Highlights
-
- Epidemiological evidence suggests a potential relationship between viruses and schizophrenia.
-
- Several viral infections can disrupt fetal brain development by activating the maternal immune system.
-
- Increases in inflammatory cytokine levels and changes in the expression of key genes may be possible links between viral infections and schizophrenia.
-
- Both immune and non-immune genes associated with schizophrenia are likely to be targets of viral proteins.
-
- Neuropsychiatric effects of antiviral drugs are common side effects that complicate viral treatment.
Introduction
The suggestion that microorganisms might provoke psychotic disorders in humans was established a long time ago. In the late nineteenth century, the Germ Theory of Disease was postulated, establishing that bacteria could be the etiologic agents related to several psychiatric disorders, including dementia praecox. At the time, the impossibility of a direct bacterial infection of the brain was verified, but it could occur indirectly through the secretion of microbial toxins in a process known as autointoxication (Yolken & Torrey, Reference Yolken and Torrey2008). However, the Theory of Autointoxication fell into disrepute after several researchers attempted to surgically remove the infection sites in various organs, often with tragic results (Noll, Reference Noll2006). After the influenza pandemic in 1918–1919, it was found that viral infection caused psychosis in many affected individuals (Menninger, Reference Menninger1994; Kępińska et al., Reference Kępińska, Iyegbe, Vernon, Yolken, Murray and Pollak2020). In the 1980s, this virus hypothesis gained renewed attention due to the occurrence of herpesvirus infections in the cerebrospinal fluid (CSF) of schizophrenic subjects (Morozov, Reference Morozov1983).
Despite controversial findings, an extensive corpus of evidence links microbial infections to neuropsychiatric disorders, with host genes influencing microbial virulence and pathogens affecting host genes, including those related to schizophrenia (Krause et al., Reference Krause, Matz, Weidinger, Wagner, Wildenauer and Obermeier2010; Pouget et al., Reference Pouget, Han, Wu, Mignot and Ollila2019; Yolken, Reference Yolken2023). Furthermore, research on the gut-brain axis shows that bacteria can influence brain function and behavior without directly replicating in the central nervous system (CNS) (Borrego-Ruiz & Borrego, Reference Borrego-Ruiz and Borrego2024a; Hashimoto, Reference Hashimoto2023). In this regard, the gut microbiome and schizophrenia exhibit a bidirectional relationship, as translocated gut microbial metabolites can negatively impact brain connectivity, inducing neuroinflammation, a critical factor in schizophrenia (Kannan et al., Reference Kannan, Gressitt, Yang, Stallings, Katsafanas and Schweinfurth2017). In fact, this mental condition can alter the function of endothelial cells and disrupt the permeability of the blood–brain barrier (BBB), causing inflammation and the translocation of inflammatory cells into the brain (Khandaker & Dantzer, Reference Khandaker and Dantzer2016; Severance & Yolken, Reference Severance and Yolken2020).
Diverse pathogenic microorganisms have been associated with psychotic symptoms, including those characteristic of schizophrenia. These pathogens encompass (i) bacteria, such as Treponema pallidum, Mycoplasma pneumoniae, Chlamydia trachomatis, and Borrelia burgdorferi; (ii) viruses, such as herpesviruses, orthomyxoviruses, rubiviruses, paramyxoviruses, picornaviruses, poxviruses, enteroviruses, arboviruses, retroviruses, and Borna disease virus; and (iii) protozoa, such as Toxoplasma gondii (Caroff, Mann, Gliatto, Sullivan, & Campbell, Reference Caroff, Mann, Gliatto, Sullivan and Campbell2001; Yolken & Torrey, Reference Yolken and Torrey2008). Although various of these pathogenic microorganisms have been extensively studied, interest has converged on a few viruses (Prasad, Watson, Dickerson, Yolken, & Nimgaonkar, Reference Prasad, Watson, Dickerson, Yolken and Nimgaonkar2012; Watson et al., Reference Watson, Prasad, Klei, Wood, Yolken and Gur2013). In addition, a better understanding of neurodevelopment and neuroinflammation has provided insights into how viruses, either alone or in combination with other microorganisms or host variables, may impact the etiopathogenesis of schizophrenia (Meyer, Reference Meyer2013; Tran, Lee, & Cho, Reference Tran, Lee and Cho2022).
Congenital infections are a major cause of childhood neurodevelopmental disabilities, contributing to structural brain abnormalities that can lead to severe impairments such as cerebral palsy, epilepsy, and neurosensory deficits (Fortin & Mulkey, Reference Fortin and Mulkey2023). Meyer, Yee, & Feldon (Reference Meyer, Yee and Feldon2007) previously suggested that infection-associated immune events during early fetal life may have a greater neurodevelopmental impact than infections occurring later in pregnancy. This is because infections in early gestation can disrupt key neurodevelopmental processes, such as cell proliferation and differentiation, and may predispose the developing nervous system to subsequent disruptions in cell migration, target selection, and synapse maturation, ultimately leading to a range of brain and behavioral abnormalities in adult offspring. In this regard, evidence suggests that exposure to infectious agents, autoimmune diseases, and other stressors during fetal and postnatal life may contribute to the development of schizophrenia and other non-affective psychoses (Benros, Nielsen, Nordentoft, Eaton, Dalton, & Mortensen, Reference Benros, Nielsen, Nordentoft, Eaton, Dalton and Mortensen2011; Brown & Patterson, Reference Brown and Patterson2011; Dickerson, Severance, & Yolken, Reference Dickerson, Severance and Yolken2017; Eaton et al., Reference Eaton, Byrne, Ewald, Mors, Chen, Agerbo and Mortensen2006; Megli & Coyne, Reference Megli and Coyne2022; Severance & Yolken, Reference Severance and Yolken2020; Yolken & Torrey, Reference Yolken and Torrey2008). Proposed mechanisms include (i) excessive inflammatory cytokine production, contributing to immune dysregulation (Di Nicola et al., Reference Di Nicola, Cattaneo, Hepgul, Di Forti, Aitchison and Janiri2013); (ii) early-life disruptions in brain development (Ahmed, Ramadan, Elbeh, & Haridy, Reference Ahmed, Ramadan, Elbeh and Haridy2024); and (iii) neurotransmitter dysfunction, including glutamate, epinephrine, dopamine, serotonin, and γ-aminobutyric acid (GABA), driven by cytokine imbalance (Murray et al., Reference Murray, Sham, Van Os, Zanelli, Cannon and McDonald2004; Potvin et al., Reference Potvin, Stip, Sepehry, Gendron, Bah and Kouassi2008). Within this context, epidemiological studies on maternal infections during pregnancy have provided contradictory results regarding microbial exposure, but have reported consistent associations with T. gondii maternal transmission (Blomström, Gardner, Dalman, Yolken, & Karslsson, Reference Blomström, Gardner, Dalman, Yolken and Karslsson2015; Cheslack-Postava & Brown, Reference Cheslack-Postava and Brown2022). Interestingly, other studies indicate that the association between early-life infection and later schizophrenia development is not completely explained by shared familial factors or by genetic liability (Karlsson & Dalman, Reference Karlsson, Dalman, Khandaker, Meyer and Jones2020; Wahbeh & Avramopoulos, Reference Wahbeh and Avramopoulos2021).
Taking into consideration the growing evidence linking viral infections to neuropsychiatric disorders, and recognizing that both viral infections and their treatments may influence brain function, this narrative review synthesizes the current understanding of how viral infections may contribute to the development of schizophrenia, addressing the underlying mechanisms. Additionally, it explores the potential therapeutic impact of antiviral drugs in neuropsychiatric conditions, emphasizing schizophrenia as the primary focus of the review.
Methods
The present work consists of a narrative review aimed at gathering and analyzing existing literature to offer a comprehensive and thorough overview of the central topic under investigation (Agarwal, Charlesworth, & Elrakhawy, Reference Agarwal, Charlesworth and Elrakhawy2023), which contemplates the link between viral infections and schizophrenia, and also the neuropsychiatric effects of the antiviral drugs, especially on schizophrenia. The authors individually conducted an exhaustive literature search within the field pertinent to the subject of study. With this aim, PubMed, Scopus, and Web of Science databases were searched between February and March 2024, using various combinations of keywords related to the research topic, such as “viral infection,” “viruses,” “schizophrenia,” “antivirals,” or “psychiatric disorders.” The search strategy involved an iterative process and also included an examination of the reference list of previous reviews and research papers. Both authors screened all eligible records separately, and each article found was individually assessed for relevance by first screening the title and abstract. Duplicates were removed, as well as articles excluded due to their subject being out of scope for this review. The full texts of the remaining articles were carefully retrieved, and relevant data were extracted for further analysis.
Section 1: Psychopathology and pathophysiology of schizophrenia
Psychological and psychiatric aspects of schizophrenia
Schizophrenia is defined as a severe mental disorder that impacts more than 21 million individuals globally, which is related to chronic disability and to impaired cognitive, emotional, and social abilities (WHO, 2018). Symptoms of schizophrenia typically appear during adolescence or early adulthood (Orsolini, Pompili, & Volpe, Reference Orsolini, Pompili and Volpe2022; Riedmüller & Müller, Reference Riedmüller and Müller2017) and include positive symptoms (delusional beliefs, hallucinations, and formal thought disorder), negative symptoms (alogia, anhedonia, avolition, and social isolation), and cognitive impairments, including deficits in areas such as attention, working memory, information processing rate, verbal and visuospatial acquisition, analytical reasoning, and cognitive adaptability (Marder & Cannon, Reference Marder and Cannon2019; McCutcheon, Reis Marques, & Howes, Reference McCutcheon, Reis Marques and Howes2020; Moritz, Silverstein, Dietrichkeit, & Gallinat, Reference Moritz, Silverstein, Dietrichkeit and Gallinat2020; Moura et al., Reference Moura, Isvoranu, Kovacs, Van Rooijen, Van Amelsvoort and Simons2022). In addition, dysfunctions in social cognition (comprising social reasoning, emotional intelligence, emotion detection from facial cues, and general emotion assessment) can markedly hinder the functional recovery of individuals with schizophrenia, due to the adverse effects on interpersonal interactions, community integration, and vocational performance (Green, Lee, & Wynn, Reference Green, Lee and Wynn2020; McCutcheon et al., Reference McCutcheon, Reis Marques and Howes2020). Concomitant occurrence with other mental disorders contributes to elevated incidence of symptom relapse, medical confinement, propensity for suicide, and family and social problems, as well as to a higher risk of negative short-term outcomes, including higher mortality rates (Correll et al., Reference Correll, Solmi, Croatto, Schneider, Rohani-Montez and Fairley2022; Drake, Xie, & McHugo, Reference Drake, Xie and McHugo2020; Plana-Ripoll et al., Reference Plana-Ripoll, Musliner, Dalsgaard, Momen, Weye and Christensen2020).
