Association of Gut Microbiome and Dipeptidyl Peptidase 4 in Immune-Mediated Inflammatory Bowel Disease: A Rapid Literature Review

1. Introduction

Immune-mediated inflammatory diseases (IMIDs) encompass a spectrum of chronic conditions characterized by aberrant immune responses and tissue inflammation. These conditions pose a significant global health burden, affecting millions of people worldwide with heavy costs for healthcare systems and impact on patient well-being [1,2]. Despite recent medical advances [3,4], the precise mechanisms underlying IMIDs pathogenesis remain elusive, with evidence of both innate and adaptive immune responses [5,6].

Disruption on the composition and function of gut microbiota (dysbiosis) has been linked to various IMIDs, including inflammatory bowel disease (IBD) [7,8]. For instance, specific microbial alterations were identified in IBD patients including decrease of bacterial diversity and in the count of Firmicutes and Bacteroidetes and increase of Proteobacteria [9]. These data result from a broad discussion concerning the importance of gut microbiome on global human health and disease [10,11,12], through the production of metabolites and signalling molecules known to interact with the immune system of the host [13,14].

In this context, dipeptidyl peptidase 4 (DPP4) emerged as a potential link between gut microbiome and IMIDs like IBD. It is a widely expressed membrane-bound glycoprotein available in human serum and fecal soluble forms [15,16]. Fungi and bacteria of the human microbiome (i.e. Prevotella, Lactobacillus, Lactococcus and Streptococcus) also exhibit DPP4 activity [17,18,19,20,21,22,23,24,25,26,27,28,29]. DPP4 has an important role in blood sugar regulation by decreasing the bioavailability of glucagon-like peptide-1 (GLP-1), which stimulates insulin secretion and suppresses glucagon release, glucagon-like peptide-2 (GLP-2), and gastric inhibitory peptide (GIP). However, it can act on a broad range of molecules, from dietary proteins to gut hormones, neuropeptides, and chemokines [30,31], with impact on the gut environment, immune responses, among other pathophysiological functions (Figure 1) [32,33,34,35].

Despite research advances, the relationship between DPP4 and the gut microbiome in immune-mediated IBD remains unclear and has shown to be complex. Research evidence seems to point to a bidirectional interaction, with changes in the gut microbiome influencing DPP4 expression/activity and modulation of the gut microbiome and immune responses by DPP4. Further research is warranted to unravel the mechanisms underlying this association and its implications for IBD pathogenesis and potential therapeutic interventions.

This rapid review intended to comprehensively evaluate the current evidence regarding the association of DPP4 and gut microbiome in immune-mediated IBD by exploring the potential influence of DPP4 on gut microbiota composition, development of dysbiosis, and IBD pathogenesis. Understanding these interactions is critical for developing novel therapeutic strategies aimed at restoring gut microbiome balance and modulating immune responses in IBD and IMIDs patients.

2. Materials and Methods

2.1. Protocol, Registration, and Literature Search

This rapid review was conducted in accordance with the Cochrane Collaboration Handbook [42] and followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines [43]. The protocol was developed according to the Preferred Reporting Items for Systematic Review and Meta-Analysis Protocols (PRISMA-P) guidelines [44] and registered in the International Prospective Register of Systematic Reviews (PROSPERO; registration number CRD42023434686).

The literature search was conducted in five databases: CENTRAL (https://www.cochranelibrary.com/central), ClinicalTrials.gov (https://clinicaltrials.gov), PubMed (https://pubmed.ncbi.nlm.nih.gov), SCOPUS (https://scopus.com), and Web of Science (https://webofscience.com), covering all reports published from inception to 21 July 2023. In addition, a gray literature database (Google Scholar; https://scholar.google.com) and reference lists of the included studies and relevant reviews were also approached to ensure the inclusion of all pertinent studies. The search strategy included a combination of controlled vocabulary terms and keywords related to our research question. In brief, queries used for the literature search comprised the intersection of the following main keywords: dipeptidyl peptidase 4, microbiome, and IBD-related immune-mediated inflammatory diseases. The search strategy used in PubMed database is presented in Supplementary Table S1.

