Molecular Psychiatry (2008) 13, 470–479 & 2008 Nature Publishing Group All rights reserved 1359-4184/08 $30.00 www.nature.com/mp PERSPECTIVE Are some cases of caused by microbial agents? A review of the evidence RH Yolken1 and EF Torrey2 1The Stanley Laboratory of Developmental Neurovirology, Department of Pediatrics, Johns Hopkins University Medical Center, Baltimore, MD, USA and 2The Stanley Medical Research Institute, Chevy Chase, MD, USA

The infectious theory of psychosis, prominent early in the twentieth century, has recently received renewed scientific support. Evidence has accumulated that schizophrenia and bipolar disorder are complex in which many predisposing genes interact with one or more environmental agents to cause symptoms. The protozoan Toxoplasma gondii and cytomega- lovirus are discussed as examples of infectious agents that have been linked to schizophrenia and in which genes and infectious agents interact. Such may occur early in life and are thus consistent with neurodevelopmental as well as genetic theories of psychosis. The outstanding questions regarding infectious theories concern timing and causality. Attempts are underway to address the former by examining sera of individuals prior to the onset of illness and to address the latter by using antiinfective medications to treat individuals with psychosis. The identification of infectious agents associated with the etiopathogenesis of schizophrenia might lead to new methods for the diagnosis, treatment and prevention of this disorder. Molecular Psychiatry (2008) 13, 470–479; doi:10.1038/mp.2008.5; published online 12 February 2008 Keywords: cytomegalovirus; toxoplasmosis; infectious; schizophrenia; genetic; environmental

Introduction cases it does last from several months to years’.2 The following year, a spirochete was discovered to be the The idea that microbial agents may cause psychotic cause of syphilis, and by the First World War, disorders has a surprisingly lengthy history. As the infectious theories had joined genetic and endocrine germ theory of diseases became established in the theories in the mainstream of psychiatric research. late-nineteenth century, several European researchers The autointoxication theory, however, fell into theorized that bacteria might be etiologically linked to disrepute following attempts by several researchers dementia praecox or other psychiatric diseases. It was to surgically remove the theoretical foci of in speculated at that time that the bacteria might not various organs, often with tragic results.3,4 Attention infect the brain directly but rather could reside in the then shifted to viruses, after it was noted that intestine or elsewhere and, by a process of autointoxi- infection with the influenza virus caused psychosis cation, affect the brain by secreting toxins. in some people.5 Such cases were extensively studied One of the earliest enthusiasts for this theory was by Karl Menninger following the 1918–1919 influenza Emil Kraepelin who, in the 1896 edition of his pandemic. After observing many cases, Menninger Psychiatrie, reasoned that dementia praecox was ‘a concluded: ‘I am persuaded that dementia praecox tangible morbid process of the brain’ caused by ‘an (schizophrenia) is at least in most instances a somata- autointoxication whose immediate causes lie some- psychosis; the psychic manifestations of an encepha- where in the body’; Kraepelin believed that the sex litis (italics in original). The acuteness or chronicity, organs were the most likely site of infection.1 The fact the benign or malignant nature of this encephalitis that psychoses occasionally accompanied known perhaps determines the degree of reversibility of the bacterial diseases such as typhoid fever, tuberculosis schizophrenia.’6 In recent years, modern imaging and diphtheria supported the infectious theory; a 1904 techniques have established that schizophrenia is, review concluded that although ‘insanity following in most cases, not associated with acute encephalitis, infection is generally of short duration, y in a few although it remains possible that it is a sequela or recurrence of a past episode of encephalitis. Correspondence: Dr RH Yolken, The Stanley Laboratory of Interest in infectious theories of psychiatric dis- Developmental Neurovirology, Department of Pediatrics, Johns orders waned in the United States and Europe in Hopkins University Medical Center, 600 N Wolfe Street, 1105 the 1930s as Freudian theories became prominent. Blalock, Baltimore, MD 21287-4933, USA. Psychiatric research on infectious agents continued E-mail: [email protected] 7,8 9 Received 21 March 2007; revised 14 June 2007; accepted 26 June sporadically, mostly in Eastern Europe and Russia, 2007; published online 12 February 2008 but after the onset of the Cold War, these studies were Psychosis and microbial agents RH Yolken and EF Torrey 471 not widely known in the West. In the 1970s, there was therapies. Thus, relatively few microbes meet epide- renewed interest in infectious agents in Western miological requirements for serious consideration for Europe10–12 and the United States,13,14 culminating causing more than sporadic cases. in a 1983 World Health Organization symposium, Another epidemiological aspect of schizophrenia ‘Research on the Viral Hypothesis of Mental Dis- and bipolar disorder is the fact that both have a orders’.15 Since that time, interest in this research area modest seasonal birth predominance in the winter has steadily increased. and spring months.19 Over 200 studies have demon- This review attempts to summarize such research. strated this, and it is, in fact, one of the most It first summarizes general evidence favoring an consistently replicated aspects of these disorders. infectious theory, and then discusses the interaction Statistical artifact and parental procreational habits of microbial agents and genetic factors. Of necessity, have been ruled out as explanations, but seasonal the review does not discuss every infectious agent differences in sunlight, temperature, rainfall, birth and every psychiatric disorder that have been complications, toxins and nutrition have all been studied. It rather focuses on psychotic disorders considered, along with microbes. If an infectious in general, and schizophrenia in particular, as agent is responsible for the seasonality, it would be examples, and on the infectious agents that appear one that occurs through the year but that has a modest to be the most promising etiological candidates: seasonal predominance. The infectious agent would Toxoplasma gondii and cytomegalovirus (CMV). also be one that does not vary markedly in incidence Finally, it concludes with a summary of the current from year to year; this makes influenza an unlikely state of this research area, including both its strengths candidate. and weaknesses. Another epidemiological fact of interest is that being born in, or raised in, an urban environment, Why should microbial agents be considered? compared to a rural environment, approximately doubles a person’s risk of later being diagnosed with There are several aspects of psychotic disorders that schizophrenia. The urban risk is clearly associated suggest a possible infectious origin. One is the century- with schizophrenia but has not been observed for old observation, noted above, that individuals bipolar disorder or other affective disorders.20 A sometimes present with the clinical symptoms of dose–response relationship has been described in- schizophrenia, bipolar disorder or depression with sofar as ‘the more years lived in the higher degree of psychosis in the course of developing, or shortly after urbanization, the greater the risk’.21 In one study, the having had, a known microbial . The list of population-attributable risk for developing schizo- infectious agents that have been reported to sporadi- phrenia due to childhood urban living was 34.6, more cally cause psychotic symptoms is long but most than six times greater than the genetic risk conferred prominently includes spirochetes such as Treponema by having a first-degree relative with schizophrenia pallidum (causing syphilis) and Borrelia burgdorferi (5.5).22 Socioeconomic status and obstetric complica- (causing Lyme disease); viruses such as herpes simplex tions have been ruled out as explanations of the urban viruses (HSV-1 and -2), Epstein–Barr virus, CMV, risk factor;23 a greater density of population must be influenza, measles, rubella, mumps, polio, vaccinia, considered, since it is associated with the rapid enteroviruses such as Coxsackie B4, arboviruses such transmission of many microbes. as Eastern equine encephalitis virus, retroviruses such An unresolved issue concerning the possible as the human immunodeficiency virus (HIV) or microbial etiology of psychosis is the timing of the endogenous human retroviruses, and Borna disease infection. In some, but not all, studies, it has been virus; and protozoa such as T. go ndi i.16–18 Such cases reported that maternal infections with rubella,24 may become evident by the appearance of neurological influenza,25 HSV-226 and T. gondii 27,28 increase the symptoms shortly after the onset of psychotic symp- incidence of psychotic disorders in the offspring. toms, or they may remain hidden until autopsy. Other studies have documented an increased rate of Epidemiological considerations, however, make it schizophrenia in adults who during childhood had unlikely that many of these microbes are candidates viral or bacterial meningitis.29 It is thus possible that for causing a large number of cases of schizophrenia microbes could cause psychosis secondary to infec- or other psychoses. Since psychoses are widespread tions in utero, in infancy or in childhood in addition in the world, localized infectious agents such as to in adulthood around the time of onset of the arboviruses, Borna disease virus and B. burgdorferi psychosis. It is also possible that an infection are unlikely to be causative agents in different areas of occurring in infancy can be reactivated in later life, the world. Schizophrenia and bipolar disorder have as is known to occur with the herpes viruses and also been described for at least 200 years, thus ruling T. gondii. The outcome of the infection may differ out infectious agents that only recently infected markedly depending on its timing, as is known to be human populations, such as HIV. In addition, the true for human infections such as those caused by the incidence of schizophrenia and bipolar disorder has rubella and polio viruses. It is also known that not decreased markedly in recent years, whereas the infections at different stages of brain development incidence of syphilis, rubella, measles and mumps result in varying degrees of lifelong changes in has done so as a result of immunization and improved behavior and cognition.

