PLoS One. 2011; 6(11): e27229. PMCID: PMC3206931 Published online 2011 November 2. doi: 10.1371/journal.pone.

0027229 Copyright Dass et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.

Protozoan Parasite Toxoplasma gondii Manipulates Mate Choice in Rats by Enhancing Attractiveness of Males
Shantala Arundathi Hari Dass,#1 Anand Vasudevan,#1 Deborah Dutta,1Linda Jing Ting Soh,1 Robert Morris Sapolsky,2 and Ajai Vyas1*
1 2

School of Biological Sciences, Nanyang Technological University, Singapore, Republic of Singapore Department of Biology, Stanford University, Stanford, California, United States of America Gordon Langsley, Editor Institut national de la sant et de la recherche mdicale - Institut Cochin, France # Contributed equally. * E-mail: avyas@ntu.edu.sg Conceived and designed the experiments: SAHD A. Vasudevan RMS A. Vyas. Performed the experiments: SAHD A. Vasudevan DD LJTS A. Vyas. Analyzed the data: SAHD A. Vasudevan A. Vyas. Wrote the paper: RMS A. Vyas. Received August 5, 2011; Accepted October 12, 2011.

Abstract
Females in various species typically avoid males infected with parasites, while parasite-free males advertise their status through conspicuous phenotypic traits. This process selects for heritable resistance and reduces direct exposure of the female to parasites. Coevolving parasites are likely to attempt to circumvent this obstacle. In this paper, we demonstrate a case of parasitic manipulation of host mate choice. We report that Toxoplasma gondii, a sexually transmitted infection of brown rats, enhances sexual attractiveness of infected males. Thus under some evolutionary niches, parasites can indeed manipulate host sexual signaling to their own advantage.

Other Sections

Introduction
Host and parasites coevolve by placing constant demand on each other for adaptations and counter-adaptations. Selection pressures acting on the host can be very severe, in fact severe enough to drive host towards bi-parental reproduction even when this is very costly to maintain non-childbearing males [1]. Hence, it is not surprising that females typically avoid and in fact show aversive response to parasitized males [212]. Aversion of females to parasitized males is likely driven by the evolutionary need for females to choose for heritable parasite resistance [13]; and/or avoid direct transmission of contagious diseases during mating [14]. Such female aversion is detrimental for parasites, particularly if they are transmitted by sexual intercourse. We posit that in this situation, parasites have an evolutionary pressure to manipulate host males in a way that overcomes the traditional female aversion.

We tested this hypothesis using a common protozoan parasite of the brown rats,Toxoplasma gondii [15]. Recently, Toxoplasma gondii has been reported to be sexually transmitted in sheep and dog [16,17]. We first tested if it is a sexually transmitted in rats; we then investigated if chronic Toxoplasma gondii infection manipulated female mate choice.

Other Sections

Results Toxoplasma gondii is a sexually transmitted infection in brown rats

Parasitic cysts containing bradyzoites could be visualized eight weeks post-infection in epididymis, a tubular structure that collects and stores sperm (estimated burden = 520 cysts per animal, 37 m in diameter, n = 5 animals; Figure 1A). These cysts were viable; feeding them resulted in sero-conversion of uninfected females (4/4 attempts; one infected epididymis/female). Infected males were mated with uninfected females. Toxoplasma gondii cysts could also be observed from vaginal lavage of females 12 hours after mating (6 out of 7 mating; Figure 1B), indicating thatToxoplasma gondii was successfully ejaculated. Mating with an infected male resulted in transmission of infection to females (4/4 mating). This was demonstrated by successful PCR amplification from female brain of a gene unique to the parasite. Using previously published primer sets, we successfully amplified B1 gene ofToxoplasma gondii from genomic DNA prepared from crude lysate of female brains four weeks after mating (see [18] for methods). Identity of PCR product was confirmed using DNA sequencing. Infection was also confirmed by serological tests showing presence of anti-parasite IgG antibodies in mated females (serum dilution 1 1000; 6/7 mating). Serum obtained from mated females contained primary antibodies against laboratorycultured Toxoplasma gondii tachyzoites (visualized using anti-rat IgG coupled with Cy3). Mated females were allowed to give birth, and pups were tested for presence of Toxoplasma gondii. Parasite cysts were detected in 43 out of 69 pups that we tested (Figure 1C). Hence Toxoplasma gondii is a sexually and vertically transmitted infection in rats.
Figure 1 Toxoplasma gondii is transmitted through intercourse.

Reproductive parameters of control and infected animals were comparable

Despite invading reproductive organs, Toxoplasma gondii did not adversely affect sexual behavior or fecundity/fertility of infected males. We tested the attraction of sexually nave control and infected males to females. Attraction was quantified by comparing time spent by a male in two opposing arms of an arena, each arm containing either an inaccessible estrus female or an unfamiliar male. Both control and infected animals exhibited comparable

attraction to females (n = 11 males in each group; exact Mann-Whitney test: Z = 0.23, p>0.8; % time spent in bisect containing receptive female = 584.8% for control and 593.7% for infected animals). Control and infected animals exhibited comparable number of mounts and intromissions during paced mating with uninfected females (n = 6 control and 7 infected males; exact Mann-Whitney test: Z<0.43, p>0.7; Table 1). Likewise, infection did not affect the number of pups, weight of newborns or sex ratio of progeny resulting from mating (Table 2).
Table 1 Despite invading reproductive tissue, Toxoplasma gondii did not affect mating performance of infected animals.

Table 2 Despite invading reproductive tissue, Toxoplasma gondii did not affect number and sex ratio of progeny.

Uninfected females preferred infected males

Next, we tested if uninfected females avoided parasitized males, as would be expected from the extensive literature on this subject [212]. Preference of an estrus female for control or infected male was determined during a two-choice preference task (12 pairs of males, 57 females per pair; total 72 trials). Contrary to expectation, sexually nave females spent more, rather than less time in the infected bisect (Wilcoxon signed ranks test: Z = 4.547, p<0.00001; 50630 s in infected bisect versus 30817 s in control; n = 72 trials). A preference score of females for infected males was computed for each trial by dividing time spent in infected bisect by that in control. During 74% of trials, females spent more time in the infected bisect (Figure 2A; Chi2 test: Chi2 = 16.1; p<0.0001; infected>control = 53 trials, control >infected = 19 trials; median preference score = 1.54).
Figure 2 Uninfected females preferred infected males.

Time spent in infected and control bisects for each unique male pair was calculated by taking median of all females tested for each male pair. In all male pairs tested, median occupancy in infected bisect was greater than median occupancy in control area (Figure 2B; exact Wilcoxon signed ranks test: Z = 3.059; p<0.0001).

The generality of this finding was confirmed using a different strain of rats in a geographically different laboratory (Long-Evans rats in Stanford University; Figure 2C; n = 36 females; Chi2 test: Chi2 = 13.4; p<0.001; infected>control = 29 females, control >infected = 7 females; median preference score = 1.30). In this case, male bedding was used instead of urine marks and males were absent from the arena during testing. Since males were not physically present during this experiment, this precludes a role for male vocalization in our observations.

Infected males gained greater reproductive opportunities

In a competitive paced mating set-up whereby female chose to mate between one control and one infected male, females allowed greater number of reproductive opportunities to infected males (Figure 2D). During two-hour trial, infected Wistar males secured more intromissions compared to controls (n = 6 pairs; exact Wilcoxon signed ranks test: Z = 2.21, p<0.05; one-sample t-test against chance i.e. 1: t = 4.36, p<0.01). Hence, Toxoplasma gondii infection altered mate choice of females in a completely unexpected direction, whereby infected male gained in terms of reproductive fitness.

