Literature Review
A review of the literature was done using the Cochrane Database. Databases searched included MEDLINE, PubMed, EMBASE, and EMBASE reviews. The search terms were autoantibody, autoimmunity, immune dysfunction, and schizophrenia. Both investigators reviewed the search results. Published studies were included in this review if they reported on objective measurements or health outcomes in patients diagnosed as having schizophrenia; extracted information included the study design, the immunologic basis for the findings, the characteristics of the study population, and the main outcome measure. One hundred eighteen articles met the inclusion criteria. Three tertiary references and one conference proceedings were also included.
Heredity and Pedigree Studies
Evidence of an association between schizophrenia and a wide range of autoimmune disorders emerged from an evaluation of data from Danish national registries on patients diagnosed with schizophrenia (n = 7704) and their parents. Nine autoimmune disorders were more prevalent among the patients with schizophrenia relative to matched comparison subjects without schizophrenia (incidence rate ratios [IRRs] for the nine disorders ranged from 1.5 to 12.9). Twelve autoimmune diseases were more common among the parents of patients with schizophrenia (calculated IRRs for the disorders ranged from 1.3 to 3.8) than among the comparison parent population. Overall, the results indicated that "having a history of any autoimmune disease was associated with a 45% increase in risk for schizophrenia" (IRR, 1.45; 95% confidence interval [CI], 1.25–1.68), the investigators reported. They speculated that the human leukocyte antigen (HLA) system may play a role in the development of schizophrenia either because of direct involvement of HLAs or because of the proximity of the loci for auto-immune disorders and schizophrenia in the HLA regions. Interestingly, the study found that rheumatoid arthritis (RA) was neither more nor less common among the patients with schizophrenia than among the comparison subjects; other research has suggested an association between RA and schizophrenia, but the evidence on this topic is inconsistent.
Wright et al. found that patients with schizophrenia who had a first-degree relative with schizophrenia were significantly more likely to have a first-degree relative with an auto-immune disease (p = 0.0003). Other research, however, has found no correlation between schizophrenia and severe immune system compromise or HIV infection.
Some etiologic studies have called into question the hypothesis that schizophrenia might be the result of neurotransmitter-receptor abnormalities, suggesting that schizophrenia might instead be the product of a different etiology that also results in neurotransmitter abnormalities. However, other research has suggested an association between schizophrenia and a number of neuroanatomical abnormalities, including changes in brain function (impaired stimulus response activation), electrical activity abnormalities, neurochemical abnormalities, and "neuronal disorganization." Steen and associates conducted a meta-analysis of studies of magnetic resonance imaging (MRI) findings in adult patients with first-episode schizophrenia. The results suggested that compared with healthy controls, the patients with schizophrenia had reduced average whole-brain volume (21 studies, p < 0.0001), reduced hippocampal volume (10 studies, p < 0.0001), and increased ventricular volume (11 studies, p < 0.0001); the limitations of MRI technology were acknowledged. Neuropathologic changes have been noted among patients with schizophrenia, suggesting that the disorder may contribute to neuronal disorganization in cortical and limbic regions of the brain. Reported neurobiological differences between patients with schizophrenia and family members may be a result of genetic differences or in utero insults (e.g., maternal infection, maternal immune system activation) or may involve neurodevelopmental abnormalities secondary to insults after birth, including immune system activation during development.
Prenatal and perinatal environmental factors linked to an increased risk of schizophrenia include maternally mediated variables (infections, nutritional deficiencies, severe stress, obstetric and perinatal complications), more advanced paternal age, and the season of birth. Several of these environmental risk factors suggest immune system dysfunction or autoimmunity as a possible etiology.
Immunogenicity Studies
The immune system has two main components: innate immunity and adaptive immunity. Innate immunity involves a rapid-response system composed of physical defense systems (e.g., skin, mucus) and specific immune cells, including monocytes, macrophages, neutrophils, mast cells, eosinophils, natural killer cells, and dendritic cells, all of which attack or process the invading pathogen (antigen) in some way, including the production of chemokines, inflammatory cytokines, and antimicrobial peptides. This antigen processing leads to the activation of adaptive immunity, which involves a response to specific antigens that provides long-term immunity. In addition, the immune system uses major histocompatibility complex (MHC) molecules to determine what is "self" and "nonself."