Schizophrenia constitutes a syndrome that encompasses a compilation of clinical features stemming from diverse etiologies, etiopathogenesis, and psychopathological processes that may be involved (Barch, Karcher, & Moran, Reference Barch, Karcher and Moran2022; Insel, Reference Insel2010; van Os, Linscott, Myin-Germeys, Delespaul, & Krabbendam, Reference van Os, Linscott, Myin-Germeys, Delespaul and Krabbendam2009). Nevertheless, numerous emerging neurobiological approaches have been recently proposed regarding the pathogenesis of schizophrenia (Chekroud et al., Reference Chekroud, Bondar, Delgadillo, Doherty, Wasil and Fokkema2021; DeLisi, Reference DeLisi2022; Nasrallah, Reference Nasrallah2022), such as the following: (i) genetic factors implicated in developmental disturbances at different periods of the fetal phase that influence brain programming, leading to the manifestation of prepsychotic features during prepubertal or pubertal stages (Ahmed et al., Reference Ahmed, Ramadan, Elbeh and Haridy2024; Boog, Reference Boog2004); (ii) the neurodevelopmental model of schizophrenia, which integrates various non-genetic determinants, including perinatal challenges, immigration status, and traumatic experiences of maltreatment and neglect during the childhood stage, potentially mediating epigenetic modifications and determining structural and functional neurodevelopmental deviations (Powell, Reference Powell2010); (iii) pathological changes in several brain areas, encompassing the frontal, parietal, temporal, and cingulate regions, as well as glial components, alongside heightened synaptic pruning and/or perturbed neuroplasticity capability (Chung & Cannon, Reference Chung and Cannon2015); (iv) immune dysfunction and neuroinflammation (Di Nicola et al., Reference Di Nicola, Cattaneo, Hepgul, Di Forti, Aitchison and Janiri2013; Kannan et al., Reference Kannan, Gressitt, Yang, Stallings, Katsafanas and Schweinfurth2017); and (v) various abnormalities in neurotransmitter pathways (Murray et al., Reference Murray, Sham, Van Os, Zanelli, Cannon and McDonald2004; Nyffeler, Meyer, Yee, Feldon, & Knuesel, Reference Nyffeler, Meyer, Yee, Feldon and Knuesel2006).
Recent findings corroborate the notion that schizophrenia is a multifactorial disorder resulting from inextricable interactions between cumulative and synergistic genetic and environmental factors (van Os, Kenis, & Rutten, Reference van Os, Kenis and Rutten2010), with a highly variable and heterogeneous clinical presentation (Gur, Reference Gur2022). Hence, owing to the absence of distinct demarcations and the large amount of involved etiological determinants, pathophysiological processes, and conjectures (Cannon, Reference Cannon2022; Taghia et al., Reference Taghia, Cai, Ryali, Kochalka, Nicholas and Chen2018; Tandon et al., Reference Tandon, Gaebel, Barch, Bustillo, Gur and Heckers2013), the perspective on schizophrenia has currently been expanded to a spectrum concept in the DSM-5-TR (APA, 2022), or as a principal psychotic disorder in the ICD-11 (Gaebel & Salveridou-Hof, Reference Gaebel and Salveridou-Hof2022; WHO, 2019).
Pathophysiology of schizophrenia
Schizophrenia affects temporal brain regions, such as the hippocampus, which undergo development during prenatal periods and over an extended time frame before the initiation of psychosis (Zaidel, Esiri, & Harrison, Reference Zaidel, Esiri and Harrison1997), impeding the multiplication of neurons and the establishment of axial linkages, and thereby resulting in limited CNS development (Kotsiri et al., Reference Kotsiri, Resta, Spyrantis, Panotopoulos, Chaniotis and Beloukas2023; Sanfilipo et al., Reference Sanfilipo, Lafargue, Rusinek, Arena, Loneragan and Lautin2000). The brain of schizophrenic subjects has shown expansions in the lateral ventricles and a diminished dimension in the frontal and the temporal lobes of the cortex, along with reduced hippocampal size (Sanfilipo et al., Reference Sanfilipo, Lafargue, Rusinek, Arena, Loneragan and Lautin2000; Thaker & Carpenter, Reference Thaker and Carpenter2001).
Schizophrenia is not considered a neuroinflammatory disease because there is no proof of astrogliosis and microglial activation (De Picker et al., Reference De Picker, Victoriano, Richards, Gorvett, Lyons and Buckland2021; Najjar, Pearlman, Alper, Najjar, & Devinsky, Reference Najjar, Pearlman, Alper, Najjar and Devinsky2013; van Kesteren et al., Reference van Kesteren, Gremmels, de Witte, Hol, Van Gool and Falkai2017), or of lymphocytic infiltration (Bogerts et al., Reference Bogerts, Winopal, Schwarz, Schlaaff, Dobrowolny and Mawrin2017; Schlaaff et al., Reference Schlaaff, Dobrowolny, Frodl, Mawrin, Gos and Steiner2020). However, several astrocytic and microglial indicators exhibit dysregulation in schizophrenia (Najjar & Pearlman, Reference Najjar and Pearlman2015; Najjar et al., Reference Najjar, Pearlman, Alper, Najjar and Devinsky2013; Ramaker et al., Reference Ramaker, Bowling, Lasseigne, Hagenauer, Hardigan and Davis2017; Toker, Mancarci, Tripathy, & Pavlidis, Reference Toker, Mancarci, Tripathy and Pavlidis2018; Trépanier, Hopperton, Mizrahi, Mechawar, & Bazinet, Reference Trépanier, Hopperton, Mizrahi, Mechawar and Bazinet2016; van Kesteren et al., Reference van Kesteren, Gremmels, de Witte, Hol, Van Gool and Falkai2017). Microarray and RNA sequencing studies of the different brain regions have consistently shown upregulated or dysregulated expression of genes in pathways related to immune response and to inflammation within schizophrenia (Bergon et al., Reference Bergon, Belzeaux, Comte, Pelletier, Hervé and Gardiner2015; Chang et al., Reference Chang, Liu, Hahn, Gur, Sleiman and Hakonarson2017; Fillman et al., Reference Fillman, Cloonan, Catts, Miller, Wong and McCrossin2013; Gamazon, Zwinderman, Cox, Denys, & Derks, Reference Gamazon, Zwinderman, Cox, Denys and Derks2019; Gandal, Zhang, et al., Reference Gandal, Zhang, Hadjimichael, Walker, Chen and Liu2018; Hess et al., Reference Hess, Tylee, Barve, de Jong, Ophoff and Kumarasinghe2016; Hwang et al., Reference Hwang, Kim, Shin, Kim, Seo and Webster2013; Kim, Hwang, Lee, & Webster, Reference Kim, Hwang, Lee and Webster2016; Lanz et al., Reference Lanz, Reinhart, Sheehan, Rizzo, Bove and James2019; Lindholm Carlström et al., Reference Lindholm Carlström, Niazi, Etemadikhah, Halvardson, Enroth and Stockmeier2021; Mistry, Gillis, & Pavlidis, Reference Mistry, Gillis and Pavlidis2013). Neuropathologic studies have shown a downregulation of microglia-related transcripts in schizophrenia, which coincides with the upregulation of sequences associated with inflammation (Gandal, Haney et al., Reference Gandal, Haney, Parikshak, Leppa, Ramaswami and Hartl2018; Snijders et al., Reference Snijders, van Zuiden, Sneeboer, Berdenis van Berlekom, van der Geest and Schnieder2021). It appears that in patients with schizophrenia, microglia could remain in a dormant state, inactive with respect to preserving cerebral homeostasis (Murphy & Weickert, Reference Murphy and Weickert2021).
The consistent upregulation of immune indicators observed in schizophrenia suggests a potential involvement of viral infection and the BBB in both the etiology and neuropathology of this disorder (Webster, Reference Webster2023). These dysregulated immune-related markers varied from study to study, although the following subset of genes is frequently upregulated. SERPINA3, a serpin peptidase inhibitor, is triggered by cytokines and possesses anti-inflammatory and antioxidant properties (Sánchez-Navarro, González-Soria, Caldiño-Bohn, & Bobadilla, Reference Sánchez-Navarro, González-Soria, Caldiño-Bohn and Bobadilla2021). Its upregulation in astrocytes at the BBB suggests that the astrocytes may be mounted against the infectious agent as a counteracting defensive mechanism, leading to an increase in interferon-induced transmembrane (IFITM) proteins within the endothelial cells (Murphy et al., Reference Murphy, Kondo, Walker, Rothmond, Matsumoto and Shannon Weickert2020). IFITM proteins are recognized for their role in defending various viruses, inhibiting viral cytosolic access, and thereby impeding an initial stage of viral replication (Ren et al., Reference Ren, Du, Xu, Li, Wu and Jin2020). GBP2 protects against viral infection in the innate immune system (Braun et al., Reference Braun, Hotter, Koepke, Zech, Groß and Sparrer2019). CD163, a cysteine-rich scavenger receptor of macrophages, has multiple functions as an innate immune sensor and as an inducer of local inflammation (Etzerodt & Moestrup, Reference Etzerodt and Moestrup2013). Interestingly, after intracerebral hemorrhage, CD163 mediates the internalization into macrophages of haptoglobin-hemoglobin complexes, which is essential for the clearance of hemoglobin from erythrocytes lysis (Garton, Keep, Hua, & Xi, Reference Garton, Keep, Hua and Xi2017). Table 1 shows the major immune-related dysregulated markers in schizophrenia.
Immune-related dysregulated markers in schizophrenia
Note: DEGs: differentially expressed genes.
Cytokines released by innate and adaptive immune cells act as signaling molecules and exert influence on the CNS and also on the peripheral immune system. In schizophrenia, changes in peripheral cytokine may affect the brain through multiple mechanisms (Khandaker & Dantzer, Reference Khandaker and Dantzer2016). Certain cytokines, such as IL-6, IL-1β, IL-8, IL-18, and TNF-α, are humoral host defense factors produced in response to infection or to tissue injury and stimulate the acute phase response, hematopoiesis, and immune processes (e.g., inflammation and interferon-γ production) (Bernhard et al., Reference Bernhard, Hug, Stratmann, Erber, Vidoni and Knapp2021; Ren & Torres, Reference Ren and Torres2009; Tanaka, Narazaki, & Kishimoto, Reference Tanaka, Narazaki and Kishimoto2014; Zhang & An, Reference Zhang and An2007). The genes responsible for these factors manifest expression in astrocytes, perivascular macrophages, and vascular endothelial cells, suggesting BBB involvement and a possible aberrant interplay with the peripheral immune system in schizophrenia (Cai et al., Reference Cai, Catts, Webster, Galletly, Liu and O’Donnell2020; Hwang et al., Reference Hwang, Kim, Shin, Kim, Seo and Webster2013; Kim, Hwang, Lee, et al., Reference Kim, Hwang, Lee and Webster2016; Murphy et al., Reference Murphy, Kondo, Walker, Rothmond, Matsumoto and Shannon Weickert2020; Purves-Tyson et al., Reference Purves-Tyson, Weber-Stadlbauer, Richetto, Rothmond, Labouesse and Polesel2021; Siegel, Sengupta, Edelson, Lewis, & Volk, Reference Siegel, Sengupta, Edelson, Lewis and Volk2014; Volk et al., Reference Volk, Chitrapu, Edelson, Roman, Moroco and Lewis2015). Murphy & Weickert (Reference Murphy and Weickert2021) postulated that the microglia do not mount a typical inflammatory response, suggesting that the immune response may be persistent and even due to previous exposure so that microglia would be depleted. Table 2 shows cytokine alterations observed in schizophrenia.
Cytokine alterations in schizophrenia
Note: sIL-2R: soluble IL-2 receptor; IL-1RA: IL-1 receptor antagonist; CSF: cerebrospinal fluid; TGF-β1: transforming growth factor beta 1. IFN-γ: Interferon-gamma; TNF-α: tumor necrosis factor-alpha.
