Transcriptome analysis suggests broad jejunal alterations in Linghu’s obesity-diarrhea syndrome: A pilot study

Abstract

BACKGROUND

Obesity is associated with a significantly increased risk for chronic diarrhea, which has been proposed as Linghu’s obesity-diarrhea syndrome (ODS); however, its molecular mechanisms are largely unknown.

AIM

To reveal the transcriptomic changes in the jejunum involved in ODS.

METHODS

In a cohort of 6 ODS patients (JOD group), 6 obese people without diarrhea (JO group), and 6 healthy controls (JC group), high-throughput sequencing and bioinformatics analyses were performed to identify jejunal mucosal mRNA expression alterations and dysfunctional biological processes. In another cohort of 16 ODS patients (SOD group), 16 obese people without diarrhea (SO group), and 16 healthy controls (SC group), serum diamine oxidase (DAO) and D-lactate (D-LA) concentrations were detected to assess changes in intestinal barrier function.

RESULTS

The gene expression profiles of jejunal mucosa in the JO and JC groups were similar, with only 1 differentially expressed gene (DEG). The gene expression profile of the JOD group was significantly changed, with 411 DEGs compared with the JO group and 211 DEGs compared with the JC group, 129 of which overlapped. The enrichment analysis of these DEGs showed that the biological processes such as digestion, absorption, and transport of nutrients (especially lipids) tended to be up-regulated in the JOD group, while the biological processes such as rRNA processing, mitochondrial translation, antimicrobial humoral response, DNA replication, and DNA repair tended to be down-regulated in the JOD group. Eight DEGs (CDT1, NHP2, EXOSC5, EPN3, NME1, REG3A, PLA2G2A, and PRSS2) may play a key regulatory role in the pathological process of ODS, and their expression levels were significantly decreased in ODS patients (P < 0.001). In the second cohort, compared with healthy controls, the levels of serum intestinal barrier function markers (DAO and D-LA) were significantly increased in all obese individuals (P < 0.01), but were higher in the SOD group than in the SO group (P < 0.001).

CONCLUSION

Compared with healthy controls and obese individuals without diarrhea, patients with Linghu’s ODS had extensive transcriptomic changes in the jejunal mucosa, likely affecting intestinal barrier function and thus contributing to the obesity and chronic diarrhea phenotypes.

Key Words: Obesity, Chronic diarrhea, Jejunum, Transcriptome, Intestinal barrier

Core Tip: This study analyzed the transcriptomic characteristics of the jejunal mucosa in patients with Linghu’s obesity-diarrhea syndrome (ODS). The jejunal gene expression profile of obese people without diarrhea was similar to that of healthy controls. However, the jejunal gene expression profile of ODS patients showed significant changes, characterized by the up-regulation of nutrient absorption, digestion, and transport and the down-regulation of rRNA processing, mitochondrial translation, antimicrobial humoral response, DNA replication, and DNA repair. In addition, although intestinal barrier damage was present in all obese individuals, it appeared to be more severe in ODS patients.

INTRODUCTION

Obesity has now reached pandemic-level rates, imposing a substantial burden on patient morbidity, mortality, and healthcare resources[1]. It is closely associated with numerous comorbidities, including type 2 diabetes mellitus, cardiovascular disease, non-alcoholic fatty liver disease, and various gastrointestinal diseases[2]. Chronic diarrhea is one of the most common obesity-related digestive symptoms with an incidence of up to 30%; however, routine examinations often fail to identify pathological changes, and the therapeutic efficacy is not satisfactory[3,4]. Given the critical role of the intestine in digestion, absorption, and metabolism[5-7], the intestinal symptoms of obese patients should be taken seriously, and modifying intestinal function may be a strategy to counter metabolic disorders. Therefore, to highlight the specific relationship between obesity and chronic diarrhea, Chinese professor Linghu proposed a novel syndrome, Linghu’s obesity-diarrhea syndrome (ODS), that is, obesity accompanied by chronic diarrhea but without organic lesions or infectious agents that can be definitively identified as causative factors for diarrhea by routine examinations[8].