2.2. Eligibility Criteria and Selection of Studies

Studies were considered eligible for inclusion in this rapid review if they reported data addressing our research question, according to the following PICO elements: i) Population: human subjects with IBD-related IMIDs and in vitro or in vivo models mimicking these diseases; ii) Intervention: not applicable; iii) Comparator: not applicable; iv) Outcomes: DPP4 and gut microbiome data. Due to the limited evidence including our outcomes of interest, no restrictions based on the research subject (i.e. in vitro, in vivo or human research) were applied to our search strategy.

Studies were excluded if they met any of the following criteria: i) not relevant to the research question; and ii) the publication type comprises review, systematic review, meta-analysis, book chapter, editorial, letter or commentary. Article language and geographical localization were not considered as exclusion criteria. The reasons for exclusion of the ineligible studies were recorded to complete the PRISMA 2020 flow diagram (Figure 2).

Rayyan [45] and Review Manager (RevMan, version 5.4.1, 2020) tools were used for the detection of duplications and selection and management of studies. To assess eligibility for inclusion, two independent reviewers screened the titles and abstracts of the retrieved records. Full-text articles of potentially eligible studies were obtained and further assessed for inclusion. Any discrepancies were resolved through discussion and, when necessary, consultation with a third reviewer.

2.3. Data Extraction

Study characteristics and outcome data were collected from the included studies: i) study details (study reference and origin); ii) study aim; iii) materials and methods (study design, sample characteristics, sample size (n), DPP4 assessment details, and gut microbiome analysis techniques); iv) IBD-related IMID; v) DPP4 outcomes; vi) gut microbiome outcomes; and vii) relevant findings. Given the heterogeneity of the methodologies used in the included studies, a meta-analysis approach was not applicable.

2.4. Risk of Bias Assessment

A critical appraisal of the included studies was performed to assess the risk of bias using the following tools: “Animal Research: Reporting of In Vivo Experiments” (ARRIVE) [46] guidelines for the in vivo studies; “STrengthening the Reporting of OBservational studies in Epidemiology” (STROBE) [47] guidelines for the cross-sectional study; and a recently developed quality assessment tool for basic science studies [48], for the in vitro study. Different criteria were independently screened by two reviewers. Discrepancies were addressed through discussion with a third reviewer. A qualitative score (i.e. low, unclear, or high risk of bias) was attributed to each criterion, enabling the estimation of the risk of bias in all included studies regardless of the type of research study.

3. Results and Discussion

3.1. Studies Characteristics

This rapid review comprehensively evaluated the association between DPP4 and gut microbiome in immune-mediated IBD. The literature search identified 1039 studies from 7 different sources; 236 records were removed after duplicates detection. Overall, 803 titles and abstracts were screened by two independent researchers based on the inclusion and exclusion criteria; 724 studies met the exclusion criteria for outcome, population, and publication type reasons, and 79 records were identified as potentially relevant to address the research question. From these, 78 full texts were retrieved and screened to identify the 7 studies included in this rapid review. The most relevant characteristics of the included studies are summarized in Table 1. The studies, published between 2018 and 2022, were performed in six different countries (Canada, China, France, Germany, Japan, and Korea). Regarding study design, one prospective cross-sectional study was performed in human patients [36], five in vivo experimental studies included mouse models [31,37,38,39,40], and one in vitro experimental study was performed on commensal bacteria from the human intestinal microbiota [41]. Concerning IMIDs, the included studies were mainly related to inflammatory intestinal conditions, such as IBD. However, primary aims and assessment techniques for the outcomes of interest diverged between studies. DPP4-related outcomes were assessed by molecular methods such as enzyme-linked immunosorbent assay (ELISA) [36,39], V-Plex and MESO QuickPlex systems [41], molecular arrays [40], reverse transcription polymerase chain reaction (RT-PCR) [37,38], and classic standard curve-based methods [31]. Regarding microbiome analysis, sequencing targeting 16S ribosomal RNA gene variable region was the most common methodology [36,37,39,40,41], but two studies approached microbiome outcomes through direct comparison of fecal content in in vivo models, using fecal transplantation strategies [31,38].

3.2. Assessment of DPP4-Related Outcomes

DPP4-related outcomes were differentially assessed in the included studies, integrating data from mRNA expression, protein activity, and protein levels of DPP4 or DPP4 substrates, including GLP-1, GLP-2, GIP, and glucagon (Table 2).