Other Sections

Discussion Further support for manipulation hypothesis

The core of parasitism is the ability of an organism to exploit its host. According to the behavioral manipulation hypothesis, a parasite may be able to alter the behavior of its host for its own selective benefit [1922]. Toxoplasma gondii has been previously described as a classical case of parasitic behavioral manipulation [2326]. It blocks the aversion of rats for cat urine, instead producing an attraction. This behavioral change is likely to increase the likelihood of a cat predating a rat. This is thought to reflect an adaptive behavioral manipulation by the parasite because it reproduces sexually only in the gut of the cat [23,24,27]. In this report, we show thatToxoplasma gondii is transmitted through sexual intercourse in brown rats, and that the parasite can manipulate mate choice of uninfected females. This behavioral change plausibly enhances transmission of the parasite from males to females and their progeny. Data presented here suggest a novel class of behavioral manipulation by Toxoplasma gondii. This runs counter to well-established observations in rodents [4,6] and in other animals [2,3,5,711] that females detect and avoid males infected with an array of viruses, bacteria, protozoa and nematodes. For example, female mice avoid males infected with Eimmeria vermiformis, a close relative ofToxoplasma gondii, and have an aversive physiological reaction to the smell of such males [6]. As a note of caution, our evidence for sexual/vertical transmission is restricted to our examination of seven matings between infected males and uninfected females, four

females post-mating and sixty-nine progenies, under laboratory conditions. This is supported by evidence of sexual transmission in sheep and dogs [16,17]. This observation is in need of verification from other research groups and in larger-scale field experiments. This has important implications because of our assumption that sexual transmission is frequent enough to produce selection pressure for a parasitic manipulation. The notion of extended phenotype refers to the situations where genotype of an organism manifests its phenotype outside the physical confines of its body [28]. Parasitic behavioral manipulation is an elegant demonstrations of extended phenotypes. In this case, genotype of the parasite induces a phenotype in terms of host behavior. The ability of the Toxoplasma gondii to not only advantageously alter the behavior and physiology of its host, but to also secondarily alter the behavior of uninfected females presents a striking example of the "extended phenotype [28]. We suggest that enhanced attractiveness of infected rats is a parasitic manipulation meant to increase chances of parasite transmission. This suggestion should be carefully tested against the ease of erroneously invoking an adaptationist explanation (see [20] for scholarly discussion of the issue). A behavioral manipulation is too well fitted to its purpose to arrive by chance. In other words, there needs to be a conformity between the observed change and a priori design specification that a engineer might use [20]. For example, both increase and decrease in locomotion of prey can be argued to lead to increased predation; it is not purposive [20]. In this regard, an infection that imparts more attractiveness to male is an a priori design specification; an obvious solution to get more parasites travel with semen. Thus, we posit that the behavioral change reported here constitutes a behavioral manipulation rather than a generalized effect of infection. Nonetheless, this view makes an important assumption that sexual transmission plays a sizeable part in life history of the parasite, an assumption that needs to be tested further in large-scale field experiments.

Exploitation of host sexual selection

Various theories have been proposed to explain the functional significance of mate choice (reviewed in [2931]); based on direct benefits to the female [32], heritable parasite resistance [13], prevention of associative transmission of infections [14] or runaway selection processes [33]. Regardless of selection pressure that maintains mate choice, it remains beneficial for a sexually transmitted parasite to manipulate it. This possibility has not been studied extensively. On related note, male chironomid midges (Paratrichocladius rufiventris) have better success at forming mating pair if infested by a mite (Unionicola ypsilophora) [34]. Mating in midges generally involves females forming a swarm and male flying through it trying to capture a female. It is not clear if female mate choice or intersexual signaling plays any meaningful role in midges. Thus, possibility of parasitic effects on mate choice has remained unaddressed in this system.

Apart from parasitic manipulation of mate choice, there are two kinds of situations where host sexual signaling and parasites could interact. In one type of examples, parasites invade reproductive tissue of host producing parasitic castration [35], aiming to block host from spending valuable metabolic currency on its own reproduction. In a second type of examples, parasites act as illegitimate receivers of sexual signals in order to locate potential hosts [36]. Both of these situations are distinct from behavioral manipulation of mate choice. Female mate choice is an important mediator of reproductive success of male rats. Several semi-naturalistic and laboratory studies have demonstrated that female rats strongly pace the sexual interaction through punctuated display of solicitation behavior (reviewed in [37], also see [3840]). Indeed neuroendocrinological changes essential for initiating pregnancy depend on intermittent nature of coital stimulation in females. These observations suggest that solicitation controlled by females is central to rat sexual behavior. Thus, it is a plausible speculation that a parasitic strategy based on female mate choice will be selected. Role of Toxoplasma gondii infection in rat mating success has been investigated before [41], using a semi-naturalistic setting. It was reported that infection did not alter mating success as defined by number of ejaculations and mounting. Reproductive success of males in this arrangement is a product of both male-male competition and mate choice. One possibility is that females in this arena had low opportunity to pace the interaction, for example by hiding in home-boxes that had an opening small enough to allow only females and not males. As indicated earlier, several studies have established that intermittent solicitation by females is regular feature of reproductive ritual in rats; and it facilitates successful pregnancy. Another possibility is that male-male competition heavily contributed to the mating success in semi-naturalistic setting, overriding influence of female mate choice. This possibility will require further experimentation in form of careful dissociation of both intra-sexual and inter-sexual behaviors. Pending that, it is difficult to ascertain if parasitic manipulation of mate choice will result in significant gain of reproductive fitness of infected males.

Mechanism
What are the proximate mechanisms? It can be safely assumed that pheromonal communication rather than acoustic signaling is involved. This is because presence of soiled bedding itself is sufficient to show difference between infected and control males. Major urinary proteins play important part in sexual signaling of house mouse [42,43]. It is possible that change in similar proteins in rats is involved in manipulation of mate choice. This possibility is supported by the fact that rat urine contains large amount of major urinary proteins. On the other hand, mouse and rat major urinary proteins have evolved separately after divergence from a common ancestor [44]. Presently, there is no unequivocal evidence

that rat major urinary proteins do participate in sexual signaling. Further experiments are needed to delineate pheromonal mechanisms involved in this behavioral change.

Cost of the parasitism

There are varieties of documented cases where parasites manipulate host behavior in order to gain selective advantage [1921]. Such behavioral manipulations usually have detrimental consequences for the host. But the atypical effects of Toxoplasma gondii on mate choice raise an intriguing speculation. Reproductive success in infected males is likely to be elevated, by virtue of their increased attractiveness. Potentially, this elevation could more than offset the decreased fitness of infected males due to their increased likelihood of being predated by cats. While difficult to test in studies of natural populations, these data raise the possibility that parasitic behavioral manipulation will at least blunt the cost of parasitism for infected males. It must be noted that precise determination of reproductive fitness is a difficult task, in view of insufficient quantitative knowledge about rodent reproduction and predation. Thus possibility that infection raises reproductive success of males itself is a speculation at this moment, albeit an interesting and plausible one (also see [41] for a contrary view). Parasitic manipulation of sexual signaling is in contrast to the idea that sexual selection results in discrimination against parasitized males. Under some evolutionary niches, parasites can indeed manipulate host sexual signaling to their own advantage.

Other Sections

Materials and Methods Animals

Wistar rats (48 days old, housed two/cage) were obtained from vivarium of National University of Singapore. The Nanyang Technological University institutional animal care and use committee reviewed and approved all procedures (ARF SBS/NIE- A0106AZ). LongEvans rats (49 days old, three/cage) were obtained from Charles River laboratories (Willmington, MA). The Stanford University administrative panel for laboratory animal care approved procedures pertaining to Long-Evans rats (APLAC#11603). Animals from this source tested serologically negative forToxoplasma gondii.

Parasites and Treatments

We used a Prugniaud strain genetically modified to constitutively express green florescent protein under GRA2 promoter. Parasites were maintained as tachyzoites by passage in human foreskin fibroblast monolayers [24,25]. Infected fibroblasts were syringe-lysed by using a 27-gauge needle to release tachyzoites. Animals were either infected with

tachyzoites (5106, i.p.) or mock-infected with sterile phosphate buffered saline. It is noteworthy here that oral ingestion of oocysts or tissue cysts, and not intraperitoneal entry of tachyzoites, constitutes a more naturalistic route of infection. Yet, we have chosen to use tachyzoites because this produces a more consistent cyst burden in our experience with minimal risk of occupational exposure to laboratory personnel. All behavioral experiments were conducted between 6 to 8 weeks post-infection, a period know to harbor chronic infection. Infected and control animals gained comparable weight during the experiment.

Visualization of Toxoplasma gondii cysts

Smears from epididymis, post-mating vaginal lavage and brain tissue of fetus were examined for the presence of Toxoplasma gondii cysts. Taking advantage of endogenous production of GFP by the transgenic parasite, cysts were observed using anti-GFP antibody (Millipore Cat # AB3080, 1 200 antibody dilution). Interaction between GFP and primary antibody was visualized using tyramide signal amplification coupled with Cy3 dye (red color; Perkin Elmer Cat # NEL744001KT).Cyst wall was stained using dolichos biflorus agglutinin (20 g/ml) coupled with fluorescein (Vector Labs Cat # FL1031, green color). Sperm nuclei were visualized using DAPI (blue color). Stained smears were imaged using a confocal microscope (LSM 710 META, Zeiss).