The jejunum is the primary site for the absorption of nutrient-derived carbohydrates, lipids, small peptides, and vitamins, and it is also the site where most immune responses to luminal antigens occur[9]. The jejunal epithelium acts as a guardian of intestinal barrier integrity, the loss of which likely facilitates the transfer of luminal antigens (food, microorganisms, toxins, etc.) across the intestinal mucosa and exacerbates immunological responses[10]. Recent studies have underscored the importance of the jejunum for the generation of low-grade inflammation and the regulation of metabolism in the pathogenesis of obesity and related metabolic complications[11-13]. Therefore, jejunal function is a potential entry point for the association between chronic diarrhea and obesity, which has not been studied. In addition, due to the difficulty of obtaining small intestine tissues, current studies on the small intestine are mainly based on obese animal models or morbidly obese patients undergoing metabolic surgery. The specific role of the small intestine in the development of human obesity and its comorbidities remains to be clarified.

This pilot study focused on the gene expression signatures of the jejunal mucosa in patients with ODS to reveal the pathophysiology of this syndrome and to help develop personalized health-preventive strategies.

MATERIALS AND METHODS

Subjects and procedures

We carried out jejunal transcriptome analysis in a cohort of 6 ODS patients (JOD group), 6 obese people without diarrhea (JO group), and 6 healthy controls (JC group) and performed serological tests to evaluate intestinal barrier integrity and permeability in another cohort of 16 ODS patients (SOD group), 16 obese people without diarrhea (SO group), and 16 healthy controls (SC group; Figure 1). The participants were recruited from the outpatient department of Gastroenterology, Chinese PLA General Hospital, following the principles of the Helsinki Declaration from February 2022 to January 2024. All subjects signed informed consent forms, and the study proposal was reviewed and approved by the Ethics Committee of the Chinese PLA General Hospital.

Figure 1 Study design. ODS: Linghu’s obesity-diarrhea syndrome; DAO: Diamine oxidase; D-LA: D-lactate; qPCR: Quantitative real-time PCR.

ODS patients who met the following criteria were included: (1) Ranged in age from 18 to 60 years; (2) were obese without significant weight loss in the past 3 months; obesity was defined as a body mass index (BMI) ≥ 30 kg/m² according to the criteria of the world health organization[14]; (3) met the diagnostic criteria of chronic diarrhea[15], including persistent alteration in stool consistency (types 5 to 7 in the Bristol Stool Form Scale) and increased frequency greater than 4 wk; (4) had neither organic disease nor infections definitively identified as the causative factor of diarrhea; and (5) had not taken diet pills, prokinetic agents, acid inhibitors, probiotics, antibiotics, or anti-inflammatory drugs within two weeks; and had not taken glucocorticoids, immunosuppressants, or psychotropic drugs taken within half a year. Patients with the following conditions were excluded: (1) Organic diseases or structural changes of the digestive system; (2) systemic diseases associated with chronic diarrhea; (3) long-term use of drugs that may cause diarrhea; (4) pregnancy or lactation; and (5) refusal to participate. Because the study was conducted during the coronavirus disease 2019 (COVID-19) epidemic, we also excluded patients with current or previous COVID-19 infection.

Obese individuals without diarrhea were recruited and matched with ODS patients in terms of gender, age, BMI, hematologic indices, and obesity-related complications. Healthy controls with BMIs between 18.5 and 24.0 kg/m² were also recruited and matched with ODS patients in terms of gender and age.

All individuals underwent clinical assessments, anthropometric measurements, biochemical determinations, and digestive endoscopy. Proximal jejunum samples were obtained from the first cohort by endoscopic biopsies and stored in RNAlater solution (Qiagen, Dusseldorf, Germany) for RNA sequencing (RNA-seq) and quantitative real-time PCR (qPCR). Serum samples were collected from the second cohort to measure diamine oxidase (DAO) and D-lactate (D-LA) concentrations. The researchers responsible for sample collection, RNA-seq, bioinformatics analysis, qPCR, and serological tests were blinded to the sample grouping.