Without any doubt, one of the most revealing neuropathologic findings within the pathophysiology of schizophrenia is the deficit of GABAergic interneurons (Dienel & Lewis, Reference Dienel and Lewis2019). Inverse correlations between elevated cytokine levels and higher IFITM mRNA levels, as well as GABAergic neuron markers in the frontal cortex, have been reported by several authors (Fillman et al., Reference Fillman, Cloonan, Catts, Miller, Wong and McCrossin2013; Siegel et al., Reference Siegel, Sengupta, Edelson, Lewis and Volk2014). Moreover, in the hippocampal region, the gene expression modules associated with immunity and inflammation also exhibit an inverse correlation with GABAergic indicators (Hwang et al., Reference Hwang, Kim, Shin, Kim, Seo and Webster2013; Kim, Hwang, Webster, & Lee, Reference Kim, Hwang, Webster and Lee2016). Several upregulated genes are linked to the response to viral infection, which suggests that the immune response induced by these pathogenic microorganisms is not exclusively related to schizophrenia, but also to a decrease in GABAergic interneurons (Kim, Hwang, Webster, et al., Reference Kim, Hwang, Webster and Lee2016). In animal models, the maternal immune activation in utero by poly(I:C) or LPS inoculation results in increased cytokine concentrations inside the fetal cerebrum (Meyer et al., Reference Meyer, Nyffeler, Engler, Urwyler, Schedlowski and Knuesel2006), which could persist postnatally (Garay, Hsiao, Patterson, & McAllister, Reference Garay, Hsiao, Patterson and McAllister2013; Purves-Tyson et al., Reference Purves-Tyson, Weber-Stadlbauer, Richetto, Rothmond, Labouesse and Polesel2021), affecting the neural circuit development process (Meyer, Nyffeler, Yee, Knuesel, & Feldon, Reference Meyer, Nyffeler, Yee, Knuesel and Feldon2008; Richetto, Calabrese, Riva, & Meyer, Reference Richetto, Calabrese, Riva and Meyer2014; Warm, Schroer, & Sinning, Reference Warm, Schroer and Sinning2022). Other upregulated genes in schizophrenia include midbrain immune-related transcripts that may contribute to the dopaminergic abnormalities observed in the disease (Purves-Tyson et al., Reference Purves-Tyson, Weber-Stadlbauer, Richetto, Rothmond, Labouesse and Polesel2021). The top genome-wide association studies implicating gene complement components, such as C4, C3, and C1qA, are also frequently identified as being upregulated in the brains of individuals with schizophrenia (Gamazon et al., Reference Gamazon, Zwinderman, Cox, Denys and Derks2019; Gandal, Zhang, et al., Reference Gandal, Zhang, Hadjimichael, Walker, Chen and Liu2018; Lindholm Carlström et al., Reference Lindholm Carlström, Niazi, Etemadikhah, Halvardson, Enroth and Stockmeier2021; Wang, Liu, et al., Reference Wang, Liu, Warrell, Won, Shi and Navarro2018). The correlation between the HLA area and schizophrenia is reinforced by the complement component C4 gene, which actively participates in synaptic trimming throughout postnatal growth (Sekar et al., Reference Sekar, Bialas, de Rivera, Davis, Hammond and Kamitaki2016).
In summary, according to Webster (Reference Webster2023), the pleiotropic effects of dysregulated immune molecules, which also contribute to various crucial processes such as cerebral development, neural plasticity, and homeostasis, may play an important role in the pathophysiology of schizophrenia compared to immune activation by microbial infection. Nevertheless, infection occurring during vulnerable periods of development in genetically predisposed subjects could indefinitely disrupt the typical expression of these pleiotropic immune molecules, thus increasing susceptibility to successive stress on brain development and function (Borrego-Ruiz & Borrego, Reference Borrego-Ruiz and Borrego2024b).
Section 2: Virus infection and schizophrenia
Schizophrenia and other psychiatric conditions may eventually be activated by different viral infections, but the precise mechanisms are yet unknown. Maternal viral infections occurring in the first stages of gestation may potentially contribute to the onset of psychiatric conditions in the progeny during later developmental stages (Kotsiri et al., Reference Kotsiri, Resta, Spyrantis, Panotopoulos, Chaniotis and Beloukas2023). We comprehensively review the literature addressing the role of several neurotropic viruses, such as herpesviruses, influenza virus, Borna disease virus, and coronaviruses, on the onset of schizophrenia.
Herpesviruses
Herpesviruses consist of double-stranded DNA viruses (family Herpesviridae) that cause persistent latent infections, which can periodically reactivate and lead to disease in susceptible populations (Croen, Reference Croen1991). The herpesviruses that are able to infect humans are divided into three groups on the basis of their genetic and antigenic characteristics: the alpha group includes herpes simplex virus 1 (HSV-1, or human herpesvirus 1, HHV-1), herpes simplex virus 2 (HSV-2, or HHV-2), and varicella-zoster virus (VZV, or HHV-3); the beta group comprises cytomegalovirus (CMV or HHV-5), human herpesvirus 6 (HHV-6), and human herpesvirus 7 (HHV-7); and the gamma group consists of Epstein–Barr virus (EBV or HHV-4), and Kaposi’s sarcoma-associated herpesvirus (HHV-8) (Gatherer et al., Reference Gatherer, Depledge, Hartley, Szpara, Vaz and Benkő2021). Among them, HSV-1, CMV, and EBV have been recognized for inducing moderate symptoms in adults, but for provoking more severe disease manifestations in neonates or immunosuppressed subjects. In such vulnerable people, immune deficiency leads to a loss of viral replication regulation, resulting in tissue harm and, in more extreme instances, terminal organ pathology and mortality (Griffiths & Reeves, Reference Griffiths and Reeves2021; Griffiths, Baraniak, & Reeves, Reference Griffiths, Baraniak and Reeves2015).
Herpes simplex virus
Herpes simplex virus (HSV) is a common infection that is spread primarily through skin-to-skin contact and can cause painful blisters or ulcers (Kriebs, Reference Kriebs2008). Its prevalence remains within populations, escalating with age, reaching 90% or higher in certain geographic areas (Prasad et al., Reference Prasad, Watson, Dickerson, Yolken and Nimgaonkar2012). There are two types of HSV: HSV-1 spreads primarily through oral contact and causes infections inside or around the mouth (oral herpes), but it can also cause genital herpes. Type 2 (HSV-2) is spread through sexual contact and provokes genital herpes.
HSV-1 often induces primary infections in human mucosal epithelial cells (Ryder, Jin, McNulty, Grulich, & Donovan, Reference Ryder, Jin, McNulty, Grulich and Donovan2009), while severe corneal keratitis and encephalitis are less common (Harkness, Kader, & DeLuca, Reference Harkness, Kader and DeLuca2014; James & Kimberlin, Reference James and Kimberlin2015; Marcocci et al., Reference Marcocci, Napoletani, Protto, Kolesova, Piacentini and Li Puma2020; Steiner, Kennedy, & Pachner, Reference Steiner, Kennedy and Pachner2007). After initial mucosal infections, which may be asymptomatic, viral particles enter sensory neurons through nerve terminals and travel retrogradely along axons, establishing a latent state in neuronal cell bodies that can persist for the entire lifetime of the host (Harkness et al., Reference Harkness, Kader and DeLuca2014; Nicoll, Proença, & Efstathiou, Reference Nicoll, Proença and Efstathiou2012; Steiner et al., Reference Steiner, Kennedy and Pachner2007). The viral DNA may endure within the neuronal cell body in an episomal state, indicating its independence from integration into the host cell chromosome, instead remaining suspended within the cytoplasm (St Leger, Koelle, Kinchington, & Verjans, Reference Leger, Koelle, Kinchington and Verjans2021). The state of latency is maintained through a complex interplay of host and viral factors, with the virus potentially reactivating later in life via sensory nerves, returning to its initial entry site (Shimomura & Higaki, Reference Shimomura and Higaki2011). Triggers recognized to initiate reactivation comprise aging, stress, immunosuppression, and co-infection with other viruses (Yan et al., Reference Yan, Luo, Li, Li, Dallmann and Kurihara2020). HSV-1 reactivation rarely causes encephalitis, but mutations in toll-like receptors and neuronal immunity increase the risk (Lafaille et al., Reference Lafaille, Harschnitz, Lee, Zhang, Hasek and Kerner2019). Newborns of mothers with HSV-1 genital infections may develop encephalitis or multi-organ infections (James & Kimberlin, Reference James and Kimberlin2015). In the cerebrum, however, HSV-1 typically results in a persistent infection regulated by the CNS immune system (Marcocci et al., Reference Marcocci, Napoletani, Protto, Kolesova, Piacentini and Li Puma2020).
Initial analyses indicated an increased prevalence of seropositivity for antibodies directed against HSV-2, cytomegalovirus, and T. gondii in subjects with diagnosed cognitive decline (Nimgaonkar et al., Reference Nimgaonkar, Yolken, Wang, Chang, McClain and McDade2016). There is a long history of research linking HSV-1 to mental disorders (Dickerson et al., Reference Dickerson, Boronow, Stallings, Origoni and Ruslanova2003; Kotsiri et al., Reference Kotsiri, Resta, Spyrantis, Panotopoulos, Chaniotis and Beloukas2023; Prasad et al., Reference Prasad, Watson, Dickerson, Yolken and Nimgaonkar2012; Yolken, Reference Yolken2004), but several studies have not found heightened HSV-1 serum antibodies from subjects with schizophrenia (Aiello et al., Reference Aiello, Haan, Blythe, Moore, Gonzalez and Jagust2006; Katan et al., Reference Katan, Moon, Paik, Sacco and Wright2013), and a current meta-analysis suggested a divergence between cases and controls, showing a small to medium effect size (Dickerson, Schroeder, Nimgaonkar, Gold, & Yolken, Reference Dickerson, Schroeder, Nimgaonkar, Gold and Yolken2020). Additional supportive evidence includes brain imaging studies showing diminished gray matter volume in HSV-1 seropositive subjects (Prasad et al., Reference Prasad, Eack, Goradia, Pancholi, Keshavan and Yolken2011; Schretlen et al., Reference Schretlen, Vannorsdall, Winicki, Mushtaq, Hikida and Sawa2010), and enhancement in the cognitive function of HSV-1 seropositive individuals with schizophrenia who underwent extended courses of high-dose antiviral treatment (Bhatia et al., Reference Bhatia, Wood, Iyengar, Narayanan, Beniwal and Prasad2018; Prasad et al., Reference Prasad, Eack, Keshavan, Yolken, Iyengar and Nimgaonkar2013). More recently, HSV-1 seropositivity was associated with suicidal behavior in a cohort of Danish blood donors (Nissen et al., Reference Nissen, Trabjerg, Pedersen, Banasik, Pedersen and Sørensen2019).
While the evidence for HSV-1 infection in schizophrenia versus control populations is inconclusive (de Witte et al., Reference de Witte, van Mierlo, Litjens, Klein, Bahn and Osterhaus2015), there is compelling data suggesting that HSV-1 infection may play a role in cognitive decline in the context of schizophrenia. Two recent systematic reviews and a meta-analysis have indicated that HSV-1 negatively impacts the neurocognitive function of schizophrenic subjects, with small to moderate effect sizes (Tucker & Bertke, Reference Tucker and Bertke2019; Dickerson et al., Reference Dickerson, Schroeder, Nimgaonkar, Gold and Yolken2020).
Cytomegalovirus
Human cytomegalovirus (CMV) belongs to the family Orthoherpesviridae (Betaherpesvirinae subfamily). Failure to achieve complete viral clearance after primary infection results in long-lasting infection marked by intervals of latency and reactivation in approximately 80% of the general population worldwide, with significant differences by age, sex, and geographic area (Goodrum, Reference Goodrum2016; Zuhair et al., Reference Zuhair, Smit, Wallis, Jabbar, Smith and Devleesschauwer2019). Primary infection with CMV frequently takes place during childhood via exposure to oral or genital secretions (Petersen, Patel, Abraham, Quinn, & Tobian, Reference Petersen, Patel, Abraham, Quinn and Tobian2021).
