Introduction

A fascinating phenomenon in behavioral biology is the ability of parasites to manipulate host behavior for their own benefit. A handful of examples are noted for insect [1], [2] and crustacean [3] hosts, but rarely so in mammals. The extraordinary effectiveness of the mammalian blood brain barrier denies most pathogens access to the privileged central nervous system, the seat of will.

Toxoplasma gondii is an obligate, single-celled protozoan parasite capable of crossing into the central nervous system of any warm-blooded vertebrate. Toxoplasma requires the cat intestine to reproduce sexually, is shed in cat feces, and must make its way from the ground to another cat host [4].

Toxoplasma manages this in part by infecting ground-dwelling rats who, remarkably, begin selectively preferring areas with cat urine [5], [6], [7]. Infected rats retain normal defensive behavior to non-feline predator odor and normal performance on memory, anxiety, fear and social tasks [6], [8]. This specific preference for cat odor is likely an adaptive manipulation by Toxoplasma, increasing infected rat predation rates and facilitating Toxoplasma transmission to the cat.

Little is known about how Toxoplasma inspires this manipulation. By two weeks post infection, Toxoplasma has settled throughout the host rat brain in spherical cysts approximately 50–70 µm in diameter [4]. Intriguingly, cysts show a slight preference for limbic system regions responsive to both predator stimuli and sexual stimuli, regions responsible for gating innate approach and avoidance behaviors [6]. We investigated the effect of Toxoplasma on the neural activity in limbic system regions involved in both ‘defensive’ and ‘reproductive’ innate behavior. Neural activity was quantified using the immediate early gene c-Fos, a proxy for neural activity (see Materials and Methods).

Results

We first confirmed limbic activity during exposure to either cat urine or an inaccessible estrous female. As expected [9], [10], [11] in uninfected rats, cat urine increased neural activity in the ‘defensive’ ventromedial hypothalamus, dorsomedial part (VMHdm) (Figure 1B and Table S1). Exposure to an estrous female rat increased activity in the ‘reproductive’ posterodorsal medial amygdala (MEApd) (Figure 1B and Table S1) [11].

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 1. Toxoplasma Infection Alters the Limbic Response in Rats Exposed to Cat Odor.

(A) Toxoplasma infected rats spent more time exploring cat urine than uninfected rats. (B) In uninfected animals, female odor induces relatively greater ‘reproductive’ MEApd activity and relatively diminished ‘defensive’ VMHdm activity, as expected. (C) In infected animals, activity levels in the MEApd are the same, regardless of whether exposed to a ‘reproductive’ or ‘defensive’ odor. (D) Neural activity in infected male rats, normalized to uninfected controls, after exposure to cat urine. Infection increased neural activity in the BLA, MEApd and VMHdm. (E) Neural activity in infected male rats, normalized to uninfected controls, after exposure to an estrous female. Infection again increased neural activity in the BLA and VMHdm, but did not change MEApd levels. (F) Toxoplasma infection increased ‘reproductive’ MEApd neural activity during exposure to cat odor, mimicking neural activity in uninfected animals exposed to a female rat. MEApd, posterodorsal medial amygdala; VMHdm, ventromedial hypothalamus, dorsomedial part; BLA, basolateral amygdala. P values listed where appropriate.

https://doi.org/10.1371/journal.pone.0023277.g001

Toxoplasma infection made rats spend more time exploring cat urine (Figure 1A) and increased neural activity in the MEApd, VMHdm and basolateral amygdala (BLA) (Figure 1C and 1D) during this exploration. During exposure to an estrous female, Toxoplasma increased activity in the VMHdm and BLA, but did not significantly alter MEApd activity (Figure 1E) relative to uninfected controls. Toxoplasma infected rats had reduced volumes of both the MEApd and the posteroventral medial amygdala (MEApv) (Figure S1 and see Discussion S1).

