# paper.pdf ## Metadata - PDFFormatVersion=1.5 - Language=en - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - Author=Bruna Maitan Santos; Jordi Estellé; Yuliaxis Ramayo-Caldas; Sead Chadi; Monica Barone; Florian Chain; Camille Kropp; Patrizia Bridigi; Philippe Langella; Rebeca Martín - CreationDate=D:20260422201345+05'30' - Creator=Arbortext Layout Developer 12.3.7125/W-x64 - Custom.EPSprocessor=PStill version 1.84.42 - ModDate=D:20260422201509+05'30' - Producer=MPS - Title=Sex modulates the long-term effects of delivery mode on microbiota–gut barrier crosstalk and colitis susceptibility in mice - dc:creator=Bruna Maitan Santos,Jordi Estellé,Yuliaxis Ramayo-Caldas,Sead Chadi,Monica Barone,Florian Chain,Camille Kropp,Patrizia Bridigi,Philippe Langella,Rebeca Martín - dc:identifier=10.1080/19490976.2026.2658276 - dc:title=Sex modulates the long-term effects of delivery mode on microbiota–gut barrier crosstalk and colitis susceptibility in mice - pdf:producer=MPS - pdfaextension:schemas=http://ns.adobe.com/pdf/1.3/pdfinternalA name object indicating whether the document has been modified to include trapping informationTrappedTextAdobe PDF Schemahttp://ns.adobe.com/xap/1.0/mm/xmpMMinternalUUID based identifier for specific incarnation of a documentInstanceIDURIinternalThe common identifier for all versions and renditions of a document.OriginalDocumentIDURIXMP Media Management Schemahttp://www.aiim.org/pdfua/ns/id/pdfuaidinternalPart of PDF/UA standardpartIntegerPDF/UA ID Schema - pdfuaid:part=1 - xmp:createdate=2026-04-22T20:13:45+05:30 - xmp:creatortool=Arbortext Layout Developer 12.3.7125/W-x64 - xmp:metadatadate=2026-04-22T20:15:08+05:30 - xmp:modifydate=2026-04-22T20:15:09+05:30 - xmpmm:documentid=uuid:8114cee6-425b-4fb1-82e2-d482741f00f0 - xmpmm:instanceid=uuid:6531e307-7fe9-4ee7-8c41-620700eca114 ## Contents ### Page 1 GUT MICROBES2026, VOL. 18, NO. 1, 2658276 ![Taylor and Francis Logo]() # RESEARCH ARTICLE ![Open access logo]() ![Cross mark logo]() # Sex modulates the long-term effects of delivery mode on microbiota–gutbarrier crosstalk and colitis susceptibility in mice Bruna Maitan Santos,a Paris-Saclay University, INRAE, AgroParisTech, Micalis Institute, Jouy-en-Josas, France 1 Equal contribution. , Jordi Estellé, b Paris-Saclay University, INRAE,AgroParisTech, GABI, Jouy-en-Josas, France , Yuliaxis Ramayo-Caldas,, c Animal Breeding and Genetics Program, Institute for Research and Technology inFood and Agriculture (IRTA), Torre Marimon, Caldes de Montbui, Spain , Sead Chadi, Monica Barone d Department of Pharmacy and Biotechnology, Universityof Bologna, Bologna, Italy ,Florian Chain , Camille Kropp, Patrizia Bridigi, Philippe Langella and Rebeca Martín;;; Contact Rebeca Martín ![Email logo]() ![Address logo]() Centre de Recherche INRAE de Jouy-en-Josas, Domaine de Vilvert, Bât 440 R-2,Jouy-en-Josas, 78352, France © 2026 INRAE, Micalis Institute.This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permitsunrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. The terms on which this article has been published allowthe posting of the Accepted Manuscript in a repository by the author(s) or with their consent. ABSTRACT Sexual dimorphism and mode of delivery are key determinants of gut physiology andmicrobiota development and may differentially affect predisposition to gut-relateddiseases. Cesarean section delivery markedly shapes early-life microbiota, predisposingindividuals to higher risk of immune and metabolic comorbidities later in life. Althoughboth sex and delivery mode are known to influence gut barrier–microbiota crosstalk,whether delivery mode modulates or counter-regulates sex-specific features of thisinteraction remains, to our knowledge, largely unexplored. Here, we investigated howsex impacts gut barrier–microbiota crosstalk shaped by delivery mode across develop-ment until adulthood by reanalyzing existing data. Using a preclinical mouse model, wecombined gut barrier analyses with differential abundance and co-occurrence networkapproaches (LinDA and NetMoss). We found that the impact of CSD on gutbarrier–microbiota crosstalk is partially dependent on sex and life stage. During thefirst days of life, delivery mode dictates immune imprinting and microbial networktopology, with only limited sex effects. However, trajectories diverged with age, withCSD males exhibiting colitis reoccurrence in adulthood. By applying integrative strate-gies to stratify data by sex and development, our study uncovers short- and long-termsex-dependent gut barrier and microbial signatures. These findings reveal that deliverymode might program sex-specific host-microbiota trajectories with consequences forgut health and disease susceptibility, highlighting the need to consider sex and early-life microbial imprinting in future microbiome-targeted interventions. ARTICLE HISTORY Received 