Primary infections in immunocompetent subjects may be subclinic or generate nonspecific symptoms like a slight fever or a mononucleosis syndrome, because CMV harbors a large number of genes that encode for evasion of the host’s innate and adaptive immune mechanisms, such as attenuation of type I interferon (IFN-I) production, downregulation of the activity of natural killer cell, inhibition of MHC class I and II antigen presentation, and interference with the proliferation of T cells (Griffiths & Reeves, Reference Griffiths and Reeves2021; Mishra, Kumar, Ingle, & Kumar, Reference Mishra, Kumar, Ingle and Kumar2020; Patro, Reference Patro2019; Ye et al., Reference Ye, Qian, Yu, Guo, Wang and Xue2020). This virus-host interaction can modify host immunity gradually, as a significant proportion of the immune system is directed against CMV antigens (Picarda & Benedict, Reference Picarda and Benedict2018), and frequent viral reactivation may detrimentally affect the immunity of the host to infections by other microorganisms (Furman et al., Reference Furman, Jojic, Sharma, Shen-Orr, Angel and Onengut-Gumuscu2015; Martinez et al., Reference Martinez, Nicol, Wedderburn, Stadler, Botha and Workman2021; Nicoli et al., Reference Nicoli, Clave, Wanke, von Braun, Bondet and Alanio2022; Wall et al., Reference Wall, Godlee, Geh, Jones, Faustini and Harvey2021). Additional evidence suggests that CMV may protect against heterologous infections and enhance vaccine efficacy through sustained immune activation, with minimal impact in healthy populations (Furman et al., Reference Furman, Jojic, Sharma, Shen-Orr, Angel and Onengut-Gumuscu2015; Forte, Zhang, Thorp, & Hummel, Reference Forte, Zhang, Thorp and Hummel2020). However, repeated episodes of CMV reactivation can result in a decrease in T cells and an increase in CMV-specific T cells, especially effector memory cells (Ford et al., Reference Ford, Teague, Bayouth, Yolken, Bodurka and Irwin2020), leading to disease or tissue damage in immunocompromised populations, and potentially in patients with psychiatric disorders (Griffiths & Reeves, Reference Griffiths and Reeves2021; Savitz & Yolken, Reference Savitz and Yolken2023).
The conceptualization that CMV might be implicated in the development of schizophrenia began to develop in the 1970s, with the first investigations based on serological studies (Yolken & Torrey, Reference Yolken and Torrey1995). Later, other serological studies included socioeconomic and geographical variables, and also individuals experiencing the initial onset or the first episode of schizophrenia (Torrey et al., Reference Torrey, Leweke, Schwarz, Mueller, Bachmann and Schroeder2006). Furthermore, a cohort study revealed that symptomatic CMV infection of the CNS in early life conferred a higher risk of developing schizophrenia (Dalman et al., Reference Dalman, Allebeck, Gunnell, Harrison, Kristensson and Lewis2008). Besides, in patients with schizophrenia, higher CMV IgG titers correlated with reduced right hippocampal volume and diminished episodic verbal memory (Houenou et al., Reference Houenou, d’Albis, Daban, Hamdani, Delavest and Lepine2014). In contrast, several studies have reported lower cognitive function in CMV-seropositive schizophrenia patients (Dickerson et al., Reference Dickerson, Stallings, Origoni, Katsafanas, Schweinfurth and Savage2014; Shirts et al., Reference Shirts, Prasad, Pogue-Geile, Dickerson, Yolken and Nimgaonkar2008). More recently, Moya Lacasa et al. (Reference Moya Lacasa, Rayner, Hagen, Yang, Marks and Kirkpatrick2021) did not find a significant difference in the prevalence of CMV seropositivity and scores in schizophrenia and other nonaffective psychotic disorders compared to healthy controls. Clinical heterogeneity and several confounding factors, including age, medication, as well as socioeconomic and habitat status, may potentially contribute to the conflicting results between studies (Bolu et al., Reference Bolu, Oznur, Tok, Balikci, Sener and Celik2016; Zheng & Savitz, Reference Zheng and Savitz2023). The detection of CMV nucleic acid in patients with schizophrenia using hybridization or polymerase chain reaction techniques has improved the sensitivity of viral detection in postmortem brain studies and strengthened the association between viral infection and schizophrenia (Cassedy, Parle-McDermott, & O’Kennedy, Reference Cassedy, Parle-McDermott and O’Kennedy2021). Nevertheless, it is noteworthy that the likelihood of detecting CMV transcripts a priori is constrained due to the typically low levels of RNA/DNA associated with latent CMV infection (Shnayder et al., Reference Shnayder, Nachshon, Krishna, Poole, Boshkov and Binyamin2018), particularly considering that postmortem investigations frequently utilize aged and fixed brain tissue. Furthermore, the nondetection of genetic material of the virus in postmortem samples does not conclusively dismiss the potential for CMV to contribute to the pathology only in the initial phases of psychiatric disorders, or for CMV to exert pathological effects through immune activation without direct cerebral infection.
Brain-derived cells, such as BBB endothelial cells, vascular pericytes, myeloid lineage cells, astrocytes, glia, and neurons, have been shown to be completely permissive to CMV infection and replication (Alcendor, Charest, Zhu, Vigil, & Knobel, Reference Alcendor, Charest, Zhu, Vigil and Knobel2012; Luo, Schwartz, & Fortunato, Reference Luo, Schwartz and Fortunato2008; Tsutsui, Kosugi, & Kawasaki, Reference Tsutsui, Kosugi and Kawasaki2005). Hence, it is plausible that CMV itself or the inflammation linked to repeated CMV reactivation can result in irregularities within cerebral morphology and function. A primary question is whether CMV can infect brain cells in vivo and induce the symptoms of psychosis, including schizophrenia. The BBB constitutes a protective barrier that plays a pivotal role in restricting the unconstrained movement of large molecules, like viral particles, from the blood stream into the CSF and the cells of the CNS. In this regard, several pathophysiological mechanisms may be involved. First, CMV particles in the circulation attain and infect brain microvascular endothelial cells, thereby gaining access to the CNS. In addition, CMV could potentially weaken the tight junctions of the BBB, predisposing the parenchyma to a pathogenesis driven by the immune system. Primary and latent CMV infections induce a systemic inflammatory response and the expression of chemokines and cytokines (e.g., MCP-1, and IFN-γ) (Hamilton, Scott, Naing, & Rawlinson, Reference Hamilton, Scott, Naing and Rawlinson2013; Lurain et al., Reference Lurain, Hanson, Martinson, Leurgans, Landay and Bennett2013; van de Berg et al., Reference van de Berg, Heutinck, Raabe, Minnee, Young and van Donselaar-van der Pant2010), which can decrease the expression of tight junction protein and increase BBB permeability (Chai, She, Huang, & Fu, Reference Chai, She, Huang and Fu2015). Upon the entry of peripheral cytokines into the brain, the function of glial cells may be impaired, resulting in axonal demyelination and neural degeneration (Hammond, Marsh, & Stevens, Reference Hammond, Marsh and Stevens2019; Mechawar & Savitz, Reference Mechawar and Savitz2016; Wohleb, Franklin, Iwata, & Duman, Reference Wohleb, Franklin, Iwata and Duman2016). Furthermore, CMV could potentially compromise the host immune response and trigger host immunosuppression, thereby potentially facilitating the cellular senescence and the neuropathological changes that resemble the aging process (Naniche & Oldstone, Reference Naniche and Oldstone2000; Salminen, Reference Salminen2021). Additionally, CMV can induce autoimmunity, and thus neuroinflammation, mediated by the autoimmune process (Halenius & Hengel, Reference Halenius and Hengel2014; Vanheusden et al., Reference Vanheusden, Broux, Welten, Peeters, Panagioti and Van Wijmeersch2017).
Although a significant and strong relationship between CMV and schizophrenia may not be fully established, CMV infection may implicate a greater risk of negatively impacting cognitive function in the context of schizophrenia.
Epstein–Barr virus
The gamma herpesvirus EBV is usually transmitted orally during childhood or adolescence, and the primary infection occurs in epithelial cells and is followed by infection of B lymphocytes (Vetsika & Callan, Reference Vetsika and Callan2004). Primary infection in infants and children is associated with nonspecific respiratory symptoms, whereas infection in adolescents and young adults manifests the symptoms of infectious mononucleosis, such as fever, adenopathy, and pharyngitis (Leung, Lam, & Barankin, Reference Leung, Lam and Barankin2024). It seems that different age-dependent symptomatology may be related to hormonal changes that occur during puberty (Rostgaard et al., Reference Rostgaard, Balfour, Jarrett, Erikstrup, Pedersen and Ullum2019). After the resolution of the acute infection, the virus can establish lifelong latency in B lymphocytes and epithelial cells, with subsequent reactivation episodes during life (Kempkes & Robertson, Reference Kempkes and Robertson2015). Long-term EBV infection has been associated with neoplastic diseases, including B-cell lymphoma, nasopharyngeal carcinoma, and gastric cancer (Dugan, Coleman, & Haverkos, Reference Dugan, Coleman and Haverkos2019). In addition, this infection leads to alterations in the host immune system and is also related to several autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus, and systemic sclerosis (Bjornevik et al., Reference Bjornevik, Cortese, Healy, Kuhle, Mina and Leng2022; Houen & Trier, Reference Houen and Trier2021). Acute infection is characterized by the production of early antigen (EA), antibodies (IgM and IgG) to viral capsid antigen (VCA), antibodies to EBV nuclear antigen (EBNA), and heterophilic antibodies (De Paschale & Clerici, Reference De Paschale and Clerici2012).
An increased incidence of psychotic experiences in adolescents who were infected with EBV during childhood supports the potential relationship between EBV infection and schizophrenia (Khandaker, Stochl, Zammit, Lewis, & Jones, Reference Khandaker, Stochl, Zammit, Lewis and Jones2014b). Individuals with schizophrenia had an aberrant response to EBV infection, characterized by increased levels of reactivity to EBV-VCA, but not to EBNA and total viral antigens. When combined with genetic risks for schizophrenia, individuals with the aberrant response to EBV were more likely to be diagnosed with schizophrenia than individuals without these risk factors (Dickerson et al., Reference Dickerson, Jones-Brando, Ford, Genovese, Stallings and Origoni2019). Moreover, schizophrenic patients with elevated levels of EBV antibodies presented lower levels of social cognition (Dickerson et al., Reference Dickerson, Katsafanas, Origoni, Squire, Khushalani and Newman2021).
In essence, EBV has been linked to schizophrenia on the basis of psychotic experiences in adolescence among individuals with a history of EBV infection in childhood, as well as the higher likelihood of schizophrenia diagnosis in those with an aberrant response to EBV.
