Schizophrenia is a deliberating and severe neuropsychiatric disease with a cluster of psychotic symptoms and cognitive impairment. Early studies have documented that maternal infection during pregnancy is implicated with increased risk of schizophrenia in the offspring (Brown and Derkits, 2010). It is known that certain clinical and genetic features are sheared between some autoimmune diseases and schizophrenia (Benros et al., 2011). Therefore, it has been proposed that immunologic abnormalities is likely to be involved in the development of schizophrenia (Tomasik et al., 2016).

Evidence from genetic studies have indicated that the concordance rates of schizophrenia for monozygotic twins are around 40–50%, and its heritability is estimated approximately 80% (Sullivan et al., 2003). As such, genetic studies of schizophrenia may provide some clues to uncover the biological mechanism of this disorder (Dhindsa and Goldstein, 2016). Several genome-wide association studies (GWASs) have consistently reported that schizophrenia susceptibility genes are located in the region of extended human major histocompatibility complex (MHC) (Shi et al., 2009; Stefansson et al., 2009). A recent meta-analysis on postmortem brain studies showed a prominent increase in the expression of pro-inflammatory genes in schizophrenia patients (van Kesteren et al., 2017). Our previous studies have identified the important roles of complement factors and pro-inflammatory cytokines in the pathophysiology of schizophrenia (Ni et al., 2015; Wang et al., 2015; Zhang et al., 2017a; Zhang et al., 2017b; Zhang et al., 2018). These results strengthened the hypothesis of immune-inflammatory processes in the pathogenesis of schizophrenia.

Based on this hypothesis, maternal immune activation (MIA) animal model is used to mimick schizophrenia-like pathological, neurochemical changes and behaviors in rodents (Zuckerman and Weiner, 2005; Wolff et al., 2011). Recent studies demonstrated that nod-like receptor pyrin domain-contraining protein 3 (NLRP3) inflammasome is involved in the schizophrenia-like behaviors of MIA model (Ventura et al., 2020; Talukdar et al., 2021). The NLRP3 inflammasome plays a key role in the innate immune system that mediates pro-inflammatory cytokines (Ventura et al., 2020). The involvement of NLRP3 in major psychiatric disorders has been reported in patients with major depressive disorder (Alcocer-Gomez et al., 2014; Tian et al., 2021) and bipolar disorder (Kim et al., 2016; Scaini et al., 2019). However, the relationship of NLRP3 with schizophrenia is not well known.

The human gene encoding NLRP3 (NLRP3) is located on chromosome 1q44, where linkage findings were reported in schizophrenia (Saviouk et al., 2005). To the best of our knowledge, there is no genetic study to identify the role of NLRP3 in schizophrenia. In this study, we hypothesized that NLRP3 may confer susceptibility to schizophrenia. First, we examined the difference of NLRP3 mRNA expression between schizophrenia patients and healthy controls. Rs10754558 is a functional single-neucleotide polymorphism (SNP) and mapped in the 3′-untranslated region of NLRP3. Evidence showed that this SNP influences the stability of NLRP3 mRNA (Hitomi et al., 2009). Thus, we secondly sought to characterize the association of NLRP3 rs10754558 polymorphism with schizophrenia in Han Chinese. Finally, we planned to conduct an eQTL (expression quantitative trait loci) analysis to detect the role of the rs10754558 polymorphism in NLRP3 mRNA expression in brain.

For the expression analysis, demographic characteristics of our sample were presented in Supplementary Table S1. Schizophrenia and control groups are well-matched with respect to age, sex and smoking. The RT-PCR results showed no significant difference in NLRP3 mRNA levels between schizophrenia and control groups (p = 0.07, Figure 1). We extracted brain NLRP3 mRNA expression data from SZDB database. We did not observe significant differences of NLRP3 mRNA expression in prefrontal cortex, hippocampus and striatum between schizophrenia patients and healthy controls (Supplementary Figure S1).

For the genetic analysis, we enrolled 823 schizophrenia patients and 859 healthy controls in this study. Demographic characteristics are well matched in both groups. No deviation from HWE was observed in genotype distribution for rs10754558 polymorphism in control group. The results of rs10754558 polymorphism were presented in Table 1. No significant differences in allele and genotype frequencies were found between the schizophrenia and control groups. With false positive rate controlled as 0.05, the statistical power to detect the odds ratio (OR) value as 1.5 for risk allele was expected to be more than 85% in our samples under the log additive model. We further examined the genetic association between the rs10754558 polymorphism and schizophrenia in PGC database, and we also did not observe significant association of rs10754558 polymorphism with schizophrenia (Supplementary Figure S2).

There were 239 drug-naïve patients with first-episode schizophrenia participated in the clinical evaluation. We examined the association of rs10754558 polymorphism with psychotic symptoms and cognitive performance in schizophrenia patients by comparing scores of the PANSS and RBANS with genotypes of SNP rs10754558. We observed significant differences in the scores of RBANS attention and total scores between the patients with different genotypes of rs10754558 polymorphism (p = 0.001 and p < 0.01, respectively), as shown in Table 2. Post-hoc comparisons showed that rs10754558 G/G or G/C carriers had lower RBANS attention and total scores than those with C/C carriers (Ps<0.01). Subsequently, we conducted an eQTL analysis to test the influence of rs10754558 polymorphism on NLRP3 mRNA expression in brain. Figure 2 showed that carriers with G allele have higher levels of NLRP3 expression than those without G allele in frontal cortex.

