This study is crucial for the field as T1DM remains a significant health challenge with limited treatment options. By exploring the role of Treg cells in modulating gut microbiota in T1DM, it advances the field by filling a knowledge gap. New information includes specific correlations between bacteria, Treg cells, and immune factors, and changes in gut microbiota profiles with Treg cell treatment. This impacts the field by offering potential novel therapeutic targets and a better understanding of T1DM pathogenesis, potentially leading to improved treatment strategies.

Type 1 diabetes mellitus (T1DM) is characterized by dysfunction and death of beta cells caused by reactive T cells, leading to abnormal synthesis and secretion of insulin. This results in glucose metabolism disorder characterized by absolute insulin deficiency and high blood sugar in patients [1]. The exact mechanism of the pathophysiology of T1DM is still unclear. Its etiology involves multiple factors, with the main known ones currently including autoimmunity, genetic susceptibility, viral infections, diet, and gut microbiota [2]. Advanced maternal age, infant obesit, premature intake of milk, high-sugar low-fiber diet are associated with an increased incidence of T1DM, and can also affect gut microbiota. With the deepening research on gut microbiota in recent years, it is suggested that gut microbiota may be a common pathway for multiple factors influencing the incidence of T1DM [3]. Treg cells in helper T cells have regulatory and anti-inflammatory functions, which can inhibit abnormal infiltration of CD8⁺ T cells in the pancreas and kidneys, thereby delaying the development of diabetes [4]. Through previous research, we have found that Treg cells can improve fasting blood glucose, body weight, 24-h urine volume, water intake, urine protein, as well as insulin and C-peptide levels in Streptozotocin-induced T1DM mice to a certain extent. They also have a role in improving renal inflammatory lesions in diabetic nephropathy, confirming the effectiveness of Treg cells in treating type 1 diabetes [5]. Existing research has confirmed that changes in the gut microbiota can induce or inhibit autoimmune responses in the pancreas, thereby affecting the occurrence of T1DM(5). Therefore, the relationship between intestinal flora and Treg cells may provide a new perspective for studying the pathogenesis and treatment of T1DM.

This study utilized intraperitoneal injection of Streptozotocin (STZ) to prepare T1DM mouse models, collected fecal samples from control group, T1DM group, and Treg cell treatment group for 16S rRNA sequencing, investigated the composition and function of gut microbiota in T1DM model mice, analyzed the impact of Treg cells on gut microbiota in T1DM mice, and further explored the potential mechanisms by which Treg cells exert their effects on the gut microbiota pathway in T1DM mice.

The gut microbiota supports host health by inhibiting pathogens, aiding mucosal repair, and maintaining immune balance. Accumulating evidence implicates it in the pathogenesis of type 1 diabetes (T1DM) [8]. Early work in germ-free non-obese diabetic (NOD) mice showed that pancreatic inflammation worsened in the absence of microbes but improved upon transfer to a conventional environment [9]. Subsequent studies found that disrupting the microbiota in NOD mice via long-term Lactobacillus acidophilus or antibiotic treatment led to intestinal barrier dysfunction, autoimmune reactions, aggravated pancreatic inflammation, and increased diabetes incidence. Conversely, probiotic intervention (e.g., VSL#3, containing bifidobacteria and lactobacilli) reduced diabetes incidence and preserved pancreatic β-cell mass [10]. These findings suggest that disrupting normal gut microbiota can induce or exacerbate T1DM. Consistently, early antibiotic treatment accelerates T1DM onset in NOD mice by altering the microbiota, metabolism, and T-cell populations [11], and dysbiosis-induced inflammation may worsen insulin resistance [12]. Conversely, drug interventions that regulate gut microbiota structure can improve pancreatic cell function, reduce the incidence of diabetes by balancing the microbiota structure [2]. Pharmacological modulation of the microbiota, such as with empagliflozin, can restore diversity, reduce lipopolysaccharide (LPS)-producing bacteria, increase short-chain fatty acid (SCFA) production, and improve diabetic outcomes by enhancing barrier function and reducing inflammation [13].