Influenza virus
The influenza virus, belonging to the family Orthomyxoviridae, represents an enveloped virus with segmented negative-sense single-stranded RNA segments and includes three genera: influenza A, B, and C (Payne, Reference Payne2023), of which influenza A is responsible for pandemics (Kępińska et al., Reference Kępińska, Iyegbe, Vernon, Yolken, Murray and Pollak2020) and has been historically associated with schizophrenia and psychotic symptoms. Influenza A viruses are categorized into subtypes on the basis of the antigenic properties of hemagglutinin (H) and neuraminidase (N). Specifically, the avian-derived strains and H5N1 virus are neurotropic, whereas H1N1 is non-neurotropic (Jang et al., Reference Jang, Boltz, Sturm-Ramirez, Shepherd, Jiao and Webster2009; Sadasivan, Zanin, O’Brien, Schultz-Cherry, & Smeyne, Reference Sadasivan, Zanin, O’Brien, Schultz-Cherry and Smeyne2015). While maternal infection with microorganisms has been reported as a risk for schizophrenia (Brown & Patterson, Reference Brown and Patterson2011), there is no clear evidence for transplacental transmission of influenza virus into the offspring brain (Egorova, Egorov, & Zabrodskaya, Reference Egorova, Egorov and Zabrodskaya2024; Lieberman, Bagdasarian, Thomas, & Van De Ven, Reference Lieberman, Bagdasarian, Thomas and Van De Ven2011; Shi, Tu, & Patterson, Reference Shi, Tu and Patterson2005), so more relevant are the effects of infection-induced maternal immune activation on the developing brain (Egorova et al., Reference Egorova, Egorov and Zabrodskaya2024; Brown & Meyer, Reference Brown and Meyer2018).
In animal models, maternal H1N1 infection provokes abnormalities in offspring at several levels: (i) gene expression; (ii) protein expression; (iii) brain structure; (iv) behavior; (v) neurotransmitter levels; and (vi) placental development. In humans, there is evidence that early maternal immune activation during pregnancy affects neonatal brain development and behavior (Lieberman et al., Reference Lieberman, Bagdasarian, Thomas and Van De Ven2011). High maternal IL-6 and cortisol levels predicted larger neonatal amygdala volume and connectivity, as well as higher internalizing behavior, which in turn predicted poorer impulse control in the early years of life (Graham et al., Reference Graham, Rasmussen, Rudolph, Heim, Gilmore and Styner2018; Reference Graham, Rasmussen, Entringer, Ben Ward, Rudolph and Gilmore2019).
Fatemi et al. (Reference Fatemi, Folsom, Rooney, Mori, Kornfield and Reutiman2012) reported that brain gene expression was altered in offspring infected at multiple embryonic times. In the first week, prenatal infection-induced gene expression changes associated with hypoxia, inflammation, and schizophrenia in the frontal cortex and hippocampus of exposed offspring. At 16 days, prenatal H1N1 infection was associated with gene changes related to myelination in the cerebellum and hippocampus, such as mag, plp1, mal, mbp, mobp, mog, ncam1, and rgs4. Several of these genes are known to be involved in schizophrenia (Fatemi et al., Reference Fatemi, Folsom, Reutiman, Abu-Odeh, Mori and Huang2009; Van Campen et al., Reference Van Campen, Bishop, Abrahams, Bielefeldt-Ohmann, Mathiason and Bouma2020). At 3 weeks, microarray data showed a high number of dysregulated genes in the frontal cortex, hippocampus, and cerebellum of virus-exposed offspring, such as Sema3a, Trfr2, and Vldlr, which have also been implicated in schizophrenia (Fatemi et al., Reference Fatemi, Reutiman, Folsom, Huang, Oishi and Mori2008). Further studies have shown that mice infected with influenza virus during pregnancy have lower hemagglutination inhibition and neutralizing antibody titers, because of a downregulation of B cell metabolism and post-translational modification systems (Swieboda et al., Reference Swieboda, Littauer, Beaver, Mills, Bricker and Esser2020).
Neuronal nitric oxide synthase (nNOS) produces nitric oxide (NO), a harmful molecule involved in fetal brain injury under hypoxia-ischemia conditions (Kneeland & Fatemi, Reference Kneeland and Fatemi2013). In the offspring of mice infected with H1N1 on day 9, nNOS was shown to increase by 147% on postnatal day 35, with a final decrease of 29% on postnatal day 56 in rostral brain areas. In addition, nNOS levels in the midbrain zones decreased by 27% at postnatal day 56 (Fatemi, Cuadra, El-Fakahany, Sidwell, & Thuras, Reference Fatemi, Cuadra, El-Fakahany, Sidwell and Thuras2000). Upregulation of NO production has been suggested as a trigger for several human neurological disorders, including Parkinson’s disease, Alzheimer’s disease, amyotrophic lateral sclerosis, and schizophrenia (Iova et al., Reference Iova, Marin, Lazar, Stanescu, Dogaru and Nicula2023). The suggested mechanisms for its neurotoxicity are mainly centered on the increased amounts of NO produced in the brain, which may lead to decoding of neuronal dysfunction associated with neuronal death. Moreover, the extracellular secretory glycoprotein reelin plays an important role during development and also in adulthood, such as maintaining synaptic function. Several neuropsychiatric disorders, including schizophrenia, share a common feature of abnormal reelin expression in the brain (Kneeland & Fatemi, Reference Kneeland and Fatemi2013). In the brains of neonatal offspring of virus-exposed mice, reelin synthesis is reduced, reflecting abnormal neuronal migration and reduced synaptic plasticity. The mechanisms for abnormal reelin expression are currently unknown, although several mechanisms have been proposed, including mutations, hypermethylation of the gene promoter, miRNA silencing, FMRP underexpression, and abnormal processing (Folsom & Fatemi, Reference Folsom and Fatemi2013).
Several brain abnormalities in mice have been associated with influenza virus infection, including increases in fractional anisotropy within the right middle cerebellar peduncle, pyramidal cell density, and brain volume, as well as decreases in area measurements within the cerebral cortex, unilateral brain hemispheres, and hippocampal region (Fatemi et al., Reference Fatemi, Earle, Kanodia, Kist, Emamian and Patterson2002; Kneeland & Fatemi, Reference Kneeland and Fatemi2013). In rhesus monkeys, maternal inflammatory responses to influenza infection reduced gray matter throughout most of the cortex, and decreased white matter in the parietal cortex of the fetal brain (Short et al., Reference Short, Lubach, Karasin, Olsen, Styner and Knickmeyer2010).
Behavioral abnormalities, such as the acoustic startle response and head twitch response, have also been observed in the offspring of virus-exposed mice (Moreno et al., Reference Moreno, Kurita, Holloway, López, Cadagan and Martínez-Sobrido2011; Shi et al., Reference Shi, Fatemi, Sidwell and Patterson2003), signs typically induced by 5-HT2A agonists. In fact, the animals had increased 5-HT2A receptor expression in the frontal cortex and reduced levels of serotonin in the cerebellum (Moreno et al., Reference Moreno, Kurita, Holloway, López, Cadagan and Martínez-Sobrido2011). In addition, several authors have reported neurotransmitter abnormalities in the offspring of virus-exposed mice, such as altered levels of taurine, serotonin, and GABA in the cerebellar region, but not in dopamine (Fatemi et al., Reference Fatemi, Reutiman, Folsom, Huang, Oishi and Mori2008; Reference Fatemi, Folsom, Liesch, Kneeland, Karkhane Yousefi and Thuras2017; Winter et al., Reference Winter, Reutiman, Folsom, Sohr, Wolf and Juckel2008). Recently, Perez-Palomar, Erdozain, Erkizia-Santamaría, Ortega, & Meana (Reference Perez-Palomar, Erdozain, Erkizia-Santamaría, Ortega and Meana2023) reported that prenatal viral infection in mice was mimicked by poly(I:C) administration, which resulted in decreased extracellular dopamine concentrations compared to controls, concluding that the poly(I:C)-based model reproduces catecholamine phenotypes reported in schizophrenia.
Placental giant trophoblast cells are producers of hormones and cytokines that can influence maternal physiology in response to the fetal allograft (Elgueta, Murgas, Riquelme, Yang, & Cancino, Reference Elgueta, Murgas, Riquelme, Yang and Cancino2022). Infected placental tissue showed cytoarchitectural disorganization (Antonson et al., Reference Antonson, Kenney, Chen, Corps, Yount and Gur2021; Van Campen et al., Reference Van Campen, Bishop, Abrahams, Bielefeldt-Ohmann, Mathiason and Bouma2020) and an increased presence of immune cells (Fatemi et al., Reference Fatemi, Folsom, Rooney, Mori, Kornfield and Reutiman2012). High levels of gene dysregulation have also been observed in the placenta of pregnant mice with influenza virus infection (Fatemi et al., Reference Fatemi, Folsom, Rooney, Mori, Kornfield and Reutiman2012; Van Campen et al., Reference Van Campen, Bishop, Abrahams, Bielefeldt-Ohmann, Mathiason and Bouma2020); these dysregulated placental genes significantly influence apoptosis, hypoxia, inflammation, immune response, and psychiatric disorders.
Both neurotropic and non-neurotropic strains of the influenza virus can induce microglial activation and contribute to inflammation (Sadasivan, Zanin, O’Brien, Schultz-Cherry, & Smeyne, Reference Sadasivan, Zanin, O’Brien, Schultz-Cherry and Smeyne2015; Wang et al., Reference Wang, Zhang, Li, Xin, Su and Gao2008). The innate immune response to influenza infection is mediated by antiviral IFN-stimulated genes (ISGs) (Schoggins, Reference Schoggins2019), including MXA (prevents virus nuclear import), IFITM3 (blocks host-virus cell membrane fusion), and viperin (blocks influenza virus release) (Iwasaki & Pillai, Reference Iwasaki and Pillai2014). In addition, innate immune mechanisms also promote disease resistance of host tissues, and prepare the microglia to respond appropriately to a novel stimulus (Kępińska et al., Reference Kępińska, Iyegbe, Vernon, Yolken, Murray and Pollak2020). Furthermore, influenza virus infection may prime microglia for increased activation, potentially increasing the risk of developing psychotic symptoms (Salam, Borsini, & Zunszain, Reference Salam, Borsini and Zunszain2018). Moreover, Littauer et al. (Reference Littauer, Esser, Antao, Vassilieva, Compans and Skountzou2017) found that influenza virus infection resulted in dysregulation of inflammatory responses and the immune responses significantly correlated with changes in hormone synthesis and regulation. Dysregulation of progesterone, COX-2, PGE2, and PGF2α expression was accompanied by remodeling of placental architecture and upregulation of matrix metalloproteinase-9, an indicator of placental spongiotrophoblast degradation, early after infection.
Thus, converging evidence suggests that influenza virus infection possesses a variety of effects on prenatal and postnatal processes that could lead to an increased risk of developing schizophrenia or acute psychosis in adulthood.
Borna disease virus
Borna disease virus (BoDV) is a negative-sense, single-stranded RNA virus of the Bornaviridae family whose specific hosts are horses, sheep, poultry, and cattle (Rubbenstroth et al., Reference Rubbenstroth, Briese, Dürrwald, Horie, Hyndman, Kuhn and Consortium2021). Several properties of BoDV make it a potential agent of human psychiatric disorders, such as: (i) it infects neurons; (ii) it has a broad host range; (iii) it presents a high tropism for the limbic circuit, which regulates behavior, memory, and emotion, and appears to play a key role in the etiopathology of several human psychiatric disorders; and (iv) experimental BoDV infection in animals can result in various symptoms related to aggression, hyperactivity, apathy, or motor symptoms that resemble core features of human psychiatric disorders, such as depression or schizophrenia (Briese, Hornig, & Lipkin, Reference Briese, Hornig and Lipkin1999). In addition, a neuronal route of BoDV transmission has been demonstrated in animals, involving the olfactory nerve in the nasal mucosa (Kupke et al., Reference Kupke, Becker, Wewetzer, Ahlemeyer, Eickmann and Herden2019).