RNA sequencing

Total RNA was extracted from jejunal mucosa using TRIzol (Thermo Fisher Scientific, Waltham, Massachusetts, United States) and then sent to Beijing Genomics Institute (BGI, Shenzhen, China) for further RNA-seq detection and analysis using the DNBSEQ platform. The poly-A-containing mRNA molecules were purified and fragmented into small pieces, which were subjected to cDNA synthesis, end repair, the addition of a single 'A' base, and subsequent adaptor ligation. After purification and enrichment of the products, a library was established and sequenced by the combinational probe-anchor synthesis sequencing method.

Data processing and analyses

Raw data were filtered using SOAPnuke (v1.5.6) and then analyzed on the Dr. Tom network platform of BGI (https://biosys.bgi.com). The data were aligned to the human reference genome (GCF_000001405.39_GRCh38.p13) using HISAT2 (v2.1.0) and gene expression was quantified using RSEM (v1.3.1)[16]. The full set of genes was first examined by principal component analysis (PCA) using the R package ‘princomp’. Genes with |log2fold change| (|log2FC|) ≥ 0.585 and P-adjusted values (Q values) < 0.05 were considered as differentially expressed genes (DEGs) using DESeq2 (v1.4.5)[17] and were visualized by volcano plot using the R packages ‘gghplot2’.

Functional and pathway enrichment analyses

Gene Ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) functional enrichment analyses of DEGs were conducted using the R package ‘phyper’. Only terms with Q values < 0.05 after Benjamini–Hochberg correction for multiple testing were deemed to be significantly enriched. Gene set enrichment analysis (GSEA)[18] (https://www.gsea-msigdb.org/gsea/index.jsp) was performed using expression sets of whole genes to identify the biological processes where the most significant changes in gene expression were concentrated. MsigDB_c5_go_bp (v6.2) was selected as the reference database. A |normalized enrichment score| (|NES|) ≥ 1, nominal P value < 0.05, and FDR Q value < 0.25 were used to indicate significantly enriched gene sets and pathways.

Weighted correlation network analysis

Weighted correlation network analysis (WGCNA)[19] was used to clarify the critical modules and hub genes that were associated with ODS using the R package ‘WGNCA’. Using the top 5000 genes with the greatest variation between the JOD group and the JC group, a correlation matrix was created and transformed into a topological overlap matrix (TOM) (parameters: power = 22, minModuleSize = 20, TOMsimilarity Threshold = 0.105). The optimal power parameter was precalculated by the scale independence and mean connectivity analysis. Co-expression networks or modules were identified by hierarchical clustering and assigned a unique color label. Then, clinical traits were associated with each module by computing their Pearson correlation with the module eigengenes. Next, pathway enrichment analysis was conducted for each module to obtain further insight into their functions.

Identification of hub genes in the key module

We sorted genes in the key module in descending order of intramodule connectivity (weight) and selected the top 100 genes to construct a network model visualized by Cytoscape (v3.8.0). Candidate genes highly interconnected with no fewer than 5 nodes in the module were selected to intersect with DEGs in the JOD group compared with the JC and JO groups using the R package ‘VennDiagram’. The overlapping genes were considered hub genes and may be functionally significant.

Quantitative real-time PCR

Total RNA was reverse transcribed into complementary DNA using the EVO M-MLV Reverse Transcription Premix Kit (Accurate Biology, Hunan, China). Then, quantitative real-time PCR (qPCR) was carried out using an mRNA/LncRNA qPCR Kit (GenePool, Beijing, China). The specific primers used are shown in Supplementary Table 1. GAPDH was used as an endogenous control to calculate relative expression with the 2^−∆∆Ct method.

Detection of serum DAO and D-LA concentrations

The serum levels of DAO and D-LA were measured using an enzyme-linked immunosorbent assay (DAO ELISA Kit, Ruixinbiotech, Quanzhou, China; D-LA ELISA Kit, Ruixinbiotech, Quanzhou, China) according to the manufacturer's protocol.

Statistical analysis

Normally distributed continuous variables are reported as mean ± SD and were compared using the ANOVA followed by Tukey's Honestly Significant Difference test for post hoc hypothesis testing. Non-normally distributed continuous variables are generally reported as the medians (25^th-75^th percentiles) and were compared by the Mann-Whitney U test. Categorical data are generally reported as number (%) and were compared by the χ² test or Fisher’s exact test. A P value < 0.05 was considered to indicate statistical significance. Statistical analyses were performed using SPSS 26.0 (SPSS Inc., Chicago, IL, United States) and GraphPad Prism 9.5 (GraphPad, La Jolla, CA, United States).