Cytokines are soluble factors secreted by both components of the immune system and have a wide range of effects on immune system function. Some cytokines are pro-inflammatory, enhancing the body's ability to fight a foreign pathogen; some are antiinflammatory, keeping the immune system response proportionate to the need, shutting down the immune response when it is no longer needed, and preventing the immune system from responding to self, or benign, pathogens.
The immune system may malfunction in numerous ways. There may be genetic or environmental alterations of normal self molecules that make them more antigenic, allowing T cells to recognize them as antigens. The body may recognize molecules from pathogens that are very similar to human molecules (molecular mimicry) and produce antibodies that recognize both the pathogen and the similar self molecule to which it is usually tolerant.
Maternal Infection during Pregnancy
A number of recently published studies investigated potential links between maternal infection during pregnancy and the development of schizophrenia in offspring. In a long-term birth-cohort study of 64 matched cases, Brown et al. found that maternal influenza during the first trimester (as determined by patient histories and analysis of archived serum samples) was associated with a sevenfold increase in schizophrenia risk in offspring. Another study found that fetal exposure to genital and reproductive system infections during the periconceptional period was associated with a fivefold higher risk of schizophrenia later in life (p = 0.001).
In a small case–control study (a total of 40 cases and controls), highly elevated levels of maternal Toxoplasma immunoglobulin G (IgG) antibodies were linked to a marginally increased risk of schizophrenia in offspring (p = 0.051). Two other studies also found an association between maternal Toxoplasma antibodies and schizophrenia and affective psychoses.
Using stored blood samples obtained at the end of pregnancy from the mothers of patients with schizophrenia (n = 27), Buka et al. linked increased schizophrenia risk to the presence of maternal antibodies to herpes simplex virus (HSV)-2 in maternal blood (p = 0.04 for HSV-2 virion antigen, p = 0.02 for HSV-2 IgG-2 glycoprotein), but another small case–control study by Brown and associates (16 cases and 24 matched controls) found no evidence of an increase in schizophrenia risk associated with maternal HSV-2 infection. Another study of mothers with serologic evidence of HSV-2 infection found a significant risk for the development of schizophrenia in their adult offspring (n = 108) relative to control subjects (n = 303) whose mothers were seropositive but who did not develop schizophrenia (odds ratio, 1.8; 95% CI, 1.1–3.0).
Two studies found an association between prenatal rubella exposure and a subsequent risk of schizophrenia-spectrum disorders, though a third study found no such association. The volume of work in this area supports the theory that immune system dysfunction may play a role in at least some cases of schizophrenia.
One proposed mechanism for the increased risk of schizophrenia after maternal infection is that maternal antibodies cross the placenta and interact with fetal brain antigens, disrupting fetal brain development. Related theories center on speculation that increased maternal cytokines secondary to maternal immune system activation may cross the placenta and damage the fetal brain. Another theory holds that fetal immune homeostasis and MHC protection from maternal immune attack may be disrupted, thereby reducing protection of the fetus from damaging inflammatory cytokines.
Theories regarding how inflammatory cytokines damage the fetal brain include mediation of the alteration of the blood-brain barrier, intravascular cell adhesion, coagulation, thrombosis, and vasoconstriction that may lead to endothelial damage and subsequent hemorrhage from germinal matrix vessels into the ventricles. Additionally, inflammatory cytokines may damage white matter in a variety of ways, including "adversely influencing oligodendrocytes, astrocytes, and myelin." This type of damage may lead to alterations in later development consistent with the findings in cases of schizophrenia. Maternal genital and reproductive infections have been postulated to directly infect the infant via contact with the perineum, vagina, or cervix, or the infection can cross the placenta to infect the infant. Viral mechanisms and increased permeability of the blood-brain barrier have been proposed as potential mechanisms. Given that the etiology of schizophrenia is probably heterogeneous, one or more of these maternal infections could play a role in its development.