GLP-1 gut peptide secretion was evaluated in intestinal neuroendocrine murine secretin tumor cell line 1 (STC-1) cells exposed to 21 commensal bacterial strains from human intestinal microbiota or butyrate (positive control), revealing a significant induced ability to stimulate the production of GLP-1 in Roseburia intestinalis AS6, Blautia obeum AS32, and Parabacteroides distasonis PF-BaE7, and AS93 bacterial strains in comparison with the GLP-1 production induced by butyrate [41]. Conversely, Doria formicigenerans AS168 showed a trend towards increased GLP-1 release, but further studies are needed to confirm statistical significance [41].

Hanawa et al. [37] investigated the effect of the artificial sweetener acesulfame potassium (ACK) on the intestinal mucosa and gut microbiota of mice and reported downregulation of GLP-1 and GLP-2 receptors mRNA expression in ACK-treated mice when compared to controls. This led to potentially worse intestinal inflammation with elevated proinflammatory cytokine expression, damage in the small intestine, and increased intestinal permeability [37].

The studies conducted by Lee et al. [38] evaluated the effect of fecal microbiota transplantation (FMT) on DDP4 expression, using fecal material from metformin-treated mice, in high-fat diet (HFD) and regular diet (RD) groups. DPP4 expression was lower in HFD metformin-treated mice compared with those in HF diet and regular diet groups. In contrast, GLP-1 expression was higher in HFD metformin-treated mice group [38].

The cross-sectional study included in this rapid review developed by Manka et al. [36] revealed that CD patients exhibit lower GLP-1 levels and increased gut motility compared with healthy subjects.

DPP4 activity was studied by Olivares et al. [31] in the cecal content of gnotobiotic mice colonized with human gut microbiota of a healthy subject and in germ-free mice (GFM). The results showed a significantly higher DPP4 activity in colonized mice and no difference in DPP4 activity and expression, suggesting that the DPP4-like activity produced by gut microbiota was the source of DPP4 activity in colonized mice [31].

Peng et al. [39] reported decreased GLP-1 serum levels in dextran sulfate sodium (DSS)-induced colitis mice. Also, GLP-1 secretion was stimulated by SCFAs via GRP43 and GRP41 activation and mRNA expression of these receptors was decreased in DSS-induced colitis model [39]. In turn, when exposed to acetic acid, propionic acid, and butyric acid, primary murine colon epithelial cells exhibited significant stimulation of GLP-1 release. However, after stimulation with feces from control mice and DSS-induced colitis mice, the cells exposed to colitis conditions released less GLP-1 [39].

In a spontaneous model of colitis mice using mucin 2 (Muc2) knock-out (KO) mice (Muc2−/− mice), it was demonstrated that circulating GLP-1, glucagon, and GIP remain unchanged compared with healthy controls [40].

Our analysis highlighted diverse mechanisms through which DPP4 modulates gut physiology and immune response in intestinal inflammatory conditions. The expression, activity, and substrate levels of DPP4 were shown to change in response to various stimuli, including artificial sweeteners [37], metformin treatment [38], and colitis induction [39,40]. Overall, the revised evidence suggested a significant role for DPP4 and its substrates, GLP-1 and GLP-2, in regulating the composition of gut microbiota, in IBD-like conditions [36,37,39,40,41]. Decreased levels of GLP-1 and GLP-2, often caused by elevated DPP4 activity, were correlated with an increased abundance of pathogenic bacteria and a decrease in beneficial microbes [41]. This imbalance is assumed to contribute to intestinal inflammation and disease progression in IBD. DPP4 activity was found to influence the abundance and function of gut microbiota, with implications for intestinal inflammation and barrier integrity, and increased DPP4 activity was associated with dysbiosis [31], evidenced by exacerbated intestinal inflammation and disease severity in IBD. This association suggests that targeting DPP-4 activity could be a therapeutic strategy for managing IBD. Furthermore, genetic variations in DPP4 were shown to be involved in IBD pathogenesis, highlighting the complexity of the underlying processes and the importance of personalized approaches in disease management.

3.3. Assessment of Gut Microbiome Outcomes

The gut microbiome data was analyzed using various methodologies to investigate the association between microbiota composition and both healthy and immune-mediated inflammatory intestinal conditions (Table 3). However, most of the included studies reported data about the diversity and the alterations detected in the composition of gut microbiota community, ranging from phylum to species level (Figure 3).