07 January 2026Revised 17 March 2026Accepted 03 April 2026 KEYWORDS Early-life; vaginal delivery;C-section; sex differences;colitis; microbiota ## Introduction Biological sex refers to the intrinsic characteristic that distinguishes males from females. Sex influences theprevalence, course and severity of the majority of common diseases and disorders, including noncommu-nicable diseases. Sexual dimorphism has been explained by differences in chromosome genes, sexhormones and social environmental behaviours (drinking, smoking, etc.), although these factors do notcompletely explain the differences found. The gut microbiota has been found to have a significant impacton the pathogenesis of several chronic diseases. Interestingly, the composition of the gut microbiota alsovaries between sexes. Numerous studies in both mice and humans have found sex differences in beta-diversity and alpha-diversity in the gut microbiota, and sex have been found among the ten major factorsthat explain taxonomic human gut microbiota variations. Nevertheless, a consistent pattern ofcommunity-level sex differentiation has not been found, and the proposal of robust and reliable microbialindicators of sex has not been made possible due todifferences among studies. Overall, while sex cancontribute to gut microbiome development, the complexity of factors that interact and influence its ![Supplementary Logo]() Supplemental data for this article can be accessed online at https://doi.org/10.1080/19490976.2026.2658276. ### Page 2 development impedes consistency across studies to confirm sex effects and characterize them. As a proof,only one human study has described that infant sex as a dominant contributor to infant gut microbialdevelopment. C-section delivery (CSD) is one of the most common surgeries in the world, accounting for approximately21% of all childbirths worldwide and being more than 40% of births in some countries, such as Brazil. In1985, the World Health Organization (WHO) recommended that CSD rates should be lower than 10%–15%.However, CSD rates have continued to rise, mainly owing to personal or medical convenience, andprojections suggest that by 2030, the global CSD rate will be nearly 30%. CSD, which is mainly elective(without labor), has been related to an increased risk of noncommunicable diseases (NCDs). Several meta-analyses have shown that infants born via CSD are at a higher risk of developing allergies, asthma, IBD andobesity, among others. Since these children have an altered bacterial community, it has been hypothesizedthat these alterations lead to differences in mucosal tolerance and disease risk later in life., It is now recognized that sex bias exists in biomedical research, which is the basis of clinical research andmedical decision making. Traditionally, female individuals have been underrepresented in research studies,both in randomized clinical trials and in preclinical animal trials. Although sex differences in murine modelsare well known, they are often mistreated by only analyzing one sex, most often males. Sex has been foundto be a factor in the CSD association with NCDs. For instance, a meta-analysis has found that asthma has ahigher incidence in CSD female offspring compared to VD ones while no differences have been found inmales. Also, the CSD risk of obesity seems higher for boys than for girls in Canada and US cohorts,, butthe opposite in a Chinese cohort. Regarding neurodevelopmental disorders, CSD effects have been associatedwith an increased risk of neurodevelopmental disorders in males and with an increased risk of motor delay infemales. Another important CSD effect characterized its association with inflammatory bowel diseases,with some studies showing potential sex effects., Taken together, understanding the sexual dimorphism indiseases is essential to investigate the pathogenesis of some chronic diseases and better deciphering the role ofCSD in their increased incidence. Here, our primary objective was to determine whether susceptibility to colitis reocurrence later in lifediffers according to sex in CSD animals. As secondary objectives, we investigated whether CSD alterationsin gut barrier integrity, immune markers, and the microbiota centrality network during early life could beassociated with sex differences in disease susceptibility later in life. Finally, we analyzed the microbialnetwork structure and abundance across multiple developmental stages to evaluate potential patterns. It isimportant to highlight that this study was conducted by reanalysing pre-existing preclinical data, inwhich sex was not considered as a biological variable. ## Materials and methods ### Animals All procedures were approved by the regional ethics committee (Comethea) and adhered to EU Directive 2010/63/EU on animal protection. Experiments were conducted under specific pathogen-free (SPF) conditions undercontrolled ambient temperature (21 °C) and humidity with water and food ad libitum. All in vivo experimentswere conducted at the National Research Institute for Agriculture, Food and Environment (INRAE), within theInfectiology of Farm, Wildlife and Laboratory Animals Platform (IERP) in Jouy-en-Josas, France. Timedpregnant RjOrl:SWISS mice (Janvier Labs, Le Genest-Saint-Isle, France) were used. ### Mode of delivery model and experimental timeline This study is based on previously published data comparing cesarean-delivered (CSD) and vaginally delivered(VD) animals, which were reanalyzed here with a focus on sex-dependent effects. On gestational day 19, damsRjOrl:SWISS (Janvier Labs; Le Genest-Saint-Isle, France) underwent hysterectomy; neonates were immediatelyfostered by surrogate dams that had delivered vaginally within the previous 12 h (C-section delivery, CSD).Control litters were obtained via spontaneous vaginal delivery (VD). To minimize potential litter effects, VDpups were cross-fostered among dams delivering in the same time window. Offspring of both sexes were group-housed and separated by sex after weaning. ### Page 3 As illustrated in the experimental timeline (), our analyses were performed in both female andmale VD or CSD offspring from the first days of life (5 d, in green) until adulthood after colitisreoccurrence (8 weeks, in pink). To better picture each stage of life, graphs followed the same colorscheme as the corresponding age in the timeline. At each life stage, the parameters analyzed for potentialsex effects were those that showed an impact of CSD at the corresponding age. ### Colitis-induced models Two models were used in this analysis. The acute colitis model involved 4-week-old mice that wereanesthetized (0.1% ketamine + 0.06% xylazine, i.p.) and received 2,4-dinitrobenzene sulfonic acid (DNBS,200 mg/kg, Sigma-Aldrich, France) intrarectally in 50 μL of 30% ethanol in sterile phosphate-based saline(EtOH) to induce colitis in young adulthood. Vehicle controls received EtOH only. A colitis reoccurrence model was used to induce chronic colonic inflammation. Four-week-old micewere submitted to chronic administration of DNBS as previously described. Briefly, mice were chal-lenged with a first dose of DNBS (200 mg/kg), the same as that used in the acute colitis model. After 21 d ofrecovery, the colitis was reactivated using DNBS at the same dose. The control mice received only EtOH. A brief explanation of the colitis induction and the parameters used to evaluate its severity can be foundin. Colonic macroscopic score measurements were conducted blindly. ### Quantification of serum lipocalin 2 and serum soluble CD14 (sCD14) levels Serum lipocalin 2 levels were measured following kit instructions mouse lipocalin-2/NGAL duo Set ELISA(R&D systems, Inc., USA & Canada). Serum soluble CD14 (sCD14) levels were measured using sCD14ELISA kit following the manufacturer's instructions (R&D systems, Inc., USA & Canada). ### Quantification of colonic myeloperoxidase (MPO) activity and colonic cytokines measurement Colonic samples were thawed and homogenized in 0.5% hexadecyltrimethylammonium bromide (HTAB)prepared in 50 mM potassium phosphate buffer (pH 6.0) using a Precellys 24 homogenizer (BertinTechnologies, France) with ceramic beads (3 × 30 s at 6,000 rpm, 4 °C). The homogenates were centrifugedat 10,000 × g for 15 min at 4 °C, and the resulting supernatants were collected for MPO and cytokinemeasurements. ![The diagram shows an experimental timeline on a horizontal black line from 0 to 8 weeks old, labeled w.o. A green dashed arrow points to 1 w.o. labeled Endpoint first days. A blue dashed arrow points to 3 w.o. labeled Endpoint weaning. An orange dashed arrow points to 4 w.o. labeled Endpoint post-weaning colitis, with a solid orange box below it labeled DNBS. A pink dashed arrow points to 6 w.o. labeled Young adult sampling, with a solid pink box below it labeled DNBS. Another pink dashed arrow points to 8 w.o. labeled Adult sampling, with a solid pink box below it labeled DNBS. A final pink dashed arrow points to 8 w.o. labeled Endpoint colitis reoccurrence in adulthood. Below the timeline, a gray line with vertical bars at 0, 1, 3, 4, and 6 w.o. is labeled Netmoss score from microbiota individual and integrative score. Below this, a black line with vertical bars at 0, 1, 3, 4, 6, and 8 w.o. is labeled LinDA analysis from microbiota.]