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 472 How genes and microbes interact areas of genetic association, it is difficult to envision how the results from these studies will solve the Despite more than a century of research, the genetic problem of the low odds ratios and the population basis of psychotic disorders is not well understood. diversity that seem to be inherent in the study of For schizophrenia, recent research has focused on the schizophrenia. In fact, it can be anticipated that these search for genetic mutations and polymorphisms as studies will result in the identification of larger causes. This research is based on multiple family numbers of potential genetic associations that will studies indicating that schizophrenia has a high need to be replicated in very large sample sizes and degree of heritability; it has been stimulated by the then integrated into current algorithms. Furthermore, wealth of genetic information that has become the analyses of larger numbers of potential loci will available as part of the human genome project. require the performance of larger numbers of multiple The search for specific genes associated with schizo- comparisons among genes, making it even more phrenia has also been fueled by findings from linkage difficult to distinguish positive findings from true studies indicating that some regions of the human associations. genome are inherited nonrandomly in family mem- For this reason, it is important to elucidate bers who have schizophrenia. nongenetic factors that might contribute to schizo- Despite a great deal of laboratory research, no single phrenia. Of particular importance are environmental gene has been discovered that has a major effect in influences that might interact with genetic factors and different populations. Although several genes have thus be applicable to a disease such as schizophrenia, been associated with schizophrenia in case-control which has a high degree of heritability. The integra- and family studies, many have not been found to be tion of environmental risks might increase the odds operant in different populations. The genes that are ratios of genetic associations in specific populations consistently associated with schizophrenia in differ- of exposed individuals and hence increase the like- ent studies have relatively low odds ratios, generally lihood of identifying true positive associations. in the range of 1.1–1.5. Studies reporting higher odds The modern era of studying how genes and ratios have generally not been reproducible in sub- microbes interact was initiated in 1948 with the sequent studies performed in other populations. An observation by Haldane31 that the gene for sickle increased understanding of human haplotype struc- hemoglobin provides protection against malaria, ture has also suggested that the strength of some undoubtedly explaining the persistence of this gene previously reported associations has been overesti- in humans living in malaria endemic areas. More mated because of haplotype misclassification. In recently, the observation that individuals vary in their addition, many of the genes that are associated with response to HIV infection has led to the discovery of a schizophrenia have similar associations with other number of genes that are associated with a variable psychiatric disorders such as bipolar disorder and response to retrovirus infection. There are many unipolar depression, as has recently been described additional infectious diseases for which a genetic for methylenetetrahydrofolate reductase.30 component has been found in repeated studies. These While the reproducible associations are of great include infections caused by Mycobacterium tuber- interest to neurobiologists interested in elucidating culosis, M. leprae (leprosy), hepatitis B and C viruses, pathways of neural transmission and other neuro- cholera, Helicobacter pylori, norovirus and Leishma- logical processes, associations at this level are of nia donovani. In addition, genes are very important in substantially less value to clinicians and epidemiol- infectious diseases involving multiple organisms, ogists, since they have low positive predictive values such as bacterial sepsis and otitis media.32 and thus provide little in the way of clinical or Currently recognized genetic determinants of diagnostic guidance. Furthermore, findings in this response to infection include receptors, transcription range are difficult to duplicate except in studies of factors, cytokines, complement components, human large sample sizes. For example, it takes a study of lymphocyte antigen and other T-cell determinants, approximately 2000 individuals to detect a genetic Toll-like receptors and other determinants of innate factor that is present in the general population at immunity. As different aspects of the immune system a rate of 10% and that has a relative risk of 1.5 with a are more precisely defined in humans and in animal power of 0.9. Even larger sample sizes are required to models, there is an ever-increasing list of genes that detect genes that are present in lower prevalence or can potentially be associated with host response to that have lower odds ratios; for example, more than infectious agents. In addition, the receptors for 6100 individuals would be required to detect a gene specific agents are being increasingly identified with a population prevalence of 5% and a relative risk through the application of molecular biological of 1.4. techniques, further increasing the number of potential Methods for genetic analysis are continually evol- genetic loci defining susceptibility to infectious ving. There is currently much interest in methods that agents. can increase the number of genetic loci that can be As discussed above, there are a number of in- evaluated in a given period of time. While such fectious agents, which have been associated with methods, which include ‘whole genome association’ schizophrenia. This review focuses on Toxoplasma studies, have given hope in terms of finding new and CMV, since they meet the criteria discussed above

