakes (Morris et al, 1995; Williamson, 1995). DNA repair in cold water systems, such as in alpine and Polar regions, is likely to be retarded and the constant high levels of UVR exposure may result in the accumulation of DNA damage (e.g., Li et al., 2002; Mitchell, 1997). The dependence of an organism on the avoidance of, versus the repair of, UV-induced DNA damage is reported throughout the literature as being species-dependent. Avoidance can be behavioral, as in the vertical migration as previously described (e.g., Leech et al., 2005), or morphological, as in the use of photoprotective compounds, such as mycosporine-like amino acids (MAA’s; e.g., Tartarotti et al., 2004) or pigmentation such as carotenoids (e.g., Hairston, 1976) or melanin (e.g., Rautio and Korhola, 2002). An organism’s ability to avoid and repair DNA damage will ultimately determine the degree of DNA damage. Pigmentation in some species, such as Daphnia, is induced in response to UVR exposure (Hessen, 1996; Gerrish and Caceres, 2003). Melanic clones of Daphnia pulex are less susceptible to UVR stress than the non-melanic clones of the same species; however, they also exhibit slower growth rates, suggesting a significant physiological cost of pigmentation (Hessen, 1996). Further, induced pigmentation is confounded by a concomitant increase in the susceptibility of an individual to visual predation (e.g., Sægrov et al., 1996; Tollrian and Heibl, 2004). Species may counteract the negative aspects of pigmentation by relying more on DNA repair mechanisms, or some combination of strategies, under otherwise favorable environmental conditions (e.g., temperature). Zagarese et al (1997) report that in three species of calanoid copepods (Boeckella) photoprotection and photorepair strategies are species-dependent, and one species relies on both mechanisms for increased survival in high UVR conditions. This approach to minimizing DNA damage is likely most effective when both avoidance and repair are present. The temperature dependence of melanin induction has not been adequately addressed in the literature (ephippia: Gerrish and Caceres, 2003). Gerrish and Caceres (2003) acclimated Daphnia pulicaria clones from five lakes to three temperatures (15, 20, and 24°C) for multiple generations. Greater variation in the pigmentation of ephippia between clones in a single lake (80.2% of variance) as compared to temperature (1.7% of total variance) was reported. Despite its overall low contribution to the ephippial pigmentation variance, temperature effects on melanin were significant, suggesting that temperature may play a critical role in species which are exposed to long durations of UVR at cold temperatures.

64 MacFadyen et al (2004) tested the effect of temperature on DNA repair rates, both NER and PER, following artificial UV-B exposure in Daphnia pulicaria at 5, 15, and 25°C by quantifying CPDs. In these experiments, Daphnia were collected from a single source pond; gravid-stage adults were isolated and acclimated to the experimental temperatures for three days prior to UVR exposure. The authors provided damage (UVB) and repair radiation (UVA/PAR) during a 12 h exposure period and report that photoprotection played a significant role in DNA damage reduction (>95%), much more so than either DNA repair process (MacFadyen et al., 2004). While not playing as great of a role as photoprotection in reducing DNA damage accumulation, MacFadyen and colleagues do report a significant effect of temperature on PER and total accumulated CPDs. They did not observe a significant effect of NER on total CPDs. While their findings give support to previously published studies demonstrating the dependence of Daphnia on light-dependent repair (PER) for survival following UVR exposure (e.g., Williamson et al, 2002), the short temperature acclimation time is not likely to be sufficient (i.e., the longer temperature acclimation post- than pre-UVR exposure) to demonstrate significant temperature effects in these organisms. The approach by MacFadyen et al (2004) that combined damage and repair radiation during the exposure period makes it difficult to isolate the direct effect of repair on the cumulative DNA damage. While this simultaneous exposure of damage and repair radiation for a prolonged period is more similar to natural conditions, it does not allow for a clear distinction between repair processes. In contrast an acute high intensity, short duration UVB exposure followed by a fixed repair period would allow for a more direct measurement of the repair processes independent of damage. Here we test the temperature dependence of two DNA repair mechanisms (i.e., PER and NER) of four freshwater Daphnia species, including D. pulicaria, D. middendorffiana, D. pulex, and D. parvula following exposure to artificial UVR at 10° and 20°C. DNA damage was quantified in all species as CPDs using RIA. Melanin was quantified at 10°, 15°, and 20°C in D. middendorffiana and D. pulicaria.

65 Materials and Methods

In order to better understand the molecular mechanisms underlying the observed differential survival, DNA damage induction and repair analyses were performed. All Daphnia were acclimated to 10, 15, or 20°C from neonate (< 24 h) to gravid stage (developmental stage temperature acclimation) prior to any UVB exposure or melanin extraction. Three experiments were conducted: (1) induction of DNA damage (measured as CPDs) using 20 kJ/m2 and the subsequent survival of D. middendorffiana, D. pulicaria, D. pulex, and D. parvula at 10 and 20°C; (2) repair of CPDs by NER and PER by D. middendorffiana, D. pulicaria, D. pulex, and D. parvula at 10 and 20°C; and (3) melanin concentration of whole animal, carapace, and ephippia of D. middendorffiana versus the non-melanic D. pulicaria at 10, 15, and 20°C.

Daphnia Daphnia were acquired from the field or a biological supplier. All Daphnia were cultured as laboratory populations for more than six months prior to the start of these experiments. The highly melanic Arctic species D. middendorffiana (courtesy of France Dufresne, Université de Québec, Québec, Canada) were initially cultured under a 24 h light regime at 15°C [Thermo Electron Corporation 818 Illuminated Incubator, Marietta, OH; PAR = 64 μEin/m2s (quantified using a spherical detector QSL-2101; Biospherical Instruments, San Diego, CA, USA)]. Three other non-melanic Daphnia species were used in this study. D. pulex (Carolina Biological Supply, NC), D. pulicaria (Dutch Springs, Bethlehem, PA; 40°41'04.99”N, 75°21'16.26”W), and D. parvula (isolated clone from Acton Lake, OH; 39°34'21.50”N, 84°45'03.76”W) were cultured under a 16:8 light:dark regime at 20°C (Precision Scientific 818 Illuminated Incubator, Chicago, IL, USA; PAR = 54 μEin/m2s). All cultures were kept in synthetic freshwater (pH 7.0; US EPA 2002) and fed Selenastrum at 6x105 cells/mL culture every 72 h. All animals were sorted and placed in the dark 24 h prior to any exposure.

UVR irradiation UVB and photoreactivating radiation (PRR) exposures were conducted in an environmentally controlled chamber (Environmental Room, Environmental Growth Chambers, Chagrin Falls, OH) at 10° or 20°C. For a complete description of the UVR-lamp system see Williamson et al,

66 2001. The UVB source consisted of three Spectronics Spectroline BLE-1T158 15W lamps (Westbury, NY, USA) yielding a spectral output of 281-405 nm (for damage radiation emission spectrum see Figure 4.1, Panel A). In contrast to Williamson et al. (2001), no acetate film was used in this experiment in order to induce higher damage during the exposure period. Quartz Petri dishes (Quartz Scientific, Fairport Harbor, OH, USA) containing the Daphnia and 30 mL synthetic freshwater were placed inside black collars on a rotating platform (2 rpm) beneath the UVB lamps to better control the UVR irradiance field and to assure uniform exposures. Samples were exposed for 15 min shielded with either a neutral density screen (stainless wire mesh, 60x60 mesh per linear inch transmitting 37.5%; McMaster Carr cat. no. 9238T336, Princeton, NJ, USA) yielding a total dose of 20 kJ/m2 or an opaque disk to produce the unirradiated controls. All exposures were done for the same length of time and in the same culture conditions with temperature as the only variable. For the repair experiments, the quartz Petri dishes were placed on the rotating platform (2 rpm) and exposed from below to PRR for variable durations. The PRR consisted of two Q-Panel UVA 340 lamps (Q-Panel Labs, Cleveland, OH, USA) and two CoolWhite lamps (Sylvania Premium CoolWhite 40W) filtered through two sheets of 12-μm thickness Mylar®D (E.I. Dupont & Co., Inc, Wilmington, DE, USA). For repair radiation emission spectrum see Figure 4.1, Panel B.

1000 UV-B Damage Repair

100 nm 2

10

Irradiance mW/m Irradiance 1

0.1 300 400 500 600 300 400 500 600 Wavelength (nm) Wavelength (nm)

Figure 4.1. Emission spectra. The emission spectra for the lamp system used to induce DNA damage (Panel A) and to photoreactivate damage after induction (Panel B).

67

DNA Dosimetry Two types of experiments were done. For the DNA damage induction experiments, animals were exposed to UVB alongside a DNA "dosimeter" consisting of purified herring testes DNA (Sigma Aldrich) diluted to 90 µg/mL sealed in a 2-cm UVR-transparent quartz tube. Dosimeters were used to quantify the potential DNA damage induced during the same exposure period in the absence of DNA repair mechanisms. Immediately following exposure, both the animals and dosimeters were frozen. For the DNA repair experiments, cultures were exposed to UVB and then either immediately placed in the dark or exposed to PRR for 0, 0.75, 1.5, 3, 6, 12, 24 and 48 h. Repair samples from each time point were immediately frozen at -20°C until extraction and analysis. DNA dosimetry is used to estimate the contribution of photoprotection to reduced DNA damage. Photoprotection herein is not representative of melanin concentration; it is the quantitative difference between damage induced in Daphnia versus that induced in purified DNA in the absence of all biological processes. This provides an estimate of the biological processes (including, but not limited to, carapace pigmentation, photoprotective compounds, and DNA repair processes) that play a role in both DNA damage induction and the rate at which DNA repair occurs. The carapace can also function to block transmission of UVR by reflectivity and refraction. As is subsequently discussed, melanin concentrations were estimated using previously published techniques (see, “Pigment analysis” in this section).

Survival Following exposure to 20 kJ/m2 UVB irradiation each species was divided into PRR and no PRR (dark) treatments and immediately placed under those conditions as described previously. All animals were maintained in treatment for 9 days with feedings of Selenastrum every 24 h. Survival was recorded at each feeding by the presence/absence of a visual heart beat. Control animals (no UVB irradiation exposure) of each species were maintained under treatment conditions for the duration of the experiment and their survival exceeded 90% for the duration of this experiment. Only 24 and 48 h survival data are presented here for direct comparison to the DNA repair experiment. Percent survival data and sample variance within a treatment (N=3 replicate dishes) are reported.

68

DNA Damage analysis Subsequent to UVR exposure and repair, samples were immediately frozen and stored at -20°C. Samples were then thawed in a 2.0 mL microcentrifuge tube (Eppendorf) containing 750 µL of lysis buffer solution (10 mM Tris-Cl, pH 8.0 (Sigma), 1 mM EDTA pH 8.0 (Sigma), and 0.25% sodium dodecyl sulfate (Boehringer-Mannheim); 21). After incubation with 10 µg/mL heat inactivated RNAse (Sigma) at 37°C for 1 h the lysates were treated with 30 µL Proteinase K (10mg/mL) (Boehringer-Mannheim) in a variable speed shaker (Eppendorf) overnight at 37°C with vigorous mixing. Following deproteination, samples were extracted sequentially with equal volumes of buffer-saturated phenol (Boehringer-Mannheim), phenol:Sevag (1:1), and Sevag (24:1 chloroform:isoamyl alcohol). Finally, the aqueous phase was transferred to a centrifuge tube containing 1/10 volume 3 M sodium acetate (pH 5.2). One volume ice-cold isopropanol was added and the mixture was inverted several times before overnight precipitation at -20°C. Following precipitation samples were centrifuged at 10,000 rpm (Eppendorf μicro, Eppendorf, Westbury, NY) for 10 min at 4°C, washed with 70% ethanol, air-dried and resuspended in 1 mL 10 mM Tris, 1 mM EDTA (pH 8.0). The DNA concentration and purity were determined by absorbance at 260/280nm using a NanoDrop spectrophotometer (NanoDrop Technologies, Wilmington, DE). CPDs were quantified using radioimmunoassay (RIA). The RIA is a competitive binding assay between very small amounts of radiolabeled and sample DNA for an antisera raised against UV-irradiated DNA. For the RIA, 50 ng of heat-denatured sample DNA was incubated with 5-10 pg of poly(dA):poly(dT) (labeled to >5 x 108 cpm/µg by nick translation with 32P-dTTP) in a total volume of 1 mL 10 mM Tris, pH 7.8, 150 mM NaCl, 1 mM EDTA, and 0.15% gelatin (Sigma). Antiserum was added at a dilution that yielded ~50% binding to labeled ligand, and after incubation overnight at 4ºC the immune complex was precipitated with goat anti-rabbit immunoglobulin (Calbiochem) and carrier serum from non-immunized rabbits (UTMDACC, Science Park/Veterinary Division, Bastrop, TX). After centrifugation, the pellet was dissolved in tissue solubilizer (NCS, Amersham), mixed with ScintiSafe (Fisher Scientific) containing 0.1% glacial acetic acid, and the 32P quantified by liquid scintillation spectrometry. Under these conditions, antibody binding to an unlabeled competitor inhibits antibody binding to the

69 radiolabeled ligand. These details, as well as those concerning the specificities of the RIAs and standards used for quantification, have been described in detail by Mitchell (1997).

Pigment analysis The temperature-dependence of melanin content of adult female D. middendorffiana and D. pulicaria reared at 10, 15, or 20°C was determined. Whole animal samples excluded individuals carrying ephippia. Ephippia and molted carapaces were analyzed separately. Samples were leached in ethanol, then dried and weighed. These were then homogenized in 5N NaOH using a sonicator. Melanin was extracted from the homogenate by five daily treatments of heating to 65°C and subsequent cooling to room temperature over a 5-day period (Hobæk and Wolf, 1991). Extracts were filtered (0.2μm Whatman nylon syringe filter) and absorbance of the filtrate was scanned (Shimadzu UV-VIS Spectrophotometer, UV1650PC, Shimadzu North America) to obtain average absorbance at 350, 400, and 450 nm. Samples were compared to a standard curve generated from synthetic melanin (M8631, Sigma Aldrich) dissolved in 5N NaOH. Whole animal “melanin” extracts of D. middendorffiana were corrected for background absorbance in non-melanic D. pulicaria. Results are reported as μg melanin/mg dry weight. Pigment analyses of other species were not conducted in this study due to their highly transparent nature and low population numbers available for the extractions.

Data analyses

RIA was used to quantify CPD frequencies at T0 and at 3, 6, 12, and 24 h post irradiation (data not shown). To estimate the relative degree of photoprotection we compared the amount of damage induced in each species with that induced in purified DNA by the same UVB exposure 2 (20 kj/m ) (Figure 4.3). We corrected the CPD frequency data to the amount of damage at T0 (T0 CPDs = 100%) because of the variable induction rates between species and used Sigma Plot (v. 9.0, SPSS Inc., Chicago, IL) to construct a 3-parameter exponential decay function y = r [y(0)] + ae-bx where a is the initial CPD frequency after exposure (amount of damaged induced by 20 2 kJ/m independent of existing damage), b is the inverse of the repair half-life (t1/2) for CPD repair in hours, x is hours post exposure, and r [y(0)] is the asymptote of the repair curve, or the amount of residual (unrepairable) CPD damage (Mitchell et al., 1999). The data derived from the

70 exponential repair curves, along with the correlation coefficients, are summarized in Table 4.2.