Discussion

Rats have separate ‘defensive’ and ‘reproductive’ pathways gating innate behavioral response to, respectively, predator or sexual stimuli [11]. Given the precipitous evolutionary pressures of both reproduction and predation, these pathways run as direct projections from the olfactory bulb to the limbic system and generate rapid and stereotyped behavioral output (see Figure 2B and Figure 2C). Although functionally distinct, the limbic ‘defensive’ and ‘reproductive’ pathways run in parallel through the medial amygdala and hypothalamus in close anatomical proximity. Previous findings of increased Toxoplasma cyst density in these areas compel the possibility that Toxoplasma is somehow perturbing its surrounding neural environment and thereby manipulating the host response to cat urine. We find, indeed, that Toxoplasma infection perturbs the ‘defensive’ pathway in the infected rat during exposure to cat urine, shifting neural activity to the nearby ‘reproductive’ pathway, specifically the MEApd (see Figure 2D).

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Figure 2. Schematic Model of Toxoplasma Induced Changes to Host Limbic System.

(A) General schematic model of limbic brain activity in a male rat after exposure to an emotionally salient odor. (B) Exposure to an inaccessible estrous female activates the ‘reproductive’ pathway, producing robust MEApd activity and evoking approach behavior. (C) Exposure to cat urine activates the ‘defensive’ pathway, producing robust activation of excitatory (+) VMHdm neurons and evoking defensive, aversion behavior. Excitatory (pointed arrow) projections from the MEApv activate inhibitory (−) VMHvl neurons, countering inhibitory (blunted arrow) projections from the MEApd and suppressing any approach behavior. (D) Proposed model for Toxoplasma rats during exposure to cat urine. Robust MEApd activity biases toward approach behavior. Aversion behavior remains, but is lessened. Raw density of c-Fos (large print) ±SEM (small print) activity is given for each region during relevant odor exposure. Model adapted from Choi 2005. AOB, accessory olfactory bulb; MEA, medial amygdala; VMH, ventromedial hypothalamus.

https://doi.org/10.1371/journal.pone.0023277.g002

The MEApd projects robustly to hypothalamic nuclei involved in sexual arousal and the generation of approach behavior. The MEApd is responsive to a variety of social odorants [12], but responds strongest to opposite-sex mating stimuli. Lesioning the MEApd reduces, specifically, attraction to opposite-sex odors [13]. Interestingly, Toxoplasma increased MEApd activity during cat urine exposure to levels mimicking uninfected rats during female exposure (Figure 1F). This suggests that the specifically increased magnitude of MEApd activity in male Toxoplasma infected rats is biasing the processing of the cat urine toward the sexual, ‘reproductive’ pathway (Figure 2D). Plausibly, this shift is altering the salience of the cat urine stimuli and mitigating the defensive response by creating, in its stead, a competing attraction to the cat urine.

Little is known about how, if at all, Toxoplasma cysts exert themselves in the host brain. Much work remains to be done, based on striking findings that Toxoplasma raises whole brain dopamine levels in mice by up to 15% [14] and that dopamine receptor antagonists block rodent host attraction to cat urine [7]. These data suggest a link between dopamine, a primary neurotransmitter in decision-making and reward, and the altered behavior. Intriguingly, the Toxoplasma genome contains a homolog of tyrosine hydroxylase [15], the rate-limiting enzyme in the vertebrate synthesis of dopamine, raising the possibility that Toxoplasma is altering dopamine levels by synthesizing its own tyrosine hydroxylase.

From the ground, Toxoplasma finds its way into other hosts besides rats, including cows, sheep, pigs and many grazing livestock. Ingestion of undercooked meat from infected livestock and the profligacy of private cat ownership are responsible for a strikingly high number of human chronic Toxoplasma infections. Approximately one-third of humans are seropositive for Toxoplasma across the world [16], and several recent studies find infection increases risk for schizophrenia [17], [18], [19] and obsessive compulsive disorder [20], diseases noted for elevated dopamine levels and disturbed amygdala function [21]. Our results are therefore of wide interest, as the ability of Toxoplasma to dramatically alter host behavior and proper amygdala functioning may extend beyond the rat into ancillary Toxoplasma hosts, including humans.

Materials and Methods

Supporting Information

Acknowledgments

The authors would like to thank Dr. John Boothroyd and Dr. Anita Koshy for valuable advice and technical help.