Studies of Cytokine Alterations
Altered cytokine activity may also play a role in the etiology of schizophrenia. One study found significantly (p = 0.008) elevated levels of interleukin (IL)-6 in patients with schizophrenia or schizoaffective disorder (n = 50) and their relatives with mood disorders (n = 48) compared with healthy matched controls (n = 48) and relatives of patients without mental disorders (n = 41). Another study (involving 40 patients and 20 controls) found elevated levels of the chemokine CCL11 in patients with schizophrenia compared with controls (p < 0.01); the investigators suggested that because CCL11 is associated with a chemokine receptor expressed on T-helper (Th) cell type 2 (Th2) lymphocytes, mast cells, and eosinophils, higher serum levels of CCL11 in patients with schizophrenia support the theory that an imbalance of Th cell type 1 (Th1) and Th2 contributes to the disorder. In addition, a meta-analysis of 62 studies of cytokine alterations involving a total of 2298 patients with schizophrenia and 1858 healthy volunteers indicated that schizophrenia is associated with increased levels of IL-1 receptor antagonist (n = 271, p = 0.0001), serum IL-2 receptors (n = 1126, p = 0.0001), and IL-6 (n = 1219, p = 0.001) in vivo and decreased levels of IL-2 (n = 1149, p = 0.023) in vitro. This body of data suggested that at least some patients with schizophrenia have altered cytokine levels that may contribute to the development and course of schizophrenia.
Inflammatory Cytokines
Th1 cytokines, such as interferon (INF)-γ and IL-2, stimulate cell-mediated immunity and are usually considered proinflammatory. Th2 cytokines, such as IL-4, stimulate humoral immunity. Some studies have documented an increase in Th2 activity in schizophrenia, with a concomitant decrease in Th1 activity. In a study involving 100 inpatients, Avgustin et al. found increases in both Th1 and Th2 activities in patients experiencing acute exacerbations of schizophrenia compared with healthy controls, with a "relative predominance of Th2 immunity." Patients in this study (n = 100) met Diagnostic and Statistical Manual of Mental Disorders, 4th Edition, Text Revision (DSM-IV-TR) criteria for schizophrenia, had a mean age of 38 years, were hospitalized for an acute exacerbation of psychosis, and had received antipsychotic medication for at least three weeks; controls (n = 34) were matched by age and sex. Exclusion criteria for both patients and controls included acute or chronic infection or inflammatory responses, other chronic illnesses known to change immune function, any neurological disorder, and alcohol or drug abuse. In vitro secretion of INF-γ by peripheral blood mononuclear cells differed significantly between patients and controls (p = 0.037), as did the secretion of IL-4 (p < 0.001). The ratio of INF-γ (proinflammatory) to IL-4 (antiinflammatory) cytokines was 1022 in patients with schizophrenia and 1432 in controls (p = 0.005).
In a prospective case–control study, Riedel et al. evaluated the type IV hypersensitivity reaction, a Th1-mediated immune response, in a group of patients with schizophrenia and a group of age- and sex-matched control subjects without the disorder; the patients were hospitalized secondary to an acute psychotic episode and had not received any anti-psychotic medication in the previous four weeks. The exclusion criteria for both groups included acute or chronic infections, skin allergies or other dermatologic abnormalities, recent immunization with live vaccines, and immunosuppressive therapy within the preceding four weeks. The investigators performed an intracutaneous antigen test on the forearms of 30 patients and 30 controls. Relative to the controls, the patients with schizophrenia had significantly fewer total positive skin reactions (p = 0.03), and significantly fewer patients than controls reacted to tetanus (p = 0.001) and diphtheria (p = 0.021) antigen tests. In aggregate, these results indicated a decreased Th1 response in patients with schizophrenia. Because all components of the immune system must work together for proper immune function, an imbalance of any kind has the potential to cause significant problems.
Autoimmune Response
Autoimmune T-cell deficiency is another form of immune dysfunction reported in patients with schizophrenia. It has been theorized that reduced T-cell response (as evidenced by decreases in T-cell-produced cytokines) in some adult patients with schizophrenia may be a result of neurodevelopmental damage secondary to maternal immune system activation. The investigators suggested that cytokine activity or inflammation in the fetal brain might disrupt the process by which T cells differentiate, leading to a long-term reduction in the availability of mature naive T cells. Although these theories provide a potential mechanism for the development of some cases of schizophrenia, they require further investigation and validation.