The t1/2 value was used to calculate the rate of CPD repair and is shown as the number of CPDs removed per megabase DNA in 1 h (Figure 4.4). The effects of temperature and repair (PER or NER) on total CPDs were evaluated using two-way ANOVA independently for species (SPSS v. 15, SPSS, Inc., Chicago, IL) (Quinn and Keough, 2002).

Results

Survival The effects of temperature on survival after exposure to 20 kJ/m2 UVB followed by 24 and 48 h incubation under PRR conditions or in the dark post UVB-exposure are shown in Figure 4.2 (treatment replicates N=3, with 9-30 individuals per replicate). Survival tended to be higher in the presence of PRR, independent of temperature and species. Under PRR conditions all species survived at 10ºC and all but D. parvula were able to survive the 20 kJ/m2 UVB dose at 20ºC for 48 h. In contrast, only D. middendorffiana and D. pulex were able to tolerate this dose at 10ºC in the absence of PRR, and only D. pulex survived in the absence of PRR at 20ºC. D. middendorffiana and D. pulex showed comparable levels of survival at 10ºC whether PRR was present or not and had the greatest tolerance to UVB. D. parvula was the most sensitive of the species tested showing low survival by 48 h under all conditions except in the presence of PRR at 10ºC.

71 D. middendorffiana D. pulex

100

80

60

% Survival 40

20

0

D. parvula D. parvula

100 10C NER 10C PER 20C NER 80 20C PER

60

% Survival 40

20

0 0 50 100 150 200 250 0 50 100 150 200 250 Time (hrs) Time (hrs)

Figure 4.2. Survival over 9 days under two repair regimes in daphniids following acute UVB exposure (20 kJ/m2). Data are shown for incubations in the presence of PRR (PER+NER) or in the absence of PRR (NER alone). % survival ± treatment variance (N=3).

Damage induction and photoprotection 2 Quantification of CPDs immediately following exposure of Daphnia to 20 kJ/m UVB (T0) show both species and temperature differences. Significantly greater DNA damage was observed in D. pulicaria and D. pulex (p<0.04 and p<0.02, respectively) than in D. middendorffiana and D. parvula. In D. parvula, D. pulicaria, and D. pulex, increasing the acclimation/exposure temperature from 10ºC to 20ºC resulted in higher levels of CPD induction. However the only significant temperature effect was observed in D. parvula (p=0.0134). These results are consistent with the reduced survival at 20°C, particularly evident under both repair regimes in D. pulicaria and D. parvula (Figure 4.3).

72 It is evident that a high degree of potential photoprotection is afforded these organisms and that photoprotection varies between individuals (Figure 4.4). The amount of CPDs induced in D. middendorffiana and D. parvula is only 10% to 20% of that induced in the DNA dosimeter, indicating that structural and biochemical mechanisms to reduce the amount of DNA damage is very effective in these species. Of the species tested, D. parvula demonstrates the highest level of photoprotection at both 10° and 20°C (p<0.04 and p<0.01, respectively). Whereas, D. pulex DNA accumulates > 60% of the CPDs induced in the DNA dosimeters as compared to D. parvula with < 20% of the dosimeter CPD induction at 20ºC (p=0.0066).

1000

800

600

400 *

200 Acute UV Induced CPDs/mb DNA CPDs/mb Induced UV Acute

0 a a x ffian icari pule rvula ndor . pul D. . pa idde D D D. m

Figure 4.3. Cyclobutane pyrimidine dimer induction. CPD frequencies in Daphnia immediately 2 after exposure to 20 kJ/m (T0) at 10º (black bars) and 20ºC (gray bars) are shown. CPDs per megabase DNA are shown with the standard error of the mean (SEM) (n = 12). Significant effects of species and temperature were calculated using two-way ANOVA (* significant temperature effect; p=0.0134; N=12).

73 100 x b x,y a a 80 a y

y,z 60

40 Photoprotection (%) Photoprotection

20

0 na ria ex la fia ca ul rvu orf uli . p pa nd . p D . de D D mid D.

Figure 4.4. Photoprotection. The degree of photoprotection for each species is expressed as the percentage of CPDs induced in the samples relative to the frequencies measured in a purified DNA dosimeter irradiated with the same fluence of UVB at the same time. Temperature data from 10º (black bars) and 20ºC (gray bars) are shown. Significant effects of species and temperature were calculated using two-way ANOVA. Data with the same letter are not statistically significant within temperature at α=0.05.

Pigmentation Whole D. middendorffiana raised at 10° and 15°C showed no difference in melanin concentration (4.7 and 4.2 μg/mg dry weight, respectively) when corrected for background using intact D. pulicaria. The analysis of whole animals raised at 20°C was unsuccessful. Analysis of the ephippia and carapaces of D. middendorffiana showed no significant difference (p>0.2) in melanin concentration at the three temperatures. Melanin concentrations ranged from 23 - 35 (ephippia) and 29 – 53 (carapaces) μg/mg dry weight at 10°C. Concentrations at 20°C ranged from 31 – 41 (ephippia) and 47 – 69 (carapace) μg/mg dry weight.

DNA repair Similar to the diversity observed in induction frequencies we see variation in PER between species and between the two test temperatures (Figure 4.5 and 4.6). The repair rates of all species were comparable in the presence of photorepair radiation at both 10° and 20°C. The half lives of

74 repair were short (< 5 h) and similar between D. parvula and D. middendorffiana at 10°C NER, 10°C PER, and 20°C PER (Figure 4.5). D. pulicaria and D. pulex showed longer repair rates (> 12 h) at 10°C NER than at 10° or 20°C PER. D. middendorffiana and D. pulex demonstrated no repair at 20°C NER. The half lives of both NER and PER in D. parvula and D. pulicaria were highly variable but generally longer at 20°C than 10°C. There was a significant effect of species by repair regime (two-way ANOVA; p<0.0321).

30 D. parvula D. pulicaria 25 D. middendorffiana D. pulex 20

15

10

5 Repair rate half life (hrs) 0

10c PER 10c NER 20c PER 20c NER

Figure 4.5. DNA repair rate half lives. Time in hours needed to repair 50% of the initial UV- induced CPDs. Rates of repair as damage removed were calculated from the t½ derived from the exponential decay values shown in Table 4.1. Significant effects of species and repair regime (temperature and repair) were calculated using two-way ANOVA.

No repair was observed in D. middendorffiana and D. pulex at 20°C NER. At 20ºC (gray bars) PER residual damage was comparable in D. middendorffiana and D. parvula, whereas it was much higher in D. pulicaria and D. pulex (Figure 4.6). At 10ºC (black bars) PER was much more

75 efficient than NER in all species except D. middendorffiana where it occurs at the lowest rate observed in all species examined. NER was much less effective than PER in these Daphnia species. However, we do see evidence of NER in three of the four species at 10°C. We observe no NER in D. middendorffiana at either temperature.

100 c A

80

b,c 60

40 b b b 20 a a a

0 aaaaa 100 B

80

60

b b 40 b Non-repairable cyclobutane pyrimidine dimers (% of total CPDs) of total (% dimers pyrimidine cyclobutane Non-repairable

20

0 na ria ex la fia ca ul rvu orf uli . p pa nd . p D . de D D mid D.

Figure 4.6. Residual damage. The percentage of total induced damage that could not be repaired was calculated from the exponential curves [y(0) in Table 4.2] for each species. Residual CPDs remaining following repair in the presence of PRR are shown in Panel A; non-repairable CPDs that remain following repair in the absence of PRR are shown in Panel B; 10ºC (black bars) and 20ºC (gray bars). Within a panel (A or B), data with the same statistical indicator letter are not statistically significant at α=0.05.

Fewer CPDs remain following NER and PER at 10ºC than at 20°C for all species except D. middendorffiana NER (Figure 4.6; Table 4.1). These data are consistent with the PER and NER data described in Figure 4.4. For example, species that showed efficient PER at 10ºC (e.g.,

76 D. parvula) and 20ºC (e.g., D. parvula and D. middendorffiana) showed lower levels of residual damage; species that showed little PER at 20ºC (e.g., D. pulex) and negligible NER at 20ºC (i.e., all species tested) or 10ºC (e.g., D. middendorffiana) showed very high levels of residual damage. In contrast, under some conditions there was no apparent correlation between the initial repair rate and the level of residual damage. For example, although we observe NER at 10°C in D. parvula, the corresponding half-life of repair (Figure 4.5) and level of residual damage were quite high (4.6 Panel B). Similarly, reduced rates of PER were observed in D. middendorffiana, D, pulicaria and D. pulex at 10ºC relative to D. parvula (p<0.04) and yet the residual damage was low in all four species.

Discussion

Daphnia species differ in their susceptibility to UVB-induced DNA damage and their dependence on PER and NER. The lethal effects of UVB are mitigated by, and more uniform following, exposure to PRR. Indeed, with the exception of D. pulex there was negligible recovery of all of the Daphnia species tested from the lethal effects of UVB at 20°C in the absence of PRR. Surprisingly, when the same survival experiment was performed in Daphnia acclimated to a lower temperature (i.e., 10 °C) all four species show high levels of survival at 24 h post-irradiation (Figure 4.2). When lethality was examined at a later time point (i.e., see 48 h in Figure 4.2) the dynamics change somewhat and we see reduced survival, particularly in the organisms incubated in the absence of PRR. The temperature effects are more pronounced in some species compared to others. For example, survival in D. middendorffiana and D. pulex was robust at 24 and 48 h regardless of the presence or absence of PRR at 10°C but falls precipitously in the absence of PRR at 20°C. Compared to the other species tested, D. parvula was not well adapted to UVB exposure. It displays the lowest survival in the presence of PRR and was unable to survive the 20 kJ/m2 dose of UVB even in the presence of PRR at 20°C. Ultimately, D. pulicaria, D. pulex, and D. middendorffiana showed high survival (>95%) at 10°C in the presence of PRR at 216 h post exposure (data not shown). This long-term survival at 10° vs. 20°C refutes the assumption that mortality induced by UVB-induced DNA damage was simply delayed at lower temperatures (Figure 4.2).

77 The high variability in response of Daphnia spp. to the environmental genotoxin UVB, specifically survival, was similar to what has been reported in other genera (Karentz et al., 1991; Joux et al., 1999; Van de Poll et al., 2002). In contrast to what has been found in other systems (e.g., Van de Poll et al., 2002), we see very significant temperature effects on survival in Daphnia following UVB exposure, particularly in the absence of PRR. The 24 h data show more pronounced differences in the lethal response compared to 48 h and may better reflect the environmental response of organisms that are exposed to 24 h diel cycles of PRR from sunlight. Survival at 48 h may be exacerbated by long-term culture effects, particularly in the dark, but do quantitatively reflect the same differences between species observed at 24 h. The observation that a pelagic organism, in this case Daphnia, displays a very robust biological response (survival) to UVB when acclimated to a lower temperature is a novel and significant finding with important environmental implications. By migrating to depth and being exposed to colder temperatures, Daphnia may be increasing their survival by increasing their rate of CPD repair. This is contrary to previous work suggesting that the enzyme kinetic hypothesis as it is related to DNA repair and temperature would hold in natural populations. Our results for DNA damage induction and repair in Daphnia are, for the most part, consistent with the observed lethal response and show similar striking temperature effects on light-dependent and nucleotide excision repair mechanisms. There is significant variation in the amount of DNA damage induced in the different species, most pronounced at 20°C. The degree of photoprotection is high in all species and appears to be a major strategy for mitigating DNA damage in response to sunlight exposure. As discussed, this variability between temperatures cannot be directly attributed to any differences of melanin concentration in the D. middendorffiana. Several mechanisms may come into play, however, that would apply to all species tested herein. First, although Daphnia is fairly transparent, it is a multicellular organism and only a limited portion of the cells within the body are exposed directly to UVR. The specific locations of the melanin may be extremely significant in protecting vital systems from UVR. Since we purify DNA from the entire animal it is probable that much of what we perceive as photoprotection is a dilution effect caused by the mixing of unirradiated with irradiated cells. In addition, Daphnia has an exoskeleton which may in itself provide some physical shielding of UVR. Our experience with fish indicates that scales are very effective shields against UVR penetration into the epidermis (Meador et al., 2000). These basic structural factors that may

78 affect UVR absorbance should be uniform in Daphnia and should not contribute to the variation we see between the different Daphnia species. We anticipate that pigment effects on reducing UVR absorption in D. middendorffiana would be greater than those observed in the other species tested. While D. middendorffiana in these experiments were notably melanic as compared to the other more pale Daphnia species, UVR-depauperate culture conditions theoretically limited the induction of high levels of melanin and concomitantly reduced the potential for additional UVR protection that is afforded natural populations of the species. However, our analysis of melanin concentration did not support the temperature dependent induction theory, showing no dependence of melanin production on temperature in D. middendorffiana (whole animal, ephippia, or carapace) under the culture conditions described herein. Anecdotally, the authors did observe increased melanotic pigmentation of D. middendorffiana when reared under higher UVR conditions (no quantitative data), and acknowledge that under natural light conditions, temperature might play a more important role in the production and sequestration of melanin. DNA repair in Daphnia is highly variable and temperature dependent. PER is the dominant repair mechanism operative in the daphniids examined in this study and is likely the dominant DNA damage removal response in all species in this genus. In three of the four species tested PER is more efficient at 10°C and in one species, D. pulex, PER is observed only at this lower temperature. In contrast, D. middendorffiana shows much more efficient PER at 20°C, presumably associated with its ability to adapt to warmer temperatures and concomitantly increase the rate of its enzymatic processes. However, D. middendorffiana is the unusual species tested with no detectable dark repair at either acclimation temperature, whereas the remaining three species exhibit no detectable NER at 20°C, yet demonstrate reasonable rates of NER at 10°C. This variation in D. middendorffiana may be attributed to its natural environment (and culture conditions) being 24 h light, potentially enhancing the PER pathway and relying almost solely on it for repair of UV-induced DNA damage. Ultimately, how these DNA damage and repair parameters relate to survival in the environment is the important question. The survival of Daphnia following UVB exposure correlates well with PER efficiency (Table 4.2). In most cases, the rate of PER correlates with the initial amount of damage, that is, when low levels of CPDs are observed the rate of PER is high (or reasonably high) and when high CPD frequencies are observed the rate of PER is reduced or very low. It is difficult to make

79 similar kinetic comparisons with NER since the rates are relatively low compared to PER and only occur at the low temperature. However, the occurrence of NER correlates, in nearly all cases, with survival in the absence of PRR. Two exceptions are noted: (1) D. middendorffiana survives well in the dark at 10°C in the absence of any detectable CPD repair, and (2) D. pulex survives well in the dark at 20°C in the absence of any detectable CPD repair. These inconsistencies suggest that other mechanisms may be at work in the recovery process. We believe that the inability to repair DNA damage and the persistence of such damage in individual organisms (Figure 4.6) will impact the survival of these organisms in response to UVB radiation in the environment. However, we have examined only one type of damage and one repair pathway. NER proceeds by two pathways, including global genome repair (GGR), which we measure with the RIA and transcription coupled repair (TCR) which comprises only a small portion of the induced damage and is not detected using RIA. It is possible that DNA damage in essential genes is efficiently repaired in the two exceptions noted above by TCR. It is also possible that the CPD is not the lethal lesion but that another photoproduct, such as the pyrimidine-pyrimidone (6-4) dimer [(6-4)PD] is actually responsible for cell killing and the failure of Daphnia to survive in response to UVR. This possibility is currently under investigation. Our results suggest that the ability to repair DNA damage at different rates at different temperatures is an important adaptation to exposure to the harmful UVB component of sunlight. They suggest that when pelagic organisms are emerging in early spring when light conditions are high but lake temperatures are at their lowest levels, NER may contribute much more significantly to UVR tolerance. In addition, the ability to repair DNA damage using NER rather than PER may be an important adaptation for organisms that migrate to lower depths in the water column where PRR conditions and temperatures are reduced compared to the surface. Our analyses also indicate that the response to UVR is complex and that different species have adapted different strategies for tolerating sunlight exposure. For instance, D. middendorffiana exhibits high survival at cold temperatures in the presence of PRR. This is comparable to its high Arctic source waters where they are most abundant in naturally cold and very clear waters during long solar photoperiods. These data provide significant evidence for UVB effects on population dynamics in conjunction with a second abiotic environmental stressor, temperature.