RESULTS

Jejunal mucosal gene expression profiles

Jejunum plays a crucial role in digestion, absorption, and metabolism and is involved in the pathological processes of many diseases, but its role in the pathogenesis of ODS has not been studied. As a first step to explore jejunal alterations in ODS, we used RNA sequencing to analyze jejunal mucosal gene expression in a cohort of 6 ODS patients (JOD group), 6 obese individuals without diarrhea (JO group), and 6 healthy controls (JC group; Supplementary Table 2). The filtered RNA-seq read-set identified approximately 18000 genes. PCA utilizing all detected genes showed the cluster of the JO group overlapped exactly with that of the JC group, whereas they displayed completely disjointed populations with the JOD group (Figure 2A). Consistent with the PCA results, only 1 gene was identified as differentially expressed between the JO and JC groups (Figure 2B). In contrast, a total of 211 DEGs were identified between the JOD and JC groups (Figure 2C), and 411 DEGs were identified between the JOD and JO groups (Figure 2D), 129 of which overlapped (Figure 2E). It is worth noting that the expression levels of 129 overlapped DEGs changed in the same direction in the JOD group, with 94 (72.87%) genes down-regulated and 35 (23.13%) genes up-regulated (Figure 2F).

Figure 2 Transcriptomic differences. A: Principal component analysis for all samples. Each point represents a sample; B: Volcano plot for the JO and JC groups; C: Volcano plot for the JOD and JC groups; D: Volcano plot for the JOD and JO groups. In the volcano plots, each gene is represented by a single node, and its expression level is indicated by color, with no significant differential expression in grey, significant up-regulation in red, and significant down-regulation in blue. The threshold of differentially expressed genes (DEGs) is |log2fold change| (|log2FC|) ≥ 0.585 and a P-adjusted value (Q value) < 0.05; E: Venn diagram for DEGs among the three groups; F: Venn diagram for DEGs in the JOD group compared with the JC and JO groups. All overlapped DEGs changed in the same direction in the JOD group, with 94 down-regulated and 35 up-regulated. PCA: Principal component analysis; DEGs: Differentially expressed genes.

Functional and pathway enrichment analyses of DEGs

For the 211 DEGs between the JOD and JC groups, the main KEGG pathways involved were DNA replication, DNA repair, biosynthesis and metabolism, and ribosome biogenesis (Figure 3A); the biological processes associated were principally antimicrobial humoral response, DNA replication, and ribosome biogenesis (Figure 3B). For the 411 DEGs between the JOD and JO groups, the main KEGG pathways involved were the digestion and absorption of various nutrients (e.g., fat, minerals, and vitamins), substance metabolism (e.g., arachidonic acid, cholesterol, amino acid, and pyrimidine), bile secretion, citrate cycle, and ribosome biogenesis (Figure 3C); the main biological processes associated were transepithelial transport, lipid metabolism, and antimicrobial humoral response (Figure 3D). For the 129 overlapped DEGs, the main KEGG pathways involved were substance metabolism (e.g. pyrimidine, amino acid, and arachidonic acid), citrate cycle, DNA repair, DNA replication, and ribosome biogenesis (Figure 3E); the main biological processes associated were antimicrobial humoral response, amino acid transport, and ribosome biogenesis (Figure 3F). The cellular components of these DEGs were mainly located in the telomeric region and mitochondria (Supplementary Figure 1). The molecular functions of these DEGs mainly included oxidoreductase activity and transmembrane transporter activity (Supplementary Figure 1).