Polymerase chain reaction and Serological confirmation of infection status

We utilized polymerase chain reaction to selectively amplify a 35-fold repetitive sequence of the gene B1, a gene that is selectively found in Toxoplasma gondii. Primer sets and PCR methods were adopted from previously published protocols (5GGAACTGCATCCGTTCATGAG-3 and 5-TCTTTAAAGCTTCGTGGT C-3) [18]. Infection was also confirmed by serological detection of anti-Toxoplasma antibodies. Toxoplasma was cultured inside human foreskin fibroblasts in 24-well plates. 24 hours post infection; the wells were aspirated, washed with PBS and fixed with 4% PFA. They were incubated with 1 ml of the serum (1 1000) overnight at 4C.If the animal had been infected with Toxoplasma the serum would contain anti Toxoplasma antibodies which would bind to Toxoplasma in culture. The bound antibodies were visualized by treating with anti Rat IgGCy3 (1 200, Millipore) (red color).

Mate choice assay

Naturally cycling females were used in all assays. Wistar: Female mate choice was quantified by comparing time spent by estrus females in two opposing arms of an arena (769 cm each; 15 cm high) during a 20 minute trial.

Opposing arms were scent marked by control and infected males. Males spent 60 minutes before the trial urine marking their respective bisects. Male were restricted to one end of bisect during actual trial, a perforated plastic sheet separating males and test female. Occupancy of female in each of the arm was quantified. Long-Evans: Male stimulus consisted of 100 g of 24 h soiled bedding, pooled from 4 cages each housing three males; placed in opposing corners of a plastic box; 92.435.3 cm, 40.1 cm high. Occupancy of estrus female in bisects containing either control or infected odor was quantified.

Paced mating
Non-competitive paced mating: Experiments were conducted in a glass arena with two chambers connected through a 5 cm semi-circular gate. One chamber housed either a control or an infected male (3045 cm). At the start of trial, estrus female was placed in the arena and observed for duration of 2 hours. Because of difference in body-size, only female could cross the gate connecting two chambers; thus allowing female to pace sexual interactions by withdrawing in chamber not occupied by the male. Intromissions were defined as a reproductive sequence starting with mount and ending in self-genital licking by males. Competitive paced mating: Arena similar to the one described above was used, but with two terminal chambers holding a control and an infected animals during same trial. Female could thus choose to have paced sexual encounter with either male or to withdraw in middle compartment where neither male was present.

Statistical analysis
Mann-Whitney test was performed to determine statistical significance when comparing between experimental groups. Wilcoxon signed-rank test were used when comparing paired differences in within-subject parameters. Wherever appropriate, chi-square test was used to test deviation from ratios predicted by random occurrence. In all statistical tests, exact statistics was used to eliminate asymptotic and approximate statistical assumptions.

Footnotes
Competing Interests: The authors have declared that no competing interests exist. Funding: This research was funded by Nanyang Technological University, Singapore, through its Nanyang Assistant Professorship program; Stanley Medical Research Institute and National Institute of Health, USA (grant AI41014). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Other Sections

References
1. Morran LT, Schmidt OG, Gelarden IA, Parrish RC, 2nd, Lively CM. Running with the Red Queen: hostparasite coevolution selects for biparental sex. Science. 2011;333:216218.[PubMed]

2.

Dawson RD, Bortolotti GR. Carotenoid-dependent coloration of male American kestrels predicts ability to reduce parasitic infections. Naturwissenschaften. 2006;93:597602.[PubMed]

3.

Deaton R. Effects of a parasitic nematode on male mate choice in a livebearing fish with a coercive mating system (western mosquitofish, Gambusia affinis). Behav Processes.2009;80:16. [PubMed]

4.

Ehman KD, Scott ME. Female mice mate preferentially with non-parasitized males.Parasitology. 2002;125:461466. [PubMed]

5.

Houde AE, Torio AJ. Effects of parasitic infection on male color pattern and female choice in guppies. Behavioral Ecology. 1992;3:346351.

6.

Kavaliers M, Colwell DD. Discrimination by female mice between the odours of parasitized and nonparasitized males. Proc Biol Sci. 1995;261:3135. [PubMed]

7.

Kennedy CEJ, Endler JA, Poynton SL, McMinn H. Parasite load predicts mate choice in guppies. Behavioral Ecology and Sociobiology. 1987;21:291295.

8.

Meril J, Sheldon BC, Lindstrm K. Plumage brightness in relation to haematozoan infections in the Greenfinch - bright males are a good bet? Ecoscience. 1999;6:1218.

9.

Milinski M, Bakker TCM. Female sticklebacks use male coloration in mate choice and hence avoid parasitized males. Nature. 1990;344:330333.

10. Mller AP. Effects of a Haematophagous Mite on the Barn Swallow (Hirundo rustica): A Test of the Hamilton and Zuk Hypothesis. Evolution. 1990;44:771784.

11. Mller AP. Parasite load reduces song output in a passerine bird. Animal Behaviour.1991;41:723730.

12. Rantala MJ, Jokinen I, Kortet R, Vainikka A, Suhonen J. Do pheromones reveal male immunocompetence? Proc Biol Sci. 2002;269:16811685. [PMC free article] [PubMed]

13. Hamilton WD, Zuk M. Heritable true fitness and bright birds: a role for parasites? Science.1982;218:384 387. [PubMed]

14. Able DJ. The contagion indicator hypothesis for parasite-mediated sexual selection. Proc Natl Acad Sci U S A. 1996;93:22292233. [PMC free article] [PubMed]

15. Dubey JP. The history of Toxoplasma gondiithe first 100 years. J Eukaryot Microbiol.2008;55:467 475. [PubMed]

16. Arantes TP, Lopes WD, Ferreira RM, Pieroni JS, Pinto VM, et al. Toxoplasma gondii: Evidence for the transmission by semen in dogs. Exp Parasitol. 2009;123:190194. [PubMed]

17. Gutierrez J, O'Donovan J, Williams E, Proctor A, Brady C, et al. Detection and quantification of Toxoplasma gondii in ovine maternal and foetal tissues from experimentally infected pregnant ewes using real-time PCR. Vet Parasitol. 2010;172:815. [PubMed]

18. Burg JL, Grover CM, Pouletty P, Boothroyd JC. Direct and sensitive detection of a pathogenic protozoan, Toxoplasma gondii, by polymerase chain reaction. J Clin Microbiol. 1989;27:17871792. [PMC free article] [PubMed]

19. Moore J. Parasites and the Behavior of Animals: Oxford University Press 2002.

20. Poulin R. "Adaptive" changes in the behaviour of parasitized animals: a critical review. Int J Parasitol. 1995;25:13711383. [PubMed]

21. Poulin R. Suppl. Vol. 116. Parasitology; 1998. Evolution and phylogeny of behavioural manipulation of insect hosts by parasites. pp. S311. [PubMed]

22. Thomas F, Adamo S, Moore J. Parasitic manipulation: where are we and where should we go? Behav Processes. 2005;68:185199. [PubMed]

23. Berdoy M, Webster JP, Macdonald DW. Fatal attraction in rats infected with Toxoplasma gondii. Proc Biol Sci. 2000;267:15911594. [PMC free article] [PubMed]

24. Vyas A, Kim SK, Giacomini N, Boothroyd JC, Sapolsky RM. Behavioral changes induced by Toxoplasma infection of rodents are highly specific to aversion of cat odors. Proc Natl Acad Sci U S A. 2007;104:6442 6447. [PMC free article] [PubMed]

25. Vyas A, Kim SK, Sapolsky RM. The effects of toxoplasma infection on rodent behavior are dependent on dose of the stimulus. Neuroscience. 2007;148:342348. [PMC free article][PubMed]

26. Webster JP. Rats, cats, people and parasites: the impact of latent toxoplasmosis on behaviour. Microbes Infect. 2001;3:10371045. [PubMed]

27. Vyas A, Sapolsky R. Manipulation of host behaviour by Toxoplasma gondii: what is the minimum a proposed proximate mechanism should explain? Folia Parasitol (Praha)2010;57:8894. [PubMed]

28. Dawkins R. Oxford: Oxford University Press; 1982. The extended phenotype.

29. Andersson M. Princeton: Princeton University Press; 1994. Sexual selection.

30. Andersson M, Iwasa Y. Sexual selection. Trends Ecol Evol. 1996;11:5358. [PubMed]

31. Andersson M, Simmons LW. Sexual selection and mate choice. Trends Ecol Evol.2006;21:296 302. [PubMed]

32. Moller AP, Jennions MD. How important are direct fitness benefits of sexual selection?Naturwissenschaften. 2001;88:401415. [PubMed]

33. Fisher RA. The evolution of sexual preference. Eugen Rev. 1915;7:184192. [PMC free article][PubMed]

34. McLachlan A. Parasites promote mating success: the case of a midge and a mite. Anim Behav. 1999;57:11991205. [PubMed]

35. Lafferty KD, Kuris AM. Parasitic castration: the evolution and ecology of body snatchers.Trends Parasitol. 2009;25:564572. [PubMed]

36. Zuk M, Kolluru GR. Exploitation of sexual signals by predators and parasitoids. The Quarterly Review of Biology. 1998;73:415438.