80 Acknowledgements

We thank the Zoology Department (Miami University) for graduate research support to SJC and Erin Overholt for laboratory assistance. We thank Lakshmi Paniker and Gui Sanchez for assistance with the RIA and Becky Brooks for administrative assistance. This work was partially supported by the National Science Foundation grant DEB-IRCEB-0552283 to CEW.

81 References

Alonso, C., V. Rocco, J. P. Barriga, M. A. Battini and H. E. Zagarese (2004) Surface avoidance by freshwater zooplankton: Field evidence on the role of ultraviolet radiation. Limnology and Oceanography. 49: 225-232.

Essen, L. O. and T. Klar (2006) Light-driven DNA repair by photolyases. Cellular and Molecular Life Science. 63: 1266-77.

Gerrish, G. A. and C. E. Caceres (2003) Genetic versus environmental influence on pigment variation in the ephippia of Daphnia pulicaria. Freshwater Biology. 48: 1971-1982.

Hairston, N.G. (1976) Photoprotection by carotenoid pigments in the copepod Diaptomus nevadensis. Proceedings of the National Academy of Sciences USA. 73: 971-974.

Hessen, D.O. (1996) Competitive trade-off strategies in Arctic Daphnia linked to melanism and UV-B stress. Polar Biology. 16: 573-579.

Hobæk, A. and H.G. Wolf (1991) Ecological genetics of Norwegian Daphnia: II. Distribution of Daphnia longispina genotypes in relation to short-wave radiation and water colour. Hydrobiologia. 225: 229-243.

Joux, F., W. H. Jeffrey, P. Lebaron and D. L. Mitchell (1999) Marine bacterial isolates display diverse responses to UV-B radiation. Applied and Environmental Microbiology. 65: 3820-7.

Karentz, D., J. E. Cleaver and D. L. Mitchell (1991) Cell survival characteristics and molecular responses of Antarctic phytoplankton to ultraviolet-B radiation. Journal of Phycology. 27: 326- 341.

Leech, D. M., A. Padeletti, C. E. Williamson (2005) Zooplankton behavioral responses to solar UV radiation vary within and among lakes. Journal of Plankton Research. 27 (5): 461-471

82

Li, S., M. Paulsson and L.O. Bjorn ( 2002) Temperature-dependent formation and photorepair of DNA damage induced by UV-B radiation in suspension-cultured tobacco cells. Journal of Photochemistry and Photobiology. 66: 67-72.

Li, Y. F., S. Kim and A. Sancar (1993) Evidence for Lack of DNA Photoreactivating Enzyme in Humans. Proceedings of the National Academy of Sciences USA, 90: 4389-4393

MacFadyen, E.J., C.E. Williamson, G. Grad, M. Lowery, W.H. Jeffrey, and D.L. Mitchell. 2004. Molecular response to climate change: temperature dependence of UV-induced DNA damage and repair in the freshwater crustacean Daphnia pulicaria. Global Change Biology. 10: 408-416.

Madronich, S., R.L. McKenzie, L.O. Bjorn, and M.M. Caldwell. (1998) Changes in biologically active radiation reaching the Earth’s surface. Journal of Photochemistry and Photobiology. 46: 5-19.

Malloy KD, Holman MA, Mitchell D, et al. (1997) Solar UVB-induced DNA damage and photoenzymatic DNA repair in Antarctic zooplankton. Proceedings of the National Academy of Sciences USA. 94: 1258-1263

McKenzie, R.L., P.J. Aucamp, A.F. Bais, L.O. Bjorn, and M. Ilyas. (2007) Changes in biologically-active ultraviolet radiation reaching the Earth’s surface. Photochemical and Photobiological Science. 6: 218-231.

Meador, J. A., R. B. Walter and D. L. Mitchell (2000) Induction, Distribution and Repair of UV Photodamage in the Platyfish, Xiphophorus signum. Photochemistry and Photobiology. 72: 260- 266.

Mitchell, D. (1997) Ultraviolet radiation damage to DNA. In Encyclopedia of Molecular Cell Biology and Molecular Medicine: Volume 6. (Edited by R.A. Meyers), pp. 149-164. Wiley Press, New Jersey.

83 Mitchell, D.L. and R.S. Nairn. (1989) The biology of the (6-4) photoproduct. Photochemistry and Photobiology. 49: 805-819.

Mitchell, D. and D. Karentz (1993) The induction and repair of DNA photodamage in the environment. In Environmental UV Photobiology. (Edited by A. R. Young, J. Moan, L. O. Bjorn and W. Nultsch), pp. 345-377. Plenum Press, New York.

Mitchell, D.L., R. Greinert, F.R. de Gruijl, K. L.H. Guikers, E. W. Breitbart, M. Byrom, M. M. Gallmeier, M. G. Lowery, and B. Volkmer (1999) Effects of chronic low dose ultraviolet-B radiation on DNA damage and repair in mouse skin. Cancer Research. 59: 2875-2884

Morris, D. P., H. Zagarese, C.E. Williamson, E.G. Balserio, B. Hargreaves, B. Modenutti, R. Moeller, and C. Queimalinos (1995) The attenuation of UV radiation in lakes and the role of dissolved organic carbon. Limnology and Oceanography. 40: 1381-1391.

Quinn, G.P and M.J. Keough. Experimental Design and Data Analysis for Biologists. Cambridge University Press. New York. 2002.

Rautio, M. and A. Korhola (2002) UV-induced pigmentation in subarctic Daphnia. Limnology and Oceanography. 47: 295-299.

Rocco, V.E., O. Oppezzo, R. Pizarro, R. Sommaruga, M. Ferraro, and H.E. Zagarese. (2002) Ultraviolet damage and counteracting mechanisms in the freshwater copepod Coeckella poppei from the Antarctic Peninsula. Limnology and Oceanography. 47: 829-836.

Sancar, A. and G.B. Sancar (1988) DNA repair enzymes. Annual Review of Biochemistry. 57: 29-67.

Sægrov, H., A. Hobæk, J. H L'Abée-Lund (1996) Vulnerability of melanic Daphnia to brown trout predation. Journal of Plankton Research. 18: 2113-2118.

84 Tartarotti, B, I. Laurion, and R. Sommaruga. (2001) Large variability in the concentration of mycosporine-like amino acids among zooplankton from lakes located across an altitude gradient. Limnology and Oceanography. 46: 1546-1552.

Tollrian, R. and C. Heibl (2004) Phenotypic plasticity in pigmentation in Daphnia induced by UV radiation and fish kairomones. Journal of Functional Ecology. 18: 497–502.

United States Environmental Protection Agency. 2002. Short-term Methods for Estimating the Chronic Toxicity of Effluents and Receiving Waters to Freshwater Organisms: 4th Edition. EPA- 821-R-02-013. October 2002.

Van de Poll, W. H., A. Eggert, A. G. J. Buma and A. M. Breeman (2002) Temperature dependence of UV radiation effects in Arctic and temperate isolates of three red macrophytes. European Journal of Phycology. 37: 59-68.

Williamson, C. E. (1995) What role does UV-B radiation play in freshwater ecosystems? Limnology and Oceanography. 40: 386-392.

Williamson, C.E., G. Grad, H.J. De Lange, and S. Gilroy. (2002) Temperature-dependent ultraviolet responses in zooplankton: implications of climate change. Limnology and Oceanography. 47: 1844-1848.

Winder, M., P. Spaak and W. M. Mooij (2004) Trade-offs in Daphnia habitat selection. Ecology, 85: 2027-2036.

Zagarese, H.E., M. Feldman, and C.E. Williamson. (1997) UV-B induced damage and photoreactivation in three species of Boeckella (Copepoda, Calanoida). Journal of Plankton Research. 19: 357-367.

Zellmer, I. D., M. T. Arts, D. Abele and K. Humbeck (2004) Evidence of Sublethal Damage in Daphnia during Exposure to Solar UV Radiation in Subarctic Ponds. Arctic and Antarctic Alpine Research. 36: 370-377.

85 Table 4.1. Estimated photoenzymatic and nucleotide excision rates over 24 h in daphniids following acute UVR exposure (20 kJ/m2). PRR conditions are as described in Table 4.1. Rates based on the exponential decay model y = r [y(0)] + ae-bx; r, correlation coefficient; a, initial damage (CPDs per mb DNA); b is the inverse of the repair half-life (t1/2) for CPD repair in hours, x is hours post exposure, and y(0) is the asymptote of the repair curve, or the amount of residual (unrepairable) CPD damage (Mitchell et al, 1999); SEM, standard error of mean repair rate; --, exponential decay model did not fit the data; n.r., no repair.

Species Temp (°C) PRR r a t1/2 y(0) SEM D. middendorffiana 20 + 0.9873 240.07 2.43 34.23 22.71 - 0.6330 -- n.r. -- 61.89 10 + 0.9940 284.79 0.91 40.23 19.58 - 0.2195 104.99 0.13 297.27 295.12 D. pulicaria 20 + 0.6547 412.78 1.37 345.00 337.37 - 0.8128 26.68 13.69 644.28 30.71 10 + 0.9835 198.44 4.84 9.05 15.04 - 0.7779 685.85 9.48 -- 271.95 D. pulex 20 + 0.7423 636.70 3.08 821.00 337.00 - 0.6372 80.30 n.r. 129.45 61.42 10 + 0.9154 133.41 2.81 12.49 22.56 - 0.8075 838.08 20.77 8.77 229.83 D. parvula 20 + 0.9245 213.74 3.48 46.72 27.08 - 0.3626 188.96 17.70 30.62 199.46 10 + 0.9749 92.56 1.68 16.47 5.96 - 0.7962 20.17 4.50 50.34 28.43

86 Table 4.2. Comparison of UVB response parameters at 10ºC and 20ºC.

Species Survival (24 h) Damage Photoprotection Repair Rate (CPD/h) Residual damage [y(0)]

PRR + - na na + - + -

Temperature (°C) 10 20 10 20 10 20 10 20 10 20 10 20 10 20 10 20

D. middendorffiana high high highlow low low high high med high low low low med high high

D. pulicaria med high high low med high med low high med med low low med low high

D. pulex high high high high med high med low med low med low low high low high

D. parvula high low highlow low low high high high high med low med med high high

87

Chapter 5

Artificial UV-B and solar radiation reduce in vitro infectivity of the human pathogen Cryptosporidium parvum

As published: Sandra J. Connelly, Elizabeth A. Wolyniak , Craig E. Williamson, and Kristen L. Jellison. 2007. Artificial UV-B and solar radiation reduce in vitro infectivity of the human pathogen Cryptosporidium parvum. Environmental Science & Technology. On line September.

88 Abstract

The potential for solar ultraviolet radiation (UVR) to act as a significant abiotic control of Cryptosporidium parvum oocysts in nature is unknown. Infectivity of C. parvum following exposure to artificial ultraviolet B (UV-B) and natural solar radiation, with and without UVR wavelengths, was tested under controlled pH and temperature conditions. Percent infectivity of exposed oocysts was determined by in vitro cell culture. Artificial UV-B exposures of 32 and 66 kJ/m2 significantly decreased oocyst infectivity by an average of 58 and 98%, respectively. Exposure of oocysts to approximately half and full intensity of full solar spectrum (all wavelengths) for a period of less than one day (10 h) in mid-summer reduced mean infectivity by an average of 67% and >99.99%, respectively. Exposure of the C. parvum oocysts to UVR- shielded solar radiation (> 404 nm) in early autumn reduced mean infectivity by 52%, while full spectrum solar radiation, exposure at all wavelengths, reduced mean infectivity by 97%. The data provide strong evidence that exposure to natural solar radiation can significantly reduce C. parvum infectivity. Direct effects of solar radiation on oocysts in nature will depend on the depth distribution of the oocysts, water transparency, mixing conditions, and perhaps other environmental factors such as temperature, pH, and stress.

89 Introduction

The presence of the obligate enteric protozoan parasite Cryptosporidium parvum in water supplies is a public health concern for both developed and developing countries. A single oocyst, the only life stage of C. parvum found in the environment, can infect a host if ingested. Rose et al. (1991) conducted a 17 State survey in the United States and estimated that Cryptosporidium is found in 55% of surface waters and 17% of drinking water supplies at average concentrations of 43 oocysts/100L under base flow conditions. Although first described in 1976, it was not until 1987 that Cryptosporidium oocysts were linked to the disease cryptosporidiosis in livestock, wildlife, and humans (presenting as severe and potentially lethal cases of gastrointestinal distress primarily in infants, the elderly, and the immunocompromised) (Rose 1997). Numerous outbreaks of cryptosporidiosis in humans have occurred over the past 20 years in the United States, the largest of which was linked to the municipal drinking water supply in Milwaukee, Wisconsin in 1993 where there were between 15,000 and 403,000 people affected and up to 100 associated deaths (MacKenzie et al., 1994; Johnson et al., 2005). More recently, 4,000 people were affected by an outbreak in a state park in Geneva, New York in August 2005. In the fall of 2005, 700 people were diagnosed with cryptosporidiosis in southwest Ohio. Both incidents were traced to recreational waters. A total of 5,659 cases of cryptosporidiosis were reported in 2005 (CDC, 2007). Outbreaks of cryptosporidiosis not only have health consequences, but result in widespread economic costs as well, including class action lawsuits against the states and/or contaminated recreational facilities (Yu, 2005) and an increased public demand for health permits and wildlife regulation in recreational areas (Yu, 2006). Given their small size (which limits the success of physical filtration) and resistance to chlorine disinfection, Cryptosporidium oocysts can persist in water supplies (Weir et al., 2002). The most effective potable water treatments for the disinfection of Cryptosporidium oocysts are ozone and germicidal ultraviolet radiation (UVR; typically UV-C, 200-280 nm). Oocyst populations have been reduced 97% from initial concentrations following pre-ozonation treatments and further ozonation has little additional effect; this treatment results in a reduction in the number, but not the infectivity, of waterborne oocysts (Hsu and Yeh, 2003). Germicidal UV-C, on the other hand, does reduce oocyst infectivity and is the most effective disinfectant of