As a neurotropic virus, BoDV infects the CNS of animals, causing neuronal degeneration and neurological disorders such as encephalitis, meningitis, various neurodevelopmental movement disorders, and behavioral disorders with schizophrenia-like manifestations (Taieb, Baleyte, Mazet, & Fillet, Reference Taieb, Baleyte, Mazet and Fillet2001). For example, Ovanesov et al. (Reference Ovanesov, Ayhan, Wolbert, Moldovan, Sauder and Pletnikov2008) reported that neonatal BoDV infection of the rat brain was associated with microglial activation and neuronal damage. Stimulated microglia expressed MHC I, MHC II, and IL-6, showing increased secretion of TNF-α and of IL-1, which have been suggested as potential molecular biomarkers for the development of schizophrenia.
Several works have studied whether BoDV disease can directly cause schizophrenic disorders in humans (Arias et al., Reference Arias, Sorlozano, Villegas, de Dios Luna, McKenney and Cervilla2012; Azami, Jalilian, Khorshidi, Mohammadi, & Tardeh, Reference Azami, Jalilian, Khorshidi, Mohammadi and Tardeh2018; Chen et al., Reference Chen, Chiu, Wei, Koong, Liu and Shaw1999; Mazaheri-Tehrani et al., Reference Mazaheri-Tehrani, Maghsoudi, Shams, Soori, Atashi and Motamedi2014; Sansom, Reference Sansom2000), or other mental conditions, such as mood-related disorders and depression (Hornig et al., Reference Hornig, Briese, Licinio, Khabbaz, Altshuler and Potkin2012; Kim et al., Reference Kim, Kim, Han, Lee, Kim and Yoon2003; Selten et al., Reference Selten, van Vliet, Pleyte, Herzog, Hoek and van Loon2000). In contrast, Iwata et al. (Reference Iwata, Takahashi, Peng, Fukuda, Ohno and Ogawa1998) measured BoDV p24 RNA in peripheral blood monocytes from psychiatric patients, and they did not find a relationship between virus infection and mood or schizophrenia. Similar results were reported in several countries, where BoDV infection and pathogenesis of schizophrenia were not found (Hornig et al., Reference Hornig, Briese, Licinio, Khabbaz, Altshuler and Potkin2012; Miranda et al., Reference Miranda, Nunes, Calvo, Suzart, Itano and Watanabe2006; Selten et al., Reference Selten, van Vliet, Pleyte, Herzog, Hoek and van Loon2000; Soltani, Mohammadzadeh, Makvandi, Pakseresht, & Samarbaf-Zadeh, Reference Soltani, Mohammadzadeh, Makvandi, Pakseresht and Samarbaf-Zadeh2016).
In summary, the inconsistency in the results may be due to different geographic and genetic factors such as HLA, which was included in different studies (Na, Tae, Song, & Kim, Reference Na, Tae, Song and Kim2009), as well as to the different techniques used for virus detection (Wolff, Heins, Pauli, Burger, & Kurth, Reference Wolff, Heins, Pauli, Burger and Kurth2006).
Coronaviruses
Coronaviridae is a family of enveloped, positive-sense single-stranded RNA viruses described in the decade of the 1960s (Mahase, Reference Mahase2020). Human coronavirus (HCoVs) 229E, OC43, HKU1, and NL63 cause upper respiratory tract infections or the “common cold” in a typical winter season (Cui, Li, & Shi, Reference Cui, Li and Shi2019; Su et al., Reference Su, Wong, Shi, Liu, Lai and Zhou2016; Woo et al., Reference Woo, de Groot, Haagmans, Lau, Neuman and Perlman2023). Since the turn of the 21st century, three highly virulent coronavirus strains have emerged. SARS-CoV, in 2002–2003, was identified in Guangdong (China), and subsequently spread to 17 countries, causing over 8000 cases (de Wit, van Doremalen, Falzarano, & Munster, Reference de Wit, van Doremalen, Falzarano and Munster2016; Zhong et al., Reference Zhong, Zheng, Li, Poon, Xie and Chan2003). Later, Middle East Respiratory Syndrome (MERS), caused by MERS coronavirus (MERS-CoV), was reported between 2012 and 2017, primarily in Middle Eastern countries (Zaki, van Boheemen, Bestebroer, Osterhaus, & Fouchier, Reference Zaki, van Boheemen, Bestebroer, Osterhaus and Fouchier2012) and later in the Republic of Korea (de Wit et al., Reference de Wit, van Doremalen, Falzarano and Munster2016); both MERS outbreaks caused a total of approximately 2500 cases. Bats and camels were suggested to be reservoirs for SARS-CoV and MERS-CoV, respectively. The newest member of the HCoVs, known as severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2), is responsible for the COVID-19 pandemic that affected more than 130 million people worldwide. While the zoonotic aspects of SARS-CoV-2 are still under investigation, bats also appear to be its animal reservoir (Ahmad et al., Reference Ahmad, Khan, Haroon, Nasir and Hui2020; Cui et al., Reference Cui, Li and Shi2019).
A key feature of these virulent HCoVs is their ability to replicate in epithelial cells and pneumocytes in the human lower respiratory tract, causing pneumonia (Cui et al., Reference Cui, Li and Shi2019; de Wit et al., Reference de Wit, van Doremalen, Falzarano and Munster2016) and, in severe cases, acute respiratory distress syndrome. However, SARS and MERS can infect the nervous system (Netland, Meyerholz, Moore, Cassell, & Perlman, Reference Netland, Meyerholz, Moore, Cassell and Perlman2008), causing peripheral neuropathies, encephalopathies, motor deficits, paralysis, and coma (Arabi et al., Reference Arabi, Harthi, Hussein, Bouchama, Johani and Hajeer2015; Wu et al., Reference Wu, Xu, Chen, Duan, Hashimoto and Yang2020), as well as psychiatric manifestations such as depression, anxiety, mood alterations, mania, and psychosis (Rogers et al., Reference Rogers, Chesney, Oliver, Pollak, McGuire and Fusar-Poli2020; Severance et al., Reference Severance, Dickerson, Viscidi, Bossis, Stallings and Origoni2011). Because the unique link between HCoVs infection and psychosis is inflammation, it is logical to think that COVID-19 infection, which induces a massive inflammatory response (Steardo, Steardo, Zorec, & Verkhratsky, Reference Steardo, Steardo, Zorec and Verkhratsky2020), could produce more psychotic symptoms (Graham et al., Reference Graham, Clark, Orban, Lim, Szymanski and Taylor2021; Mazza et al., Reference Mazza, Palladini, De Lorenzo, Magnaghi, Poletti and Furlan2021; Taquet, Geddes, Husain, Luciano, & Harrison, Reference Taquet, Geddes, Husain, Luciano and Harrison2021).
SARS-CoV-2 has a similar neurotropism to SARS-CoV and enters into the CNS through the circulatory route (Baig, Reference Baig2020) and via peripheral neurons, to reach the brain (Wu et al., Reference Wu, Xu, Chen, Duan, Hashimoto and Yang2020). Another pathway, the cribriform plate of the ethmoid bone near the olfactory bulb, has also been suggested (Pourfridoni & Askarpour, Reference Pourfridoni and Askarpour2022). In addition, the isolation of SARS-CoV-2 from CSF may suggest a direct infection of the neurological system causing nerve damage (Rentero et al., Reference Rentero, Juanes, Losada, Álvarez, Parra and Santana2020). The two SARS-CoV viruses share the same adsorption host cell receptor, angiotensin-converting enzyme 2 (ACE2), extensively expressed in neurons and glial cells (Steardo et al., Reference Steardo, Steardo, Zorec and Verkhratsky2020), although the binding affinity for SARS-CoV-2 is 10 to 20 times higher than that of SARS-CoV (Hoffmann et al., Reference Hoffmann, Kleine-Weber, Schroeder, Krüger, Herrler and Erichsen2020; Wrapp et al., Reference Wrapp, Wang, Corbett, Goldsmith, Hsieh and Abiona2020). Binding of SARS-CoV-2 to ACE2 and subsequent viral endocytosis dysregulate the angiotensin system, leading to the loss of ACE2-mediated health protection and adverse systemic effects by releasing TNF-α, IL-6, and other cytokine mediators, which lead to the cytokine storm in COVID-19 (Davies, Adlimoghaddam, & Albensi, Reference Davies, Adlimoghaddam and Albensi2021; Gheblawi et al., Reference Gheblawi, Wang, Viveiros, Nguyen, Zhong and Turner2020).
A number of pathways associated with COVID-19 overlap with those that may be involved in the pathophysiology of schizophrenia, including cellular response to dopamine, IL-1 signaling, T helper cell differentiation, and immunological synapse formation (Moni, Lin, Quinn, & Eapen, Reference Moni, Lin, Quinn and Eapen2021). In both schizophrenia and COVID-19, as a consequence of the inflammatory process, there is an increase in levels of kynurenic acid, which acts as an anti-excitotoxic factor (Collier, Zhang, Scrutton, & Giorgini, Reference Collier, Zhang, Scrutton and Giorgini2021). In addition, COVID-19 may induce psychosis through the development of the anti-N-methyl-D-aspartate (NMDA) receptor encephalitis, according to the theory of molecular mimicry of the SARS-CoV-2 (Payus et al., Reference Payus, Jeffree, Ohn, Tan, Ibrahim and Chia2022; Vasilevska et al., Reference Vasilevska, Guest, Bernstein, Schroeter, Geis and Steiner2021), and increase the activity of the midbrain dopamine neurons (Erhardt, Schwieler, Nilsson, Linderholm, & Engberg, Reference Erhardt, Schwieler, Nilsson, Linderholm and Engberg2007). Interestingly, activation of monocytes by toll-like receptor 3 (TLR-3) and depletion of natural killer cells associated with an early antiviral response has been shown to occur in both schizophrenia and anti-NMDA receptor encephalitis (Karpiński, Samochowiec, Frydecka, Sąsiadek, & Misiak, Reference Karpiński, Samochowiec, Frydecka, Sąsiadek and Misiak2018; Müller & Schwarz, Reference Müller and Schwarz2010; Müller et al., Reference Müller, Wagner, Krause, Weidinger, Wildenauer and Obermeier2012). In COVID-19, the TLR-3 pathway is linked to a tissue-destructive senescence-associated secretory phenotype that is related to short- and long-term complications (Tripathi et al., Reference Tripathi, Nchioua, Prata, Zhu, Gerdes and Giorgadze2021).
C-reactive protein (CRP) has been studied as a potential peripheral measure of immunologic activation that may have a causal or precipitating function in schizophreniform psychosis (Fond, Lançon, Auquier, & Boyer, Reference Fond, Lançon, Auquier and Boyer2018). HCoVs possess neuroinvasive properties as a result of either autoimmune responses or viral replication, leading to the hypothesis that they cause neuroinflammation and CNS penetration in psychotic disorders (Ferrando et al., Reference Ferrando, Klepacz, Lynch, Tavakkoli, Dornbush and Baharani2020). In fact, COVID-19 is involved in several inflammatory changes, including disruption of the Th1/Th2 and IL-2 balances, and alteration of Tregs (Aleebrahim-Dehkordi et al., Reference Aleebrahim-Dehkordi, Molavi, Mokhtari, Deravi, Fathi and Fazel2022).