Figure 3 Functional and pathway enrichment analyses. A: Kyoto Encyclopedia of Genes and Genomes (KEGG) pathway analysis based on the 211 differentially expressed genes (DEGs) between the JOD and JC groups; B: Gene Ontology-biological process (GO-BP) enrichment analysis based on the 211 DEGs between the JOD and JC groups; C: KEGG pathway analysis based on the 411 DEGs between the JOD and JO groups; D: GO-BP enrichment analysis based on the 411 DEGs between the JOD and JO groups; E: KEGG pathway analysis based on the 129 overlapped DEGs; F: GO-BP enrichment analysis based on the 129 overlapped DEGs. Bubble charts show the top 20 significantly enriched terms. The color of the bubbles represents the Q value of each enriched term, and the size of the bubbles represents the number of DEGs enriched in the term. The horizontal axis indicates the rich ratio of the number of DEGs enriched in the term to the total number of human genes annotated to this term. KEGG: Kyoto Encyclopedia of Genes and Genomes; GO-BP: Gene Ontology-biological process.

In addition, we performed GSEA to further elucidate the overall expression trends of functional gene sets. The results suggested the gene sets involved in intestinal absorption, especially lipid digestion and absorption, cholesterol storage, and adaptive immunity response were up-regulated in the JOD group (Figure 4). In contrast, the gene sets involved in ribosome function, mitochondrial function, antimicrobial humoral response, DNA replication, and DNA repair showed a decreasing trend in the JOD group (Figure 5).

Figure 4 Gene set enrichment analysis of up-regulated biological processes in the JOD group. A: Lipid digestion; B: Intestinal lipid absorption; C: Cholesterol storage; D: Adaptive immunity response. Each figure is divided into three parts. The top half is the line diagram of the gene enrichment score (ES). The horizontal axis represents the genes in the gene set, and the vertical axis represents the ES value corresponding to each gene. The peak value > 0 indicates that the biological process is up-regulated. The middle part is the hits chart, and each bar represents a gene. The lower half is the rank value. The Signal2noise ratio corresponding to each gene is shown in the grey area map.

Figure 5 Gene set enrichment analysis of down-regulated biological processes in the JOD group. A: Ribosome biogenesis; B: Mitochondrial translation; C: Oxidative phosphorylation; D: Antimicrobial humoral response; E: Cell cycle DNA replication; F: DNA damage response detection of DNA damage. Each figure is divided into three parts. The top half is the line diagram of the gene enrichment score (ES). The horizontal axis represents the genes in the gene set, and the vertical axis represents the ES value corresponding to each gene. The peak value < 0 indicates that the biological process is down-regulated. The middle part is the hits chart, and each bar represents a gene. The lower half is the rank value. The Signal2noise ratio corresponding to each gene is shown in the grey area map.

Weighted co-expression network construction and key module identification

To clarify the functional gene clusters in ODS, we performed WGCNA on the 12 samples from the JOD and JC groups. With each module assigned a color, a total of 16 modules were identified in the present study, each containing a group of coordinately expressed genes that were potentially involved in similar biological processes (Figure 6A, Supplementary Figure 2).

Figure 6 Identification of modules associated with the clinical traits of Linghu’s obesity-diarrhea syndrome. A: Cluster dendrogram of co-expression network modules. The branches represent highly interconnected clusters of genes. The colors in the horizontal bar represent the co-expression modules; B: Heat map of the relationships between the module eigengenes and clinical traits. The horizontal axis corresponds to clinical traits. The colors of the vertical axis represent the co-expression modules. The color of each cell indicates the degree of correlation between the module and the clinical trait. In each cell, the number in the first row represents the Pearson correlation coefficient, and the number in the second row represents the P value of the correlation coefficient. ODS: Linghu’s obesity-diarrhea syndrome; BMI: Body mass index; MetS: Metabolic syndrome.

Subsequently, we calculated the correlations between the module eigengenes and the clinical traits (ODS, BMI, metabolic syndrome, fecal mucus, abdominal pain associated with defecation, and diarrhea duration). As shown in Figure 6B, the black module containing 127 genes, the pink module containing 123 genes, and the green-yellow module containing 181 genes had strong negative correlations with ODS, BMI, metabolic syndrome, and diarrhea duration, but were not significantly correlated with fecal mucus or abdominal pain. In contrast, the brown module containing 252 genes was positively correlated with abdominal pain but not with other traits. The 47 genes in the light-cyan module were positively correlated with fecal mucus but not with other traits.