90 Cryptosporidium (Zimmer et al., 2003). UV-C disinfection is routinely built into new treatment plants, but it can be extremely expensive to retrofit UV-C into existing plants. Numerous papers have been published on the use of UVR as a disinfectant of Cryptosporidium in water treatment (e.g., Craik et al., 2001; Linden et al., 2001; Zimmer et al., 2003). Previous research has focused on the inactivation of viable oocysts from the potable water supply, calculating UVR dose for specific UVR lamps to optimize oocyst disinfection in water treatment (Clancy et al., 2004). Low pressure (LP), high output mercury lamps with a monochromatic output (254 nm) provide the greatest disinfection rates with the lowest temperature variation (bulb temperature 100°C). In contrast, the medium pressure (MP) polychromatic UVR lamps, which have a broader spectrum output (185-1367 nm), are a less efficient germicide based on electrical energy inputs and require a tenfold increase in power (increasing the temperature variation; bulb temperature 600 – 900°C) as compared to LP lamps (Clancy et al., 2004). The inactivation of C. parvum with low doses of UVR using LP or MP mercury lamps in the lab (Craik et al., 2001) are very specific to the water treatment industry and give limited insight into the potential effects of natural solar radiation on oocysts. The sensitivity of Cryptosporidium oocysts to germicidal lamps is reported in the disinfection literature as log10 inactivation. Oocyst infectivity rates are reduced two log10 at very low levels of artificial UV-C (< 0.1 kJ/m2) using LP and MP lamps (Linden et al., 2001; Morita et al., 2002; Rochelle et al., 2005; Hijnen et al., 2006). By comparison, incident daily summer solar UVR (290-400 nm) at 40°N latitude in Pennsylvania is >1,300 kJ/m2 (Williamson et al., 2001). However, the shorter wavelengths of UV-C from LP and MP lamps are much more damaging than longer wavelengths of solar UVR on a per-photon or per-unit-energy basis. While UVR tolerance cannot be extrapolated from the lab to the field based on energy alone, previous research shows that inactivation of oocysts by 290 nm UV-B, the shortest wavelength of solar UVR reaching the Earth’s surface, is less than 10% as effective as that observed at shorter germicidal wavelengths from artificial lamps (Linde et al., 2001). Thus, the potential for solar UVR to influence Cryptosporidium in surface waters exposed to sunlight remains in question. A better understanding of the potential natural regulators of this pathogen is needed because of the medical risks associated with exposure to Cryptosporidium and the difficulties and expense associated with its removal and/or disinfection in water supplies. Extreme temperature (Pokorny et al., 2002) and water quality changes have the potential to significantly

91 reduce the percent of infectious Cryptosporidium oocysts in water supplies. Low pH would be the primary water quality concern due to its impact on the hydrophobicity of the oocyst surface (Brown et al., 1996), although other abiotic stressors, including turbidity, water depth (which influences settling rates; Searcy et al., 2005), mixing, and alkalinity, may also impact oocyst fate and transport. Extreme pH is not likely to be an important factor in most reservoirs (Campbell et al., 2002), however, due to buffering of inputs by the surrounding watershed, and the observed temperature fluctuations in nature do not approach the extremes reported as effectively reducing oocyst infectivity in the lab. For these reasons, solar UVR, which has not been adequately addressed in previous studies, is one of the most likely natural abiotic factors influencing the infectivity of Cryptosporidium in surface waters. Several lines of evidence suggest a plausible effect of solar radiation on the infectivity of C. parvum oocysts in surface waters. Extensive laboratory and field experiments have demonstrated that a wide variety of freshwater protozoa, rotifers, and crustacean plankton succumb to levels of solar UVR that are the equivalent of <1 day of field exposure (Williamson et al., 1994; Leech and Williamson, 2000; Williamson et al., 2001; Sanders et al., 2005). While many studies have reported germicidal UV-C to be one of the most effective disinfectants against C. parvum oocysts, recent studies have begun to test the effects of solar-simulating lamps. McGuigan et al. (2006) have demonstrated that oocysts are not infective to CD-1 suckling mice following a 30 kJ total optical dose with a xenon arc lamp (solar simulator), and Mendez- Hermida et al. (2005) report complete loss of oocyst infectivity in neonatal Swiss CD-1 mice after 12-h exposures using UVR lamps at intensities similar to natural sunlight. In addition, Johnson and colleagues (1997) reported a 90% reduction in the excystation (i.e., the ability to release sporozoites) of C. parvum oocysts in marine waters following a 3-day exposure to direct sunlight (no test of infectivity was reported), while King and Monis (2007) have recently reported >90% inactivation of Cryptosporidium oocysts by solar radiation. These studies of the susceptibility of oocysts to UV-C, in addition to some natural and simulated solar UVR work, suggest that C. parvum oocysts are likely to be inactivated by UV-B and natural levels of solar UVR. Infectivity is the only accurate test for the effects of UVR on Cryptosporidium. Some studies of Cryptosporidium and UVR have used viability and/or excystation as infectivity proxies to determine the potential for inactivation of the oocysts. Oocysts which are not viable

92 (by vital dye staining) and/or unable to excyst can be considered non-infectious, as opposed to viable and/or excysted oocysts which may not necessarily be infective. Recent studies have demonstrated that viability and excystation assays do not correlate well to infectivity of Cryptosporidium oocysts following UVR exposure, often overestimating the percent of infective oocysts. In a study of excystation and infectivity methods, Morita et al. (2002) report that a 200- fold increase in UV-C (254 nm) dose is required to decrease oocyst excystation to comparable infectivity levels. This demonstrates a significant overestimation of infectivity by excystation assays, giving further support for the use of infectivity assays, rather than viability or excystation techniques, to test the effect of UVR on oocysts. None of these previous studies have reported a direct test of natural solar radiation under otherwise controlled conditions on the infectivity of Cryptosporidium oocysts. Here we focus on the direct effects of artificial UV-B and natural solar radiation, under controlled pH and temperature conditions, on Cryptosporidium parvum infectivity by in vitro cell culture. The effects of (i) artificial UV-B, (ii) full spectrum solar radiation, and (iii) solar radiation with and without UVR wavelengths on C. parvum oocyst infectivity are all presented.

Methods and Materials

Pathogen maintenance C. parvum oocysts (Iowa isolate) derived from mice were acquired live from Waterborne, Inc. (New Orleans, LA) within 48 h of being shed. The oocysts were stored in Hanks balanced salt solution (1x HBSS; pH 7) at 4°C in the dark for less than two weeks prior to use in these experiments. The same oocyst lot (# 060806) was used for the artificial UV-B exposures and the solar exposure with and without UVR. A different oocyst lot (# 070706) was used for the full spectrum solar radiation exposures (manipulated intensity).

Artificial UV-B exposure The C. parvum oocyst stock suspension was diluted to 5000 oocysts in 20 mL of 1x phosphate buffered saline (PBS; pH=7.4) in quartz Petri dishes with quartz lids for exposure. Nine replicate quartz Petri dishes were placed at random into three UV-B exposure treatments (0, 32, and 66 kJ/m2 UV-B). The UV-B exposure of the oocysts was conducted in an environmentally

93 controlled chamber using a UV-lamp phototron (Williamson et al., 2001) at 10°C for 12 h. In the lamp phototron, the quartz Petri dishes containing oocyst suspensions were placed inside black collars (to reduce UV-B scattering) on a rotating wheel (2 rpm) and were exposed to UV-B from a single lamp from above (UV-B lamp: Spectronics Spectroline BLE-1T158 15W lamp, Westbury, NY). The 66 kJ/m2 treatment was full exposure of the dishes to the UV-B lamp with a spectral output of 290-360 nm (Williamson et al., 2001). The 32 kJ/m2 treatment was established by placing a single layer of wire mesh screening (48% transmittance) on top of the quartz dishes to reduce the UV-B intensity. The 0 kJ/m2 treatment was established by placing a black disk on top of the corresponding replicate dishes so that the oocysts had no UV-B exposure but were otherwise maintained under identical conditions. Following exposure, oocyst suspensions from each dish were transferred to 50 mL tubes and concentrated by centrifugation to 1 mL (1100xg at 10°C for 15 min). The 1 mL oocyst suspension was stored at 4°C in the dark for 24 h prior to the infectivity assay. This hold time was necessary due to handling and shipping of the oocysts for the infectivity studies. Lab control stock solutions accompanied the oocysts during all shipping and handling. No studies are known that demonstrate evidence of dark DNA repair mechanisms that result in restored infectivity of C. parvum oocysts following UVR exposure. Oguma et al. (2001) present evidence of dark repair in C. parvum by citing a reduction of UV-induced DNA damage sites, but do not show evidence of restored infectivity of the oocysts.

Solar exposure Two solar experiments were conducted in July and September 2006 to study the effect of natural solar radiation on C. parvum oocyst infectivity. For both experiments, the C. parvum oocyst stock was diluted to 5000 oocysts in 20 mL of 1x PBS (pH = 7.4) in black Pyrex Petri dishes with quartz lids. The black dishes reduce light scattering, while the quartz lids transmit all solar radiation. Nine replicate Petri dishes were placed randomly into three exposure treatments for each experiment (July treatments: dark, approximately half full solar intensity (UV320 = 3.85 kJ/m2; PAR = 5,484 kJ/m2), and full solar intensity (UV320 = 7.7 kJ/m2; PAR = 10,968 kJ/m2); September treatments: dark, full spectrum solar (UV320 = 4.19 kJ/m2; PAR = 6,958 kJ/m2), and solar >404 nm (i.e., no UVR wavelengths; PAR = 6,958 kJ/m2)). The solar radiation exposure of the oocysts was administered using a Solar Phototron (Williamson et al., 2001). The Solar Phototron consists of Petri dishes placed in a shallow Plexiglas box connected to a recirculating

94 water bath that regulates temperature during the experiment. The experiments were run for 10 h surrounding solar noon. Percent solar dose was manipulated in the first solar experiment (July 2006). C. parvum oocysts were exposed to 0, 48, or 100% full spectrum solar radiation at an average temperature of 14°C, using the same wire mesh screening as previously described in the artificial UV-B exposure experiments. Dark controls (0% solar exposure) were created by covering the dishes with a black disk while maintaining all other treatment conditions. The second solar experiment (September 2006) exposed C. parvum oocysts to three treatments of solar radiation: (i) zero solar exposure (dark controls), (ii) solar exposure including UVR wavelengths (+UVR), and (iii) solar exposure with the UVR wavelengths blocked (-UVR) at an average temperature of 20°C. The dark treatment Petri dishes were again covered with a black disk, and the +UVR treatment dishes were covered only with quartz lids, permitting essentially full solar exposure. The –UVR treatment was established by placing a sharp cutoff 404 nm long wavepass Schott filter on top of the Petri dishes to remove all wavelengths less than 404 nm. Temperature was recorded every ten minutes during all solar exposures using i-Button temperature loggers (Maxim Integrated Products and Dallas Semiconductor, Sunnyvale, CA). Temperature data were more variable for the July experiments of solar radiation intensity (July 15 range 10.0 - 18.0°C; avg±SE: 13.7±0.38°C; July 16 range 9.0 - 20.0°C; avg±SE: 14.1±0.33°C) than for the September experiments that manipulated the presence and absence of solar UVR (Sept 7 range 19.5 - 21.5°C; avg±SE: 20.3±0.07°C; Sept 10 range 19.0 - 22.5°C; avg±SE: 20.7±0.11°C). The effect of temperature fluctuations on oocyst infectivity are corrected for by using the dark treatment (no UV-B or solar exposure) for a given experiment as the control. Due to the rapid turnover of water in this recirculating design, we do not anticipate that the temperature of the dark dishes would be different from the light exposed dishes. Solar spectrum data [305, 320, 380 nm, and Photosynthetically Active Radiation (PAR) 400-700 nm] were recorded for the duration of the experiments using a BIC Internal Logger 2100 (Biospherical Instruments, Inc., San Diego, CA). Solar spectrum data were integrated for each wavelength band to calculate the maximum solar exposure per experiment at four wavelengths (305, 320, 380 nm, and PAR). Data were plotted as a function of time. The area under the curve was integrated, resulting in a total exposure dose for a given wavelength for the

95 duration of the experiment (kJ/m2). These exposures were then corrected for treatment conditions as necessary within each experiment (e.g., 48% solar exposure on July 16, 2006 = 0.48 * 7.70 kJ/m2 = 3.69 kJ/m2 at 320 nm) to determine the actual light energy dose to the C. parvum oocysts. Exposure days were calculated for 320 nm as this has been shown to be the most biologically effective wavelength present in solar radiation for aquatic invertebrates (Williamson et al., 2001) and, therefore, most likely to have a significant effect on C. parvum oocysts. Solar data were collected during all exposure periods of C. parvum oocysts in July and September, 2006. Wavelength band dose data from the BIC Internal Logger are presented for 305, 320, 380 nm, and PAR, in addition to the equivalent 320 nm exposure days for each solar experiment (Table 5.1).

Table 5.1: Maximum solar radiation exposure (dose) for each experiment at four wavelengths (305, 320, 380 nm, and Photosynthetically Active Radiation [PAR]) and the equivalent exposure days for the most biologically effective wavelength of 320 nm.

Experiment Date (2006) 305 nm 320 nm 380 nm PAR 320 nm Dose* Dose* Dose* Dose* Exposure Days§

Solar Exposure July 15 0.46 3.88 10.57 5126 0.36

Solar Exposure July 16 1.11 7.70 20.91 10968 0.71

Solar Exposure September 7 0.40 3.56 10.69 5955 0.33 +/- UVR

Solar Exposure September 10 0.51 4.19 12.39 6958 0.38 +/- UVR

* Data collected using a BIC Internal Logging Radiometer (μW cm-2 nm-1) and converted to dose for the entire exposure period (kJ/m2). § A value of 10.9 kJ/m2 represents one 320 nm exposure day, or the amount of solar UVR at 320 nm received during one day of full sunlight (no clouds) at the water surface during summer solstice and average ozone conditions at 41° N latitude (Cooke and Williamson, 2006).

96 Following all solar exposure experiments, the oocyst suspensions were moved immediately to the dark and transferred to 50 mL tubes. The suspensions were concentrated by centrifugation to 1 mL (1100xg at 10°C for 15 min) and stored at 4°C in the dark for 24 h prior to the infectivity assay. Transport and handling of the oocysts in the natural solar experiments were conducted identically to the UV-B lamp experiments in the laboratory.

Infectivity assay Percent infectivity of exposed oocysts was determined by in vitro cell culture infection of human ileocecal adenocarcinoma (HCT-8) cells grown in eight-well chamber slides according to Slifko et al. (1997; 1999). Exposed oocysts (concentrated to 1 mL) were pretreated with a 10% sodium hypochlorite solution, and the concentration of the treated oocyst suspension was determined using a hemacytometer. An infection volume containing 100 oocysts was calculated and added to each of three wells on the chamber slide (3 replicate infections per treatment replicate * 3 treatment replicates per UV-B/solar exposure = 9 analytical replicates per treatment). Laboratory controls of the C. parvum oocysts maintained at 4°C in the dark during the exposure experiments were also tested for infectivity by cell culture. Two wells on each slide were left uninfected to serve as negative controls and to monitor the health of the cell monolayer. Infected chamber slides were incubated at 37°C and 5% CO2 for 48 h and then stained according to the Sporo- GloTM antibody protocol (Waterborne, Inc.). Foci of infection were counted using a Nikon epifluorescence microscope (Nikon, Inc., USA), and the percent infective oocysts was calculated by dividing the number of infection foci per well by 100 (the total number of intact oocysts inoculated onto each well). Percent reduction in infectivity was calculated using the mean (N=9) treatment infectivity.