In line with this, our 16S rRNA sequencing revealed significant differences in α- and β-diversity among control, T1DM, and Treg cell-treated T1DM mice, indicating marked microbial alteration in T1DM that can be further modulated by therapy. Altered composition may contribute to pathogenesis, while Treg treatment may act by enriching beneficial taxa and reducing harmful ones, thereby restoring community structure. Interestingly, overall microbial richness was lower in Treg-treated mice than in untreated T1DM mice, a pattern differing from the partial diversity recovery often reported. This may reflect a pronounced suppression of pathogenic bacteria that is not fully offset by the expansion of beneficial taxa (e.g., Ruminococcaceae) within the 14-day intervention window. Although beneficial taxa were enriched post-treatment, their functional maturation (e.g., SCFA production) may require more time. Given the small sample size and exploratory design, we cannot establish causality but conclude that specific microbial shifts coincide with Treg administration and metabolic improvement in this model.

Based on heat maps and ANOVA analysis, we found that Verrucomicrobiota is significantly reduced in the diseased group compared to the normal group, suggesting that the decrease in Verrucomicrobiota is related to T1DM. Akkermansia is a representative genus of Verrucomicrobiota. Previous studies have found that Akkermansia can help reduce the risk of diabetes and inflammation [14]. Shi M and others found that it also can reduce LPS entering the bloodstream by protecting the intestinal mucus layer [15]. In addition, Akkermansia indirectly promotes the production of butyrate by producing high levels of acetate [16, 17]. The [Eubacterium] siraeum group was more abundant in T1DM mice at baseline, consistent with its reported negative correlation with glucose tolerance [18]. We observed a higher abundance of Helicobacter (phylum Campylobacterota) in the Day14-DM group compared to the Day14-Med-Treg group, with H. pylori being the most prevalent species. The relationship between H. pylori and diabetes is complex and may involve indirect effects on insulin resistance via metabolic and inflammatory pathways [19]. On the other hand, the hyperglycemic environment in diabetic patients can regulate immune cell function, stimulate pro-inflammatory cytokine production, leading to neurodegeneration, delayed gastric emptying, reduced gastric acid secretion, and increased risk of H. pylori infection [20, 21].

We observed a differential abundance of Helicobacter between T1DM and Treg-treated groups. This may be attributable to a direct modulatory effect of Treg cells on the gut microbiota or secondary to improved systemic glucose control. Notably, no significant difference in Helicobacter abundance was detected between T1DM and control groups at the genus level. This lack of difference could stem from the limited statistical power of our small sample size. Alternatively, it may reflect a weak or absent direct association between H. pylori and T1DM pathogenesis, as mechanistic links involving H. pylori (e.g., promotion of insulin resistance) are more established in type 2 diabetes [22]. Some scholars believe that in adolescents and children with T1DM, levels of glycated hemoglobin and poor blood sugar control are positively correlated with H. pylori infection [23]. Delitala and others found that highly toxic strains of H. pylori may induce immune-mediated diabetes [24]. In addition, alterations in the abundance of Tyzzerella, Mycoplasma, and the [Eubacterium] ruminantium group were observed in the T1DM group compared with the Medium-dose Treg group. Following Treg cell therapy, the abundance of Tyzzerella was reduced relative to the T1DM group. Given that Tyzzerella can produce substantial propionate—which downregulates cytokine-induced VCAM-1 expression and impairs insulin secretion [25, 26]—the decrease in Tyzzerella abundance induced by Treg cells may contribute to glycemic control through a concomitant increase in VCAM-1. Mycoplasma, commonly linked to respiratory infections, also colonizes the gastrointestinal tract. Previous studies indicate that Treg cells can modulate T-cell responses to Mycoplasma by enhancing IFN-γ and IL-17 production, thereby restraining destructive immune reactions in the respiratory tract [27]. The Gut–Lung Axis (GLA) theory posits a bidirectional interplay between intestinal and pulmonary immune systems and microbiota [28]. In the present study, Mycoplasma abundance was lower in the Treg-treated group than in the T1DM group, suggesting that Treg-mediated suppression of excessive immune responses may extend to intestinal Mycoplasma colonization. Moreover, through the GLA [29], Treg cells could indirectly regulate gut immune function via pulmonary immune modulation. Zhang et al. reported that Eubacterium promotes bladder cancer progression by upregulating extracellular matrix protein 1 (ECM1), which influences matrix metalloproteinase-9 (MMP9) expression in bladder tissue [30]. In tumor microenvironments, excessive Treg cell activation can inhibit anti-tumor immunity and facilitate “immune escape” [31]. Thus, the attenuated immune clearance associated with Treg cells in this study may potentially relate to changes in the [Eubacterium] ruminantium group.

