Methamphetamine (METH), an amphetamine-type stimulant, is the most commonly abused drug in China, with 1.18 million officially registered drug users reporting methamphetamine use, accounting for 55.5% of all drug users nationwide by the end of 2019 (Office of China National Narcotics Control Commission, 2020). Individuals with methamphetamine use disorder (MUD) often suffer from serious neuropsychological symptoms including psychosis, depression, anxiety and cognitive deficits, even after extended periods of abstinence (London et al., 2004; Paulus and Stewart, 2020). Despite of the adverse health consequences and economic burden ascribed to MUD (United Nations Office on Drugs and Crime, 2021), there are currently no FDA-approved pharmacological interventions available for treating MUD and limited evidence for the effectiveness of cognitive-behavioral therapy (Paulus and Stewart, 2020), highlighting the urgent need for more effective treatment options.

In recent years, there has been an increasing interest in the gut microbiota as a potential target for novel therapeutics, due to its tremendous impact on the brain, behavior, and health (Cryan et al., 2019). While the majority of work concerning the role of the gut microbiota in neuropsychiatric conditions has focused on autism spectrum disorder (Sharon et al., 2019; Wang et al., 2019), Alzheimer’s disease (Vogt et al., 2017; Bulgart et al., 2020), Parkinson’s disease (Petrov et al., 2017; Aho et al., 2019), depression (Bastiaanssen et al., 2020; Yang et al., 2020), anxiety (Jiang et al., 2018) and schizophrenia (Zhu et al., 2019; Golofast and Vales, 2020), the evidence of interactions between the gut microbiota and substance use has been relatively limited (Qin et al., 2021; Salavrakos et al., 2021), and even less is known in the context of METH use.

Evidence of METH-induced dysbiosis largely comes from observations using rodent models. Significant changes in the composition of gut microbiota have been found in rats and mice under different regimens of METH exposure (Ning et al., 2017; Angoa-Pérez et al., 2020; Chen et al., 2021; Forouzan et al., 2021; Wang et al., 2021). Further, the alteration of Akkermansia has been identified to be associated with the rat’s behavioral response to METH in a conditioned place preference (CPP) model (Yang C. et al., 2021), suggesting a possible role of gut microbiota in modulating METH-induced behavior and vulnerability to MUD. However, clinical evidence in relation to MUD remains scarce. So far, only two recently published clinical studies have addressed this issue (Cook et al., 2019; Yang Y. et al., 2021). A study among men who have sex with men (MSM) found that METH use in the past six months was associated with higher levels of pro-inflammatory bacteria and lower levels of butyrate-producing genera in rectal swab samples (Cook et al., 2019). While the study cohort was comprised entirely of MSM, all of whom were engaging in anal intercourse, findings from this study might not generalize to the heterosexual population since it has been suggested that sexual behavior might influence the composition of gut microbiota (Noguera-Julian et al., 2016). Another study involving 16 male methamphetamine abusers and 14 healthy subjects also reported alterations in gut microbiota diversity and composition associated with MUD (Yang Y. et al., 2021). However, this study was limited by a relatively small sample size and inablility to adjust for possible confounders, highlighting the need for additional investigation.

Multiple mechanisms, including immune, endocrine and neuronal pathways have been described to be involved in the microbiota-gut-brain signaling (Agirman and Hsiao, 2021). For MUD, the interaction between gut microbiota and the host immune system has been assumed to be one of the key mechanisms. It was hypothesized that METH-induced pro-inflammatory microbiota profile and gut barrier dysfunction might allow the translocation of bacteria and bacterial components into the circulation, thus exacerbate systemic inflammation, abnormal immune response and neuroinflammation, leading to aberrant brain function and behaviors (Prakash et al., 2017; Qin et al., 2021), but no definitive evidence currently exists to prove this hypothesis. Indeed, studies have shown that METH elicits dysregulation in the innate and adaptive immune response, causing changes to both pro- and anti- inflammatory cytokines (Papageorgiou et al., 2019). However, no previous study has investigated the possible connection between gut microbiota and systemic inflammation in MUD. Therefore, the present study is set out to characterize the gut microbiota in MUD and preliminary analyze the correlation between the altered gut microbiota, the markers of systemic inflammation and clinical features in MUD.