Statistical analyses Treatment controls were compared prior to any statistical analyses and were found to vary less than 5% from the laboratory controls. All UVR treatments were compared to the respective dark control (no UVR or solar exposure) within a single experiment, providing an internal control for oocyst handling and processing. Logistic regression was used to analyze percent infectivity of the C. parvum oocysts following UVR exposure relative to the dark control treatment within

97 each experiment (Regression; SPSS, 2005, SPSS Inc., Release 13.0). Data were analyzed as treatment replicates (N=3) to conserve the variability in infectivity observed in this study.

Results and Discussion

Laboratory exposure of C. parvum oocysts to 32 kJ/m2 and 66 kJ/m2 UV-B at 10°C significantly reduced mean in vitro infectivity by 58% (0.38 log reduction; p<0.001) and 98% (1.70 log reduction; p<0.001), respectively, compared to the zero kJ/m2 treatment (Table 5.2).

98 Table 5.2. In vitro infectivity of C. parvum oocysts following a 12-h artificial UV-B exposure, a 10-h percent solar exposure, and a 10-h solar exposure with and without UV wavelengths.

Artificial UVR-B Experiment

C. parvum Infectivity 0 kJ/m2 32 kJ/m2 66 kJ/m2 Lab Control (Dark Control) (48%) (100%)

% Infectivity* 4.0% 1.7% 0.1% 2.7% Total Infections Mean ± S.E. 4.0 ± 0.4 1.7 ± 0.2 0.1 ± 0.1 2.7 ± 0.9 Range 3-6 1-3 0-1 1-4

% Solar Transmittance Experiment (July 15)

0 kJ/m2 320 nm 1.9 kJ/m2 320 nm 3.9 kJ/m2 320 nm Lab Control (Dark Control) (48%) (100%)

% Infectivity* 12.9% 8.2% 0.0% 11.3% Total Infections Mean ± S.E. 12.9 ± 2.1 8.2 ± 2.8 0 ± 0 11.3 ± 0.7 Range 6-23 0-26 0 10-12

% Solar Transmittance Experiment (July 16)

0 kJ/m2 320 nm 3.7 kJ/m2 320 nm 7.7 kJ/m2 320 nm Lab Control (Dark Control) (48%) (100%)

% Infectivity* 22.4% 0.3% 0.0% 13.3% Total Infections Mean ± S.E. 22.4 ± 6.6 0.3 ± 0.2 0 ± 0 13.3 ± 4.1 Range 6-59 0-1 0 5-18

Solar ± UVR Experiment (September 7)

0 kJ/m2 320 nm 0 kJ/m2 320 nm 3.6 kJ/m2 320 nm Lab Control (Dark Control) (-UVR) (+UVR)

% Infectivity* 3.1% 1.2% 0.0% 21.3% Total Infections Mean ± S.E. 3.1 ± 0.2 1.2 ± 0.3 0 ± 0 21.3 ± 5.5 Range 2-4 0-2 0 14-32

Solar ± UVR Experiment (September 10)

0 kJ/m2 320 nm 0 kJ/m2 320 nm 4.2 kJ/m2 320 nm Lab Control (Dark Control) (-UVR) (+UVR)

% Infectivity* 10.2% 5.9% 0.7% 11.3% Total Infections Mean ± S.E. 10.2 ± 0.6 5.9 ± 0.4 0.7 ± 0.2 11.3 ± 2.4 Range 8-13 5-8 0-2 8-16

*Data are reported as percent infectivity (i.e., the number of infection foci per 100 intact oocysts inoculated onto the HCT-8 cell monolayer, multiplied by 100).

S.E. = Experimental standard error by linear regression of a given treatment where N=3.

99 Exposure of C. parvum oocysts to 48% natural solar radiation (full spectrum) at 10°C did not significantly reduce the average in vitro infectivity (0.19 log reduction; p=0.346) on July 15, however, the same exposure on July 16 did significantly reduce the average in vitro infectivity by 99% (2.0-log reduction; p=0.001) from 0% solar exposure. Exposure of C. parvum oocysts to 100% natural solar radiation (full spectrum) at 10°C reduced the average in vitro infectivity by >99.99% on July 15, 2006 (4.0-log reduction; p<0.001) and July 16, 2006 (4.0-log reduction; p=0.001) in comparison to the dark controls (0% solar exposure; Table 5.2). Exposure of C. parvum oocysts to natural solar radiation, with and without UVR wavelengths, significantly reduced mean in vitro infectivity by >99.99% (4- log reduction; p<0.001) and 61% (0.41 log reduction; p<0.001), respectively, on September 7, 2006 compared to the dark controls. Mean infectivity was reduced by 93% (1.15 log reduction; p<0.001) with UVR wavelengths and 42% (0.24 log reduction; p<0.001) without UVR wavelengths present in the September 10, 2006 solar exposure, compared to the associated dark controls (Table 5.2). This is the first study to demonstrate a significant inactivation of C. parvum by UV-B and natural solar radiation under controlled conditions using in vitro infectivity. Both artificial UV-B and solar radiation significantly reduced the percent infectivity of C. parvum oocysts under otherwise environmentally controlled conditions, indicating a direct negative effect of solar wavelengths on this waterborne pathogen. Even with high variability in natural solar irradiance in the July experiments, a 36-99% reduction in mean infectivity of C. parvum oocysts was still observed at only about one half of full solar exposure levels. Further, oocyst infectivity was reduced in the absence of solar UVR wavelengths, suggesting long wavelength (blue) inactivation of oocysts is possible. Variability in oocyst infectivity in dark controls is high for these experiments (even within an oocyst lot), but the resulting percent decrease in infectivity with exposure to UV-B and solar radiation is similar. Variations within and among oocyst lots have been reported by other labs (Sivaganesan and Sivaganesan, 2005) and are recognized as a source of experimental error. We suggest that some of this variability may be attributed to slight difference in oocyst age or experimental temperature conditions. In addition, some of the variability may be attributed to incomplete HCT-8 cell monolayers (preventing the identification of infectious oocysts in regions of the chamber slide where the monolayers fail to grow) or oocyst clumping in the infection inoculums (leading to overlapping infection foci which may be counted as one single infectious oocyst).

100 Water transparency and natural mixing processes are likely to be important factors for the inactivation of C. parvum oocysts by solar radiation in nature. Land use patterns, climate change, and watershed characteristics will influence surface water transparency and mixing, concomitantly affecting solar exposure of oocysts. Human-induced environmental changes, such as alteration of land use and the hydrologic cycle, are likely to affect chromophoric, or color- absorbing, dissolved organic matter (CDOM), the major regulator of underwater UVR environments in lakes (Morris et al., 1995). In low CDOM freshwater systems, solar UVR (305 and 320 nm) has been reported at depths greater than 10m (Morris et al., 1995; Williamson, 1995). These high transparency systems are likely to have greater potential for natural disinfection of oocysts than a low transparency system with high CDOM concentrations. Watershed variability, such as type and quantity of vegetation in the watershed, could play a significant role in water clarity and ultimately control the number of infectious oocysts present in water supplies. The presence of wetlands in the watersheds can contribute significantly to the CDOM inputs to freshwater systems (Canham et al., 2004). Wetlands have higher rates of dissolved organics exported annually (Xenopoulos et al., 2003) but, depending on the percentage of wetland coverage in a watershed, may not contribute as substantially to inputs as other upland areas, particularly following the clear-cutting of forests, etc. for development. All of these factors, in conjunction with rainfall events, have the potential to significantly impact the CDOM concentrations (UVR attenuation of aquatic system), and ultimately the organism’s UVR exposure, in these freshwater systems. The UVR sensitivity of Cryptosporidium also has important implications for reservoir engineering and the epilimnetic versus hypolimnetic withdrawal of water from reservoirs for human consumption. Settling rates of free oocysts are reported to be 0.76μm/sec (36), or the equivalent of 0.07m/day. Their slow sinking rates suggest that wind-driven mixing will maintain oocysts in the surface waters for extended periods of time. Epilimnetic water withdrawal may increase the likelihood that waterborne oocysts would have been inactivated by solar UVR exposure; conversely, oocysts which have settled into the cooler, deeper hypolimnion are more likely to have been sheltered from solar UVR exposure, and thus, more likely to remain infectious for longer periods of time. Seasonal variability in the presence of oocysts in surface waters must be considered in conjunction with seasonal variability of solar UVR to better understand the potential for solar

101 inactivation of pathogens in surface water supplies. This study shows significant effects of UVR exposure in both July and September on Cryptosporidium oocysts. Tsushima et al. (Tsushima et al., 2003) report significant fluctuations in the numbers of oocysts detected in natural water supplies annually, with the highest numbers of oocysts present in late summer (July) through early autumn (September) and the lowest levels (no oocysts detected) in winter and spring. These findings are recognized as being regional and potentially driven by agricultural inputs. However, it is during this period annually (July – September) that photobleaching of CDOM is maximal in the northern hemisphere, increasing the UVR transparency of natural waters (Morris and Hargreaves, 1997) and providing the greatest potential for natural solar disinfection. Summer solstice (mid-June) is the annual peak of incident solar UVR, and exposure to 320 nm UV-B at >7.5 m in the midsummer (July) is equivalent to exposure at depths as shallow as 1.8 m in the spring (Morris and Hargreaves, 1997). These extreme seasonal fluctuations in solar UVR penetration could play a significant role in the natural disinfection of C. parvum. A better understanding of oocyst transport cycles within watersheds is needed to apply land management strategies that will take full advantage of the corresponding solar UVR cycles and maximize the potential for solar UVR disinfection of these pathogens prior to water treatment. Although we have shown significant reductions in oocyst infectivity after solar exposure at both 10°C and 20°C, future studies might focus on the interaction of variable temperature extremes and changing solar UVR, in light of climate change and global warming scenarios, to better understand oocyst infectivity reductions in nature. Interactions between solar UVR, temperature, pH, and microbial community structures are likely to have the greatest impact on the persistence of pathogens in natural systems around the globe.

Acknowledgements

We thank the Zoology Department (Miami University) for graduate research support to SJC and the National Water Research Institute Fellowship Program for partial funding of this research. This work was partially supported by a National Science Foundation CAREER grant to KLJ (award # 0545687) and by NSF DEB-IRCEB-0552283 to CEW.

102 References

Brown, S.M.A., V. McDonald, H. Denton, and G.H. Coombs. 1996. The use of a new viability assay to determine the susceptibility of Cryptosporidium and Eimeria sporozoites to respiratory inhibitors and extremes of pH. FEMS Microbiology Letters. 142: 203-208.

Campbell, A.T., L.J. Robertson, and H.V. Smith. 2002. Viability of Cryptosporidium parvum oocysts: correlation of in vitro excystation with inclusion or exclusion of fluorogenic vital dyes. Applied and Environmental Microbiology. 58: 3488-3493.

Canham, C. D., M.L. Pace, M.J. Papaik, A.G.B. Primack, K.M. Roy, R.J. Maranger, R.P. Curran, and D.M. Spada. 2004. A spatially explicit watershed-scale analysis of dissolved organic carbon in Adirondack lakes. Ecological Applications. 14: 839-854.

Centers for Disease Control and Prevention Morbidity and Mortality Weekly Report (MMWR). 2007. Summary of Notifiable Diseases – United States, 2005. Centers for Disease Control and Prevention. 54: 2-92.

Clancy, J. L., M.M. Marshall, T.M. Hargy, and D.G. Korich. 2004. Susceptibility of five strains of Cryptosporidium parvum oocysts to UV light. Journal of the American Water Works Association. 96: 84-93.

Cooke, S. L. and C.E. Williamson. 2006. Positive effects of UV radiation on a calanoid copepod in a transparent lake: do competition, predation, or food availability play a role? Journal of Plankton Research.28: 171-179.

Craik, S. A., D. Weldon, G.R. Finch, J.R. Bolton, and M. Belosevic. 2001. Inactivation of Cryptosporidium parvum oocysts using medium- and low-pressure ultraviolet radiation. Water Research. 35: 1387-1398.

Hijnen, W.A.M., E.F. Beerendonk, and G.J. Medema. 2006. Inactivation credit of UV radiation for viruses, bacteria, and protozoan (oo)cysts in water: A review. Water Research. 40: 3-22.

Hsu, B.M. and H.H. Yeh. 2003. Removal of Giardia and Cryptosporidium in drinking water treatment: a pilot-scale study. Water Research. 37: 1111-1117.

103 Johnson, D.C., C.E. Enriquez, I.L. Pepper, T.L. Davis, C.P. Gerba, and J.B. Rose. 1997. Survival of Giardia, Cryptosporidium, poliovirus and Salmonella in marine waters. Journal of Water Science and Technology. 35: 261–268.

Johnson, A.M., K. Linden, K.M. Ciociola, R. De Leon, G. Widmer, and P.A. Rochelle. 2005. UV inactivation of Cryptosporidium hominis as measured in cell culture. Applied and Environmental Microbiology. 71: 2800-2802.

King, B.J. and P.T. Monis. 2007. Critical processes affecting Cryptosporidium oocyst survival in the environment. Parasitology. 134: 309-323.

Leech, D. M. and C.E. Williamson. 2000. Is tolerance to UV radiation in zooplankton related to body size, taxon, or lake transparency? Ecological Applications. 10: 1530-1540.

Linden, K.G., G.A. Shin, and M.D. Sobsey. 2001. Relative efficacy of UV wavelengths for the inactivation of Cryptosporidium parvum. Water Science and Technology. 43: 171-174.

MacKenzie, W.R., N.J. Hoxie, M.E. Proctor, S. Gradus, K.A. Blair, D.E. Peterson, J.J. Kazmierczak, D.G. Addiss, K.R. Fox, J.B. Rose, and J.P. David. 1994. A massive outbreak in Milwaukee of Cryptosporidium infection transmitted through the public water supply. New England Journal of Medicine. 331: 161-167.

McGuigan K.G. F. Mendez-Hermida, J.A. Castro-Hermida, E. Ares-Mazás, S.C. Kehoe, M. Boyle, C. Sichel, P. Fernández-Ibáñez, B.P. Meyer, S. Ramalingham, and E.A. Meyer. 2006. Batch solar disinfection inactivates oocysts of Cryptosporidium parvum and cysts of Giardia muris in drinking water. Applied and Environmental Microbiology. 101: 453–463.

Mendez-Hermida, F., J.A. Castro-Hermida, E. Ares-Mazas, S.C. Kehoe and K.G. McGuigan. 2005. Effect of batch-process solar disinfection on survival of Cryptosporidium parvum oocysts in drinking water. Applied and Environmental Microbiology. 71: 1653-1654.

Morita, S., A. Namikoshi, T. Hirata, K. Oguma, H. Katayama, S. Ohgaki, N. Motoyama, and M. Fujiwara. 2002. Efficacy of UV irradiation in inactivating Cryptosporidium parvum oocysts. Applied and Environmental Microbiology. 68: 5387-5393.

104 Morris, D. P., H. Zagarese, C.E. Williamson, E.G. Balserio, B. Hargreaves, B. Modenutti, R. Moeller, and C. Queimalinos. 1995. The attenuation of UV radiation in lakes and the role of dissolved organic carbon. Limnology and Oceanography. 40: 1381-1391.

Morris, D. P. and B.R. Hargreaves. 1997. The role of photochemical degradation of dissolved organic carbon in regulating the UV transparency of three lakes on the Pocono Plateau. Limnology and Oceanography. 42: 239-249.