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 473 and have been associated with schizophrenia in United States, the rate is approximately 25%.33 Of different populations interest in relationship to schizophrenia, the peak ages of seroconversion is between ages 15 and 35;34 T. gondii and schizophrenia adolescent males become infected earlier than females35 and there are inconsistent suggestions of a T. gondii is a coccidian protozoan of the apicomplexa seasonal occurrence (least in summer)36,37 and greater family, first described in 1908. Felines, including urban risk.38 domestic cats, are its definitive host, and the organ- Genetically, numerous animal studies have identi- ism can only complete the sexual part of its life cycle fied genetic determinants of response to T. gondii within feline hosts. As shown in Figure 1, T. gondii infection, especially in immune-compromised hosts. oocysts are excreted in the feces of cats at the time Host genes associated with susceptibility or resistance they are initially infected. The oocysts may then to Toxoplasma infections include ones that regulate the become aerosolized and infect humans who are expression of T-cell determinants,39 nuclear factor-kB changing cat litter boxes, gardening or playing in transcription,40 interferon gamma,41 lymphotoxina,42 sandboxes. The cat may also deposit feces on the other cytokines,43 Toll-like receptors44 and chemokine ground or in animal feed in barns; domestic animals receptors,45 as well as genes of an as yet undefined may eat it, producing T. gondii tissue cysts in their effect on macrophage function.46 The genes that muscles, which then infect humans who eat under- determine the response to Toxoplasma in immune cooked meat. Felines can also contaminate water, competent humans have not yet been defined but are with subsequent infection of humans through drink- the subject of ongoing investigations. ing water, eating vegetables washed with contami- Clinically, most human infections with T. g ond ii had nated water or eating undercooked fish that had lived been thought to be asymptomatic. However, recent in the contaminated water. The relative importance of studies of individuals infected with Toxoplasma- different modes of transmission depends upon local contaminated water have indicated that a majority of geographic, cultural and socioeconomic factors. infected immune-competent individuals have clini- T. gondii is distributed worldwide and infects cally apparent symptoms, such as headache, fever, between 10 and 80% of the adult population; in the malaise, myalgias, adenitis, anorexia and arthralgias.47

Oocysts deposited in many places

KITTYLITTER The ground Animal feed in barn Water supply Gardensands and boxes Eaten by animals

Pregnant woman inhales Washed Eaten by mice and oocysts vegetables and rats, whose fish are People work behavior may be contaminated in gardens, altered by the play in Toxoplasma, making People eat sandboxes them easier prey Fetus is affected undercooked (congenital meat toxoplasmosis) Theyare then eaten by other cats, beginning another cycle

Figure 1 Life cycle of T. gondii.

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 474 Occasional cases have been described with delusions schizophrenia has been the lack of a teleological and hallucinations.48 The microbe is known to be framework relating to why, from an evolutionary neurotrophic and to infect both neurons and glia.49 point of view, infection with a microbial agent might Serologically, the first research linking schizophrenia lead to the altered behavior that is the hallmark of and other psychoses to an increase in antibodies to schizophrenia. However, recently it has been recog- T. g ond ii was published in 1953; since then, at least nized that infectious agents can alter host behavior in 41 other studies have been carried out. A recent a manner than enhances the survival and reproduc- meta-analysis on 23 of these studies reported an odds tion of the infecting agent. T. gondii provides a prime ratio for schizophrenia risk associated with sero- example of this type of evolutionary adaptation. As logical evidence of T. gondii infection as 2.73 (95% noted above, felines constitute the natural hosts for confidence interval, 2.10–3.60; P < 0.000001).50 While the protozoan. If a Toxoplasma organism infects a cat still in the range of moderate association, this odds or other member of this family, it can complete its ratio is substantially higher than that found in most entire life cycle, including sexual reproduction. If, meta-analyses of specific genes associated with schizo- however, the organism infects another animal, for phrenia risk. example, through the ingestion of Toxoplasma organ- These serological findings are supported by other isms shed by infected cats, the non-feline host is a studies linking T. gondii to schizophrenia. A study ‘dead end’ for the Toxoplasma, in that the microbe of newborn sera and a study of maternal sera of cannot complete the sexual part of its life cycle and individuals who later developed schizophrenia and thus remains trapped in cyst form within the tissues schizophrenia spectrum disorders both reported more of this incomplete host. The organism trapped in a T. gondii antibodies in cases than in controls.27,28 In non-feline host has another way out, however; if it addition, preliminary analysis of a cohort of indivi- can get the host to be eaten by a feline, it will be able duals in the US Military indicates that increased to complete its life cycle in the new feline host. levels of Toxoplasma antibodies can be found in Since all felines are carnivores, the possibility that individuals prior to the onset of symptoms, obviating another animal will be consumed is not remote; if the the possibility that the finding of increased levels of host is a rodent or other small animal, the possibility antibodies is an epiphenomenon associated with is even greater. Recent research has indicated that exposure occurring after the onset of disease.51 T. gondii can increase the likelihood of having its There are a number of reasons why interest in the rodent intermediate host eaten by a cat by altering the study of the role of Toxoplasma and other infectious motor behavior of the rodent.54 Specifically, the agents in the etiology of schizophrenia has, until rodent loses its inherent fear of novel stimuli, recently, been limited to a small number of investiga- including its fear of cats. Behavioral experiments tors. First, until recently, there was little understand- have documented that Toxoplasma-infected rats lose ing of the molecular biology of T. g ond ii and the their aversion to cat urine without other alterations in pathophysiological mechanisms by which Toxoplasma motor behavior, such as lethargy, which might make infection causes disease. Recently, the techniques that the rat less desirable prey to felines. In the real world, led to the sequencing of the human genome have been this alteration is likely to lead to a markedly increased applied to the sequencing of protozoan genomes as consumption of Toxoplasma-infected rodents by cats, well, resulting in a corresponding increase in the resulting in an infection of the cats and completion of characterization and understanding of the molecular the Toxoplasma life cycle. Since humans are gener- basis of infection. For example, the characterization of ally not liable to be consumed by felines (hunters and the genome of T. gondii, which was completed in 2005, trainers of large cats excepted), the effect of Toxo- has led to an increased understanding of the biology of plasma on humans is largely vestigial. Since the Toxoplasma infections and to the development of Toxoplasma organism has limited ability to modulate assays for antibodies to specific Toxoplasma proteins. its response in different intermediate hosts, however, The application of these assays has revealed that it can be postulated that all hosts can be affected by Toxoplasma organisms, which had previously been similar pathways. In fact, in humans, two studies thought to be relatively uniform in terms of biological have independently reported that young adults with properties, exist in multiple strains with differing serological evidence of Toxoplasma infection have degrees of pathogenicity. Of particular importance has increased rates of automobile accidents compared to been the recent elucidation of specific Toxoplasma age-and gender-matched controls, a behavior that is protein kinases in some strains, which alter signal likely to be associated with increased risk-taking transduction in infected hosts and thus modulate behavior.55,56 humoral immune function. These protein kinases The neurobiological mechanism by which T. gondii appear to be the major determinants of the pathogeni- causes psychiatric symptoms and exerts its effect on city of individual Toxoplasma strains.52,53 Further human behavior is unknown. One possibility is a studies of these and other markers of pathogenicity direct effect of the organism on neurons and/or glia; should lead to improved assays for defining the risk of such an effect may occur at the time of infection or disease in Toxoplasma-infected individuals. much earlier in life. There are animal models of Another factor that has impeded an understanding infections known to occur in utero or early in life that of the role of infectious agents as etiological factors in do not become manifest with behavioral changes until