To the best of our knowledge, the current study is the first to provide evidence for a link between altered fecal microbiota and systemic inflammation in MUD. Consistent with previous studies (Cook et al., 2019; Yang Y. et al., 2021), our results demonstrated that subjects with MUD exhibited differences in the relative abundance of several microbial taxa. Importantly, the order Lactobacillales, containing several anti-inflammatory bacterial species, decreased in MUD and correlated positively with the plasma level of anti-inflammatory cytokine IL-10, as well as the duration of METH abstinence, suggesting a potential role of the gut-immune-brain axis (Fung et al., 2017) in MUD.

In our study, no difference was observed between the MUD and HC groups regarding the alpha diversity of fecal microbiota, which is in agreement with previously published literature (Cook et al., 2019). With respect to beta diversity, recent studies reported a significant variation in overall microbial composition between METH users and non-users (Fulcher et al., 2018; Cook et al., 2019), which was not replicated in the present study. This inconsistency may be due to the longer duration of abstinence in our study population. The METH-users in the aforementioned study reported METH use within 6 months (Cook et al., 2019), whereas the median duration of METH withdrawal in our study was 8.22 months. A recent study conducted in rats found that the significant difference in beta diversity caused by METH exposure was no longer evident after 7 days of cessation (Forouzan et al., 2021). Therefore, the similarity of the overall microbial structure between the MUD and HC groups in the present study might be explained by the restoration during prolonged abstinence.

According to our results, Collinsella, a genus belonging to the family Coriobacteriaceae and the phylum Actinobacteria, was recognized as the most robustly associated with MUD and positively related to the total years of METH use. In humans, Collinsella has been linked to poor metabolic states including obesity (Gomez-Arango et al., 2016; Frost et al., 2019), type 2 diabetes mellitus (Candela et al., 2016), hypercholesterolemia (Khan et al., 2018) and nonalcoholic steatohepatitis (Astbury et al., 2020). Additionally, Collinsella was considered as a biomarker in patients with symptomatic atherosclerosis (Karlsson et al., 2012). Since METH use is closely related to enhanced atherosclerotic plaque formation (Kevil et al., 2019), enrichment of Collinsella in MUD may indicate an increase susceptibility for cardiovascular disease, which is among the leading causes of death associated with MUD (Paulus and Stewart, 2020). Higher abundance of Collinsella has also been observed in patients with schizophrenia (Shen et al., 2018; Li et al., 2020). Given the considerable overlap in the clinical symptoms of METH-induced psychosis and schizophrenia (Wearne and Cornish, 2018), as well as the usage of METH in inducing animal models of schizophrenia-like phenotype (Machiyama, 1992; Oka et al., 2020), it is therefore likely to be a connection between Collinsella and psychotic symptoms in MUD, which was not investigated in this study. Furthermore, Collinsella has been associated with negative affect assessed by the Positive Affect and Negative Affect Schedule (PANAS) in healthy adults (Lee et al., 2020). However, no evidence of correlations between Collinsella and depression or anxiety was detected in the current study, possibly attributed to the differences in the scales used for measurement and the study population. With respect to substance use, Fulcher et al. (2018) found that decreased abundance of Collinsella was associated METH use during chronic HIV-infection among young men who have sex with men, which is inconsistent with our results. As participants with evidence of HIV infection were ruled out in our study, this contradiction may be due to the influence of uncontrolled confounders in Fulcher et al.’s study. Besides, a reduction of Collinsella has been indicated in alcohol use disorder (Litwinowicz et al., 2020). Notably, our results supported the negative association between Collinsella and alcohol use by incorporating the AUDIT score as a covariate ( Supplementary Table 1 ). More importantly, Collinsella has been shown to influence the epithelial production of IL-17A and contribute to hyper-permeability of the gut by reducing the expression of the tight junction protein ZO-1 in research of rheumatoid arthritis (Chen et al., 2016). According to these data, we may infer that the observed higher abundance of Collinsella in MUD might further contribute to raise the host inflammatory level although no correlations between Collinsella and levels of inflammatory factors were found in our study.