Oguma, K.,H. Katayama, H. Mitani, S. Morita, T. Hirata, and S. Ohgaki. 2001. Determination of pyrimidine dimers in Escherichia coli and Cryptosporidium parvum during UV light inactivation, photoreactivation, and dark repair. Applied and Environmental Microbiology. 67: 4630-4637.

Pokorny, N.J., S.C. Weir, R.A. Carreno, J.T. Trevors, and H. Lee. 2002. Influence of temperature on Cryptosporidium parvum oocyst infectivity in river water samples as detected by tissue culture assay. Journal of Parasitology. 88: 641-643.

Rochelle, P.A., S.J. Upton, B.A. Montelone, and K. Woods. 2005. The response of Cryptosporidium parvum to UV light. Trends in Parasitology. 21: 81-87.

Rose, J.B., C.P. Gerba, and W. Jakubowski. 1991. Survey of potable water supplies for Cryptosporidium and Giardia. Environmental Science and Technology. 25: 1393–1400.

Rose, J. 1997. Environmental ecology of Cryptosporidium and public health implications. Annual Review of Public Health. 18: 135-161

Sanders, R.W.; A.L. Macaluso, T.J. Sardina, and D.L. Mitchell. 2005. Photoreactivation in two freshwater ciliates: differential responses to variations in UV-B flux and temperature. Aquat. Microbial Ecology. 40: 283-292.

Searcy, K.E.; A.I. Packman, E.R. Atwill, and T. Harter. 2005. Association of Cryptosporidium parvum with suspended particles: impact on oocyst sedimentation. Applied and Environmental Microbiology. 71: 1072-1078.

105 Sivaganesan, M. and S. Sivaganesan. 2005. Effect of lot variability on ultraviolet radiation inactivation kinetics of Cryptosporidium parvum oocysts. Environmental Science and Technology. 39: 4166-4171.

Slifko, T. R., D. Friedman, J.B. Rose, and W. Jakubowski. 1997. An in vitro method for detecting infectious Cryptosporidium oocysts with cell culture. Applied and Environmental Microbiology. 63: 3669-3675.

Slifko, T. R., D.E. Huffman, and J.B. Rose. 1999. A most-probable number assay for enumeration of infectious Cryptosporidium parvum oocysts. Applied and Environmental Microbiology. 65: 3936-3941.

Tsushima, Y., P. Karanis, T. Kamada, L. Makala, X. Xuan, Y. Tohya, H. Akashi, and H. Nagasawa. 2003. Seasonal change in the number of Cryptosporidium parvum oocysts in water samples from the rivers in Hokkaido, Japan, detected by the ferric sulfate flocculation method. Journal of Veterinary Medical Science. 65: 121-123.

Weir, S.C., N.J. Pokorny, R.A. Carreno, J.T. Trevors, and H. Lee. 2002. Efficacy of common laboratory disinfectants on the infectivity of Cryptosporidium parvum oocysts in cell culture. Applied and Environmental Microbiology. 68: 2576-2579.

Williamson, C.E., H.E. Zagarese, P.C. Schulze, B.R. Hargreaves, and J. Seva. 1994. Impact of short-term exposure to UV-B radiation on zooplankton communities in north temperate lakes. Journal of Plankton Research. 16: 205-218.

Williamson, C.E. 1995. What role does UV-B radiation play in freshwater ecosystems? Limnology and Oceanography. 40: 386-392.

Williamson, C.E.; P.J. Neale, G. Grad, H.J. De Lange, and B.R. Hargreaves. 2001. Beneficial and detrimental effects of UV on aquatic organisms: implications of spectral variation. Ecological Applications. 11: 1843-1857.

Xenopoulos, M.A.; D.M. Lodge, J. Frentress, T.A. Kreps, S.D. Bridgham, E. Grossman, and C.J. Jackson. 2003. Regional comparisons of watershed determinants of dissolved organic carbon in

106 temperate lakes from the Upper Great Lakes region and selected regions globally. Limnology and Oceanography. 48: 2321-2334.

Yu, R. 2005. Crypto lawsuit hits N.Y.; Ohio. Aquatic International. www.allbusiness.com/periodicals/article/842538-1.html.

Yu, R. 2006. N.Y. sprayparks require UV: some manufacturers question the effectiveness and cost of the stipulation. Aquatic International. www.allbusiness.com/periodicals/article/878405- 1.html.

Zimmer, J.L., R.M. Slawson, and P.M. Huck. 2003. Inactivation and potential repair of Cryptosporidium parvum following low- and medium-pressure ultraviolet irradiation. Water Research. 37: 3517-3523.

107

Chapter 6

Impact of zooplankton grazing on the excystation, viability, or infectivity of the protozoan pathogens Cryptosporidium parvum and Giardia lamblia

Connelly, SJ, EA Wolyniak, KL Dieter, CE Williamson, KL Jellison. 2007. Impact of zooplankton grazing on the excystation, viability, or infectivity of the protozoan pathogens Cryptosporidium parvum and Giardia lamblia. Applied and Environmental Microbiology. 73(22): 7277-7282.

108 Abstract

Very little is known about the ability of the zooplankton grazer Daphnia pulicaria to reduce populations of Giardia lamblia cysts and Cryptosporidium parvum oocysts in surface waters. The potential for D. pulicaria to act as a biological filter of C. parvum and G. lamblia was tested under three grazing pressures (1, 2, or 4 D. pulicaria per 66 mL). (Oo)cysts (1x104 per 66 mL) were added to each grazing bottle along with the algal food Selenastrum (6.6x104 per 66 mL) to stimulate normal grazing. Bottles were rotated (2 rpm) to prevent settling of (oo)cysts and algae for 24 h (16:8 light:dark cycle) at 20 °C. The impact of D. pulicaria grazing on (oo)cysts was assessed by (i) (oo)cyst clearance rates, (ii) (oo)cyst viability, (iii) (oo)cyst excystation, and (iv) oocyst infectivity in cell culture. Two D. pulicaria significantly decreased the total number of C. parvum oocysts by 52% and G. lamblia cysts by 44%. Furthermore, two D. pulicaria significantly decreased C. parvum excystation and infectivity by 5% and 87%, respectively. Two D. pulicaria significantly decreased the viability of G. lamblia cysts by 52%, but analysis of G. lamblia excystation was confounded by observed mechanical disruption of the cysts after grazing. No mechanical disruption of the C. parvum oocysts was observed, presumably due to their smaller size. The data provide strong evidence that zooplankton grazers have the potential to substantially decrease the population of infectious C. parvum and G. lamblia in freshwater ecosystems.

109 Introduction

The obligate enteric protozoan parasites Cryptosporidium parvum and Giardia lamblia pose significant human health risks, with the most severe and potentially lethal cases of cryptosporidiosis and giardiasis occurring in infants, the elderly, and the immunocompromised. Although Cryptosporidium was not recognized as a waterborne source of human disease until 1987 (Rose, 1988), cases of what is now known as cryptosporidiosis have been documented since 1976 (CDC, 2005). Cryptosporidiosis is most commonly acquired by fecal-oral transmission through contaminated water and food supplies. The largest outbreak of cryptosporidiosis, linked to a municipal drinking water supply, occurred in Milwaukee, Wisconsin in 1993 and affected over 403,000 people (CDC, 2005). Following parasite exposure, the incidence of cryptosporidiosis in the United States approaches 40% in at-risk populations (Smith and Rose, 1990) compared to 4-12% for giardiasis (US EPA, 1998). While medical treatment has been previously limited to the alleviation of disease symptoms (i.e., diarrhea and dehydration), antiprotozoal treatments that are successful in ridding the body of Giardia are now available in developed countries (e.g., nitroimidazoles, with the most common being metronidazole; Gardner and Hill, 2001); these treatments are less successful for the eradication of Cryptosporidium. The risk of outbreaks and limited treatment options even in developed countries makes it important to investigate natural control mechanisms for these parasites. While the effects of some abiotic ecological determinants on protozoans in surface water systems have been identified (Pokorny et al., 2002; Hsu and Yeh, 2003; Brookes et al., 2004; Campbell et al., 2002), very little work has addressed the biotic effects of cohabitating freshwater zooplankton as grazers of these pathogens. Prior studies have shown that protozoa can ingest human bacterial and protozoan pathogens, including Cryptosporidium (Trout et al., 2002; Stott et al., 2003; Agasild and Noges, 2005). Stott et al. (2003) report the ingestion of Cryptosporidium oocysts by ciliated protozoa, amoebas, and rotifers at concentrations of 186 oocysts in 4 μl. Rotifera, a phylum of small invertebrates that includes many zooplankton species, have also been shown by gut imaging to ingest G. lamblia cysts (Trout et al., 2002). However, essentially no information is available on the rate at which zooplankton can clear (oo)cysts from the surrounding water or whether feeding by zooplankton renders the (oo)cysts non-viable. While rotifers have been shown to ingest Cryptosporidium oocysts and Giardia

110 cysts, they are generally selective feeders and not prone to ingesting particles larger than 9 μm (Armengol et al., 2001). The effect of larger invertebrate crustaceans, such as Daphnia pulicaria, which are highly efficient and non-selective filter feeders of particulates in the size range of both C. parvum (4 – 8 μm) and G. lamblia (7-14 μm), is unknown. Daphnia are more intensive grazers than rotifers, are widespread and abundant in many surface waters, and have the potential as a community to filter the equivalent of the entire volume of a lake in a given day (Cyr and Pace, 1992). While these traits imply that Daphnia could be a potentially significant natural control of C. parvum and G. lamblia, their role in the fate and transport of (oo)cysts has never been tested. This study focuses on the direct effect of a zooplankton grazer on the density, viability, and infectivity of the protozoan pathogens C. parvum and G. lamblia. The potential for D. pulicaria to act as an effective natural biological filter of C. parvum and G. lamblia was tested in laboratory-based grazing experiments by estimating Daphnia clearance rates for the pathogens. Since ingestion alone may or may not destroy the (oo)cysts, pathogen integrity was also assessed here with DAPI/PI stains for viability, excystation bioassays, and (for Cryptosporidium only) oocyst infectivity in human cell culture.

Methods

Daphnia culturing Daphnia pulicaria (source: Dutch Springs, Bethlehem, PA; 40°41'04.99”N, 75°21'16.26”W) were maintained in culture at 20 °C on a 16:8 light:dark cycle for at least three months prior to these experiments and fed Selenastrum (typical green algae used in Daphnia culturing) at a rate of 1000 cells mL-1 every two days (high food density). D. pulicaria cultures were kept in synthetic freshwater (US EPA, 2002) which was changed every five to seven days, depending on the population density.

Pathogen maintenance C. parvum oocysts derived from mice (Iowa isolate Lot # 060518) and G. lamblia cysts (Human isolate H-3 Lot # 060328) derived from gerbils were acquired live from Waterborne, Inc (New

111 Orleans, LA, USA). (Oo)cysts were shipped within 48 h of being shed and stored in 1x Hanks balanced salt solution (HBSS; pH 7) at 4 °C in the dark for less than two weeks prior to use in the grazing experiments.

Gut content analysis An (oo)cyst wall-specific antibody (MeriFluor®, Meridian Scientific, Cincinnati, OH, USA) was applied to C. parvum and G. lamblia prior to D. pulicaria grazing experiments to visualize grazed (oo)cysts within the D. pulicaria gut. Initial tests were conducted to ensure that the MeriFluor® antibody did not cross-react with Selenastrum by attempting to stain Selenastrum directly. No fluorescence of Selenastrum, beyond the natural green coloring of the algal cells, was observed microscopically using a FITC filter cube (excitation 420-490 nm, emission 520 nm). One drop of MeriFluor® Detection Reagent was added to each 1 mL suspension of 1x104 (oo)cysts in 1x HBSS (pH 7.0) and incubated at room temperature in the dark for 30 minutes. Following incubation, the entire suspension of stained (oo)cysts, along with 1x105 Selenastrum (to induce grazing), were fed to 10 D. pulicaria in 30 mL of synthetic freshwater. Images were acquired using an Olympus AX-70 Multimode Microscopy System (Olympus America Inc., Center Valley, PA, USA). Ingestion of the pathogens by D. pulicaria was observed by fluorescence microscopy using a FITC filter cube (Figure 6.1A). The G. lamblia cyst damage following mechanical digestion by D. pulicaria was observed using phase contrast microscopy (Figure 6.1B).

112

A B

5μm

Figure 6.1. (A) Gut content analysis. MeriFluor®-stained C. parvum and G. lamblia observed in the gut tract of D. pulicaria. C. parvum oocysts (circle) and G. lamblia cysts (square) are identified by specific fluorescence. Note that the green appearance of the D. pulicaria body and gut is an artifact of the FITC microscopy and is not a cross reaction with the MeriFluor® antibody. (B) G. lamblia were fed in moderate concentration to D. pulicaria for two h. The upper left cyst is intact following the grazing period; the cyst on the bottom right sustained damage to the upper left quadrant of the outer wall.

Grazing experiments Twelve hours prior to the grazing experiments, adult D. pulicaria were isolated from culture, placed in synthetic freshwater (pH 7.6-7.8) in 66 mL Pyrex bottles, and starved to clear their guts. Five replicates of three grazing pressure treatments were established (1, 2, or 4 D. pulicaria per 66 mL bottle). An additional treatment (five replicates) with zero D. pulicaria per bottle was established as the control. The bottles were stored under normal culture conditions (16:8 light:dark cycle at 20 °C) prior to the experiment. C. parvum and G. lamblia were tested independently. At the start of each experiment, C. parvum oocysts or G. lamblia cysts were added to each bottle (seeding methods below), along with Selenastrum (1000 mL-1) to stimulate grazing, and the bottles were filled to 66 mL total volume with synthetic freshwater. Structural differentiation of Selenastrum, C. parvum, and G. lamblia was observed using differential interference contrast (Nomarski contrast) microscopy (Figure 6.2).

113 5μm

Figure 6.2. Differential interference contrast (Nomarski contrast) microscopy was used to observe the structural differentiation of the three food items used in this study: Selenastrum, C. parvum, and G. lamblia (clockwise from left). The similar size of C. parvum and Selenastrum should be noted, while G. lamblia is distinctly larger. Image credit: M. Duley (Miami University).

(Oo)cysts were obtained as commercial stock solutions (Waterborne, Inc., New Orleans, LA, USA) of 1x107 (oo)cysts in 8 mL. A working solution of 2x105 (oo)cysts was made by diluting 160 μl of each thoroughly-mixed stock solution to 10 mL (1x HBSS, pH 7.0). This concentration was confirmed by hemacytometry (three replicate counts of the working dilution were within 5% of the calculated 2x105 (oo)cysts in 10 mL). The grazing bottles were seeded with 500 μl of this working stock, yielding 1x104 (oo)cysts per grazing bottle (66ml total volume), or 151 (oo)cysts mL-1. Laboratory controls of C. parvum and G. lamblia working stocks were maintained at 4 °C and compared to the zero D. pulicaria treatment to identify any confounding effects of (oo)cyst handling in the experimental setup. Grazing bottles were placed on a rotating wheel (2 rpm) for 24 h at 20 °C on a 16:8 light:dark cycle. Immediately following the 24-h grazing period, the D. pulicaria were removed using a transfer pipet (minimizing liquid withdrawal) and the remaining suspension was concentrated by centrifugation (1500 x g, 10 min) to 1 mL for analysis. Aliquots of each treatment suspension were taken from the 1 mL concentrate for (i) calculation of clearance rates (Table 6.1) and (ii) excystation, viability, and infectivity analyses (Table 6.2).