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 475 the animal reaches maturity.57 In the case of T. gondii, infection with a range of clinical consequences, it is known that the organism may affect signal including encephalitis, pneumonia, hepatitis, enter- transduction pathways and that it also encodes itis and retinitis. proteins with hemology to tyrosine hydroxylase In immune-competent individuals, most initial and the mammalian D2 receptor, suggesting that it CMV infections are asymptomatic. If symptomatic, may interfere with dopamine synthesis pathways in the most common syndrome is fever and enlarged human hosts.58,59 There are also models of infectious lymph nodes, and CMV is thought to be responsible agents that affect dopamine pathways in one part of for approximately 10% of mononucleosis-like syn- the brain but not in others.60 Alternatively, infectious dromes in young adults. Occasional cases of ence- agents such as T. gondii may cause psychiatric phalitis have been reported. In one case, a 30-year-old symptoms by the effect of antibodies against the male developed chronic CMV encephalitis with organism or by cytokines elicited by the infection.61 ‘deficits in concentration, memory, manipulation of Studies of such mechanisms are ongoing. knowledge, humor and emotional expression’.71 Si- The development of hallucinations and other milarly, a study of CMV antibodies in 323 individuals clinical symptoms in an individual infected with with schizophrenia reported that those who had Toxoplasma is thus likely to be dependent on both the primarily deficit symptoms (n = 88) were significantly human and the microbial genomes. In this sense, more likely to have CMV antibodies compared to schizophrenia might be considered not merely a those who had nondeficit symptoms (n = 235; genetic disease but rather a disease resulting from P = 0.006).72 Another study of individuals with schi- the interaction of multiple genomes. In this context, it zophrenia reported that ‘higher levels of CMV anti- is likely that study of both the human and the bodies were associated with decreased performance microbial genomes will be required to come to a on baseline measures of verbal learning and mem- complete understanding of complex neuropsychiatric ory.73 Also of interest is a case study of a young disorders such as schizophrenia. woman who presented with auditory hallucinations, delusions, tangential thinking and flattened affect and Cytomegalovirus and schizophrenia was diagnosed with schizophrenia; while hospita- lized, she died and was then diagnosed with CMV Cytomegalovirus is another example of an infectious encephalitis on the basis of retrospective antibody agent that has been linked to a specific psychiatric titers and post-mortem findings.16 disorder, schizophrenia. It is a b-herpesvirus initially Neuropathologically, CMV is known to have an isolated in 1956 from children with congenital mental affinity for the limbic system74 and to evoke a strong retardation and hearing loss. CMV occurs worldwide glial response, producing cytokines and chemo- and is spread by personal contact, including saliva, kines.75 It is unclear whether CMV’s effects on the blood, urine, genital secretions and breast milk. It brain are primarily due to the direct effects of the virus thus spreads more quickly under conditions of poor or are indirectly mediated by the immune response. hygiene; in developed countries, approximately 60% The histopathologic hallmark of CMV infections are of adults are infected, while in developing countries, intranuclear inclusion bodies and scattered glial the figure approaches 100%. Some observers have nodules. Most neuropathological studies of CMV have said that CMV infections are more common in the been done on immune-compromised individuals, and winter and spring,62 but others have not observed any it is unclear whether immune-competent individuals seasonal difference. exhibit similar findings. In animal studies, infecting It is known that resistance to CMV infection is rats with a rat CMV shortly after birth was said to associated with variations in the gene encoding tumor produce ‘a deficit in sensorimotor gating upon necrosis factor-a;63 variations in this gene have also dopamine stimulation, supporting a possible link been associated with an increased risk for schizo- between viral infection and schizophrenia’.76 phrenia in some studies64,65 but not others.66 Varia- Many studies have looked for evidence of CMV tions in the gene-encoding interleukin-10 have also infection in the blood, CSF and brain tissue of been associated with susceptibility to both CMV individuals with schizophrenia. Fourteen serological infections and schizophrenia.67,68 studies carried out prior to 1994, using less sensitive The two most prevalent CMV clinical syndromes assays and mostly patients with chronic schizophre- are congenital infection and systemic infections in nia, were negative (reviewed in Yolken and Torrey77). immunosuppressed individuals. Congenital infec- More recently, five studies using more sensitive tions may produce mental retardation, hearing loss, assays and patients with a more recent onset of visual impairment and other deficits in the newborn. schizophrenia have all reported a higher rate of CMV In many cases, the infection is initially asymptomatic seropositivity in the patients compared to controls but later decreases the child’s IQ.69 Once infected (reviewed in Torrey et al.78). with CMV, individuals remain latently infected for Of particular note was the study by Leweke et al.79 life, although they may also become secondarily in Germany. They studied 36 first-episode, never- infected with a different CMV strain.70 Suppression treated individuals with schizophrenia; 10 indivi- of the immune system, as occurs in transplantation duals with schizophrenia who had been treated in the and HIV infection, often leads to reactivation of CMV past but were medication-free at the time of the study;

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 476 2.5 2.0 was noted for patients who were seropositive for other * Never therapy (n=36) herpes viruses. *** Past therapy (n=10) Current therapy (n=39) In the other treatment study, 50 outpatients with an 2.0 Controls (n=39) acute exacerbation of schizophrenia were treated with * 1.5 ** adjunctive celecoxib in a double-blind trial. Those 1.5 receiving the drug had a significant improvement in their symptoms, especially between 2 and 4 weeks 1.0 (P = 0.001).83 Celecoxib is an inhibitor of cyclooxy- 1.0 genase (COX-2),84 and it is known that inhibition of COX-2 blocks the replication of CMV.85 Based on the CSF IgG antibody level Serum IgG antibody level 0.5 results of these two treatment trials, additional tests 0.5 are underway.