In the current study, MUD was characterized by a reduction in the relative abundance of genera Faecalibacterium, Blautia, Dorea and Streptococcus, most of which are in line with previously reported data (Xu et al., 2017; Cook et al., 2019), although these associations did not remain significant after adjusting for confounders. Decreased abundance of Faecalibacterium has also been found in METH-users from a large cohort study (Cook et al., 2019). Faecalibacterium prausnitzii, the only identified species in the genus Faecalibacterium, is one of the most abundant butyrate-producing bacteria in the gastrointestinal tract and has demonstrated anti-inflammatory effects via NF-κB inhibition and IL-10 induction (Miquel et al., 2013). Thus, it is probable that the reduction in the abundance of Faecalibacterium may exacerbate inflammation in MUD. In contrast to our results, mice exposed to escalating dose-multiple binge METH presented a significantly higher abundance of Blautia (Chen et al., 2021), probably because of the species differences between human and rodent model. Blautia has been shown to play an important role in maintaining environmental balance in the intestine and preventing inflammation by upregulating intestinal regulatory T cells and producing short chain fatty acids (Kim et al., 2014). Nonetheless, it remains unclear whether Blautia is good for human health given that higher abundance of Blautia was detected under unhealthy conditions such as irritable bowel syndrome (Rajilić–Stojanović et al., 2011) and ulcerative colitis (Nishino et al., 2018) as well. In fact, it has been supposed that different species of Blautia may exert beneficial or adverse effects (Liu et al., 2021). In our study, Blautia was positively correlated with the plasma level of pro-inflammatory cytokine IL-2. Although the simultaneous increase of the relative abundance of Blautia and the level of IL-2 has been reported (Qi et al., 2019), further investigations are required to identify the relationship between Blautia and IL-2. In accord with our findings, a study comparing subjects with substance use disorder (SUD) with healthy controls reported a depletion of the genus Dorea, the genus Streptococcus and upper-level taxa including the family Streptococcaceae, the order Lactobacillales and the class Bacilli in the SUD group (Xu et al., 2017). Enrichment of the genus Megasphaera in the HC group was also reported (Xu et al., 2017); however, our study found the opposite. The discrepancy might be explained by the fact that participants with MUD made up only for 30% of the SUD group in the study.

Perhaps the most notable finding is that the order Lactobacillales, exhibiting lower abundance in the MUD group, was positively correlated with the duration of METH abstinence and the plasma level of anti-inflammatory cytokine IL-10. Lactobacillales, a dominant order in class Bacilli and phylum Firmicutes, is considered as a beneficial commensal. Bacteria in the order Lactobacillales produce lactic acid as a major end-product of carbohydrate fermentation and are among the most common microorganisms used as probiotics (Gareau et al., 2010; Lahtinen et al., 2011). Correlation analysis in our study indicated that the relative abundance of Lactobacillales increased as the withdrawal period progressed, suggesting a tendency towards restoration during abstinence. Bacterial strains belonging to Lactobacillales are known to have potential to modulate cytokine release (Gareau et al., 2010; Vemuri et al., 2017). Specifically, administration of Lactobacillus casei suppresses proinflammatory responses by increasing IL-10 levels (So et al., 2008b; So et al., 2008a). Interestingly, our results also revealed that Lactobacillales was positively correlated with IL-10 levels. While the mechanism for MUD-associated decrease in Lactobacillales remains unclear, the correlation between Lactobacillales and IL-10 would seem to suggest a potential role of the altered gut microbiota in MUD by modulating the systemic immune system of the host.

Though not the primary object of our study, it is somehow surprising to find that the frequency of METH use was conversely correlated with plasma levels of IL-6. In contrast, Luo et al. (2021) found a positive association between IL-6 and frequency of METH use among chronic METH users with an average abstinence time of about 40 days, much shorter than that of our subjects. Another study in which alcohol-dependent subjects hospitalized for a 3-week detoxification program reported a positive correlation between IL-10 and alcohol craving at onset of withdrawal (Leclercq et al., 2012). After 3 weeks of abstinence, however, IL-10 was found negatively relating to depression, anxiety and craving (Leclercq et al., 2012). As in our study, IL-10 was positively correlated with craving for METH while negatively correlated with anxiety measured by GAD-7. In addition, CRP, a typical inflammation marker, presented negative correlations with the frequency of METH use and craving in our study. Many of findings were contrary to expectations that higher pro-inflammatory markers would relate to more frequent METH use, stronger craving and more severe negative affect (i.e., depression and anxiety) while higher anti-inflammatory cytokines would relate to less frequent METH use, lower craving and less severe negative affect. These rather disturbing results might indicate a more complex relationship between immune system activity and addiction as mentioned by (Stamatovich et al., 2021) in a previous study of cocaine use disorder, possibly distinct across different periods of substance use.