114 Calculation of densities and clearance rates Selenastrum, C. parvum, and G. lamblia were counted in the zero-grazer treatments following the 24-h grazing period; the zero-grazer treatment after the 24-h grazing rotation, centrifugation, and collection of the (oo)cyst pellet accounted for algal and (oo)cyst losses through handling and was a better initial control for this study. Therefore, counts of recoverable (oo)cysts from the zero-grazer treatments were used as the initial seeding densities for the 1, 2, and 4 D. pulicaria grazing experiments. Counts were conducted using a 1 mL Sedgwick-Rafter cell (SR cell), which permitted the counting of all (oo)cysts present in an entire 20 μl sub sample of the 1 mL concentrate post-grazing (compared to the 0.1 μl volume that would be counted in typical hemacytometry). Two SR cell counts were conducted for each sample, using 20 μl of the post- graze concentrate for each count diluted to 1 mL (to fill the SR cell). Five replicates of each grazing treatment were analyzed. Final recovered (oo)cyst and algal cell densities for the 1, 2, and 4 D. pulicaria grazing experiments were counted using the same SR cell method for all replicates of each grazing treatment. The recoverable (oo)cysts in the zero-grazer treatment were used to calculate the clearance rates (Table 6.1). Clearance rates (F) were calculated according to the formula: ρ ln( C ) ρ F = V E NT where V = grazing volume (0.066 L), ρC = density of recoverable food (i.e., Selenastrum or

(oo)cysts) in the zero-grazer control treatment, ρE = density of recoverable food (i.e., Selenastrum or (oo)cysts) in a given grazing treatment, N = total number of grazers in a given grazing treatment, and T = experimental duration (1 day). Clearance rates are reported as mL grazer-1 day-1 (per food item) (modified from Williamson and Reid, 2001) (Table 6.1).

Determining grazing effects on pathogens Three methods for determining the direct grazing effect of D. pulicaria on C. parvum oocysts and G. lamblia cysts were included in the present study. C. parvum viability, excystation, and in vitro infectivity and G. lamblia viability and excystation were assessed as described below. Viability assesses the permeability of the (oo)cyst wall by the uptake of the vital dyes 4’,6- diamidino-2-phenylindole (DAPI) and propidium iodide (PI). (Oo)cysts that are intact and viable

115 will stain with DAPI but will be impermeable to PI, whereas damaged (oo)cyst walls will permit uptake of PI and subsequent staining of the nucleic acids with both DAPI and PI. Excystation assesses the ability of an oocyst or cyst to excyst, releasing sporozoites or trophozoites, respectively, in the presence of bile acid (taurocholic acid). In vitro infectivity assesses the oocysts’ ability to infect a human cell monolayer and develop intracellular life stages. Given these distinctions, it is possible that damage which does not affect the permeability of the outer wall, but does render the (oo)cysts unable to excyst or reproduce would yield very different results in these three assays.

Vital dye staining (i.e., viability assay) Viability was evaluated based on standard DAPI/PI vital dye staining techniques for C. parvum (4) and G. lamblia (US EPA, 2001) using two 20 μl sub samples of the 1 mL concentrate post- grazing for each bottle (since five replicate bottles were included for each grazing pressure (1, 2, or 4 D. pulicaria), 10 DAPI/PI counts were analyzed for each grazing pressure). Viability was assessed for all (oo)cysts present in each sub sample (up to 100 (oo)cysts), and the average of two counts for each treatment replicate was used for the statistical analysis. Percent viability of C. parvum and G. lamblia was calculated as a percent change from zero grazers for each treatment (Figure 6.3).

Excystation assay Excystation of C. parvum and G. lamblia was evaluated by standard protocol (Robertson et al., 1993; Rice and Schaefer, 1981, respectively). Positive excystation counts of (oo)cysts include both partially and fully excysted (oo)cysts as a percentage of the total (oo)cysts (intact, partially and fully excysted). Excystation counts were conducted using two 20 μl sub samples of the 1 mL concentrate post-grazing for each grazing bottle (since five replicate bottles were included for each grazing pressure (1, 2, or 4 D. pulicaria), 10 excystation counts were analyzed for each grazing pressure). Taurocholic acid (50 μL of 0.75% solution) was added to each 20 μl sub sample. 930 μl of distilled water was added to the taurocholic acid and (oo)cyst suspension to dilute the (oo)cysts for counting in an SR cell. Up to 100 (oo)cysts in each replicate excystation sample were classified as intact, partially excysted, or totally excysted, and the average of two counts for each treatment replicate was used for the statistical analysis (Table 6.2). Percent

116 excystation of G. lamblia cysts and C. parvum oocysts were calculated as a percent change from zero grazers for each treatment (Figure 6.3).

Infectivity assay Percent infectivity of grazed C. parvum oocysts was calculated by in vitro cell culture infection of human ileocecal adenocarcinoma (HCT-8) cells grown in 8-well chamber slides according to Slifko et al. (1997; 1999). The oocysts remaining in the 1 mL post-grazing concentrate (after sub samples for density counts, viability staining, and excystation assays were removed) were pretreated with a 10% NaOCl solution and the concentration of oocysts in the concentrate was determined by hemacytometry. An infection volume containing 100 oocysts from each experimental concentrate was calculated and three replicate wells with confluent cell monolayers on the chamber slide were seeded with 100 oocysts from the post-grazing concentrate (i.e., 15 replicate wells were seeded for each grazing pressure (1, 2, or 4 D. pulicaria) since five replicate bottles were included for each grazing pressure and three replicate wells were infected per grazing bottle). Two wells on each slide were left uninfected to serve as negative controls.

Infected chamber slides were incubated at 37 °C and 5% CO2 for 48 h and then stained according to the Sporo-GloTM antibody protocol (Waterborne, Inc., New Orleans, LA, USA). Foci of infection were counted using a Nikon 50I upright epifluorescence microscope (Optical Apparatus, Ardmore, PA, USA). Infection foci were counted as positive only if a cluster of three or more distinct life stages were present. The percent infective oocysts was calculated by dividing the number of infection foci by 100 (the total number of oocysts used to infect). The average percent infectivity and standard error were calculated from the five experimental replicates per grazing treatment (Table 6.2). Percent change in infectivity of HCT-8 cells in vitro by C. parvum oocysts was compared to the zero-grazer control to limit the confounding effects of oocyst handling and recovery (Figure 6.3). (Oo)cyst lab controls were tested simultaneously for infectivity.

Statistical analyses For all pathogen analyses following grazing, the recoverable (oo)cysts and Selenastrum from the zero D. pulicaria treatment were used as the controls (to eliminate any effect of handling and processing on pathogen health). The grazing data of one D. pulicaria on C. parvum oocysts and

117 G. lamblia cysts are not presented due to the unrealistic conditions as compared to a natural system; therefore, these analyses address the impact of high intensity grazing in a 24-h period by two and four D. pulicaria. The impact of grazing pressure on the probability of an (oo) cyst being viable was modeled using logistic regression (SPSS Release 13.0, SPSS Inc., Chicago, IL, USA; α=0.05).

Results

C. parvum oocysts and G. lamblia cysts were observed for mechanical disruption of the outer wall. High mechanical disruption of the G. lamblia cysts was noted (Figure 6.1B), whereas little or no outer wall disruption of the C. parvum oocysts was observed in any of the grazing treatments. In the control (zero-grazer) treatment, recoverable C. parvum and G. lamblia (oo)cyst counts were 97-122 (mean ± experimental standard error; 107±2.7) mL-1 and 113-265 (189±23.9) mL-1, respectively, in 66 mL. Selenastrum counts in the zero-grazer treatment with C. parvum and G. lamblia were 1114-1230 (1159±12.3) mL-1 and 1628-1796 (1701±28.4) mL-1, respectively, in 66 mL. For comparison, final recoverable C. parvum oocysts and G. lamblia cysts in the two-grazer treatment ranged from 38-62 (51±2.5) mL-1 and 76-189 (106±22.1) mL-1, respectively, and Selenastrum with C. parvum and Selenastrum with G. lamblia were 632-662 (646±3.5) mL-1 and 1114-1220 (1162±19.4) mL-1, respectively, in 66 mL. Final recoverable C. parvum oocysts and G. lamblia cysts in the four-grazer treatment ranged from 10-58 (29±5.1) mL-1 and 38-114 (76±11.9) mL-1, respectively, and Selenastrum with C. parvum and Selenastrum with G. lamblia were 81-160 (126±6.7) mL-1 and 796-939 (855±28.3) mL-1, respectively, in 66 mL (Table 6.1). Clearance rates of two D. pulicaria on (i) G. lamblia alone, (ii) Selenastrum in the presence of G. lamblia, and (iii) C. parvum alone were 19, 13, and 24 mL grazer-1 day-1, respectively (not significantly different from the clearance rates of four D. pulicaria at 15, 11, and 22 mL grazer-1 day-1, with p=0.441, p=0.133, p=0.827, respectively) (Table 6.1). There was a significant increase in the clearance of Selenastrum in the presence of C. parvum from two to four D. pulicaria (19 mL grazer-1 day-1 to 37 mL grazer-1 day-1; p<0.001) (Table 6.1).

118 Table 6.1. Clearance rates of C. parvum oocysts, G. lamblia cysts, and Selenastrum following a 24-h grazing period by D. pulicaria, including average recovered (oo)cysts and algae per grazing treatment.

Food # D. pulicaria Ctrl Food Density Expt Food Density Avg Clearance Rate Grazers (avg cells mL-1) (avg cells mL-1) (mL grazer-1 day-1) C. parvum 2 107 51 24

4 107 29 22

G. lamblia 2 189 106 19

4 189 76 15

Selenastrum (w/ 2 1159 646 19 C. parvum)

4 1159 126 37

Selenastrum (w/ 2 1701 1162 13 G. lamblia)

4 1701 855 11

Average clearance rates (mL grazer-1 day-1) for all food particles (C. parvum, G. lamblia, Selenastrum with C. parvum, and Selenastrum with G. lamblia) in these grazing experiments based on clearance rate formula. Calculations based on 24-h experimental duration in 66 mL grazing volume. Control (zero-grazer to account for recoverable (oo)cysts) and experimental food densities are average cells mL-1 (in 66 mL) in a given grazing treatment (N=5).

Mean percent infectivity (± standard error) of six working stock controls based on 100 intact C. parvum oocysts from the same lot used in the grazing experiment was 28±11 (Table 6.2). The zero-grazer treatment mean percent infectivity of 100 intact C. parvum oocysts was 15±1.8 (Table 6.2), demonstrating a 46% mean decrease in infectivity with handling during this experiment. Two D. pulicaria grazers significantly decreased C. parvum mean excystation and mean infectivity by 5% (p=0.001) and 87% (p<0.001), respectively (Table 6.2, Figure 6.3), compared to the zero-grazer control. Four D. pulicaria grazers significantly decreased C. parvum mean excystation and infectivity by 13% (p<0.001) and 87% (p<0.001), respectively (Table 6.2, Figure 6.3), compared to the zero-grazer control. Given the extremely low numbers of oocysts counted after vital dye staining (presumably due to oocyst losses during staining and rinsing of the slides), oocyst viability data was inconclusive.

119 Two D. pulicaria grazers significantly decreased G. lamblia mean viability by 52% (p<0.001), but significantly increased mean excystation by 20% (p<0.001) following grazing (Figure 6.3). Four D. pulicaria grazers significantly decreased G. lamblia mean viability by 42% (p<0.001), but significantly increased mean excystation by 28% (p<0.001) (Table 6.2, Figure 6.3).

120 Table 6.2. Viability, excystation, and infectivity counts of C. parvum oocysts and G. lamblia cysts following a 24-h grazing period by D. pulicaria.

# D. pulicaria grazers

0 2 4 Lab Control

C. parvum Viability§ % Viable Oocysts (Mean) 97% 100% 92% 100% % Viable Oocysts (Range) 83-100% 100% 67-100% 100% # Oocysts Counted (Range) 2-6 2-4 1-4 18 # Oocysts Counted (Mean ± Expt 3 ± 0.8 3 ± 0.4 2 ± 0.6 18 ± 0.0 St Err)

G. lamblia Viability§ % Viable Cysts (Mean) 84% 40% 49% 86% % Viable Cysts (Range) 80-87% 30-47% 44-59% 84-87% # Cysts Counted (Range) 92-102 84-119 78-98 100-102 # Cysts Counted (Mean ± Expt St 97 ± 1.8 96 ± 4.3 88 ± 2.7 101 ± 0.3 Err)

C. parvum Excystation# % Excysted Oocysts (Mean) 100% 95% 87% 100% % Excysted Oocysts (Range) 99-100% 94-98% 85-89% 100% # Oocysts Counted (Range) 99-102 99-101 99-102 100-102 # Oocysts Counted (Mean ± Expt 100 ± 0.3 100 ± 0.2 100 ± 0.2 101 ± 0.5 St Err)

G. lamblia Excystation# % Excysted Cysts (Mean) 74% 89% 95% 78% % Excysted Cysts (Range) 68-79% 80-95% 89-98% 76-80% # Cysts Counted (Range) 112-136 108-121 88-132 100-116 # Cysts Counted (Mean ± Expt St 120 ± 4.2 114 ± 2.6 110 ± 7.1 107 ± 3.3 Err)

C. parvum Infectivity* % Infectivity (Mean ± Expt St 15% ± 1.8% 2% ± 0.5% 2% ± 0.5% 28% ± 11.0% Err) % Infectivity (Range) 9-20% 0-3% 0-3% 2-61%

Because of the intensive handling of the (oo)cysts in this experiment, the number of pathogens counted in any analysis is highly variable. Here were report this variability as range (min and max) and mean ± experimental standard error (N=5 replicates). § Viability counts are the range of total (oo)cysts (min and max) and mean number of (oo)cysts ± experimental standard error (N=5) counted in the DAPI/PI procedure. Viable (oo)cysts stained DAPI+/PI-. Due to the low numbers of C. parvum oocysts counted in the DAPI/PI viability assay (presumably due to oocyst losses during staining and rinsing of the slides), these data are not statistically significant but are presented for completion of the data set. # Excystation counts are the total number of (oo)cysts expressed as range (min and max) and mean excysted ± experimental standard error (N=5). * Counts of infectivity are per 100 intact oocysts and are reported as mean number of infections ± experimental standard error (N=5) and as range of total infection foci (min and max).

121

Cryptosporidium parvum Giardia lamblia

60 60

40 40

20 20

0 0

-20 -20

Viability (% (% change) Viability -40 -40

-60 -60 40 024 40 X Data 20 20

0 0

-20 -20 Excystation (% change) (% Excystation

-40 -40 100 024 # D. pulicaria grazers 50

0

-50 Infectivity (% change) (% Infectivity

-100 024 # D. pulicaria grazers

Figure 6.3. Average percent change from zero-grazer control of excystation, viability, and infectivity of C. parvum and G. lamblia by zero-, two-, and four-D. pulicaria grazers in a 24-h experiment. Error bars represent experimental standard error (N=5).