Although it has been suggested that an increased risk of schizophrenia in the offspring is associated with maternal infection in earlier trimesters (Brown et al., Reference Brown, Schaefer, Wyatt, Goetz, Begg and Gorman2000), data on maternal and neonatal outcomes of COVID-19 infection are limited (Ashraf et al., Reference Ashraf, Keshavarz, Hosseinpour, Erfani, Roshanshad and Pourdast2020; Islam et al., Reference Islam, Poly, Walther, Yang, Wang and Hsieh2020; Kulaga & Miller, Reference Kulaga and Miller2021). Because so little time has elapsed since the beginning of the COVID-19 pandemic, it is too early to link cases of schizophrenia to SARS-CoV-2. This possibility is more likely for people who were exposed to this virus during the prenatal and perinatal periods, a similar dynamic that occurred in such a way after influenza epidemics when individuals developed schizophrenia due to exposure to the viral infection.
SARS and MERS lead to severe respiratory illnesses, and several studies have investigated the acute and post-illness neuropsychiatric outcomes of these diseases (Rogers et al., Reference Rogers, Chesney, Oliver, Pollak, McGuire and Fusar-Poli2020). The mental health impact of the SARS outbreaks in 2002–2003 caused significant psychological distress and morbidities among patients, healthcare workers, and the general public in affected regions (Chau et al., Reference Chau, Wong, Ramakrishnan, Chan, Wong and Li2021). The chronic psychiatric issues observed in SARS survivors underscore the potential long-term mental health complications that COVID-19 patients may face. Thus, in addition to the biological effects of COVID-19 that may increase the risk of schizophrenia, the COVID-19 pandemic may contribute to an increase in reactive psychotic disorders due to its stressful nature (Kulaga & Miller, Reference Kulaga and Miller2021; Valdés-Florido et al., Reference Valdés-Florido, López-Díaz, Palermo-Zeballos, Martínez-Molina, Martín-Gil and Crespo-Facorro2020). This may be explained by the diathesis-stress hypothesis. In this theory, stressful experiences have always been considered one of the most important factors in the development and exacerbation of mental disorders, independently of the effects of nervous system infections. One of the longest-standing pathoetiological hypotheses for schizophrenia is the interaction between external stimuli and internal vulnerability (Pruessner, Cullen, Aas, & Walker, Reference Pruessner, Cullen, Aas and Walker2017). According to this theory, psychosocial stress can cause the microglia to become pathologically activated, leading to excessive synaptic pruning and the loss of cortical gray matter. Therefore, damage to the stress-sensitive area may lead to immediate negative cognitive symptoms. Additionally, a loss of cortical control may prevent subcortical dopamine from acting, resulting in the positive symptoms of psychosis (Ma et al., Reference Ma, Jiang, Huang, Li, Zhang and Liu2021).
Section 3: Neuropsychiatric adverse effects of antiviral drugs
Within the context, neuropsychiatric adverse effects refer to brain-related symptoms developed during the treatment of pre-existing neurologic or psychiatric disorders, which range from mild symptoms, such as irritability and insomnia, to severe complications, including depression, psychosis, and painful peripheral neuropathy (Zareifopoulos, Lagadinou, Karela, Kyriakopoulou, & Velissaris, Reference Zareifopoulos, Lagadinou, Karela, Kyriakopoulou and Velissaris2020). However, it is very complex to distinguish whether these symptoms are due to the viral infection, to the immune response, or to the antiviral treatment.
In the case of the influenza virus, the most commonly used antivirals are neuraminidase inhibitors (such as oseltamivir or zanamivir), which prevent the viral spread and infection in the respiratory tract. However, a causal relationship between oseltamivir treatment and abnormal behavior has not yet been established, although the risk of abnormal behavior may be due to the viral infection, not to the antiviral drug (Ueda et al., Reference Ueda, Umetsu, Abe, Kato, Nakayama and Kato2015). Zanamivir treatment does not appear to cause neurological effects in humans. Table 3 shows the neuropsychiatric side effects and mechanisms of neurotoxicity of the major antiviral drugs.
Neuropsychiatric side effects and mechanisms of neurotoxicity of antivirals
Note: MAO-A: monoamine oxidase-A; CMMG: carboxymethoxymethylguanine; IFN: interferon; DAAs: direct-acting antivirals; NRTIs: nucleoside reverse transcriptase inhibitors; NNRTIs: non-nucleoside reverse transcriptase inhibitors; GABA: γ-aminobutyric acid; 5HT2A: serotonin; HPA: hypothalamic–pituitary–adrenal.
Acyclovir and valacyclovir are the most commonly used antivirals for herpes infections. Their neuropsychiatric adverse effects include agitation, altered state of consciousness, confusion, dysarthria, and hallucinations (Brandariz-Nuñez, Correas-Sanahuja, Maya-Gallego, & Martín Herranz, Reference Brandariz-Nuñez, Correas-Sanahuja, Maya-Gallego and Martín Herranz2021). In the case of famciclovir, potential neurogenic adverse effects include ataxia, dementia, dysarthria, encephalopathy, and tremor (Fang, Zhou, Han, Chen, Guan, & Li, Reference Fang, Zhou, Han, Chen, Guan and Li2024). The prevalence of nephrotoxic side effects ranged from 13% to 21%, while the prevalence of neurotoxicity was not clearly defined (Aboelezz & Mahmoud, Reference Aboelezz and Mahmoud2024). However, Brandariz-Nuñez et al. (Reference Brandariz-Nuñez, Correas-Sanahuja, Maya-Gallego and Martín Herranz2021) reported that neurotoxicity associated with acyclovir and valaciclovir was 73.9% and 29.4%, respectively. Adverse effects related to famciclovir were more common in women (59.72%) than in men (34.49%) (Fang et al., Reference Fang, Zhou, Han, Chen, Guan and Li2024).
Acute psychosis was reported in individuals with AIDS and older adults with poor renal function treated with gancyclovir (Zareifopoulos et al., Reference Zareifopoulos, Lagadinou, Karela, Kyriakopoulou and Velissaris2020). Several authors have reported that valacyclovir significantly improved verbal memory, working memory, and visual object learning, as well as schizophrenia symptoms (Bhatia et al., Reference Bhatia, Wood, Iyengar, Narayanan, Beniwal and Prasad2018; Deshpande & Nimgaonkar, Reference Deshpande and Nimgaonkar2018; Prasad et al., Reference Prasad, Eack, Keshavan, Yolken, Iyengar and Nimgaonkar2013; Tsai et al., Reference Tsai, Chen, Kuo, Hung, Tseng and Lai2020). However, other authors did not find significant improvement with this antiviral on the psychotic symptoms and cognitive function measures (Breier et al., Reference Breier, Buchanan, D’Souza, Nuechterlein, Marder and Dunn2019; Jonker et al., Reference Jonker, Doorduin, Knegtering, Van’t Hag, Dierckx and de Vries2023).
Antiviral drugs most used for the treatment of hepatitis B are nucleoside and nucleotide analogs (Kayaaslan & Guner, Reference Kayaaslan and Guner2017). Several of these drugs produce mitochondrial DNA depletion and have been associated with myopathy and peripheral neuropathy (Kamara et al., Reference Kamara, Smith, Ryom, Reiss, Rickenbach and Phillips2016). The combination of pegylated IFN and ribavirin was the treatment of choice for hepatitis C infection until the approval of direct-acting antivirals (DAAs) that presented improved patient tolerability and lower side effects (Gutiérrez-Rojas et al., Reference Gutiérrez-Rojas, de la Gándara Martín, García Buey, Uriz Otano, Mena and Roncero2023; Sakamaki et al., Reference Sakamaki, Kamimura, Fukui, Watanabe, Sakai and Tominaga2019). Psychiatric disorders were a common adverse outcome during the treatment with the former drugs (Davoodi et al., Reference Davoodi, Masoum, Moosazadeh, Jafarpour, Haghshenas and Mousavi2018; Sakamaki et al., Reference Sakamaki, Kamimura, Fukui, Watanabe, Sakai and Tominaga2019). However, Cheng et al. (Reference Cheng, Hu, Chang, Lin, Ku and Chien2021) demonstrated that the hepatitis C-associated risk of schizophrenia could be reversed by interferon-based antiviral therapy.
Nucleoside reverse transcriptase inhibitors (NRTIs) are nucleoside analogs lacking a 3-OH group, which are processed by the retroviral enzymes as unaltered nucleotides. The NRTIs incorporation into the newly synthesized viral DNA strand induces early termination of transcription, thereby inhibiting retroviral replication (Hirnschall, Harries, Easterbrook, Doherty, & Ball, Reference Hirnschall, Harries, Easterbrook, Doherty and Ball2013). Efavirenz is a non-nucleoside reverse transcriptase inhibitor (NNRTI) also used to treat HIV infection, usually in combination with NRTIs, with its clinical use discussed due to multiple psychiatric side effects (Clifford et al., Reference Clifford, Evans, Yang, Acosta, Goodkin and Tashima2005; Kenedi & Goforth, Reference Kenedi and Goforth2011). Several studies establish that efavirenz is a potent psychotropic drug with a higher affinity for GABA-A, 5-HT2A, and 5-HT2C receptors, while it also acts as a monoamine oxidase inhibitor and as a dual serotonin/dopamine reuptake inhibitor (Gatch et al., Reference Gatch, Kozlenkov, Huang, Yang, Nguyen and González-Maeso2013). Nevirapine is another NNRTI that has rarely been associated with neuropsychiatric adverse effects (Wise, Mistry, & Reid, Reference Wise, Mistry and Reid2002), although van Griensven et al. (Reference van Griensven, Zachariah, Rasschaert, Mugabo, Atté and Reid2010) reported its association with hepatotoxicity and peripheral neuropathy.
Protease inhibitors can also be used as part of active antiretroviral therapy. One of these, ritonavir, is an inhibitor of cytochrome P450 3A4 (CYP3A4) and is synergistic with drugs that rely on CYP3A4 metabolism for inactivation. However, its use has been associated with a lipodystrophy syndrome characterized by insulin resistance, dyslipidemia, central obesity, and an increased risk of cerebrovascular disease (Duval et al., Reference Duval, Journot, Leport, Chêne, Dupon and Cuzin2004; Gupta et al., Reference Gupta, Knight, Losso, Ingram, Keller and Bruce-Keller2012). Other HIV antivirals, such as raltegravir, elvitegravir, and dolutegravir, inhibit retroviral integrase, the enzyme that allows the integration of the viral DNA transcript into the host cell genome (Hoffmann & Llibre, Reference Hoffmann and Llibre2019). Side effects similar to those of efavirenz have been reported mainly for dolutegravir, but no evidence has been reported that provides insight into the mechanisms of integrase inhibitor-induced neuropsychiatric effects (Zareifopoulos et al., Reference Zareifopoulos, Lagadinou, Karela, Kyriakopoulou and Velissaris2020). The most recently approved antiretrovirals are maraviroc and enfuvirtide, two agents included in the class of entry and fusion inhibitors, which are not considered first-line agents and are therefore used in combination with other antivirals. Clinical data indicate that maraviroc induces an increase in the CD4+ count of patients (Llibre et al., Reference Llibre, Rivero, Rojas, Garcia Del Toro, Herrero and Arroyo2015), but the side effects of these substances have not yet been established.