Several limitations of this study need to be addressed. First, the present work should be interpreted as exploratory owing to the relatively small sample size consisting of male subjects only and the imbalance of the size between the two groups. Besides, the cross-sectional design of our study does not allow for conclusions about the causality of the demonstrated correlations. Thus, prospective research with a larger cohort is warranted. Further studies involving in vitro and in vivo host-microbe interaction are also needed to understand causal mechanisms. Second, plasma levels of systemic inflammatory factors and the severity of negative affect were not measured in the HC group; thus, whether these indicators were abnormal in the MUD group remains unclear in this study. Although higher levels of depression and anxiety in METH users, as well as the METH-associated changes in inflammatory factors have been reported (Huckans et al., 2015; Papageorgiou et al., 2019), results concerning these metrics in our study should be interpreted cautiously. Third, despite the fact that this study was controlled for factors affecting the gut microbiota, including age, BMI, alcohol use and smoking, other confounding factors such as diet, physical activity and socio-economic variables such as economic status and level of education should be fully considered in future studies. Fourth, this study was limited to a panel of only five markers of systemic inflammation. A greater scope of cytokines, chemokines and other molecular factors related to METH use (e.g., IL-1β, IL-8, IL-12, IL-18, interferon-gamma and Intercellular adhesion molecule 1; Papageorgiou et al., 2019) should be evaluated in further investigations. Moreover, our study included only a brief battery of self-reported questionnaires. Although we did find significant correlations between fecal microbiota, immune factors and clinical variables, future studies could apply a more comprehensive assessment involving cognitive function and psychotic symptoms, which is also an issue of concern. Fifth, the current study used only VAS as a subjective evaluation to assess craving for METH, which may lead to reporting bias and social desirability bias, especially among subjects undergoing compulsory rehabilitation in this study. Although VAS has been commonly used for craving assessment in multiple studies (Rosenberg, 2009; Kleykamp et al., 2019), future research could apply objective methods (e.g., electroencephalograph) to improve the reliability of craving measurement (Parvaz et al., 2016). Finally, the use of 16S rRNA sequencing only allowed us to identify taxa down to the genus level, and specific variation at the species level was therefore not evaluated. It is now generally accepted that different bacterial species, or even strains, of the same genus may exert completely different effects on the host (Mihatsch et al., 2012). Therefore, metagenomics studies are necessary to pursue in-depth understanding in this regard.

In summary, our findings provide evidence for alterations of fecal microbiota in male subjects with MUD. Furthermore, we demonstrated, for the first time, that altered microbial taxa were correlated with the levels of specific systemic inflammatory factors and clinical characteristics in MUD. As inflammation is one of the proposed mechanisms through which gut microbiota could potentially regulate aberrant brain functions in MUD, further revealing the precise interactions between gut microbiota and systemic inflammation may provide mechanistic insights and help to develop new therapeutic approaches for individuals with MUD.

The data presented in the study are deposited in the National Center for Biotechnology Information (NCBI) BioProject database with project number PRJNA766176. http://www.ncbi.nlm.nih.gov/bioproject/766176.

The studies involving human participants were reviewed and approved by the Institutional Review Board and the Ethics Committee of Shanghai Mental Health Center. The patients/participants provided their written informed consent to participate in this study.

Conceptualization, DD and MZ. Data curation, DD, HS, YS, TC, and QS. Formal analysis, DD. Funding acquisition, MZ. Investigation, DD, HS, YS, and TC. Project administration, QS. Supervision, HJ and MZ. Writing—original draft, DD. Writing—review & editing, DD and MZ. All authors contributed to the article and approved the submitted version.

This work was supported by the National Nature Science Foundation of China (81771436, 82130041), Shanghai Municipal Science and Technology Major Project (2018SHZDZX05), Shanghai Shenkang Hospital Development Center (SHDC2020CR3045B), Shanghai Clinical Research Center for Mental Health (19MC1911100), Shanghai Engineering Research Center of Intelligent Addiction Treatment and Rehabilitation (19DZ2255200) and Shanghai Key Laboratory of Psychotic Disorders (13DZ2260500).

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.