Functional enrichment analyses of the key modules

We performed GO enrichment analysis for key modules (black, pink, green-yellow, brown, and light-cyan) to obtain further insight into the biological processes in which these co-expression genes were involved. The black module was principally associated with rRNA processing, mitochondrial translation, antimicrobial humoral response, and DNA replication (Figure 7A). The pink module was principally associated with cytoplasmic translation and rRNA processing (Supplementary Figure 3A). The green-yellow module was principally associated with DNA replication and repair (Supplementary Figure 3B). The genes in the brown module were mainly related to the immune response, especially the adaptive immune response (Supplementary Figure 3C), while those in the light-cyan module were related to cellular response to stimulus and signal transduction (Supplementary Figure 3D).

Figure 7 Functional enrichment analysis and identification of hub genes in the black module. A: Gene Ontology-biological process enrichment analysis; B: Network of the 65 genes with the top 100 weighted values in the black module; C: Two major interaction networks. The blue nodes represent the candidate genes with a degree ≥ 5. The rhombic nodes represent the hub genes that are differentially expressed in the JOD group compared with the JC and JO groups. GO-BP: Gene Ontology-biological process.

Identification of hub genes in the black module

We analyzed the interplay between genes and screened for key regulatory genes in the black module. A total of 65 genes with the top 100 weight values were selected for network visualization by Cytoscape (Figure 7B). Two major interaction networks were ultimately identified (Figure 7C): The largest network was mainly related to rRNA processing, mitochondrial translation, and DNA replication, and the smallest network was associated with mucosal innate immunity. Fifteen candidate genes had a degree ≥ 5, 8 of which (CDT1, NHP2, EXOSC5, EPN3, NME1, REG3A, PLA2G2A, and PRSS2) were differentially expressed in the JOD group compared with the JC and JO groups and were eventually considered hub genes (Figure 7C). These hub genes were all down-regulated in the JOD group and might play a regulatory role in jejunal epithelial dysfunction in ODS.

Validation of DEG expression in jejunal mucosa by qPCR

The mRNA expression of the above 8 hub genes and two immune genes (DEFA5 and DEFA6) among the three groups were validated by qPCR (Figure 8). Consistent with the RNA-seq results, qPCR results showed that mRNA levels of these DEGs in the JOD group were significantly lower than those in the JC and JO groups (P < 0.001). Except for the significant decrease of PRSS2 and DEFA5 in the JO group (P < 0.01), there was no significant difference in the expression of these genes between the JO and JC groups (P > 0.05).

Figure 8 Validation of gene expression in jejunal mucosa by quantitative real-time PCR. A-J: The relative mRNA expression of CDT1 (A), NHP2 (B), EXOSC5 (C), EPN3 (D), NME1 (E), REG3A (F), PLA2G2A (G), PRSS2 (H), DEFA5 (I), and DEFA6 (J) in jejunal mucosa from the JC group (n = 6), JO group (n = 6) and JOD group (n = 6). Data were expressed as mean ± SD. ^aP < 0.01, ^bP < 0.001. NS: Not significant.

Alteration of intestinal barrier function in ODS

To verify the intestinal barrier integrity and permeability of ODS, we tested serum DAO and D-LA concentrations in the second cohort of 16 ODS patients (SOD group), 16 obese individuals without diarrhea (SO group), and 16 healthy controls (SC group) (Supplementary Table 2). Compared with the SC group, the serum DAO and D-LA levels in both SOD and SO groups were significantly increased (DAO: 12.34 ± 0.71 ng/mL vs 8.71 ± 0.56 ng/mL, P < 0.001; 11.13 ± 0.73 ng/mL vs 8.71 ± 0.56 ng/mL, P < 0.001; D-LA: 38.35 ± 5.65 µmol/L vs 25.27 ± 2.89 µmol/L, P < 0.001; 30.73 ± 3.73 µmol/L vs 25.27 ± 2.89 µmol/L, P < 0.001). Moreover, the serum DAO and D-LA levels in the SOD group were significantly higher than those in the SO group (12.34 ± 0.71 ng/mL vs 11.13 ± 0.73 ng/mL, P < 0.001; 38.35 ± 5.65 µmol/L vs 30.73 ± 3.73 µmol/L, P < 0.001; Figure 9). The results suggested that the integrity and permeability of the intestinal mucosal barrier were impaired in obese subjects, but the impairment was more significant in ODS patients than in obese subjects without diarrhea.