0 0 Discussion Figure 2 Pathways of transmission of Toxoplasma gondii T. gondii and CMV are examples of infectious agents from cats to humans. that may be plausibly linked to schizophrenia. They are a good fit epidemiologically, including a world- wide distribution and a peak exposure. They are 39 individuals with schizophrenia with recent onset known to be neurotrophic and to alter human but who were on medication at the time of the study behavior. They are capable of infecting individuals and 73 unaffected controls. As shown in Figure 2, in early life and are thus consistent with neurodeve- CMV immunoglobulin G antibody levels were sig- lopmental theories of schizophrenia. They are also nificantly higher in both the serum (P < 0.001) and capable of reactivation in early adulthood, the peak cerebrospinal fluid (CSF) (P < 0.004) in individuals time for the onset of the symptoms of schizophrenia. with schizophrenia who had never been treated And they are known to be suppressed, to some degree, compared to the unaffected controls. Noteworthy by antipsychotic medications. They thus represent was the stepwise gradient of antibody levels in both examples of microbes that may have an etiological serum and CSF from those who had never been relationship to a major psychiatric disorder. treated, those who had been treated in the past, those There are two major limitations to any infectious who were presently being treated and the unaffected theory of schizophrenia: timing and causality: Since controls, suggesting that antipsychotic medication most studies rely on the detection of infection in may have been decreasing the antibodies. individuals who already have schizophrenia, there Other CMV studies of the CSF in individuals with has been difficulty in distinguishing whether an schizophrenia have been less conclusive. Since 1980, increased rate of exposure to an infectious agent is a eight studies have been done; four reported signifi- cause or an effect of the altered behavior that is cantly increased CMV antibodies in the CSF of patients characteristic of schizophrenia. For example, it might compared to controls, while four others did not be argued that some of the concomitants of schizo- (reviewed in Torrey et al.78). Nine studies have also phrenia, such as hospitalization, homelessness and attempted to identify the CMV genome in post-mortem the administration of medications, might alter the brain tissue from individuals with schizophrenia, using host immune response and result in an increased hybridization or PCR techniques. Eight of the studies exposure to an infectious agent independent of the were unable to detect CMV in the brain tissue factors that led to the initiation of the disease itself. (reviewed in Torrey et al.78); the other reported that ‘a This problem is generic to all potential associations clear hybridization signal was detected with DNA from between infection and chronic disease defined in the temporal cortex of a young man with the full cross-sectional case control studies, since it can be picture of schizophrenia’.80The Leweke et al.study, difficult to determine if the measured exposure cited above, suggests that treating individuals with occurred before, during or after the initiation of the schizophrenia with antipsychotic drugs may reduce disease process. The problem can be compounded by antibodiestoCMV,presumablybysuppressingthe the difficulty in identifying unaffected controls who infection. This observation is consistent with studies do not have the disease but who are likely to be reporting that some antipsychotic drugs suppress the similar to the cases in terms of social and demo- replication of some infectious agents.81 Basedonthis graphic factors that are related to exposures to observation, two studies have also been carried out infectious agents. using adjunct medications that suppress CMV replica- The problem of timing has been partly addressed by tion in individuals with schizophrenia. In the first, the evaluation of individuals with ‘recent onset’ valacyclovir, a known antiviral drug, was used for a schizophrenia who have not been previously hospi- 16-week, double-blind trial for 65 outpatients with talized or, in some cases, have not received medica- schizophrenia. Among the 21 patients who were CMV tions. Several studies in such populations have found seropositive, there was ‘a significant improvement in increased evidence of exposure to T. gondii as well as overall symptoms (P < 0.0005)’.82 No such improvement to other infectious agents. These studies, however,

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 477 might also be affected by exposures occurring around samples documented an increased level of antibodies the time of disease onset, particularly in individuals to T. gondii in samples obtained approximately with prolonged periods of undiagnosed and untreated 2 years prior to the diagnosis of schizophrenia. This psychosis. study, combined with the perinatal studies and the For this reason, there has been an increased studies of recent-onset individuals cited above, recognition of the importance of prospective cohort strongly suggests that a substantial part of the studies for the evaluation of the risk of environmental exposure to Toxoplasma in individuals with schizo- exposure in complex disorders. Such cohorts have phrenia occurs prior to the onset of schizophrenia and proven to be invaluable in addressing the long-term is not solely a result of post hoc exposures. Such effects of diet, smoking and other toxic exposures on studies provide the framework for attempts to prevent chronic diseases such as heart disease, stroke and schizophrenia through the prevention of Toxoplasma cancer. Of particular importance in terms of the study infection or the treatment of infected individuals. of the role of infectious agents in schizophrenia are The other major limitation is related to the cohorts in which blood samples have been obtained difficulty of proving causality. The interactions and stored, since such samples can be retrieved after between infectious agents and host factors resulting the diagnosis of the disease and used to assess pre- in complex disorders such as schizophrenia generally disease exposure to infectious agents by the measure- do not follow the rules laid down by Robert Koch and ment of specific antibodies. his co-workers at the end of the nineteenth century. One type of cohort study that has been employed to These rules, generally known as Koch’s postulates, examine the role of exposures in the subsequent require that there be a one-to-one correspondence development of schizophrenia has involved the study between an infectious agent and a disease process. of pregnant women and their offspring. These have It is clear, however, that the relationship between either been cohorts specifically set up for the study of infectious agents and chronic diseases often does not diseases or population-based cohorts that have made follow the straightforward associations defined by use of neonatal blood samples obtained for the Koch’s postulates but rather follows a more compli- analysis of neonatal metabolic diseases and then cated course comprising multiple pathways leading stored for future studies. In two prospective neonatal to diverse clinical outcomes. These pathways can cohorts, it was found that perinatal exposure to involve more than one infectious agent resulting in T. gondii resulted in an increased risk of schizophre- the same disease, as well as many individuals who nia in later life. Additional perinatal exposures that are infected but who do not develop the target disease have been found to be associated with increased risk in their lifetime. This latter phenomenon is based on a of subsequent schizophrenia in cohort studies number of factors, including the timing of infection, include rubella virus, influenza viruses, enteroviruses the nature of the infecting strain and the genetic and HSV-2 in some populations but not others. makeup of the host. Increased risk of schizophrenia has also been asso- Guidelines for evaluating the relationship between ciated with nonspecific markers of inflammation such an infectious agent and a chronic disorder to supplant as cytokines and with other inflammatory conditions Koch’s postulates have been devised by a number of such as pre-eclampsia. authors. The most widely cited are those devised by Perinatal cohort studies have been important in British statistician Austin Bradford Hill published in identifying exposures in the prenatal period asso- 1985, which identify quantitative degrees of associa- ciated with risk of schizophrenia in later life. tion between exposure and disease and incorporate However, these studies cannot evaluate exposures biological plausibility and coherence into the criteria.86 beyond the neonatal period. Similarly, cohorts of The extent to which the association between Toxo- older adults, such as the Framingham or Cache plasma infection and schizophrenia fulfill the Hill County studies, are directed largely at the study of criteria has been the subject of a recent review.87 diseases that occur in later life and enroll the majority Specifically, with regard to T. gondii, the association of their subjects after the age of risk of schizophrenia. between Toxoplasma infection and schizophrenia Fortunately, there are cohorts of young adults that can fulfills many of the criteria, including strength, be employed to study diseases such as schizophrenia consistency, temporality, plausibility and analogy. with onset in young adulthood. The most widely The main factor that is missing is direct evidence that employed to date has been the cohort of individuals the replication of Toxoplasma organisms is associated in the US Military maintained by the Army Medical with ongoing symptoms of schizophrenia. The most Surveillance Activity. This cohort has been employed straightforward way to test this hypothesis would be to study the role of Epstein–Barr infection in multiple to demonstrate that the inhibition of Toxoplasma sclerosis as well as other associations with infectious replication by pharmacological means would result in diseases. Recently, researchers have identified indi- a decrease in the symptoms of schizophrenia and an viduals who developed schizophrenia while they alteration in the clinical course of the disease. In the were in the military and measured antibodies con- past, this approach has not been feasible because of tained in archived blood samples obtained both the lack of a safe and effective method for the before and after the onset of their symptoms. As treatment of Toxoplasma that would be suitable for noted previously, preliminary analysis of these administration to individuals with schizophrenia.