37. Erskine MS. Solicitation behavior in the estrous female rat: a review. Horm Behav.1989;23:473 502. [PubMed]

38. McCIintock MK, Anisko JJ. Group mating among Norway rats. I. Sex differences in the pattern and neuroendocrine consequences of copulation. Anim Behav. 1982;30:398409.

39. McClintock MK, Anisko JJ, Adler NT. Group mating among Norway rats. II. The social dynamics of copulation: Competition, cooperation, and mate choice. Anim Behav.1982;30:410425.

40. Paredes RG, Vazquez B. What do female rats like about sex? Paced mating. Behav Brain Res. 1999;105:117127. [PubMed]

41. Berdoy M, Webster JP, Macdonald DW. Parasite-altered behaviour: is the effect of Toxoplasma gondii on Rattus norvegicus specific? Parasitology. 1995;111(Pt 4):403409.[PubMed]

42. Hurst JL. Female recognition and assessment of males through scent. Behav Brain Res.2009;200:295 303. [PubMed]

43. Roberts SA, Simpson DM, Armstrong SD, Davidson AJ, Robertson DH, et al. Darcin: a male pheromone that stimulates female memory and sexual attraction to an individual male's odour. BMC Biol. 2010;8:75. [PMC free article] [PubMed]

44. Logan DW, Marton TF, Stowers L. Species specificity in major urinary proteins by parallel evolution. PLoS One. 2008;3:e3280. [PMC free article] [PubMed]

PLoS One. 2011; 6(9): e23866. PMCID: PMC3177840 Published online 2011 September 21. doi: 10.1371/journal.pone.0023866 Copyright This is an open-access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication.

The Neurotropic Parasite Toxoplasma Gondii Increases Dopamine Metabolism

Emese Prandovszky,1 Elizabeth Gaskell,1 Heather Martin,1 J. P. Dubey,2Joanne P. Webster,3 and Glenn A. McConkey1*
1

Institute of Integrative and Comparative Biology, Faculty of Biological Sciences, University of Leeds, Leeds, United Kingdom 2 Animal Parasitic Diseases Laboratory, USDA, ARS, ANRI, BARC-East, Beltsville, Maryland, United States of America 3 Department of Infectious Disease Epidemiology, Faculty of Medicine, Imperial College, London, United Kingdom Paulo Lee Ho, Editor Instituto Butantan, Brazil * E-mail: G.A.McConkey@Leeds.ac.uk Conceived and designed the experiments: GAM EG EP. Performed the experiments: EG EP HM JPD. Analyzed the data: EP GAM. Contributed reagents/materials/analysis tools: GAM JPD. Wrote the paper: GAM EP JPW JPD. Received May 17, 2011; Accepted July 26, 2011.

Other Sections

Abstract
The highly prevalent parasite Toxoplasma gondii manipulates its host's behavior. In infected rodents, the behavioral changes increase the likelihood that the parasite will be transmitted back to its definitive cat host, an essential step in completion of the parasite's life cycle. The mechanism(s) responsible for behavioral changes in the host is unknown but two lines of published evidence suggest that the parasite alters neurotransmitter signal transduction: the disruption of the parasite-induced behavioral changes with medications used to treat psychiatric disease (specifically dopamine antagonists) and identification of a tyrosine hydroxylase encoded in the parasite genome. In this study, infection of mammalian dopaminergic cells with T. gondiienhanced the levels of K+-induced release of dopamine several-fold, with a direct correlation between the number of infected cells and the quantity of dopamine released. Immunostaining brain sections of infected mice with dopamine antibody showed intense staining of encysted parasites. Based on these analyses, T. gondiiorchestrates a significant increase in dopamine metabolism in neural cells. Tyrosine hydroxylase, the rate-limiting enzyme for dopamine synthesis, was also found in intracellular tissue cysts in brain tissue with antibodies specific for the parasite-encoded tyrosine hydroxylase. These observations provide a mechanism for parasite-induced behavioral changes. The observed effects on dopamine metabolism could also be relevant in interpreting reports of psychobehavioral changes in toxoplasmosis-infected humans.

Other Sections

Introduction

A complex range of interactions exist between a pathogen with its host, which may include manipulation of the host for the pathogen's own advantage. There are several examples of viruses, such as rabies virus [1], and parasites, includingAcanthocephala spp. [2] and Toxoplasma gondii [3], that influence host behavior to increase their transmission efficiency. For years, scientists have been intrigued by the association between T. gondii infection and altered aversive behavior. The underlying mechanism(s) responsible for this behavior change are presently unknown. The aim of our study was to identify a possible explanation for this phenomenon. T. gondii is a common, global protozoan parasite, which requires both a definitive host and an intermediate host to complete its life cycle. Although felines are the only definitive host of T. gondii, any warm-blooded animal, including humans, can be infected [4]. It is estimated that one quarter of the population (over 12 years of age) in the United States is positive for T. gondii infection (Center for Disease Control, USA, 2008). Prevalence in some areas can be as high as 95% in older populations. Latent, chronic infection, which is characterized by parasite encystment in the host muscle and brain cells (particularly neurons and glial cells), persists following the resolution of acute infection and continues with seropositivity throughout the host's lifetime [4]. Due to its high prevalence in the human population, it is critical to better understand the effects of T. gondii infection in the brain. During the chronic stage of infection, infected rodents, which are a key intermediate host for T. gondii, exhibit a distinct repertoire of specific behavioral changes, including a loss of aversion to cat odors [3], [5]. Infected rodents are, conversely, attracted to these odors, and this may be responsible for increased predation and for an increase in successful transmission of the parasite to the feline host; as cats are the only animal that can shed the environmentally-resistant stage of the parasite known as oocysts. This behavior change in infected rodents during the chronic stage of infection appears highly specific to feline odor, as a similar change is not evoked by other predators and has no effect on conditioned fear and anxiety [6], [7]. The underlying mechanism(s) responsible for this behavior change still remain unclear, however, it has been revealed that anti-psychotic (haloperidol) and moodstabilizing medication (valproic acid) can prevent the development of these behavior changes[7] in addition the dopamine uptake inhibitor GBR12909 modifies behavioral responses associated with latent toxoplasmosis in infected rodents [8]. Furthermore, we have recently identified an enzyme with tyrosine hydroxylase activity encoded in the Toxoplasma genome whose expression is induced during differentiation to tissue cyst stages [9]. Several studies have suggested that T. gondii infection in humans can have serious neurological effects [10]. Associations have been identified between T. gondii seroprevalence and schizophrenia [11][13]. The schizophrenia-associated risk factors of T. gondii infection have been found to be greater than the risk factors associated with an individual's genes and with other environmental factors [13], [14]. Schizophrenia

affects approximately 1% of the adult population and in most cases is a lifelong disease with exacerbations. Although schizophrenia is a multifactorial disease, pharmacological and genetic evidence suggest that dysregulation of dopamine metabolism is involved in schizophrenia [15], [16]. Thus, it is crucial to examine whether dopamine metabolism is affected by T. gondiiinfection, particularly based on evidence of a tyrosine hydroxylase encoded by T. gondii. To address these questions, dopamine metabolism was monitored in vivo in the brains of chronically infected mammals and monitored in vitro during infection of neural cells.

Other Sections

Methods Ethics statement

All animal work was performed according to national and international guidelines following approved animal procedures by the Beltsville Area Animal Care Committee, United States Department of Agriculture (Protocol no. 09-010Toxoplasmosis in mice; approved June 4, 2009). This protocol is reviewed annually, and any amendments are approved separately.

Growth of parasites and host cells

T. gondii strains were maintained in human foreskin fibroblasts (HFFs) as previously described [9]. PC-12 cells obtained from ECACC (Salisbury) were maintained as described by the supplier.

Mouse strains
Female Swiss Webster mice infected with T. gondii VEG strain were used for histology.