Autoantibody production, the hallmark sign of many autoimmune diseases, has been documented in patients with schizophrenia. Three recent studies revealed differences in levels of anticardiolipin antibodies (aCL) and other phospholipids between patients with schizophrenia and control subjects. One of the studies involved 23 patients with schizophrenia who were hospitalized for an acute psychotic episode and a control group of 20 age- and sex-matched healthy volunteers. The study patients had not received antipsychotic medication for three months before the study, but neuroleptic medications were started on admission; the exclusion criteria included any psychiatric comorbidity and acute or chronic immune pathology. Blood samples were taken on hospital admission (t1) and 42 days later (t2). The investigators found that significantly more patients than controls tested positive for immunoglobulin (Ig) M aCL at both t1 (p = 0.039) and t2 (p = 0.014); there was no significant difference between the groups with regard to IgG aCL. In addition, patients with schizophrenia had higher levels of autoantibodies to both IgM aCL (at t1, p = 0.014; at t2, p = 0.0003) and IgG aCL (at t1, p = 0.0001; at t2, p = 0.0001) relative to the controls.
Other studies have linked auto-antibodies to nerve growth factor (NGF), a neurospecific protein, to "positive symptoms" of schizophrenia (i.e., symptoms that appear to reflect an excess or distortion of normal function). One study investigated leukocyte elastase (LE) activity and antibodies to NGF in the serum of patients with acute-stage schizophrenia. In that study, two groups of Russian patients with a mean duration of illness of 16 months (n = 53) or 89 months (n = 18) who had not received psychotropic medications for two weeks before the study comprised the patient group. Positive symptoms of schizophrenia were associated with high levels of antibodies to NGF, and negative symptoms were associated with higher LE activity.
Other antibodies for which positive associations with schizophrenia have been reported include platelet-associated antibodies; muscarinic acetylcholine-receptor (mAChR) antibodies, specifically M1 and M2 antibodies; and antibodies to the α-7 subunit of the acetylcholine nicotinic receptor (α7AChNR) in patients with schizophrenia as compared with matched controls. The α7AChNR receptors modulate γ-aminobutyric acid and glutamate neurotransmitter release, and reduced levels of those neurotransmitters are common findings in schizophrenia. According to Gallowitsch-Puerta and Tracey, α7AChNR also plays a role in balancing the proinflammatory and antiinflammatory pathways of the immune system and reinforces the theory of Th1–Th2 imbalance in schizophrenia. A variety of antibodies may play a role in the pathophysiology of schizophrenia, but more research is necessary to replicate the research findings to date and provide additional evidence of a connection between immune system dysfunction and schizophrenia.
Genetic Studies
According to Tandon et al., schizophrenia "heritability is high, there is no 'major' gene locus that could explain a substantial portion of the heritability though a large number of candidate susceptibility genes may contribute, no gene appears to be either sufficient or necessary for the development of schizophrenia, and inconsistent study replication prevents the consideration of any single allelic variant as a gene for schizophrenia with absolute certainty at this time." However, three recent genome-wide association studies (GWAS) found a significant association between schizophrenia and the MHC on chromosome 6. The studies involved a total sample of 8,008 cases of schizophrenia and 19,077 controls of European ancestry. According to the investigators in those studies, his-tones neutralize bacterial endotoxins in the human placenta; the disruption of that infection barrier might increase susceptibility to infection and increase susceptibility to schizophrenia, they theorized. Additionally, the MHC region has been associated with other immune-related disorders such as type 1 diabetes, RA, and Crohn's disease.
Genetic changes in the IL-2 and IL-4 genes are associated with schizophrenia, which gives further support to the Th1–Th2-imbalance theory of schizophrenia etiology. In a study involving 230 unrelated Caucasian patients with schizophrenia and a control group of 251 healthy Caucasian subjects, the patients and controls differed with respect to both the IL-2 single nucleotide polymorphism (SNP) of the TT genotype (p = 0.024) and the IL-4 SNP of the CC genotype (p = 0.026). Another GWAS indicated significant associations between schizophrenia and cytokine-receptor abnormalities within the colony-stimulating-factor receptor 2-α gene (p = 0.016) and the IL-3 receptor-α gene (p = 0.027). These findings lend credence to the immune-dysfunction theory of schizophrenia etiology.