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 478 Recently, however, it has been shown that derivatives 22 Mortensen PB, Pedersen CB, Westergaard T, Wohlfahrt J, Ewald H, of artemisinin, which has proven to be safe and Mors O et al. Effects of family history and place and season of effective for the treatment of malaria, also inhibit the birth on the risk of schizophrenia. N Engl J Med 1999; 340: replication of Toxoplasma, which shares a substantial 603–608. 88 23 Harrison G, Fouskakis D, Rasmussen F, Tynelius P, Sipos A, portion of its genome with malaria organisms. Gunnell D. Association between psychotic disorder and urban A number of clinical trials examining the efficacy of place of birth is not mediated by obstetric complications or artemisinin derivatives in individuals with schizo- childhood socio-economic position: a cohort study. Psychol Med phrenia are in progress. Their successful conclusion 2003; 33: 723–731. 24 Brown AS, Cohen P, Greenwald S, Susser E. Nonaffective would provide the most direct evidence for a role of psychosis after prenatal exposure to rubella. Am J Psychiatry Toxoplasma in the etiology of schizophrenia and 2000; 157: 438–443. would open a new era in methods for the diagnosis, 25 Brown AS, Begg MD, Gravenstein S, Schaefer CA, Wyatt RJ, prevention and treatment of this devastating illness. Bresnahan MA et al. Serologic evidence for prenatal influenza in the etiology of schizophrenia. Arch Gen Psychiatry 2004; 61: 774–780. 26 Buka SL, Tsuang MT, Torrey EF, Klebanoff MA, Bernstein D, Yolken RH. Maternal infections and subsequent psychosis among References offspring. Arch Gen Psychiatry 2001; 58: 1032–1037. 27 Brown AS, Schaefer CA, Quesenberry Jr CP, Liu L, Babulas VP, 1 Noll R. Historical review: autointoxication and focal infection Susser ES. Maternal exposure to toxoplasmosis and risk of theories of dementia praecox. World J Biol Psychiatry 2004; 5: schizophrenia in adult offspring. Am J Psychiatry 2005; 162: 66–72. 767–773. 2 Ireland WW. Review of: on insanity after acute and chronic 28 Mortensen PB, Nørgaard-Pedersen B, Waltoft BL, Sørensen T, infectious diseases. J Mental Sci 1904; 50: 772–773. Hougaard D, Torrey EF et al. Toxoplasma gondii as a risk factor for 3 Scull A. Madhouse: A Tragic Tale of Megalomania and Modern early-onset schizophrenia: analysis of filter paper blood samples Medicine. Yale University Press: New Haven, 2005. obtained at birth. Biol Psychiatry 2007; 61: 688–693. 4 Noll R. Infectious insanities, surgical solutions: Bayard Taylor 29 Abraho AL, Focaccia R, Gattaz WF. Childhood meningitis Holmes, dementia praecox and laboratory science in early increases the risk for adult schizophrenia. World J Biol Psychiatry 20th-century America. Part 2. Hist Psychiatry 2006; 17: 299–311. 2005; 6(Suppl 2): 44–48. 5 Althaus J. On psychoses after influenza. J Mental Sci 1893; 39: 30 Gilbody S, Lewis S, Lightfoot T. Methylenetetrahydrofolate 163–176. reductase (MTHFR) genetic polymorphisms and psychiatric 6 Menninger KA. Reversible schizophrenia: a study of the implica- disorders: a HuGE review. Am J Epidemiol 2007; 165: 1–13. tions of delirium schizophrenoides and other post-influenzal 31 Haldane JBS. The rate of mutation of human genes. Hereditas syndromes. Am J Psychiatry 1922; 1: 573–588. 1948; 35(Suppl): 267–273. 7 Kozar Z. Badania nad toksoplazmoza wsrod umyslowo chorych. 32 Burgner D, Jamieson SE, Blackwell JM. Genetic susceptibility to Bull Inst Mar Trop Med Gdansk 1953; 5: 146–173 (Summary in infectious diseases: big is beautiful, but will bigger be even better. English: Studies of toxoplasmosis among the mentally sick. Lancet Infect Dis 2006; 6: 653–663. Bull Inst Mar Trop Med Gdansk 1953; 5: 142–145. 33 Jones JL, Kruszon-Moran D, Wilson M, McQuillan G, Navin T, 8 Vojtechovska M, Vojtechovsky M, Petru M. Some problems of McAuley JB. Toxoplasma gondii infection in the United States: parasitology in mental patients (in Czechoslovakian). Casopis seroprevalence and risk factors. Am J Epidemiol 2001; 154: Lekaru Ceskych 1956; 95: 559–566. 357–365. 9 Morozov VM. On the problem of the viral aetiology of schizo- 34 Jackson MH, Hutchison WM. The prevalence and source of phrenia. J Neuropath Psychiat, Korsakov 1954; 54: 732–734. Toxoplasma infection in the environment. Adv Parasitol 1989; 10 Rimon R, Halonen P. Herpes simplex virus infection and 28: 55–105. depressive illness. Dis Nerv Sys 1969; 30: 338–340. 35 Beverley JKA, Fleck DG, Kwantes W, Ludlam GB. Age-sex 11 Crow TJ, Ferrier IN, Johnstone EC, Macmillan JF, Owens DGC, distribution of various diseases with particular reference to Parry RP et al. Characteristics of patients with schizophrenia or toxoplasmic lymphadenopathy. J Hyg (Camb) 1976; 76: 215–228. and virus-like agents in cerebrospinal fluid. 36 Bannister B. Toxoplasmosis 1976–80: review of laboratory reports Lancet 1979; 1: 842–844. to the Communicable Disease Surveillance Centre. J Infect 1982; 5: 12 Chacon C, Monro M, Harper I. Viral infection and psychiatric 301–306. disorders. Acta Psychiat Scand 1975; 51: 101–103. 37 Tizard IR, Fish NA, Quinn JP. Some observations on the 13 Torrey EF, Peterson MR. Slow and latent viruses in schizophrenia. epidemiology of toxoplasmosis in Canada. J Hyg (Camb) 1976; Lancet 1973; 2: 22–24. 77: 11–21. 14 Torrey EF, Yolken RH, Winfrey CJ. Cytomegalovirus antibody in 38 Ades AE, Parker S, Gilbert R, Tookey PA, Berry T, Hjelm M et al. cerebrospinal fluid of schizophrenic patients detected by enzyme Maternal prevalence of toxoplasma antibody based on anonymous immunoassay. Science 1982; 216: 892–894. neonatal serosurvey: a geographical analysis. Epidemiol Infect 15 Morozov PV (ed). Research on the Viral Hypothesis of Mental 1993; 110: 127–133. Disorders. Karger: Basel, 1983. 39 Lutjen S, Soltek S, Virna S, Deckert M, Schluter D. Related articles, 16 Torrey EF. Functional psychoses and viral encephalitis. Integr organ- and disease-stage-specific regulation of Toxoplasma gondii- Psychiatry 1986; 4: 224–236. specific CD8-T-cell responses by CD4T cells. Infect Immun 2006; 17 Little JT, Sunderland T. Psychosis secondary to encephalitis and 74: 5790–5801. encephalopathies. Semin Clin Neuropsychiatry 1998; 3: 4–11. 40 Tato CM, Mason N, Artis D, Shapira S, Caamano JC, Bream JH et al. 18 Caroff SN, Mann SC, Gliatto MF, Sullivan KA, Campbell EC. Opposing roles of NF-kappaB family members in the regulation of Psychiatric manifestations of acute viral encephalitis. Psych NK cell proliferation and production of IFN-gamma. Int Immunol Annals 2001; 31: 193–204. 2006; 18: 505–513. 19 Torrey EF, Miller J, Rawlings R, Yolken RH. Seasonality of births in 41 Lang C, Algner M, Beinert N, Gross U, Luder CG. Diverse schizophrenia and bipolar disorder: a review of the literature. mechanisms employed by Toxoplasma gondii to inhibit Schizophr Res 1997; 28: 1–38. IFN-gamma-induced major histocompatibility complex class II 20 Eaton WW, Mortensen PB, Frydenberg M. Obstetric factors, gene expression. Microbes Infect 2006; 8: 1994–2005. urbanization and psychosis. Schizophr Res 2000; 43: 117–123. 42 Schluter D, Kwok LY, Lutjen S, Soltek S, Hoffmann S, Korner H 21 Pedersen CB, Mortensen PB. Evidence of a dose–response et al. Both lymphotoxin-alpha and TNF are crucial for control of relationship between urbanicity during upbringing and schizo- Toxoplasma gondii in the central nervous system. J Immunol 2003; phrenia risk. Arch Gen Psychiatry 2001; 58: 1039–1046. 170: 6172–6182.