Cuffaro et al. [41] revealed that specific bacterial strains exhibit a combination of beneficial functions for a next generation of probiotic candidates: i) epithelial barrier protection (A. soehngenii AS170, B. coprocola AS101, B. uniformis PF-BaE13 / PF-BaE8, L. sabbureum AS4, and P. distasonis AS93); ii) anti-inflammatory effect (B. coprocola AS101, B. intestinihominis AS13, B. obeum AS32, B. ovatus AS171, B. xylanisolvens AS99, B. xylanisolvens AS146, B. uniformis PF-BaE13 / PF-BaE8, D. formicigenerans AS168, P. distasonis AS93 / PF-BaE7, P. merdae AS106, and R. intestinalis AS6); iii) butyrate production (A. soehngenii AS170, L. sabbureum AS4, and R. intestinalis AS6); iv) and tolerance to gastric stress conditions (B. fragilis PF-BaE4, B. intestinihominis AS13, B. ovatus AS171, B. vulgatus AS15 and PF-Ba10, B. xylanisolvens AS146, and P. distasonis PF-BaE7), in addition to the ability to induce GLP-1 production. Among the seven strains that combined two or three probiotic properties, P. distasonis AS93 and R. intestinalis AS6 exhibited all three functional activities, i.e. barrier protection, anti-inflammatory, and GLP-1 secretion [41]. Bacteroides coprocola AS101, Bacteroides uniformis PF-BaE8, and Bacteroides uniformis PF-BaE13 displayed anti-inflammatory profile and the capacity to enhance the epithelial barrier [41]. In turn, B. obeum AS32 and P. distasonis PF-BaE7 exhibited anti-inflammatory properties and the ability to induce GLP-1 production. Overall, only P. distasonis PF-BaE7 revealed tolerance to gastric stress conditions [41].

Alfa(α)-diversity in gut microbiota was lower in inflammatory intestinal conditions triggered by artificial sweeteners like ACK [37]. In the same study, β-diversity analysis showed a different distribution pattern in ACK-treated mice at the phylum (differences in the proportion of Actinobacteria, Bacteroidetes, Deferribacteres, Proteobacteria, and Verrucomicrobia), family (increased proportion of Erysipelotrichacecae and decreased proportion of Clostridiaceae, Lachnospiraceae, and Ruminococcaceae), and genus levels [37]. The administration of ACK resulted in gut microbiome alterations; however, transferring fecal material from ACK-treated mice to recipient mice did not replicate intestinal damage [37].

In addition, one study showed that the relative abundance of Akkermansia, Bacteroides, and Butyricimonas was not significantly different between HFD and HFD metformin-treated mice [38].

The analysis of the altered gut microbiota composition in CD patients showed that 99% of the total bacterial community belonged to the phyla Bacteroidetes, Actinobacteria, Firmicutes, Proteobacteria and Fusobacteria [36]. These findings evidenced a lower α-diversity in gut microbiota of CD patients contrasting with more complex communities in healthy patients [36]. Firmicutes were progressively reduced from control to CD-TNF to CD groups and Proteobacteria were increased in control and CD-TNF conditions compared with CD group [36]. The most representative families in CD were Ruminococcaceae, Bacteroidaceae, Enterobacteriaceae, Veillonellaceae, Acidaminococcaceae, Lachnospiraceae, Rickenellacea, Prevotellaceae, and Porphyromonadaceae [36]. Also, Enterobacteriaceae progressively increased from control to CD-TNF to CD groups and Ruminococcaceae declined when comparing controls with CD-TNF and CD groups [36].

Olivares et al. [31] identified gut microbiota as a source of significant DPP4-like activity in colonized mice.

According to the studies performed by Peng et al. [39], the reestablishment of gut microbiota in colitis mice by BLG is associated to an increase of SCFA-producing bacteria (Akkermansia and Prevotellaceae_UCG-001) and decrease of other bacteria (Eubacterium_xylanophilum_group, Ruminococcaceae_UCG-014, Intestinimonas, and Oscillibacter).

The colitis mouse model developed by Ye et al. [40] exhibited a dysbiotic gut microbiome and was characterized by an increased prevalence of several families, including Prevotellaceae, Pophyromonadacea, Deferribateraceae, Peprostreptococcaceae, Ruminococcaceae, and genera such as Clostridium spp., Mucispirillum spp., and Bilophilia spp.. Furthermore, colitis mice exhibited a reduction in key members responsible for producing SCFAs, including Odoribacter and Butryivibrio spp. [40].