Immunofluorescence assay of brain sections

Immunofluorescence against multiple targets was performed on paraformaldehyde-fixed, paraffin-embedded mouse brain sections. Female Swiss Webster mice were infected with T. gondii VEG strain oocysts 68 weeks prior to processing. Tissues were collected, formalin-fixed and paraffin-embedded using standard protocols and following approved guidelines. Slides were deparaffinized and rehydrated with an alcohol descending row, which was then followed by epitope retrieval in 10 mM sodium citrate buffer (pH 6.0) overnight at 60C following sectioning. Slides were blocked with 2% normal goat sera for 1 h at room temperature. TRITC-conjugated lectin from Dolichos biflorus (Cat # L9658, Sigma, St. Louis) was introduced to the slides for 4 h at room temperature, diluted 1 200 in primary staining solution (1% BSA, 0.1% cold fish skin gelatine, 0.5% Triton X-100 in 0.1 M

PBS pH 7.2). Next, samples were washed (310 min) in wash buffer (TBS pH 8.4 with 0.1% Triton X-100 and 1% fish skin gelatin) and blocked using a biotin-streptavidin blocking kit (Cat # SP-2002, Vector Labs, Peterborough) according to the manufacturer's protocol. Samples were incubated with primary antibody (raised in rabbit) against dopamine (Cat # ab8888, Abcam, Cambridge, MA) (diluted 1 200) or tyrosine hydroxylase (Cat # ab112, Abcam) (diluted 1 500) overnight at 4C. Samples were rinsed with wash buffer and incubated for 1 h with biotinylated anti-rabbit IgG secondary antibody (Cat # B-1000, Vector Labs) diluted 1 500 in secondary antibody solution (0.05% Tween in 0.1 M PBS pH 7.2). Sections were rinsed and incubated with FITC-conjugated streptavidin (Cat # SA-5001, Vector Labs) diluted 1 100 according to the manufacturer's guidelines in secondary antibody buffer at room temperature. Finally, slides were rinsed in wash buffer containing DAPI and double-distilled water prior to mounting in Fluoromount G (Southern Biotech, Birmingham). All incubation and blocking steps were carried out in a wet chamber. All slides were kept at 4C in the dark before imaging using a Zeiss LSM510 META laser scanning inverted AxioVert 200M confocal microscope with DIC optics. 3D reconstructions of serial sections were generated with the same equipment using the LSM imaging software for the 3D deconvolution. To assess the specificity of dopamine staining, sections were incubated either without primary antibody or with primary anti-dopamine antibody in the presence of freshly prepared dopamine or serotonin for 30 min prior to and for overnight following addition to the sections. A T. gondii tyrosine hydroxylase antibody custom antibody (Genscript, Piscataway) was developed to assess the parasite enzyme in animals. The affinity purified antibody is directed against a unique sequence (CIRSSPDPLDLRKMT) in the amino terminal domain that is not found in mammalian tyrosine hydroxylase and has no significant similarity to any protein in the predicted mammalian proteome or other proteins of the T. gondii proteome. The specificity of the antibody for T. gondiityrosine hydroxylase was confirmed by Western analysis. Total protein from half mouse brains was isolated in 20 volumes (wt/vol)lysis buffer (20 mM Tris-HCl pH 8; 137 mM NaCl; 10% glycerol; 1% Triton X; 2 mM EDTA and protease inhibitors (cOmplete Mini EDTA-free cocktail, Roche)) and quantified using Bradford reagent (Sigma) as per manufacturer's instructions. Expression and purification of T. gondiityrosine hydroxylase was as previously described [9]. SDS-PAGE was following standard protocols with 220 g protein separated on a 12% sodium dodecylsulphatepolyacrylamide gel. The proteins were transferred to nitrocellulose membrane, blocked with 5% non-fat dried milk in PBS containing 0.05% Tween-20 (vol/vol) for 1 hour. Incubation with the custom antibody (1 2500) 4C overnight was followed by washing in PBS-Tween (0.05%) and incubation with an anti-rabbit (1 5000) conjugated horseradish peroxidase antibody (Sigma) at room temperature for 1 hour. Blots were then washed as above and developed using Supersignal West Pico Chemiluminescent kit (Pierce, Rockford, IL). Bands

were visualised with an X-Omat film system. The membrane was stripped and re-probed with mouse anti--actin (1 25,000; Sigma) overnight at 4C followed by anti-mouse (1 10,000) conjugated horseradish peroxidase antibody (AutogenBioclear, Wiltshire, UK) at room temperature for 1 hour and subsequent visualisation. The anti-T. gondii tyrosine hydroxylase antibody was used for immunofluorescence (diluted 1 500) following a similar protocol as described above. Immunofluorescence of tyrosine hydroxylase in cultured parasites was performed with paraformaldehyde-fixed cell cultures. Cultures of T. gondii stably expressing RFPconjugated GRASP protein (kindly donated by Manami Nishi from David Roos laboratory, University of Philadelphia, USA) in human foreskin fibroblasts grown on polylysine-coated coverslips were alkaline induced for differentiation as published and differentiation monitored by counting the number of parasites in the vacuoles in the normal and pH shifted cultures. These conditions yielded bradyzoite forms as shown by RT-qPCR with the bradyzoite markers BAG1 and SAG4 [9]. After five days, coverslips were paraformaldehyde-fixed and probed with tyrosine hydroxylase antibody (Cat # ab112, Abcam) (diluted 1 500) with visualisation as described above.

Immunohistochemical assay of brain sections

For immunohistochemical assays, sections were treated as described above, except for the following steps: for washing and dilution buffers, 0.1 M PBS supplemented with 0.1% Tween was used. After antigen retrieval, slides were incubated in 0.3% H2O2 (in 0.1 M PBS) to quench endogen peroxidases. Following secondary antibody treatment, 5 g/ml HRPconjugated streptavidin (Cat # SA-5004, Vector Labs) was applied, and next, the peroxidase substrate kit (Vector Labs, ImmPACTDAB, Cat # SK-4105) was used according to the manufacture's protocol. Prior to mounting, sections were stained with haematoxylin to visualize cell nuclei. Imaging of slides was performed using a Zeiss Axioplan microscope equipped with DIC optics. Photomicrographs were collected with a Photometrics CoolSNAP camera and Improvision Openlab software.

Glyoxylic acid staining

A cytochemical method was used to assess the dopamine staining of tissue cyst-containing neural cells in infected mice brain. Glyoxylic acid reacts with catecholamines in a gaseous reaction to form fluorescent products. Dopamine reacts with glyoxylic acid to form a product that specifically emits at 478480 nm[17].

Dopamine accumulation and release from dopaminergic cells

The dopaminergic cell line PC12 (ECACC) was infected with Prugniard strain of T. gondii tachyzoites that had been alkaline shocked to induce bradyzoite differentiation. Liberated tachyzoites were incubated in RPMI media at pH 8 with 1% FCS at 37C and

ambient CO2 for 1618 h, then rinsed with DMEM and returned to standard PC12 cell culturing conditions. PC12 cultures were infected 2.51057.5105parasites and cultured for five days prior to assay. The cultures infected with higher numbers of parasites had parasitemia of 4050%. Prior to assay, samples were normalized to equivalent numbers of cells (2.5106) per assay. One set of cultures was harvested by centrifugation and lysed by sonication in 0.1 M perchlorate for total dopamine measurement by HPLC with electrochemical detection. A parallel set of cultures were equilibrated with wash buffer with low KCl containing buffer (140 mM NaCl, 4.7 mM KCl, 1.2 mM MgCl2, 2.5 mM CaCl2, 11 mM dextrose, 10 mM HEPES, pH 7.4) for 30 min followed by incubation with two volumes high KCl containing buffer (40 mM NaCl, 100 mM KCl, 1.2 mM MgCl2, 2.5 mM CaCl2, 11 mM dextrose, 10 mM HEPES, pH 7.4) for 2 min to induce dopamine release as previously described [18]. During the assay, samples were taken from the media, washing buffer, and high KCl containing buffer and immediately supplemented with 0.3 volumes 0.1 M perchlorate. Three independent experiments were performed with a representative experiment shown. Following centrifugation, cell homogenates and media were assayed by HPLC-ED. Reverse phase chromatography, combined with electrochemical detection, was performed with a Dionex HPLC system consisting of a P580 Pump (Dionex) and Ultimate 3000 Autosampler Column Compartment with a C18 Acclaim 120 column (5 m, 4.6150 mm) and an ESA Coulochem III cell, equipped with a glassy carbon electrode used at 700 mV versus Ag/AgCl reference electrode for detection of monoamines. The mobile phase consisted of degassed 57 mM anhydrous citric acid (Fisher Scientific, Loughborough), 43 mM sodium acetate (Dionex, Sunnyvale) buffer containing 0.1 mM EDTA (Sigma Aldrich), 1 mM sodium octanesulphonate monohydrate, and 10% methanol. The pH was adjusted to 4. The mobile phase was delivered at a flow rate of 1.5 ml/min, and the column temperature was set at 40C. Applied standards (dopamine, L-DOPA) were dissolved in 0.1 M perchlorate for chromatography. The concentration of compounds was determined using Chromeleon software.