Spearman correlation analysis indicated a positive correlation between Firmicutes abundance and Treg levels, and a negative correlation with IFN-γ. Firmicutes may influence host immune status and metabolic balance through metabolites such as short-chain fatty acids, which can modulate CD4⁺ T-cell differentiation [32] and thereby exert systemic effects. Treg cells are known to suppress IFN-γ secretion, reduce adipose tissue macrophage activation, and ameliorate insulin resistance [33]. Leite, A. Z and others observed a significant correlation between IFN-γ and the relative abundance of Firmicutes in patients with type 2 diabetes [34]. This could be attributed to Firmicutes-associated dysbiosis, which may increase intestinal permeability, promote LPS translocation, and activate the TLR4/NF-κB pathway, ultimately elevating inflammatory cytokines including IFN-γ and contributing to insulin resistance. In contrast, the Cyanobacteria phylum showed a negative correlation with Treg levels and a positive correlation with IFN-γ. Certain cyanobacteria such as spirulina can stimulate macrophages to produce IL-12, a cytokine that promotes IFN-γ synthesis by NK cells [35]. Additionally, cyanobacteria may downregulate IL-4, which normally inhibits spontaneous apoptosis of Tregs; reduced IL-4 could therefore enhance Treg cell death [36, 37]. Mesenchymal stem cells (MSCs) have been shown to suppress IFN-γ secretion by T and NK cells, attenuate adipose tissue macrophage activation, and promote Treg generation [38], thereby improving insulin sensitivity and supporting β-cell proliferation [39]. Consistent with this, earlier studies reported that gingival MSCs downregulated IL-17 and IFN-γ in CD4⁺CD8⁺ T cells, increased Treg numbers, and alleviated hyperglycemia and pancreatic inflammation in T1DM mice [5, 40]. In the present study, changes in C-peptide were used to assess pancreatic β-cell function. Overall, Firmicutes abundance correlated positively with C-peptide regulation, whereas Cyanobacteria exhibited a negative association.

Firmicutes are major producers of short-chain fatty acids (SCFAs) such as acetate, propionate, and butyrate, which contribute to gut barrier integrity, exert anti-inflammatory effects, and modulate glucose and lipid metabolism, blood pressure, and gut–brain axis signaling [41, 42]. On the other hand, Bifidobacterium—a key taxon linked to branched-chain amino acid biosynthesis and insulin resistance [18]—can also influence leptin signaling; its increased abundance may reduce energy intake and alter carbohydrate fermentation and lipopolysaccharide metabolism [43], thereby playing a beneficial role in glycemic control and T1DM prevention. The ratio of Firmicutes to Bacteroidetes (F/B ratio) is critical for polysaccharide fermentation. An elevated F/B ratio has been associated with inflammatory responses, increased body mass index, and diabetes development [44, 45]. In line with previous findings showing a higher duodenal F/B ratio in T1DM patients [46], our study observed an increased F/B ratio in the T1DM group compared with controls, while Treg treatment lowered this ratio. These results suggest that Treg cells may help restore microbial balance, improve metabolic regulation, and support weight management in T1DM.