() ### Page 4 MPO activity was determined spectrophotometrically by mixing 10 μL of the homogenate with 200 μLof reaction buffer containing o-dianisidine dihydrochloride and hydrogen peroxide (H₂O₂). The change inabsorbance was measured at 450 nm for 5 min at 25 °C using a microplate reader. MPO activity wasexpressed as units per gram of tissue (U/g) as previously described. Colonic IL-1α, IL-1β, IL-6, IL-10, IL-12p70, IL-17A, IL-23, IL-27, MCP-1, IFN-β, IFN-γ, TNF-α, andGM-CSF cytokines were determined with LEGENDplex™ mouse inflammation panel according to themanufacturer's instructions (Biolegend, USA). ### Histological analyses Histological features were assessed using Alcian blue (AB) staining in accordance with standard proto-cols, in the Histology facility of a Bridge platform of UMR 1313 GABI. Images were blindly analyzedwith Panoramic Viewer software. ### Splenic T cells populations Splenic cell suspensions from the first days of life at the pups' endpoint were obtained by mechanicalextrusion through a 40-μm nylon cell strainer (BD, Switzerland). The cells were strained using 1 mL ofDulbecco's modified Eagle's medium (DMEM, Gibco, France) supplemented with 10% fetal bovine serum(FBS, Gibco, France). Erythrocytes were lysed using buffer Hybri-Max (Sigma-Aldrich, USA) followingspecifications. The cell pellets were resuspended in 1 × 106 cells and blocked with anti-CD16/32 andstained with panels CD3+CD4+Tbet+and CD3+CD4+ROR-γ+ using eBioscience antibodies (France). Thesamples were subsequently analyzed using an Accuri C6 cytometer (BD). Data were acquired on an AccuriC6 flow cytometer (BD) and analyzed using CFlowSampler. ### Using chamber experiments Segments of the ileum and colon were opened along the mesenteric border and mounted in Ussing chambers(P2300, 0.2 cm2 exposed area) with oxygenated Krebs buffer (glucose on the serosal side; mannitol on themucosal side) at 37 °C. Transepithelial conductance and short-circuit current were recorded under a voltageclamp. To assess paracellular permeability, TRITC (4 kDa, TdB Labs, Sweden) was added to the mucosalchamber (0.4 mg/mL). Serosal samples were collected every 15 min for 2 h and analyzed enzymatically. Trans-epithelial conductance was measured by clamping the voltage and recording the change in the short-circuitcurrent (Isc) Carbachol (100 μM) was added serosally at the end of the experiments to verify tissue viability. ### Microbial DNA extraction and 16S sequencing analysis Microbial DNA extraction was performed with standard procedures as previously reported In brief, DNAwas isolated from colon tissue (postnatal day 5), colonic content (endpoint), and fecal samples collected at ![The diagram illustrates colitis induction and severity parameters. A horizontal arrow labeled Colitis points from Day 0 to Day 4. Below Day 0, a mouse figure is shown with text Anesthesia: 0.1 percentage ketamine and 0.06 percentage xylazine intraperitoneally. Two arrows originate from the mouse. The upper arrow points to a box with text DNBS: 200 milligrams per kilogram in 50 microliters of vehicle \(EtOH\) in PBS intrarectally. The lower arrow points to a box with text EtOH \(vehicle\): 50 microliters of 30 percentage ethanol in PBS intrarectally. To the right, under Day 4, a box labeled Severity parameters lists four bullet points: Sickness symptoms \(more body weight loss\), local damage \(high colonic macroscopic score\), mucus production \(decrease in AB plus cells percentage\), and leaky gut barrier \(increased FITC-dextran\).]() ### Page 5 intermediate time points and from maternal feces, skin, and vaginal swabs using the QIAGEN Mouse StoolMini Kit (QIAGEN, Hilden, Germany) according to the manufacturer's instructions. The V3–V4 region ofthe 16S rRNA gene was amplified with the primers PCR1F_343 and PCR1R_784, and the Illumina MiSeqplatform (Illumina, USA) was used to sequence. The raw sequence data were deposited in the Sequence ReadArchive under Bioproject PRJNA876103 (accessions SAMN30638467–SAMN30638852). Bioinformatic analysis of 16S sequencing data was performed using DADA2 v1.28.0 R packagefollowing the author's recommendations. In brief, we applied the standard DADA2 pipeline with defaultparameters for (1) quality filtering of the raw reads, (2) denoising DADA2 algorithm, (3) mergingpaired‑end reads, (4) removing chimeric sequences, and (5) quantifying amplicon sequence variants(ASVs) in each sample. The taxonomic classification of each ASV was performed with the DADA2naive classifier using the SILVA v.138.1 database as the reference taxonomy. Finally, the abundance tableat the genera level was constructed by using the tax_glom funtion in the phyloseq v1.46.0 R package. ### NetMoss analysis The resulting genus abundance table was used for diversity analyses and aggregated to the genus level forco-occurrence Sparcc network inferences between the VD and CSD groups with NetMoss. Networkgraphical representation was created with CytoScape, and the topological parameters of the network and“node” centrality values were calculated using the CentiScaPe plugin. Centrality measures, such asdegree, quantify the importance of each bacterial genus within the network, reflecting its potentialinfluence on microbial community structure and intermodule