Molecular Psychiatry Psychosis and microbial agents RH Yolken and EF Torrey 479 43 Kasper L, Courret N, Darche S, Luangsay S, Mennechet F, Minns L schizophrenia. Eur Arch Psychiatry Clin Neurosci 2002; 252: et al. Toxoplasma gondii and mucosal immunity. Int J Parasitol 232–234. 2004; 34: 401–409. 67 Hurme M, Haanpaa M, Nurmikko T, Wang XY, Virta M, Pessi T 44 Yarovinsky F, Kanzler H, Hieny S, Coffman RL, Sher A. Toll-like et al. IL-10 gene polymorphism and herpesvirus infections. JMed receptor recognition regulates immunodominance in an anti- Virol 2003; 70(Suppl 1): S48–S50. microbial CD4 þ T cell response. Immunity 2006; 25: 655–664. 68 Chiavetto LB, Boin F, Zanardini R, Popoli M, Michelato A, Bignotti S 45 Khan IA, Thomas SY, Moretto MM, Lee FS, Islam SA, Combe C et al. Association between promoter polymorphic haplotypes of et al. CCR5 is essential for NK cell trafficking and host survival interleukin-10 gene and schizophrenia. Biol Psychiatry 2002; 51: following Toxoplasma gondii infection. PLoS Pathog 2006; 2: e49. 480–484. 46 Buzoni-Gatel D, Werts C. Toxoplasma gondii and subversion of the 69 Hanshaw JB, Scheiner AP, Moxley AW, Gaev L, Abel V, Scheiner B. immune system. Trends Parasitol 2006; 22: 448–452. School failure and deafness after ‘silent’ congenital cytomegalo- 47 de Moura L, Bahia-Oliveira LMG, Wada MY, Jones JL, Tuboi SH, virus infection. N Engl J Med 1976; 295: 468–470. Carmo EH et al. Waterborne toxoplasmosis, Brazil, from field to 70 Boppana SB, Rivera LB, Fowler KB, Mach M, Britt WJ. Intrauterine gene. Emerg Infect Dis 2006; 12: 326–329. transmission of cytomegalovirus to infants of women with 48 Kramer W. Frontiers of neurological diagnosis in acquired preconceptional immunity. N Engl J Med 2001; 344: 1366–1371. toxoplasmosis. Psychiatr Neurol Neurochir 1966; 69: 43–64. 71 Duchowny M, Caplan L, Siber G. Cytomegalovirus infection of the 49 Estran C, Brenier-Pinchart MP, Pelletier L, Cesbron-Delauw MF, adult nervous system. Ann Neurol 1979; 5: 458–461. Pelloux H. Toxoplasma gondii in human astrocytes in vitro: 72 Dickerson F, Kirkpatrick B, Boronow J, Stallings C, Origoni A, interleukin (IL)-12 and IL-10 do not influence cystogenesis. Yolken R. Deficit schizophrenia: association with serum J Parasitol 2006; 92: 1108–1110. antibodies to cytomegalovirus. Schizophr Bull 2006; 32: 396–400. 50 Torrey EF, Bartko JJ, Lun Z-R, Yolken RH. Antibodies to 73 Hoff AL, DeLisi LE, Yolken RH. Relationship of viral infection to Toxoplasma gondii in patients with schizophrenia: a meta- progressive MRI and neuropsychological change in a 10 year analysis. Schizophr Bull 2007; 33: 729–736. follow-up study of first episode schizophrenia (abstract). Neuro- 51 Niebuhr DW, Millikan AM, Cowan DN, Yolken R, Li Y, Weber NS. psychopharmacology 2004; 29(Suppl 1): S164. Selected infectious agents and risk of schizophrenia among U.S. 74 Hanshaw JB. Cytomegalovirus. In: Remington JS, Klein JO (eds). Military personnel. Am J Psychiatry 2008; 165: 99–106. Infectious Diseases of the Fetus and Newborn Infant. Saunders: 52 Saeij JP, Coller S, Boyle JP, Jerome ME, White MW, Boothroyd JC. Philadelphia, 1976, p 127. Toxoplasma co-opts host gene expression by injection of a 75 Lokensgard JR, Cheeran MC-J, Hu S, Gekker G, Peterson PK. polymorphic kinase homologue. Nature 2007; 445: 324–327. Glial cell responses to herpesvirus infections: role in defense 53 Taylor S, Barragan A, Su C, Fux B, Fentress SJ, Tang K et al. A and immunopathogenesis. J Infect Dis 2002; 186(Suppl 2): secreted serine-threonine kinase determines virulence in the S171–S179. eukaryotic pathogen Toxoplasma gondii. Science 2006; 314: 76 Rothschild DM, O’Grady M, Wecker L. Neonatal cytomegalovirus 1776–1780. exposure decreases prepulse inhibition in adult rats: implications 54 Webster JP, Lamberton PH, Donnelly CA, Torrey EF. Parasites as for schizophrenia. J Neurosci Res 1999; 57: 429–434. causative agents of human affective disorders? The impact of 77 Yolken RH, Torrey EF. Viruses, schizophrenia, and bipolar anti-psychotic, mood-stablizer and anti-parasite medication on disorder. Clin Microbiol Rev 1995; 8: 131–145. Toxoplasma gondii’s ability to alter host behaviour. Proc Biol Sci 78 Torrey EF, Leweke MF, Schwarz