122 Discussion

This is the first quantitative demonstration of the potential for zooplankton grazers to reduce C. parvum and G. lamblia in freshwater systems. The actual reduction of human pathogens by zooplankton in surface waters will depend on many factors including the species, density, and depth distribution of the grazers and the pathogens. In nature, D. pulicaria and other species of Daphnia can reach maximum densities approaching 100 L-1, although average densities in lakes are more often on the order of 10-20 L-1 or less, depending on lake or reservoir productivity. At clearance rates even as low as 11 mL grazer-1 day-1 (the minimum clearance rate observed in the current study), high densities of Daphnia have the potential to clear the entire population of human pathogens from the water in a single day under favorable conditions. Variations in the vertical distribution of pathogens and grazers and other environmental factors are likely to substantially reduce these impacts, but these data do indicate the potential for zooplankton to act as a significant natural biological filter of human protozoan pathogens. Grazing by D. pulicaria significantly decreased the total numbers of intact pathogens following a 24-h grazing period in the presence of an alternate food source, Selenastrum. Both C. parvum and G. lamblia were cleared from the water by D. pulicaria at rates that are similar to or greater than the culture food, Selenastrum. The similarity in size of the Selenastrum, C. parvum, and G. lamblia (Figure 6.2) is believed to contribute to the insignificant differences in clearance rates in this study. The one significant clearance rate between two and four grazers (Selenastrum in the presence of C. parvum) may have occurred because of clumping of the Selenastrum with the oocysts or itself, or simply because of the increased number of grazers and the potential for repeated filtering of food particles. The grazing methods used herein do not permit distinction between (oo)cysts that were never ingested by D. pulicaria and those that were ingested multiple times during the 24-h period. Given the relatively small grazing volume (66 mL), the high densities of both pathogens and Daphnia and the clearance rates per grazer (Table 6.1), potential repeated ingestion rates can be calculated. We estimate that two grazers will filter 26-48 mL of water every 24 h, or 39-73% of the grazing vessel volume (based on the range of calculated clearance rates of 13-24 mL grazer-1 day-1 food item-1). We estimate that four grazers will filter 44-148 mL, or 67-224% of the grazing vessel volume, every 24 h (based on the range of calculated clearance rates of 11-37

123 mL grazer-1 day-1 food item-1). Although this experimental design does not allow confirmation of the repeated ingestion of the (oo)cysts, the clearance rates of this experimental design (26-59 mL grazer-1 day-1 of a pathogen and algae) suggest that one grazer could pass all food items in the grazing vessel in a 27-61 h period (Table 6.1). The high likelihood for repeated ingestion in this study may cause mechanical damage to the (oo)cyst walls, resulting in decreased viability (due to PI penetration through the damaged wall), decreased infectivity, but potentially increased excystation counts (as the release of sporozoites or trophozoites due to excystation conditions may be difficult to differentiate from their release due to mechanical damage to the (oo)cyst wall). Upon visual observation of the (oo)cysts following grazing, C. parvum showed, in most cases, no outer wall damage of the oocysts that would be attributed to mechanical disruption in the grazing treatments. In contrast, the outer wall of many G. lamblia cysts appeared to be mechanically disrupted after their ingestion and excretion by D. pulicaria (Figure 6.1B). This observation supports our findings that G. lamblia excystation counts increased, but viability decreased, with increased grazing (Table 6.2, Figure 6.3). Reduced C. parvum oocyst excystation with increased numbers of grazers suggests some influence of the D. pulicaria on the ability of the oocysts to initiate the life cycle and develop intracellular life stages. These data were supported by the significant reduction in infectivity of C. parvum by in vitro cell culture following D. pulicaria grazing. This marked decrease in infectivity following a 24-h grazing period by even two grazers strongly supports the biotic control of infectious human pathogens by zooplankton. In the absence of an established in vitro cell culture assay for G. lamblia, viability and excystation were used here as infection proxies. D. pulicaria grazing significantly decreased the percent viability of intact G. lamblia cysts. This decreased viability of G. lamblia following grazing supports the theory that mechanical digestion by the D. pulicaria, likely with repeated ingestion and excretion in a 24-h period, disrupts the outer wall. This outer wall disruption made it appear that excystation was increasing, although in reality, it is likely that many of the cysts that appeared to have excysted were actually damaged by mechanical digestion. Although viability staining may overestimate the percent of infectious cysts in the population (Thiriat et al., 1998), we predict a significant decrease in the infectivity of G. lamblia cysts following grazing by D. pulicaria based on the striking decrease in viability observed here. Future studies

124 that utilize an animal model to determine infectivity of G. lamblia following grazing should include an analysis of both excystation and viability to confirm mechanical interference with the outer wall as a mechanism for reduced survival of the cysts (as opposed to any intrinsic difference in the outer wall construction of G. lamblia versus C. parvum that might be impacted by pH or other abiotic influences of the Daphnia gut). Multiple factors may influence the impact of zooplankton grazers on the presence and infectivity of human pathogens in natural systems, and interactions of abiotic and biotic factors may play a significant role in pathogen control. Zooplankton grazers are often present in freshwater systems that contain pathogens (lakes, reservoirs, etc.) and can have a significant impact on these waterborne pathogen populations. The data presented in this study provide strong evidence that under conditions in which zooplankton grazers flourish, and where settling rates of the oo(cysts) are low, D. pulicaria grazers can significantly decrease the numbers of infectious pathogens in freshwater systems.

Acknowledgements

SJC and CEW thank the Zoology Department (Miami University) for graduate research support to SJC. SJC thanks M. Duley (Miami University) for DIC imaging and microscopy support on this project. This work was partially supported by CAREER Award # 0545687 from the National Science Foundation to KLJ.

125 References

Agasild, H. and T. Nõges. 2005. Cladoceran and rotifer grazing on bacteria and phytoplankton in two shallow eutrophic lakes: in situ measurement with fluorescent microspheres. Journal of Plankton Research. 27(11): 1155-1174.

Armengol, X., L. Boronat, A. Camacho, and W.A. Wurtsbaugh. 2001. Grazing by a dominant rotifer Conochilus unicornis Rousselet in a mountain lake: In situ measurements with synthetic microspheres. Hydrobiologia. 446/447(1): 107-114.

Brookes, J.D., J. Antenucci, M. Hipsey, M.D. Burch, N.J. Ashbolt, and C. Ferguson. 2004. Fate and transport of pathogens in lakes and reservoirs. Environmental International. 30:741- 759.

Campbell, A.T., L.J. Robertson, and H.V. Smith. 2002. Viability of Cryptosporidium parvum oocysts: correlation of in vitro excystation with inclusion or exclusion of fluorogenic vital dyes. Applied and Environmental Microbiology. 58(11): 3488-3493.

Centers for Disease Control and Prevention. December 2005 posting date. Laboratory Identification of Parasites of Public Health Concern: Parasites and Health: Cryptosporidiosis. [Online.] http://www.dpd.cdc.gov/dpdx/

Cyr, H. and M.L. Pace. 1992. Grazing by zooplankton and its relationship to community structure. Canadian Journal of Fisheries and Aquatic Science. 49: 1455-1465.

Gardner, T.B. and D.R. Hill. 2001. Treatment of giardiasis. Clinical Microbiology Reviews. 14(1): 114-128.

Hsu, B.M. and H.H. Yeh. 2003. Removal of Giardia and Cryptosporidium in drinking water treatment: a pilot-scale study. Water Research. 37: 1111-1117.

126 Pokorny, N.J., S.C. Weir, R.A. Carreno, J.T. Trevors, and H. Lee. 2002. Influence of temperature on Cryptosporidium parvum oocyst infectivity in river water samples as detected by tissue culture assay. Journal of Parasitology. 88: 641-643.

Rice, E.W. and F.W. Schaefer III. 1981. Improved in vitro excystation procedure for Giardia lamblia cysts. Journal of Clinical Microbiology. 14: 709-710.

Robertson, L.J., A.T. Campbell, and H.V. Smith. 1993. In vitro excystation of Cryptosporidium parvum. Parasitology. 106(Pt 1): 13-19.

Rose, J. 1988. Occurrence and significance of Cryptosporidium in water. Journal of the American Water Works Association. 2: 53-58.

Slifko, T.R., D. Friedman, J.B. Rose, and W. Jakubowski. 1997. An in vitro method for detecting infectious Cryptosporidium oocysts with cell culture. Applied and Environmental Microbiology. 63(9): 3669-3675.

Slifko, T.R., D.E. Huffman, and J.B. Rose. 1999. A most-probable number assay for enumeration of infectious Cryptosporidium parvum oocysts. Applied and Environmental Microbiology. 65(9): 3936-3941.

Smith, H.V. and J.B. Rose. 1990. Waterborne cryptosporidiosis. Parasitology Today. 6(1): 8- 12.

Stott, R., E. May, E. Ramirez, and A. Warren. 2003. Predation of Cryptosporidium oocysts by protozoa and rotifers: implications for water quality and public health. Water Science and Technology. 47(3): 77-83.

Thiriat, L., F. Sidaner, and J. Schwartzbrod. 1998. Determination of Giardia cyst viability in environmental and fecal samples by immunofluorescence, fluorogenic dye staining and differential interference contrast microscopy. Letters of Applied Microbiology. 26: 237-242.

127 Trout, J.M., E.J. Walsh, and R. Fayer. 2002. Rotifers ingest Giardia cysts. Journal of Parasitology. 88: 1038-1040.

US Environmental Protection Agency. 1998. Giardia: Human Health Criteria Document. Office of Water. Washington, D.C., USA.

US Environmental Protection Agency. 2001. Method 1623: Cryptosporidium and Giardia in water by filtration/IMS/FA. Washington, D.C., USA.

US Environmental Protection Agency. 2002. Short-term Methods for Estimating the Chronic Toxicity of Effluents and Receiving Waters to Freshwater Organisms: Fourth Edition. Washington, D.C., USA.

Williamson, C.E. and J.W. Reid. 2001. In: Thorp, J.H. and A.P. Covich (eds). Ecology and Classification of North American Freshwater Invertebrates: 2nd Edition. Academic Press, USA.

128

Concluding remarks

129 Freshwater ecosystems are especially susceptible to climatic change, including anthropogenic induced changes, as they are directly influenced by the atmosphere and terrestrial ecosystems. A major environmental factor that potentially affects every element of an ecosystem, directly or indirectly, is ultraviolet radiation (UVR). UVR has been shown to negatively effect the DNA of aquatic organisms by the same mechanism, that is by the formation of cyclobutane pyrimidine dimer (CPDs) as well as other photoproducts, as in humans. Following UV-induced DNA damage, an organism may repair the damage by two primary means, photoenzymatic repair (light driven, UV-damage specific repair not present in all organisms) or nucleotide excision, or dark, repair (metabolically driven, general DNA repair present in all organisms). The species and/or temperature dependence on the ability of an organism to repair the UV-induced damage and ultimately survive is not well understood. The induction of CPDs by solar UVR was quantified across six temperate ecosystems (aquatic and terrestrial). Data show significant variation in CPD formation not only between aquatic and terrestrial ecosystems but also within a single ecosystem and between seasons. Variable induction of DNA damage across an ecosystem may be used to explain species distribution patterns and community structure. The direct effect of solar radiation and artificial UVB on Daphnia spp. was tested. The ability of Daphnia species to repair UV-induced DNA damage at depth in natural lake systems is unknown. DNA damage (CPDs) was quantified in Daphnia spp. of two temperate lake systems of variable transparency. Data show DNA damage induction beyond the initial exposure in the surface waters and decreased damage with depth (repair). These data were collected monthly from early spring (April) through late summer (August), but were significantly confounded by rain events and the subsequent decreased transparency of the lakes. Artificial UV was used to induce DNA damage in four Daphnia species in the lab. The induction was quantified as DNA photoproducts (CPDs) produced after five different UVR exposures in the four species. Data show significant differences in induced DNA damage between species. Further, there is little quantitative data on the effectiveness of various DNA repair mechanisms in freshwater invertebrates and what is available is somewhat contradictory. From the induction experiments, the lethal dose for approximately 50% of each population (LD50; 20 kJ/m2) was determined and used to induce DNA damage for repair kinetic experiments at two temperatures in the four Daphnia species. Three of the Daphnia species show

130 nucleotide excision repair at 10°C, but none of the species show photoenzymatic repair at the same temperature. Further, survival is higher under both repair regimes in the 10°C acclimated Daphnia as compared to the 20°C acclimated Daphnia after 216 h. These data may suggest a mechanism by which sensitive aquatic species are able to tolerate the high UV, cold temperature of high alpine and Arctic systems. In natural systems, freshwater pathogens may be affected by both biotic and abiotic environmental stressors which may provide act as a natural control. UV is used as a disinfectant in water treatment, however, effects of natural UV on human protozoan pathogens are not known. The environmental abiotic stressor solar radiation significantly reduced the infectivity of Cryptosporidium parvum under otherwise controlled environmental conditions. Additional study of UVB in the laboratory demonstrated a similar reduction in this pathogens infectivity, suggesting solar radiation may be a low-cost water treatment option. Grazing of the human protozoan pathogens, Cryptosporidium parvum and Giardia lamblia, by a freshwater filter- feeding zooplankton, Daphnia pulicaria, significantly impacted the excystation, viability, and/or infectivity of the pathogens. Together, these abiotic and biotic factors may play a significant role in the natural control of pathogen persistence in natural waters. We have tested and supported the following hypotheses: (i) there is a significant increase in photoproduct concentrations (CPDs) in systems with increased ultraviolet radiation (UVR) exposure (surface of aquatic systems, low DOC aquatic systems, open terrestrial systems, etc.) and there is significant seasonal variability in UVA (actinometers) and UVB (actinometers and dosimeters) in the six ecosystems tested; (ii) four Daphnia spp. show variable induction and repair of UV-induced DNA damage in the laboratory and repair rates (albeit PER or dark) are temperature; (iii) artificial UVB and naturally occurring levels of solar radiation significantly reduced the infectivity of Cryptosporidium parvum; and (iv) grazing of Cryptosporidium parvum oocysts and Giardia lamblia cysts by Daphnia pulicaria had significant effects on excystation, viability, and/or infectivity of the (oo)cysts. Our study of Daphnia catawba exposed to fixed UVB and natural repair radiation was inconclusive as was presented herein. However, the work on the four Daphnia spp. in the lab provides substantial evidence for the ability of Daphnia to repair significant UV-induced DNA damage under both PER and dark repair regimes at varying temperatures.

131 These studies have supported many of aspects of the proposed hypotheses. Further, they have provided novel insights into many areas of research. We show high variability in UV- induced DNA damage in six temperate ecosystems, both aquatic and terrestrial. There is significant variability in UV-DNA damage induction in the four species of Daphnia tested. Further, there is a significant direct effect of temperature on the repair rate and efficiency (long- term survival) in four Daphnia species. When considering the distribution of human pathogens in potable and other freshwater systems, we found a direct negative effect of natural solar radiation and artificial UVB radiation on Cryptosporidium parvum at exposures equivalent to less than 1 solar day. In addition, we provide evidence for a direct negative effect of Daphnia pulicaria on excystation, viability, and infectivity of Cryptosporidium parvum oocysts and Giardia lamblia cysts. This work set out to address five environmentally-based hypotheses using almost purely molecular techniques. By this, these works have bridged two unique disciplines and have focused on applying interdisciplinary approaches to otherwise very divergent fields of study. Techniques such as UV-induced DNA damage assessment, environmental controls of human pathogens and the like used herein provide a more comprehensive approach to understanding biotic and abiotic forcers in nature. Approaches such as these can be applied to many systems and questions throughout both disciplines of study, and are techniques that are easily applicable across disciplines.

132