Figure 9 Intestinal barrier function measurements. A and B: Comparison of serum diamine oxidase (A) and D-lactate (B) levels among the SC group (n = 16), SO group (n = 16), and SOD group (n = 16). Data were expressed as mean ± SD. ^aP < 0.001. DAO: Diamine oxidase; D-LA: D-lactate.

DISCUSSION

In this pilot study, we report the human jejunal transcriptome in obese patients with unexplained chronic diarrhea, named Linghu’s ODS by Professor Linghu, for the first time[8]. We described the gene expression profile and enrichment pathways of jejunal mucosa associated with ODS and identified hub genes that may be involved in the pathogenesis of the syndrome. Finally, changes in the intestinal barrier were evaluated.

The upper small intestine, particularly the jejunum, serves as the initial and crucial site for food digestion and nutrient absorption and is therefore more exposed to environmental antigens (food, microorganisms, toxins, etc.) than the colon, which may trigger cell damage and apoptosis[6]. The intestinal epithelium forms a selective barrier, permeable to ions, small molecules, and macromolecules, contributing to nutrient uptake and protecting the sterile underlying tissue from the gut luminal antigens[10]. Intestinal epithelial cells (IECs) undergo continuous cell renewal to replace damaged cells, maintaining epithelial integrity and barrier function[20]. The balance between epithelial cell proliferation and exfoliation is tightly regulated by multiple signaling pathways, and disruption of this regulation compromises intestinal homeostasis. Our findings suggested down-regulation of genes associated with DNA replication in the jejunal mucosa of ODS patients, which may diminish intestinal regeneration and self-repair after damage. CDT1, an essential component for the assembly of a pre-replicative complex[21,22], was down-regulated in these patients and network interaction analysis suggested that this gene plays a role in the development of ODS. A recent study reported an up-regulation of DNA replication licensing factors and robust acceleration of intestinal epithelial proliferation after glutamine (Gln) supplementation[23]. Oral Gln supplementation has also been shown to reduce body weight, waist circumference, and serum lipopolysaccharide levels in overweight/obese individuals, possibly through modulation of the intestinal microbiota and intestinal barrier[24,25]. Accordingly, Gln supplementation may improve intestinal symptoms in ODS patients and facilitate weight control.

Inevitable replication errors during rapid cell renewal, as well as environmental genotoxic factors in the gut lumen, pose a challenge to the genomic integrity of IECs[26-28]. DNA repair mechanisms exist to maintain cellular viability and genomic stability in response to DNA damage[29]. Obesity has been shown to induce DNA damage and inhibit DNA repair through oxidative stress and inflammation in different tissues[30-32]. Accumulation of DNA damage can induce tissue inflammation, which disturbs the homeostasis of systemic metabolism and contributes to the development of obesity-related disease[33,34]. Moreover, non-corrected errors during DNA replication can lead to mutations and thus malignant transformation. Overweight and obesity have been linked to several cancers, including colorectal cancer[35-40]. However, the effects of obesity on the small intestine genome remain unclear. In this study, we observed no significant transcriptomic changes in obese individuals without diarrhea, but down-regulation of DNA repair-related genes in the jejunal epithelium of ODS patients, suggesting that ODS patients may be at increased risk for intestinal tumors and metabolic disease. Further studies to identify the relationship between ODS and the long-term risk of cancer are of enormous significance.

The intestinal epithelium requires large amounts of energy directly provided by mitochondria for rapid epithelial turnover, ATPase-dependent transporters, and intestinal homeostasis maintenance[41,42]. Mitochondrial dysfunction results in excessive production of reactive oxygen species, triggering oxidative stress that leads to lipid peroxidation, irreversible DNA and protein damage, and ultimately epithelial structural and functional disorders[42,43]. In addition, mitochondrial dysfunction has been reported to be involved in the pathogenesis of obesity and its associated metabolic diseases by impairing adipose tissue function and is also considered a consequence of obesity[44-48]. High-fat diet consumption has been suggested to impact colonic homeostasis through a mechanism potentially involving mitochondria[42,49,50]. The relationship between obesity and jejunal mitochondrial function has not yet been demonstrated. According to our preliminary results, the jejunal mucosa of ODS patients showed down-regulation of gene sets involved in ribosome synthesis and translation, especially mitochondrial translation, suggesting possible mitochondrial dysfunction. However, no such changes were observed in obese people without diarrhea. Therefore, we speculate that mitochondria-related intestinal homeostasis imbalance may be one of the key processes associated with obesity and chronic diarrhea.