MJ, Mueller N, Bachmann S, 2006; 273: 1023–1030. Schroeder J et al. Cytomegalovirus and schizophrenia. CNS Drugs 55 Yereli K, Balciog˘lu IC, O¨ zbilgin A. Is Toxoplasma gondii a 2006; 20: 879–885. potential risk for traffic accidents in Turkey? Forensic Sci Int 79 Leweke FM, Gerth CW, Koethe D, Klosterkotter J, Ruslanova I, 2006; 163: 34–37. Krivogorsky B et al. Antibodies to infectious agents in individuals 56 Flegr J, Havlicek J, Kodym P, Maly M, Smalhel Z. Increased risk with recent onset schizophrenia. Eur Arch Psychiatr Clin Neurosci of traffic accidents in subjects with latent toxoplasmosis: 2004; 254: 4–8. a retrospective case control study. BMC Infect Dis 2002; 2: 11–16. 80 Moises HW, Ru¨ ger R, Reynolds GP, Fleckenstein B. Human 57 Mohammed AK, Magnusson O, Maehlen J, Fonnum F, Norrby E, cytomegalovirus DNA in the temporal cortex of a schizophrenic Schultzberg M et al. Behavioural deficits and serotonin depletion patient. Eur Arch Psychiatr Neurol Sci 1988; 238: 110–113. in adult rats after transient infant nasal viral infection. 81 Jones-Brando L, Torrey EF, Yolken RH. Drugs used in the treatment Neuroscience 1990; 35: 355–363. of schizophrenia and bipolar disorder inhibit the replication of 58 Carruthers VB, Suzuki Y. Effects of Toxoplasma gondii infection Toxoplasma gondii. Schizophr Res 2003; 62: 237–244. on the brain. Schizophr Bull 2007; 33: 745–751. 82 Dickerson FB, Boronow JJ, Stallings CR, Origoni AE, Yolken RH. 59 Radke JR, Behnke MS, Mackey AJ, Radke JB, Roos DS, White MW. Reduction of symptoms by valacyclovir in cytomegalovirus- The transcriptome of Toxoplasma gondii. BMC Biol 2005; 3:26. seropositive individuals with schizophrenia. Am J Psychiatry 60 Czub S, Koutsilieri E, Sopper S, Czub M, Stahl-Hennig C, Mu¨ller JG 2003; 160: 2234–2236. et al. Enhancement of central nervous system pathology in early 83 Mueller N, Riedel M, Scheppach C, Brandstatter B, Sokullu S, simian immunodeficiency virus infection by dopaminergic drugs. Krampe K et al. Beneficial antipsychotic effects of celecoxib add- Acta Neuropathol 2001; 101: 85–91. on therapy compared to risperidone alone in schizophrenia. Am J 61 Ashdown H, Dumont Y, Ng M, Poole S, Boksa P, Luheshi GN. The Psychiatry 2002; 159: 1029–1034. role of cytokines in mediating effects of prenatal infection on the 84 Martelius TJ, Wolff H, Bruggeman CA, Hockerstedt KA, Lautens- fetus: implications for schizophrenia. Mol Psychiatry 2006; 11: 47–55. chlager IT. Induction of cyclo-oxygenase-2 by acute liver allograft 62 Hanshaw JB. Congenital cytomegalovirus infection: a fifteen year rejection and cytomegalovirus infection in the rat. Transpl Int perspective. J Infect Dis 1971; 123: 555–561. 2002; 15: 610–614. 63 Hurme M, Helminen M. Resistance to human cytomegalovirus 85 Zhu H, Cong JP, Yu D, Bresnahan WA, Shenk TE. Inhibition of infection may be influenced by genetic polymorphisms of the cyclooxygenase 2 blocks human cytomegalovirus replication. Proc tumour necrosis factor-alpha and interleukin-1 receptor antagonist Natl Acad Sci USA 2002; 99: 3932–3937. genes. Scand J Infect Dis 1998; 30: 447–449. 86 Dickerson MC, Johnston J, Delea TE, White A, Andrews E. The 64 Schwab SG, Mondabon S, Knapp M, Albus M, Hallmayer J, causal role for genital ulcer disease as a risk factor for transmission Borrmann-Hassenbach M et al. Association of tumor necrosis of human immunodeficiency virus. An application of the Bradford factor alpha gene-G308A polymorphism with schizophrenia. Hill criteria. Sex Transm Dis 1996; 23: 429–440. Schizophr Res 2003; 65: 19–25. 87 Yolken RH, Torrey EF. Infectious agents and gene-environmental 65 Boin F, Zanardini R, Pioli R, Altamira CA, Maes M, Gennarelli M. interactions in the etiopathogenesis of schizophrenia. Clin Association between -G308A tumor necrosis factor alpha gene Neurosci Res 2006; 6: 97–109. polymorphism and schizophrenia. Mol Psychiatry 2001; 6: 79–82. 88 Jones-Brando L, D’Angelo J, Posner GH, Yolken R. In vitro 66 Riedel M, Kro¨nig H, Schwarz MJ. No association between the inhibition of Toxoplasma gondii by four new derivatives of G308A polymorphism of the tumor necrosis factor-alpha gene and artemisinin. Antimicrob Agents Chemother 2006; 50: 4206–4208.

Molecular Psychiatry