Other Sections

Results Dopamine metabolism in infected neural cells in brain tissue

A previous study found that the global content of dopamine in the brains of mice chronically infected with T. gondii was increased by 14% (114% of uninfected (P<0.01)), whereas other neurotransmitters were unchanged [19]. The localized effects of T. gondii infection on dopamine metabolism in tissue cysts have not been examined. T. gondii forms intracellular tissue cysts in neurons with each tissue cyst containing hundreds of bradyzoites (slowly

dividing stage) that may remain in situthrough the host's lifetime [4]. Formaldehyde-fixed brain sections from mice chronically infected with T. gondii were probed with dopamine antibody (Abcam). Dopamine antibody staining was readily apparent in infected cells (Fig. 1). Surprisingly, the localization was primarily within the T. gondii tissue cysts containing the parasites visualized as intensely stained cysts (Figs. 1, ,2),2), rather than the host neural cell. The dopamine antibody staining in tissue cysts was punctate. Image rotation illuminated staining throughout the tissue cyst with most concentrated staining near the periphery. The antibody was raised against dopamine glutaraldehyde conjugated to bovine serum albumin (BSA). The antibody also labelled neurons in the amygdala and hippocampus in uninfected and infected mice (data not shown); areas with a high concentration of dopaminergic neurons. Intracellular tissue cysts were identified based on morphology (for immunohistochemistry) and by labeling the periphery of the tissue cysts with fluorescently-tagged lectin (for immunofluorescence) [20]. DAPI counterstaining of nuclei visualized individual parasites in the tissue cysts, highlighting the hundreds of bradyzoites within each tissue cyst.
Figure 1 Dopamine in tissue cysts of T. gondii in brain tissue sections.

Figure 2 Specificity of dopamine staining T. gondii tissue cysts.

The presence of dopamine in tissue cyst-containing neural cells was confirmed by cytochemical staining and competition assays. Glyoxylic acid staining, a classic method for detection of dopamine-containing cells by chemical reaction of glyoxylic acid with dopamine to produce a fluorescent product [17], was applied. Interestingly, the tissue cyst infected cells fluoresced blue and white, with the entire cell body of the infected cell displaying fluorescent staining (Fig. 2A). Staining of the encysted parasites within cells and neural cell nuclei are black due to the presence of parasite and host nuclear chromatin. The lack of cytosolic staining in the immunofluorescent images with dopamine antibody are likely to be

due to saturation of the image with the very intense cyst staining. The specificity of the dopamine antibody was confirmed by competition assays. Primary dopamine antibody staining of tissue sections was performed in the presence of exogenous dopamine followed by secondary staining with fluorescein labelled antibody. This eliminated staining as visualized by loss of fluorescence (Fig. 2B). In contrast, addition of exogenous serotonin (another catecholamine neurotransmitter) did not disrupt staining with the dopamine antibody. This verifies that the dopamine antibody is detecting dopamine. It remains possible that the dopamine antibody is also detecting the metabolic precursor to dopamine, L-DOPA, although manufacturer (Abcam) tests show a >400-fold higher affinity for dopamine compared to L-DOPA using conjugates to BSA. Competition assays exhibited some decrease in staining with exogenous L-DOPA although this was not quantifiable (data not shown).

T. gondii infected cells release high amounts of dopamine

To assess whether the dopamine detected in T. gondii tissue cysts could affect neurotransmission, the effect of T. gondii infection on dopamine release from dopaminergic neural cells in vitro was determined. PC12 cells were utilized as this cell line is the most commonly used in vitro model of dopaminergic neurons. Dopaminergic PC12 cells were infected with T. gondii parasites incubated under conditions that induce differentiation, and dopamine content and release were monitored by HPLC-ED. Conditions were used (as described in the Materials andMethods) for stage conversion of tachyzoites (the rapidly dividing stage of T. gondii) into the tissue cyst stages (ie. bradyzoites) with alkaline pH and decreased CO2content as described by others [4]. The total dopamine in infected PC12 cultures was measured to determine whether infection increases the amount of dopamine synthesized in dopaminergic cells. Cultures were infected with different numbers of alkaline-induced T. gondii and total dopamine was quantitated by HPLC-ED following washes with low KCl buffer. We found that infected cultures accumulated significantly greater levels of dopamine and the increase correlated with infection rate (Fig. 3). Infection led to greater than three-fold increase in total dopamine content compared to mock-treated, uninfected cells.
Figure 3 Elevated dopamine from T. gondii infected dopaminergic cells.

Dopamine release assays were performed with cultures of T. gondii-infected PC12 cells to assess effects of infection on dopamine signalling. The cultures infected with different numbers of alkaline-induced T. gondii were induced to release dopamine with potassium as K+ causes release of dopamine in vesicles following methods in other studies [18]. As a result of infection, dopamine release increased in infected cultures in a dose-dependent

manner with the number of parasites in the culture correlating with the amount of dopamine released (Fig. 3). Dopamine release in infected cells was up to 350% greater compared to dopamine release in uninfected cells. Dopamine release was specific for high KCl induction since wash buffer (Fig. 3B) and media alone (data not shown) did not induce the release of detectable amounts of dopamine in infected or uninfected cultures. The low KCl wash ensures that the dopamine released is induced by potassium and not due to dopamine released by cell lysis of infected cells. The dopamine release reported here is the minimum amount increased by T. gondii infection as less than or equal to half of the cells in the cultures were infected. Normalizing for the infection rate results in a seven-fold increase in dopamine release in infected cells relative to uninfected PC12 cells. Taken together, an increase in dopamine content and an increase in dopamine release were observed in neural cells as a direct response to T. gondii infection. The enhanced dopamine release observed in infected cells in this study is likely to be an underestimate of the effect on dopamine release in vivo, cultured parasites contain few bradyzoites per vacuole compared to brain tissue cysts that contain hundreds of bradyzoites.

Tyrosine hydroxylase is expressed in bradyzoites

Tyrosine hydroxylase is the rate-limiting enzyme in dopamine biosynthesis. Tyrosine hydroxylase localization in the brain sections of mice chronically infected with T. gondii was determined to examine the expression of this crucial enzyme in infected neural cells. Significant levels of tyrosine hydroxylase were localized within T. gondiitissue cysts in the brain sections of infected mice (Fig. 4). As expected, tyrosine hydroxylase was also found in the cytosol of neurons in the expected areas of the brain in both infected and uninfected mice (data not shown). Staining was not apparent in control sections that were treated with only secondary antibody. It is intriguing that both tyrosine hydroxylase and dopamine staining were localized in the tissue cysts of infected mouse brains, displaying similar staining patterns (Figs. 1, ,4).4). Thus, the rate limiting enzyme for dopamine synthesis and the product itself were both found in T. gondii tissue cysts in the brain.
Figure 4 Dopamine enzyme tyrosine hydroxylase in intracellular T. gondii.

It is possible that the tyrosine hydroxylase expression observed within the tissue cyst could be either the T. gondii-encoded tyrosine hydroxylase or neuronal tyrosine hydroxylase that has been imported from the host. T. gondii has complex interactions with its host cell and

co-ops several host proteins (e.g. calpains), and hence could potentially import neuronal tyrosine hydroxylase into the tissue cyst [21]. Alternatively, T. gondii could provide an enzyme with tyrosine hydroxylase activity. We previously described a T. gondii encoded tyrosine hydroxylase that could be expressed in the brain tissue cysts [9]. T. gondii has two copies of the tyrosine hydroxylase gene encoding nearly identical proteins (97.5%) with one gene induced in bradyzoite-stage parasites. The parasite tyrosine hydroxylase has a high degree of homology (53% identity) with mammalian tyrosine hydroxylases. Unique for tyrosine hydroxylases, the parasite orthologue enzyme contains a putative signal sequence that could permit the enzyme to be trafficked to an organelle or secreted by T. gondii. Additionally, it was observed that the commercial tyrosine hydroxylase antibody used in these studies recognizes the T. gondii encoded orthologue, as well as the mammalian tyrosine hydroxylases (unpublished observations). Indeed, in vitrocultivated parasites under alkaline conditions that induce formation of bradyzoites bind commercial tyrosine hydroxylase antibody (Fig. 4B). The tyrosine hydroxylase antibody stains the parasitophorous vacuole and also stains the periphery of parasites. To specifically identify parasite-encoded tyrosine hydroxylase within brain tissue, a custom antibody for T. gondii tyrosine hydroxylase (TgTH) was developed. The target sequence of this custom antibody is located in the amino-terminal domain of TgTH, which is unique and divergent from mammalian tyrosine hydroxylases. T. gondiityrosine hydroxylase was localized within tissue cysts in neural cells in chronically-infected mouse brains (Fig. 5A). Staining was only detectable in tissue cysts in infected neurons. Western analysis was performed to validate the specificity of the TgTH antibody. The antibody recognized recombinant TgTH but did not bind mouse brain proteins, confirming the specificity of this antibody for T. gondii tyrosine hydroxylases (Fig. 5B). Hence the intense dopamine antibody staining and TgTH are both found in T. gondii brain tissue cysts.
Figure 5 Expression of a parasite-encoded tyrosine hydroxylase in brain tissue cysts.