Based on LEfSe analysis, the untreated disease group showed a distinct microbiota profile, characterized by dominance of the phyla Firmicutes and Desulfobacterota, and enrichment of the genera Lachnospiraceae NK4A136 group, Lactobacillus, Eubacterium_xylanophilum_group, Ruminococcus, Oscillibacter, and Colidextribacter. The increase in Firmicutes is consistent with reports linking a higher Firmicutes-to-Bacteroidetes ratio to a pro-inflammatory gut state. The rise in Desulfobacterota—a sulfate-reducing phylum—could further promote the production of hydrogen sulfide, a metabolite that impairs colonic epithelial cells and is associated with gut inflammation [47]. In support, Wei et al. observed a positive correlation between the relative abundance of Desulfobacterota and TNF-α levels [48]. In contrast, Treg cell therapy led to notable enrichment of several beneficial taxa. These included the short-chain fatty acid (SCFA)-producing genera Eubacterium_ruminantium_group and Quinella, as well as the family Atopobiaceae, which is regarded as potentially beneficial in the mammalian gut. Most Atopobiaceae species produce metabolites such as lactic acid and SCFAs, which help maintain intestinal homeostasis and attenuate inflammation by modulating Th17/Treg differentiation, autophagy, and inflammatory cell death pathways, thereby protecting against inflammatory diseases [49]. Additionally, the family Eggerthellaceae—which can convert ellagic acid into urolithins [50]—was enriched. Urolithins have been shown to mitigate high-fat-diet-induced weight gain via microbiota modulation [51], suggesting a possible role of Eggerthellaceae in improved weight regulation in our model. Collectively, these shifts indicate that Treg cells may promote metabolic benefits, at least in part, through the shaping of a more favorable gut microbiota composition.

To explore the functional consequences of gut microbiota alterations, we predicted KEGG pathways from 16S rRNA data and compared the metabolic profiles of the three groups at day 14. Compared with healthy controls, T1DM mice showed significant enrichment in pathways related to fructose and mannose metabolism, pentose phosphate pathway, methane metabolism, two-component system, and ABC transporters. This pattern corresponds to known metabolic dysregulation in T1DM. In line with this, intestinal methane producers have been reported to display impaired glucose tolerance under high carbohydrate loads and a greater tendency toward hyperglycemia [52], offering a plausible basis for the observed methane metabolism enrichment in T1DM mice. The two-component system has also been linked to diabetes pathology; for example, diabetes may promote nasal colonization of virulent Staphylococcus aureus—a common pathogen in diabetic foot infections—through modulation of this system [52]. ABC transporters were most enriched in the T1DM and Treg-treated groups, with levels significantly higher than in controls. ABC transporters are involved in various disease processes, including diabetes-related changes in bile formation due to altered renal tubular expression in STZ-induced diabetic rats [53, 54]. Within the immune system, ABC transporters—particularly ABCG1—contribute to Treg cell biology. Treg development depends on cholesterol homeostasis, where ABCG1-mediated cholesterol transport helps regulate intracellular cholesterol levels, thereby influencing Treg differentiation and function [55].

The limitations of our study should be acknowledged with emphasis on the exploratory nature of the current analyses, which is consistent with the pilot study design with n = 3 per group per time point. First, the small sample size inevitably compromises statistical power, increasing the randomness and uncertainty of the results. Second, as noted, the reliance on OTU-based clustering and PICRUSt predictive functional profiling further limits the robustness of taxonomic identification, correlation analyses, and KEGG pathway inferences. Thus, all conclusions derived from these analyses should be interpreted as exploratory rather than confirmatory. In addition, the lack of consistency in the timeframe of the study subjects does not rule out errors caused by individual differences in mice. It should also be clarified that the current results only represent changes in the gut microbiota of STZ-induced T1DM mice and cannot be directly generalized to clinical populations or other T1DM models. To validate and extend our observations, future studies should include larger cohorts and collect samples at multiple time points to delineate temporal dynamics in the gut microbiota. Such designs would help clarify whether the microbial shifts we observed are consistent and causally linked to disease progression or treatment response. Despite these limitations, our findings suggest that Treg cell therapy modifies the gut microbiota composition in STZ-induced T1DM mice. This supports the notion that the therapeutic effect of Treg cells may be partly mediated through the regulation of intestinal microbial communities.