Discussion
Schizophrenia appears to be the result of the influence of hereditary and environmental factors. Some environmental contributors include birth complications, maternal nutritional deficiencies, medication use, stress during pregnancy, and viral infections. Prenatal exposure to stressors, especially early in pregnancy, is critical for fetal hippocampal development, which influences an individual’s likelihood of developing a psychotic disorder (Brown & Derkits, Reference Brown and Derkits2010) and schizophrenia (Brown, Reference Brown2011; Reference Brown2012). Epidemiological research has identified viral infection as one of the environmental risk factors for schizophrenia (Khandaker, Zimbron, Lewis, & Jones, Reference Khandaker, Zimbron, Lewis and Jones2013). Some viral infections can disrupt normal fetal CNS development by activating the maternal immune system. Therefore, changes in the expression of key genes and increased levels of inflammatory cytokines are thought to be the connecting link between viral infection and schizophrenia (Huang, Zhang, & Zhou, Reference Huang, Zhang and Zhou2022). Figure 1 shows hypothetical interactions between viral infection and the development of schizophrenia (according to the Inflammatory Model and to Kępińska et al., Reference Kępińska, Iyegbe, Vernon, Yolken, Murray and Pollak2020).
Hypothetical interactions between viral infection and the development of schizophrenia (according to the Inflammatory Model and to Kępińska et al., Reference Kępińska, Iyegbe, Vernon, Yolken, Murray and Pollak2020).
Inflammation seems to be associated with several processes, such as microbial infections, obesity, tobacco smoking, and autoimmune diseases. Macrophages play a pivotal role as the major cellular component of the adipose tissue regulating chronic inflammation and modulating the secretion and differentiation of various pro- and anti-inflammatory cytokines (Savulescu-Fiedler et al., Reference Savulescu-Fiedler, Mihalcea, Dragosloveanu, Scheau, Baz and Caruntu2024). Elisia et al. (Reference Elisia, Lam, Cho, Hay, Li, Yeung, Bu, Jia, Norton, Lam and Krystal2020) evaluated the effects of smoking on inflammatory markers, and they found that plasma samples from heavy smokers had significantly higher CRP, fibrinogen, IL-6, and carcinoembryonic antigen levels than non-smoking controls. A possible mechanism involving many pro-inflammatory cytokines has been proposed to explain the causal relationship between the virus and schizophrenia (Figure 1). These modulators may act either directly on neurons or indirectly via neurotransmission, for example, through TNF-α (Xiu et al., Reference Xiu, Man, Wang, Du, Yin and Zhang2018). Moreover, the TNF-α gene is located at a locus previously associated with genetic susceptibility to schizophrenia. Additionally, increased complement protein activity, particularly C1q, C3, and C4, was found to contribute to the accelerated synapse pruning (Presumey, Bialas, & Carroll, Reference Presumey, Bialas and Carroll2017). These authors reported that the complement system stimulates synapse loss in the early stages of neurodegenerative diseases. Similar pathways can also be activated in response to inflammation, such as in West Nile virus infection, where peripheral inflammation can promote microglia-mediated synapse loss (Stonedahl, Clarke, & Tyler, Reference Stonedahl, Clarke and Tyler2020).
In the present review, we have examined the influence of viral infections on schizophrenia, establishing that some of these are neurotropic and have the potential to provoke neurological disorders (Bauer et al., Reference Bauer, Laksono, de Vrij, Kushner, Harschnitz and van Riel2022; Capendale, Wolthers, & Pajkrt, Reference Capendale, Wolthers and Pajkrt2023; Ludlow et al., Reference Ludlow, Kortekaas, Herden, Hoffmann, Tappe and Trebst2016; Marcocci et al., Reference Marcocci, Napoletani, Protto, Kolesova, Piacentini and Li Puma2020; Meyding-Lamadé, Craemer, & Schnitzler, Reference Meyding-Lamadé, Craemer and Schnitzler2019). In addition, some of these viruses induce latent infections in the host that can be reactivated by psychological stress (Cliffe et al., Reference Cliffe, Arbuckle, Vogel, Geden, Rothbart and Cusack2015; Griffiths & Reeves, Reference Griffiths and Reeves2021; Klopack, Reference Klopack2023) or by an inflammatory response to the viral infection (Cuddy et al., Reference Cuddy, Schinlever, Dochnal, Seegren, Suzich and Kundu2020; Savitz & Harrison, Reference Savitz and Harrison2018), factors closely linked to the onset of several psychiatric disorders (Goldsmith & Rapaport, Reference Goldsmith and Rapaport2020; Kendler & Gardner, Reference Kendler and Gardner2016; Pape, Tamouza, Leboyer, & Zipp, Reference Pape, Tamouza, Leboyer and Zipp2019; Savitz & Yolken, Reference Savitz and Yolken2023). Furthermore, in this review, we have also explored the neuropsychiatric impact of antiviral drugs. Although several studies have established that antiviral therapy reduces the risk of schizophrenia (Breier et al., Reference Breier, Buchanan, D’Souza, Nuechterlein, Marder and Dunn2019; Jonker et al., Reference Jonker, Doorduin, Knegtering, Van’t Hag, Dierckx and de Vries2023; Tsai et al., Reference Tsai, Chen, Kuo, Hung, Tseng and Lai2020), the negative effects of these antivirals have been poorly studied. On the basis of shared consequences of antivirals and schizophrenia, we can hypothesize that some antivirals provoke side effects compatible with schizophrenia symptoms, such as hallucinations, delusional beliefs, cognitive impairments, and psychosis, to name a few. Nevertheless, the studies reviewed present several significant limitations that must be considered when interpreting their findings. Many studies were observational or cross-sectional, which limits the capacity to establish causal relationships between viral infections, antiviral treatments, and neuropsychiatric outcomes, particularly in the context of schizophrenia. Small sample sizes were a common issue, undermining the statistical power of the studies and potentially leading to type II errors. Inconsistent reporting of variables, including variations in diagnostic criteria, medication types, and outcome measurement tools, introduced significant heterogeneity that reduced the robustness of the conclusions. Additionally, many studies did not adequately account for confounding factors such as genetic predispositions, comorbidities, or lifestyle factors like smoking, which could have influenced the results. Retrospective designs, reliance on clinical diagnoses rather than structured interviews, and incomplete or missing data (e.g., viral load, liver function tests, or genetic markers) further compromised the validity of the findings. Some studies also faced difficulties in standardizing methodologies, including the arbitrary selection of CRP cut-off values, the use of dichotomized data, and inconsistent cytokine panels. This lack of standardization, combined with publication and selection biases, limits the potential to generalize the results. Furthermore, the absence of long-term follow-up data and the failure to include diverse patient populations (e.g., regional biases) complicates the interpretation of the long-term effects of viral infections and antiviral treatments on schizophrenia. The limited focus on specific mental health diagnoses, as well as the under-recording of certain symptoms such as delirium in COVID-19 patients, may have also affected the outcomes. These methodological weaknesses point to the demand for well-designed, large-scale longitudinal studies with standardized protocols, better control of confounding factors, and more inclusive and diverse cohorts to clarify the complex relationships between viral infections, antiviral treatments, and neuropsychiatric conditions.
In addition to the key role of viral infections in the etiology of schizophrenia, it is essential to highlight the societal implications of this particular condition. Public perception often associates schizophrenia with aggressive tendencies, a belief reinforced by media coverage of violent incidents involving individuals identified as mentally ill (Wehring & Carpenter, Reference Wehring and Carpenter2011). Studies indicate that schizophrenia is among the most stigmatized mental illnesses (de Jacq, Norful, & Larson, Reference de Jacq, Norful and Larson2016; Reisinger & Gleaves, Reference Reisinger and Gleaves2023; Mannarini, Taccini, Sato, & Rossi, Reference Mannarini, Taccini, Sato and Rossi2022; Valery & Prouteau, Reference Valery and Prouteau2020), fostering prevalent beliefs about dangerousness and incompetency that contribute to a poor prognosis and increased social distance, further isolating those affected and hindering access to support and resources (Valery & Prouteau, Reference Valery and Prouteau2020). Specifically, the perception of individuals with schizophrenia as dangerous influences the inclination for social distance, with beliefs about biogenetic causes and appropriate medical treatment also shaping these perceptions (Mannarini et al., Reference Mannarini, Taccini, Sato and Rossi2022). Moreover, social exclusion is affected by the presence of negative symptoms and diagnosis awareness, where increased knowledge of the diagnosis leads to greater social distance when symptoms are absent, and decreased distance when symptoms are present (Zahid & Best, Reference Zahid and Best2021). Perceived discrimination and stigma consciousness negatively impact psychological well-being, diminishing self-esteem and social functioning, while disrupting daily activities and leading to heightened mental health adverse outcomes and reduced quality of life (Magallares, Perez-Garin, & Molero, Reference Magallares, Perez-Garin and Molero2016; Lampropoulos, Fonte, & Apostolidis, Reference Lampropoulos, Fonte and Apostolidis2019). Furthermore, higher levels of self-stigma are observed in individuals with schizophrenia spectrum disorders compared to those with depressive disorders, despite similarities in overall quality of life, suggesting that the severity of the mental disorder significantly influences self-stigmatization (Holubová et al., Reference Holubová, Prasko, Matousek, Latalová, Maracková and Vrbová2016). In addition, bullying victimization has been identified as a major risk factor for the increased incidence of schizophrenia (Jester et al., Reference Jester, Thomas, Sturm, Harvey, Keshavan and Davis2023), suggesting that a key prevention strategy could involve studying the psychological experiences of victims, with particular focus on the impact of negative emotions such as humiliation (Borrego-Ruiz & Fernández, Reference Borrego-Ruiz and Fernández2024). Therefore, individuals with schizophrenia often face significant barriers in several aspects of life, which underscores the pressing need to enhance mental health education and assistance services, and also to dispel the uncertainty around the complex interaction of genetic and environmental factors that underlie schizophrenia, thereby clarifying its causes and raising awareness about the experiences of people suffering from it. Ultimately, understanding the diverse influencing factors of schizophrenia could lead to more targeted intervention approaches, potentially reducing some of the social and individual burdens associated with the disorder.
This review presents various potential limitations that should be appropriately acknowledged: (i) it only considers specific infections that are central to the topic under study, but the pathophysiology of schizophrenia may involve more complex and multifactorial processes in which immune dysregulation could constitute a contributing factor rather than being attributable to a single infectious agent; (ii) the potential for prevention through specific treatments is not addressed, as the primary aim of this review is to delineate underlying pathological mechanisms rather than propose direct therapeutic interventions; (iii) the samples used in some of the studies reviewed, including both experimental and control groups, exhibit heterogeneity, which may affect the comparability of findings within the broader context of neuropsychiatric disorders; (iv) the contributions of maternal infections, in utero exposures, and postnatal infections are not separately analyzed, which may limit the clarity regarding their distinct roles in the pathogenesis of schizophrenia; and (v) the impact of confounding factors, including body mass index, drug use, genetic predisposition, and socioeconomic status, may introduce bias and affect the interpretation of the results.
Conclusions
Epidemiological evidence suggests a potential relationship between viruses and schizophrenia. Some viral infections, such as the influenza virus, can disrupt fetal brain development by activating the maternal immune system. Increases in inflammatory cytokine levels and changes in the expression of key genes observed in several viral infections (e.g., coronaviruses) may constitute potential links between these viral infections and schizophrenia. In addition, immune and non-immune genes associated with schizophrenia are likely to be targets of viral proteins.
Neuropsychiatric effects caused by antiviral drugs are common and represent significant adverse outcomes for viral treatment. From the data presented in this review, it can be concluded that antivirals may affect the CNS, although for most drugs, their action mechanisms are still unclear, and a strong relationship between antivirals and schizophrenia has not yet been established. Therefore, further research is required to elucidate the mechanisms underlying the neuropsychiatric effects of antiviral drugs.
Funding statement
No funding was needed for the development of this article. The open access publication charge was funded by the University of Málaga/CBUA.
Competing interest
The authors declare no competing interests exist.
Open access