Other Sections

Discussion
Changes in behavior of the intermediate host that could lead to increased transmission of a parasite to its definitive host are likely to be positively selected as these changes would provide a significant benefit in completion of the parasite's life cycle. T. gondii induces

behavioral alterations in infected rodents that would facilitate the transmission of the parasite to its definitive feline host, however, the mechanism responsible for these changes remains unclear. Our study provides a mechanism for these changes. Previous studies showing that anti-dopaminergic drugs can prevent the development of the behavior changes in rodents suggest that dopamine regulation altered by T. gondii infection of mammals [7]. The altered behavior may be a direct effect or an indirect effect of T. gondii infection. In this study, significant levels of dopamine was detected by immunohistochemistry in T. gondii tissue cysts in the brain (Fig. 1), as well as, increased dopamine release from dopaminergic cells infected with T. gondii (Fig. 3). Based on these novel findings, this is the first study to suggest that a parasite can directly alter dopamine signalling to mediate host behavior changes. These results provide a potential mechanism for T. gondii-induced host behavioural changes. In our study, localizing the changes in dopamine metabolism during infection was crucial, as the location of dopamine metabolic changes in the brain is likely to be a critical factor for its effect on host behavior. Encysted T. gondii have been observed in functional neurons with intact synapses [22], [23]. Tissue cysts have been detected throughout the brain, although higher percentages of cysts were reported in the amygdala and nucleus accumbens [5], [24]. These limbic brain regions are well known to contain dopamine that plays important functions in the control of movements (basal ganglia), reward to stimuli, pleasure, dependency (nucleus accumbens and hippocampus), motivation and cognition, and species and stimuli specific fear (amygdala). Altered dopamine levels induced by T. gondii in tissue cysts in these regions of the brain could have significant harmful consequences on a variety of brain functions, possibly leading to an array of behavioral changes and possible neurological malfunctions. The observed intense dopamine staining within the T. gondii tissue cysts in brains was unexpected. Dopamine in neurons is synthesized in the cytosol, packaged into vesicles, and transported along axons [25]. Thus, dopamine staining in neurons is primarily detected within vesicles. Indeed, cytosolic dopamine can induce cell apoptosis if it is not properly packaged into vesicles [26]. Packaged dopamine in neurons is rapidly transported away from the cell body to the axon terminal. In our brain sections, any dopamine released from the cyst into the cell body of the neuron would be packaged and transported by the efficient dopaminergic vesicle transport along axons. This may explain the apparent lower level of dopamine in the host cell body compared to the tissue cyst (Figs. 1, ,2).2). Alternatively, the observed staining by the dopamine antibody could be due to detection of L-DOPA within the T. gondiitissue cyst that escapes the cyst and is metabolised into dopamine in the host cytosol. This interpretation is coherent with the observed cytosolic staining of infected

neurons using glycoxylic acid that yields a specific product with dopamine (Fig. 2). The parasite provides tyrosine hydroxylase, the rate-limiting enzyme in dopamine synthesis but the source of DOPA decarboxylase required for conversion of L-DOPA to dopamine needs further investigation. DOPA decarboxylase present in the cytosol of dopaminergic neurons could provide this enzyme. A hypothetical nutrient pore expressed in the parasitophorous vacuole membrane of T. gondii tachyzoites that permits the passage of metabolites (<1300 Da) from the host cell cytosol into the parasitophorous vacuole could allow passage of small compounds from the vacuole into the host cytosol [27]. If the pore is expressed in bradyzoites then it could provide a means for dopaminergic metabolites (L-DOPA, dopamine) to exit the vacuole and enter the host cytosol where L-DOPA would be converted to dopamine by cytosolic DOPA decarboxylase and dopamine would be packaged into secretory vesicles. The generation of T. gondii mutants that will provide a conclusive dissection of the role of the parasite's tyrosine hydroxylase in the dopamine synthesis and release are in progress, but with either site of dopa decarboxylase action, the increased dopamine metabolism has important implications on the host neurochemistry. In addition to dopamine, neurotransmitters such as serotonin and glutamate need to be considered in T. gondii-induced behavioral changes. Prior studies have proposed that the host immune response to T. gondii infection may lead to altered neurotransmitter levels [28]. Immunocompetent hosts control chronic T. gondiiinfection with a T-lymphocyte driven defense [29]. Infection of mice with T. gondiielicits a dominant Th1 response involving interferon-gamma (IFN-), interleukin-12 (IL-12), IL-18, and tumor necrosis factor alpha (TNF-). TNF- induction has a serious impact on T. gondii induced pathology at early stages of infection. Th2-associated cytokines, such as IL-4 and IL-10, appear relatively late after infection and may limit immune pathology. To resolve acute infection, IFN- induces indoleamine 2,3-dioxygenase (IDO) release, resulting in tryptophan degradation and kynurenic acid accumulation [30]. Tryptophan depletion is thought to be responsible for suppression of the growth of the acute stage tachyzoites. Changes in serotonin levels were not observed in mice with T. gondii chronic infections although there may be localized undetected changes [19]. Kynurenic acid accumulation in the CNS could potentially alter dopamine metabolism due to its NMDA antagonistic properties [12]. Thus, the host immune response to T. gondii infection could contribute to alterations in neurotransmitter levels that could affect behaviour in conjunction with the increased dopamine mediated by the parasite. Further studies are essential to investigate these possibilities. Behavioral changes associated with T. gondii infection may contribute to serious neurological disorders in humans. Several studies have observed an association between T. gondii seroprevalence with schizophrenia [10], [13]. Since T. gondiiinfection has been found to last throughout the lifetime of the host, seroprevalence is likely to reflect

chronic infection [4]. Dopamine dysregulation is proposed to play a central role in schizophrenia, potentially in combination with glutamate metabolism. How dopamine dysregulation plays a role in schizophrenia, however, is still unknown. The principal antipsychotic drug that has been used to treat schizophrenia, dopamine antagonist haloperidol, can also block the development of behavior changes in T. gondii infected rodents. It is possible that the increased dopamine accumulation and release observed during T. gondii infection may contribute to T. gondii associated schizophrenia. Dopamine metabolite concentrations have been inversely correlated with gray matter volume in schizophrenia patients, and recent MRI evidence found that the majority of volume reduction is in those patients seropositive for T. gondii, suggesting that T. gondii infection leads to an increase in dopamine metabolite concentrations [31], [32]. It would be of interest to analyze the ability of other pathogens associated with schizophrenia, and other neurological disorders, to directly alter dopamine metabolism to see if other pathogens have this ability or if this phenomena is unique to T. gondii. Malfunctions of dopamine metabolism have a serious impact on human behavior. Dopamine dysfunction has been associated with a variety of neurological disorders including schizophrenia, attention deficit hyperactivity disorder, tic disorders, Tourette's syndrome, and dyskinesias. The novel findings of this study, that demonstrate T. gondii's ability to directly alter dopamine levels will not only help to better understand the relationship between schizophrenia and T. gondiiseroprevalence, but these findings may be critical for understanding the mechanism(s) involved in a variety of pathogen-associated neurological disorders[10], [13]. Thus, it is crucial to determine if other pathogens associated with neurological disorders also have the ability to directly alter dopamine levels. It is also critical to determine the possible contributions of T. gondii infection to other dopamine-related diseases [33], [34].

Acknowledgments
We would like to acknowledge Dr. Oliver Kwok and Gareth Howell for technical assistance and Prof. Elwyn Isaac and Drs. Sophie Bamps and Christopher D. O'Donnell for helpful discussions and comments on the manuscript.