In this study, another notable feature of ODS patients was the down-regulation of key genes associated with the antimicrobial humoral response, including REG3A, PLA2G2A, PRSS2, DEFA5, and DEFA6. Trillions of commensal and pathogenic microorganisms inhabit the gastrointestinal tract, contributing to host digestion, energy regulation, immunity, and metabolism[9,51]. The gut microbiome has a symbiotic relationship with the immune system by maintaining an immune tolerance state. Reduced secretion of antimicrobial peptides disrupts this state, causing bacteria to invade the intestinal mucosa, resulting in intestinal epithelial damage and inflammation, and subsequent intestinal symptoms such as diarrhea[52]. In addition, a deficient antimicrobial humoral response can disrupt the balance of the intestinal microbial ecosystem and cause changes in the composition and function of the intestinal microbiota. Accumulating evidence suggests that alterations in the intestinal flora participate in the occurrence and development of obesity and its metabolic complications in various manners, such as by affecting dietary digestion and absorption, inducing metabolic disorders, and causing systemic inflammation[53,54]. The risk of small intestinal bacterial overgrowth (SIBO) in obese people has been reported to be almost double that in non-obese people[55], but the exact mechanism involved remains unknown. Down-regulation of antimicrobial humoral response-related genes in the jejunal mucosa may be one of the pathological mechanisms leading to SIBO.

Based on the above analysis, we speculated that the transcriptomic changes observed in the jejunum of ODS patients might impair intestinal barrier function and increase intestinal permeability, which was supported by the elevation of serum DAO and D-LA. Intestinal barrier dysfunction may promote the penetration of bacterial antigens and endotoxins into the intestinal mucosa, resulting in an unbalanced response of mucosal immune cells and an elevated level of systemic endotoxin. This condition may then contribute to systemic and tissue-specific chronic inflammation, oxidative stress, insulin resistance, and adipocyte hyperplasia, all of which are involved in the pathogenesis of obesity and obesity-related changes[12,13,56]. Studies in both human cohorts and animal models have confirmed the association between a leaky gut and obesity, yet few have investigated the impact of these pathological processes on bowel habits. Our study suggested obese individuals with diarrhea may have more severe intestinal dysfunction than those without diarrhea. Future studies are needed to explore the relationship between diarrhea and metabolic complications in obese people.

As the first report on the jejunal transcriptomic profile of ODS patients, our pilot study has several limitations. First, we studied a small cohort due to the difficulty of recruiting patients during the COVID-19 pandemic and the need to rule out COVID-19 infection. However, all subjects were from a city in China, which likely minimized environmental differences. Second, jejunal samples were collected in a fasting state; however, the intestinal function might vary over time (e.g., after meal assumption). Third, we did not fully adjust for all clinical covariates associated with clinical response. Some uncontrolled non-genetic factors, such as exercise and diet, may have influenced the results. However, we designed strict inclusion and exclusion criteria to reduce bias and eliminate the effects of drugs and infections. Finally, the described gene expression profile requires further metabolomics and proteomics validation.

CONCLUSION

In summary, this study provides the first evidence of altered jejunum transcriptome in ODS patients, characterized by up-regulation of genes involved in nutrient absorption, digestion, and transport and down-regulation of genes involved in rRNA processing, mitochondrial translation, antimicrobial humoral responses, DNA replication, and DNA repair. Intestinal barrier function is more impaired in ODS patients than in obese individuals without diarrhea, which may be due in part to the jejunal transcriptomic changes. These findings pave the way for the development of personalized therapeutic approaches to modulate intestinal barrier function and improve weight control in ODS patients.