On the other hand, the assessment of gut microbiome outcomes provided insights into the composition and function of gut microbiota in IBD. Studies revealed dysbiosis characterized by reduced microbial diversity and altered community composition, with distinct shifts in the abundance of key microbial taxa implicated in IBD pathogenesis [36,37,40]. These findings may be explained by the reduction of certain bacterial strains that exhibit probiotic properties, such as epithelial barrier protection, anti-inflammatory effects, and butyrate production unveiling potential therapeutic approaches for modulating intestinal inflammation and restoring gut homeostasis in IBD [41]. In addition, gut bacteria have been shown to exhibit DPP4-like activity, which may influence the degradation and inactivation of hormones like GLP-1 and GLP-2 in the gut [31]. However, it has not been proved that this bacterial DPP4-activity directly reaches host tissues.

3.4. DPP4 and Gut Microbiome: A Dynamic Interplay Shaping Disease Outcomes

The findings from DPP4-related and gut microbiome assessments compiled in this rapid review highlight the bidirectional relationship between DPP4 and gut microbiota in the context of IBD. In fact, the included studies evidenced that DPP4 activity influences gut microbiota composition and function, while microbial dysbiosis reciprocally affects DPP4 expression and activity, creating a feedback loop that perpetuates intestinal inflammation and disease progression in IBD. Furthermore, the review suggests that the development of advanced microbiome-based therapeutics, targeting specific microbial taxa with probiotic properties, constitutes a potential strategy to mitigate intestinal inflammation and restore gut microbiome balance in IBD patients.

3.5. Risk of Bias Assessment

The scores (low, high, or unclear risk of bias), attributed for different criteria by each quality tool, are summarized in Supplementary Table S2. Overall, animal studies scored low risk of bias in 56.0% of the applied criteria, but most of the studies were identified with high risk of bias regarding randomization and blinding domains. The cross-sectional study included in this rapid review met 18 out of 22 (81.8%) quality parameters, with a low risk of bias score. Regarding the included basic science study, 64.3% of the applied criteria to assess quality were identified with a low risk of bias score.

3.6. Limitations

This rapid review presents limitations that deserve to be discussed. First, the number of included studies is low. In fact, despite the numerous preclinical and clinical studies elucidating DPP4 mechanisms and gut microbiome independently, the published empirical evidence on the association of DPP4 and gut microbiome in the context of IMIDs is scarce. To attenuate this limitation, we adopted an inclusive selection process to collect more comprehensive literature from seven different sources, using an extensive search query, followed by a rigorous screening system based on our research question and inclusion and exclusion criteria. Second, rapid review studies present a risk of bias that cannot be neglected. Our quality assessment attributed “unclear risk of bias” to at least two parameters, in each included study, and identified the lack of randomization and blinding in animal studies as main methodological limitations. Third, the heterogeneity found in study design, methodologies, and outcome measures hinders broader generalization and comparison of the main findings.

4. Conclusions

This rapid review provides a comprehensive overview of the association between DPP4 and the gut microbiome in the context of IBD. The findings suggest that DPP4 plays a pivotal role in modulating immune responses and gut physiology, with its elevated activity linked to dysbiosis, exacerbated inflammation, and disease progression in IBD. Furthermore, the review underscores the bidirectional relationship between DPP4 and gut microbiota, where DPP4 activity affects microbial composition and, reciprocally, gut microbiota dysbiosis influences DPP4 expression and function, perpetuating intestinal disease. These insights open new avenues for therapeutic interventions targeting DPP4, either through pharmacological inhibition or microbiome modulation, to restore gut homeostasis and reduce inflammation in IBD patients. Additionally, the potential of using specific probiotic strains to enhance epithelial barrier integrity and modulate immune responses represents a promising approach for managing IBD.

Despite its limitations, this review provides valuable foundational knowledge and identifies gaps for future research. To the best of our knowledge, this is the first attempt to systematically review literature associating DPP4 and the gut microbiome in immune-mediated IBD, offering valuable insights into intestinal disease pathogenesis and potential therapeutic interventions. Targeting DPP4, either through pharmacological inhibitors or microbiome modulation, holds promise for improving outcomes in IBD patients. Further research is warranted to elucidate the underlying mechanisms driving DPP4-gut microbiome interactions and validate the therapeutic efficacy of targeting these pathways in the management of IBD and other IMIDs.