There is a growing body of evidence implying that schizophrenia is attributed to a dysregulation of innate and adaptive immune system (Michel et al., 2012). Our previous studies have provided supportive evidence for the involvement of innate immune response in the pathogenesis of schizophrenia (Ni et al., 2015; Wang et al., 2015; Zhang et al., 2017b). NLRP3 is an intracellular protein involved in the initiation of the innate immune response (Bonar et al., 2012). It is known that NLRP3 mutation confers susceptibility to immune disorders (Pontillo et al., 2010). Accordingly, we employed this study to identify the role of NLRP3 in schizophrenia.

In this study, we found that peripheral NLRP3 mRNA level has no difference between patients with schizophrenia and healthy controls. We also did not observe the difference of NLRP3 mRNA level in brain between schizophrenia and controls. It has been well-established that MIA is associated with an increased risk of schizophrenia in offspring (Scola and Duong, 2017). Recent preclinical studies have found that activation of NLRP3 when LPS counts may be one of the potential mechanisms driving schizophrenia-like behavior in offspring in MIA model (Ventura et al., 2020; Talukdar et al., 2021). However, a postmortem study compared the expression of NLRP3-related protein in brain tissue from patients with schizophrenia and bipolar disorder in which increased level of NLRP3 was observed in patients with bipolar disorder, but not schizophrenia (Kim et al., 2016). Another comparative study also showed that peripheral level of NLRP3 is increased in patients with bipolar disorder, but not schizophrenia (Garcia-Alvarez et al., 2018). As such, our findings implied that NLRP3 mRNA level is not a risk factor for schizophrenia.

Another aim of this study is to detect the association of NLRP3 functional SNP rs10754558 with schizophrenia. SNP rs10754558 is located in the 3′-untranlated region (3′-UTR) of NLRP3 and affects expression of NLRP3 mRNA through altering its stability (Hitomi et al., 2009). This SNP has been reported to confer susceptibility to disorders of the central nervous system, such as Alzheimer’s disease (Tan et al., 2013), multiple sclerosis (Imani et al., 2018) and ischemic stroke (Lv et al., 2020). However, our results did not support the significance for rs10754558 polymorphism in the susceptibility to schizophrenia in either Han Chinese or European ancestry from PGC GWAS. This suggests that NLRP3 SNP rs10754558 is probably not a risk SNP for schizophrenia.

Schizophrenia is known to have variable clinical presentation. Therefore, we further sought to determine the role of rs10754558 in the expression of clinical features in a sample of drug-naïve patients with first-episode schizophrenia. We observed a significant association of rs10754558 polymorphism with RBANS attention and total scores in such patients, but not psychotic symptoms. This suggested that rs10754558 polymorphism may result in cognitive impairment, especially attention deficit in schizophrenia. We further carried out an eQTL analysis to detect the effect of rs10754558 polymorphism on NLRP3 mRNA expression in brain. Data showed that rs10754558 polymorphism is significantly associated with the NLRP3 mRNA expression in frontal cortex, and the risk allele G is associated with increased level of NLRP3 mRNA. There is accumulating evidence from magnetic resonance imaging (MRI) studies showing that frontal regions are critical for cognitive impairment in schizophrenia (Glascher et al., 2019). Our recent findings indicated that dysfunctional connectivity of frontal cortices is associated with cognitive impairment in patients with schizophrenia (Yu et al., 2021). North et al. (2021) found that elevated level of peripheral inflammation is a risk factor for cognitive deficits and brain structure in schizophrenia, especially reduced attention and cortical thickness. Meta-analysis also showed that attention deficit has a strongest relationship with peripheral inflammation in all cognitive domains tested in patients with schizophrenia (Bora, 2019). Our previous studies have demonstrated that elevated levels of pro-inflammatory cytokines are associated with cognitive impairment in schizophrenia (Zhang et al., 2017a; Zhang et al., 2018; Xu et al., 2019). NLRP3 is a critical regulator of pro-inflammatory cytokines production by activated caspase-1 to initiating an inflammatory form of cell death (Feng et al., 2021). Recent literature documented that enriched NLRP3 may lead to cognitive deficits, and inhibition of NLRP3 could ameliorate such impairment (Fang et al., 2021). As such, our results implied that increased level of NLRP3 mRNA may result in cognitive impairment, especially attention deficit in patients with schizophrenia, possibly through triggering a cascade of inflammatory responses in frontal cortex.

Some limitation in this study should be mentioned. First, we evaluated the NLRP3 mRNA in peripheral level, not in brain. It is uncertain whether the peripheral level of NLRP3 mRNA is correlated with the level in the central nervous system. Second, the lack of a significant association of rs10754558 polymorphism and schizophrenia may be caused by the modest sample size, whereas the power reached statistical significance. Accordingly, our findings should be considered only preliminary and require further investigations for validation in independent samples. Third, this study employed samples from Eastern China and may not be representative of the Han Chinese population in general, nor other closely related populations in the area. Finally, we have analyzed only one functional SNP in NLRP3 to evaluate the genetic relationship with schizophrenia. This means further analyses need to be carried out with more variants in the NLRP3 for validation of our results.

In conclusion, we performed a comprehensive analysis to detect the association of NLRP3 with schizophrenia in Han Chinese. Although our findings did not support NLRP3 confer susceptibility to schizophrenia, NLRP3 may be a risk factor for cognitive impairment, especially attention deficit in this disorder. Further investigations are required for validation of our results in larger and independent sample across various ethnicities.