Footnotes
Competing Interests: The authors have declared that no competing interests exist. Funding: This project was funded by the Stanley Medical Research Institute (to JPW, GAM) and the USDA CRIS 126532000-076 (to JPD). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.

Other Sections

References
1. Lagrue C, Poulin R. Manipulative parasites in the world of veterinary science: implications for epidemiology and pathology. Veterinary journal. 2010;184:913. 2. Lefevre T, Adamo SA, Biron DG, Misse D, Hughes D, et al. Invasion of the body snatchers: the diversity and evolution of manipulative strategies in host-parasite interactions. Adv Parasitol.2009;68:4583. [PubMed] 3. Berdoy M, Webster JP, Macdonald DW. Fatal attraction in rats infected with Toxoplasma gondii.Proc Biol Sci. 2000;267:15911594. [PMC free article] [PubMed] 4. Dubey JP. Toxoplasmosis of animals and humans. Boca Raton,, Florida: CRC Press; 2010. 5. Vyas A, Kim SK, Giacomini N, Boothroyd JC, Sapolsky RM. Behavioral changes induced byToxoplasma infection of rodents are highly specific to aversion of cat odors. Proc Natl Acad Sci U S A.2007;104:64426447. [PMC free article] [PubMed] 6. Lamberton PH, Donnelly CA, Webster JP. Specificity of the Toxoplasma gondii-altered behaviour to definitive versus non-definitive host predation risk. Parasitology. 2008;135:11431150. [PubMed] 7. Webster JP, Lamberton PH, Donnelly CA, Torrey EF. Parasites as causative agents of human affective disorders? The impact of anti-psychotic, mood-stabilizer and anti-parasite medication onToxoplasma gondii's ability to alter host behaviour. Proc Biol Sci. 2006;273:10231030.[PMC free article] [PubMed] 8. Skallova A, Kodym P, Frynta D, Flegr J. The role of dopamine in Toxoplasma-induced behavioural alterations in mice: an ethological and ethopharmacological study. Parasitology. 2006;133:525535.[PubMed] 9. Gaskell EA, Smith JE, Pinney JW, Westhead DR, McConkey GA. A unique dual activity amino acid hydroxylase in Toxoplasma gondii. PLoS One. 2009;4:e4801. [PMC free article] [PubMed] 10. Brown AS, Derkits EJ. Prenatal infection and schizophrenia: a review of epidemiologic and translational studies. Am J Psychiatry. 2010;167:261280. [PubMed] 11. Brown AS, Schaefer CA, Quesenberry CP, Jr, Liu L, Babulas VP, et al. Maternal exposure to toxoplasmosis and risk of schizophrenia in adult offspring. The American journal of psychiatry.2005;162:767773. [PubMed] 12. Mortensen PB, Norgaard-Pedersen B, Waltoft BL, Sorensen TL, Hougaard D, et al. Early infections of Toxoplasma gondii and the later development of schizophrenia. Schizophrenia bulletin.2007;33:741744. [PMC free article] [PubMed] 13. Torrey EF, Bartko JJ, Lun ZR, Yolken RH. Antibodies to Toxoplasma gondii in patients with schizophrenia: a meta-analysis. Schizophr Bull. 2007;33:729736. [PMC free article] [PubMed] 14. Purcell SM, Wray NR, Stone JL, Visscher PM, O'Donovan MC, et al. Common polygenic variation contributes to risk of schizophrenia and bipolar disorder. Nature. 2009;460:748752. [PubMed] 15. Howes OD, Kapur S. The dopamine hypothesis of schizophrenia: version IIIthe final common pathway. Schizophrenia bulletin. 2009;35:549562. [PMC free article] [PubMed] 16. Seeman P, Lee T. Antipsychotic drugs: direct correlation between clinical potency and presynaptic action on dopamine neurons. Science. 1975;188:12171219. [PubMed] 17. Lent CM. Fluorescent properties of monoamine neurons following glyoxylic acid treatment of intact leech ganglia. Histochemistry. 1982;75:7789. [PubMed]

18. Yamboliev IA, Smyth LM, Durnin L, Dai Y, Mutafova-Yambolieva VN. Storage and secretion of beta-NAD, ATP and dopamine in NGF-differentiated rat pheochromocytoma PC12 cells. The European journal of neuroscience. 2009;30:756768. [PMC free article] [PubMed] 19. Stibbs HH. Changes in brain concentrations of catecholamines and indoleamines inToxoplasma gondii infected mice. Ann Trop Med Parasitol. 1985;79:153157. [PubMed] 20. Coppin A, Dzierszinski F, Legrand S, Mortuaire M, Ferguson D, et al. Developmentally regulated biosynthesis of carbohydrate and storage polysaccharide during differentiation and tissue cyst formation in Toxoplasma gondii. Biochimie. 2003;85:353361. [PubMed] 21. Chandramohanadas R, Davis PH, Beiting DP, Harbut MB, Darling C, et al. Apicomplexan parasites co-opt host calpains to facilitate their escape from infected cells. Science. 2009;324:794797. [PubMed] 22. Melzer TC, Cranston HJ, Weiss LM, Halonen SK. Host Cell Preference of Toxoplasma gondiiCysts in Murine Brain: A Confocal Study. Journal of neuroparasitology. 2010;1 23. Ferguson DJ, Graham DI, Hutchison WM. Pathological changes in the brains of mice infected with Toxoplasma gondii: a histological, immunocytochemical and ultrastructural study. International journal of experimental pathology. 1991;72:463474. [PMC free article] [PubMed] 24. Gonzalez LE, Rojnik B, Urrea F, Urdaneta H, Petrosino P, et al. Toxoplasma gondii infection lower anxiety as measured in the plus-maze and social interaction tests in rats A behavioral analysis.Behav Brain Res. 2007;177:70 79. [PubMed] 25. Cartier EA, Parra LA, Baust TB, Quiroz M, Salazar G, et al. A biochemical and functional protein complex involving dopamine synthesis and transport into synaptic vesicles. The Journal of biological chemistry. 2010;285:19571966. [PMC free article] [PubMed] 26. Ogawa N, Asanuma M, Miyazaki I, Diaz-Corrales FJ, Miyoshi K. L-DOPA treatment from the viewpoint of neuroprotection. Possible mechanism of specific and progressive dopaminergic neuronal death in Parkinson's disease. J Neurol. 2005;252(Suppl 4):IV23IV31. [PubMed] 27. Schwab JC, Beckers CJ, Joiner KA. The parasitophorous vacuole membrane surrounding intracellular Toxoplasma gondii functions as a molecular sieve. Proceedings of the National Academy of Sciences of the United States of America. 1994;91:509513. [PMC free article] [PubMed] 28. Webster JP. Rats, cats, people and parasites: the impact of latent toxoplasmosis on behaviour.Microbes Infect. 2001;3:10371045. [PubMed] 29. Denkers EY, Butcher BA, Del Rio L, Bennouna S. Neutrophils, dendritic cells and Toxoplasma.International journal for parasitology. 2004;34:411421. [PubMed] 30. Silva NM, Rodrigues CV, Santoro MM, Reis LF, Alvarez-Leite JI, et al. Expression of indoleamine 2,3dioxygenase, tryptophan degradation, and kynurenine formation during in vivo infection withToxoplasma gondii: induction by endogenous gamma interferon and requirement of interferon regulatory factor 1. Infection and immunity. 2002;70:859868. [PMC free article] [PubMed] 31. Breier A, Buchanan RW, Elkashef A, Munson RC, Kirkpatrick B, et al. Brain morphology and schizophrenia. A magnetic resonance imaging study of limbic, prefrontal cortex, and caudate structures. Archives of general psychiatry. 1992;49:921926. [PubMed]

32. Horacek J, Flegr J, Tintera J, Verebova K, Spaniel F, et al. Latent toxoplasmosis reduces gray matter density in schizophrenia but not in controls: Voxel-based-morphometry (VBM) study. 2011. The world journal of biological psychiatry: the official journal of the World Federation of Societies of Biological Psychiatry. 33. Brynska A, Tomaszewicz-Libudzic E, Wolanczyk T. Obsessive-compulsive disorder and acquired toxoplasmosis in two children. European child & adolescent psychiatry. 2001;10:200204. [PubMed] 34. Miman O, Kusbeci OY, Aktepe OC, Cetinkaya Z. The probable relation between Toxoplasma gondii and Parkinson's disease. Neuroscience letters. 2010;475:129131. [PubMed]