Autism spectrum disorder (ASD) refers to early-onset neurodevelopmental disorders with core deficits in social interactions and restricted repetitive behavior and interest (RRBI) (Lord et al., 2018). Increasing evidence supports the notion that the gut-brain axis plays a non-negligible regulatory role in autism, while intestinal microbes are critical bidirectional communication components between the central nervous system (CNS) and our gut (Qin et al., 2010; Cryan and O’Mahony, 2011; Dinan and Cryan, 2013). Our brain can modify gastrointestinal dynamics and local blood flow, increasing or decreasing the secretion of gastrointestinal hormones, and thereby affecting the local immune response (Mayer, 2011; Dinan and Cryan, 2017). Simultaneously, external environmental signals are also mediated by the intestinal microbiome and interact in various ways with the CNS (De Vadder et al., 2014; Sherwin et al., 2018).

Recent studies have shown that the intestinal flora may directly affect our brain function by producing neuroendocrine metabolites such as short chain fatty acids (SCFAs) and tryptophan, or cause interference to other pathways that affect human physiological functions and behavioral performance, including stressful behavior, social interaction, eating, and addiction (Cryan and Dinan, 2012; Cussotto et al., 2018). Research has revealed that one of the gut microbes, Lactobacillus rhamnosus, diminishes anxiety-related behavior in rodent models depending on vagus signal transmission (Bravo et al., 2011). Social behavior deficits in germ-free (GF) mice are also considered critical evidence supporting the assertion that changes in the intestinal microbiome can affect brain function as well as behavioral changes in the individual (Desbonnet et al., 2014; Buffington et al., 2016).

Restricted repetitive behavior and interest (RRBI) is considered a key feature of ASDs and may be related to compositional changes of the intestinal flora (Wolff et al., 2014; McKinnon et al., 2019; Osadchiy et al., 2019), but few studies have addressed this domain (Ho et al., 2020). In this cross-sectional study, we hypothesized that the ASD gut microbiome would be reflected by similar community groups and the severity of their repeated, stereotypic behaviors would vary between clusters. Previous studies on intestinal flora alpha diversity in children with ASD have not yet reached a consensus on this (Liu et al., 2019). Thus, we hypothesized that alpha diversity may be related to one or more RRBI sub-domains.

We used 16S rRNA high-throughput sequencing technology to obtain the abundance of gut microbial genes with large inter-individual variations and examined the gut microbiota composition correlated with ASD phenotypic features quantitatively for the Chinese population living in southeast-China. Our data suggested that two major subgroups, which were characterized either by the increased abundance of Veillonella or Ruminococcus, might be correlated with RRBI symptom severity in children with ASD. Cluster 1, with a relatively high abundance of Veillonella and a low abundance of Ruminococcus, showed more severe RRBI symptoms. However, Cluster 2, with a high abundance of Ruminococcus and a low abundance of Veillonella, presented milder symptoms of repetitive behaviors.

To our knowledge, this is the first study to show that the composition of the intestinal microbiome is associated with repetitive, stereotyped behaviors in children with ASD. Participants in each group were defined based on the composition of their gut microbiome. Although the use of clustering to analyze the microbiome remains a controversial topic (Knights et al., 2014), this study provides a considerably important translational starting point to reflect the underlying mechanism of autism symptom severity and to explore potential treatment methods.

The genus, Ruminococcus, is recognized as the core gut microbiome and is abundant in the human intestinal system. Compared to typically developed children, the abundance of Ruminococcus torques is significantly elevated in stool samples from children with autism (Williams et al., 2012; Wang et al., 2013). However, other researchers have reported a decrease in Ruminococcus abundance in autism stool samples (Finegold et al., 2010). Ruminococcus plays an essential role in boosting intestinal movement and food digestion by degrading mucus. It has been proven that mucin in the large bowel can facilitate the growth of Ruminococcus torques, which is also associated with gastrointestinal (GI) disturbance in patients with inflammatory bowel disease (IBD) (Wang et al., 2013). Groups with higher Ruminococcus abundance showed less severe repetitive, stereotypical behaviors in our study. The role of Ruminococcus spp. in affecting RRBI symptoms and GI symptoms is not clear. Veillonella is known to metabolize carbohydrates in the human digestive system and its abundance is decreased in children with ASD in older age groups (Strati et al., 2017). Our study found that toddlers with ASD who carry a higher abundance of Veillonella exhibited more severe repetitive, stereotypic behaviors. Thus, further studies are needed to explore the underlying mechanism using metagenome sequencing to describe species in the genera Ruminococcus and Veillonella quantitatively.

Alpha diversity is an indicator of the richness of species and the diversity of the gut microbiome in a single ecosystem and it reflects local diversity. Lower α-diversity commonly represents a less mature gut microbiome and is often related to unhealthy conditions (Abrahamsson et al., 2014; Kostic et al., 2015). We observed that Cluster 1 had lower alpha diversity and more severe RRBI symptoms (measured by the observed_OTUs method). In addition, one sub-domain of RRBI, apparently compulsively adhering to non-functional routines or rituals, had a negative correlation with alpha diversity in children with autism (measured by the Shannon and Simpson indices), which is in line with the previously obtained outcomes. Hence, the alteration of alpha-diversity in relation to autism or other developmental delays is still not clear. Some studies have shown that alpha-diversity decreases in patients with ASD (Kang et al., 2013, 2017, 2018), while others show no significant changes (Gondalia et al., 2012; Kushak et al., 2017; Strati et al., 2017), or an increase (Finegold et al., 2010; De Angelis et al., 2013). Further research is needed to target clinical heterogeneity, and a larger population would provide further evidence of the relationship between alpha-diversity and ASD phenotypes.

This is a preliminary study to explore the relationship between composition and alpha-diversity in the gut microbiome and repetitive behavior in ASD. Because of the nature of the unsupervised clustering method, we need to explain the outcomes cautiously and be aware of generalization.

In summary, we demonstrated that the difference in intestinal microbiome composition and altered alpha diversity can be associated with repetitive, stereotypic behavior in autism. The role of Ruminococcus and Veillonella in ASD is not yet understood.

The data presented in the study are included in the article/Supplementary Material, further inquiries can be directed to the corresponding authors.

The studies involving human participants were reviewed and approved by the Ethics Committee of Third Hospital of Sun Yat-sen University. Written informed consent to participate in this study was provided by the participants’ legal guardian/next of kin.

BC contributed to sample preparation, data collection, and took the lead in writing the manuscript. NY contributed to the interpretation of the results. BP and XH contributed to data analysis. XZ contributed to the leading of all research steps and interpretation of the results. All authors provided critical feedback and helped shape the research, analysis, and manuscript.

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.