interactions. High-centrality nodes areinterpreted as key taxa that may coordinate community dynamics, while low-centrality nodes representmore peripheral members. Modules within these networks were identified using the Weighted Gene Co-Expression Network Analysis. This step captures microbial taxa that cooperate within the same modulewhile maintaining competitive interactions between modules. The contribution of each node (bacterialgenus) to the transition between VD and CSD networks was calculated as follows:where A and B represent the VD and CSD networks, respectively. D denotes the differential moduledistance matrix, Neighbors A include all neighboring modules in the VD network, and Neighbors Brepresent all neighboring modules in the CSD network. Microbial network approaches assume that co-occurrence patterns represent biologically meaningfulinteractions and that the resulting network topology reflects underlying dependencies among taxa. Toensure robustness, analyses were performed separately for females and males, and key driver taxa wereidentified using conservative NetMoss score thresholds (≥0.5 for weaning, ≥0.6 for early life). Theintersection modules represent the stable elements during the transition, where the transition modulesresulted in alterations in the network structure. In this way, the model quantifies the shift in modulestructure, highlighting microbial taxa that drive network changes between the experimental conditions.Nodes with high NetMoss scores correspond to genera that contribute most to network restructuringbetween VD and CSD conditions. Their centrality values provide additional biological insight: taxa withhigh centrality are likely influential in maintaining community stability or mediating interactions betweenmodules, suggesting mechanistic roles in microbiota assembly. Receiver operating characteristic (ROC)curves and area under the curve (AUC) metrics were used to assess the discriminatory power of these taxain classifying delivery mode. ### LinDA differential abundance analysis Differential abundance analyses were performed using Linear Models for the differential abundanceframework in the LinDA v0.2.0 R package. This method fits linear regressions on centered log-ratio(CLR) transformed abundance data, identifies a bias term owing to the compositional nature of micro-biome data and corrects this bias using the mode of regression coefficients across taxa. The same raw reads ### Page 6 were reprocessed using p-values derived from the bias-corrected coefficients and adjusted for multipletesting by Benjamini–Hochberg. The results are summarized as effect sizes and adjusted p-values.Differential abundance analyses were used in three models to test: (1) the effect of sex (male vs. female)corrected by delivery mode, (2) the effect of delivery mode (VD vs. CSD) corrected by sex, and (3) theinteraction effect between sex and delivery mode (i.e., a differential abundance model comparing fourlevels: VD_male vs. VD_female vs. CSD_male vs. CSD_female). In all three models, the effect of DNBStreatment was included as a cofactor for time points after the start of the first treatment. The addition ofadditional co-factors such as dams or adoptive dams was explored but not deemed necessary. ### Statistical analyses of mouse phenotypes The strategy of this study to evaluate sex effects in VD and CSD animals was achieved by re-using pre-existing data with only parameters affected by CSD. The data distribution was evaluated using theShapiro–Wilk test and Prism's distribution-likelihood analysis. Two- or three-way ANOVA was applied toevaluate the effects of delivery, sex, and DNBS analyses for normally distributed data with equal variances,followed by Fisher's test with multiple comparisons. When assumptions of normality or variancehomogeneity were not met, the Kruskal–Wallis test was used, followed by Dunn's post hoc test withmultiple comparisons. Both analyses were performed in Prism 11 (GraphPad Software, USA). The numberof animals is referenced in each graph legend and detailed with attrition in Sup. Tables 1 and 2. p-values < 0.05 were considered statistically significant. ## Results ### Sex-dependent differences in colitis reoccurrence in CSD animals during adulthood To determine whether susceptibility to colitis reoccurrence differs according to sex and delivery mode, weevaluated body weight loss as the primary endpoint of colitis severity, together with mucin production andinflammatory markers following DNBS rechallenge in adulthood. CSD males lost more weight compared to VD counterparts with colitis reoccurrence (). Thiseffect was supported by a decrease in acid mucin (AB+) cells (), together with increased MPOactivity () and Il-1α levels (), reinforcing a stronger inflammatory response in CSDmales after colitis reoccurrence than VD counterparts. In contrast, females had no weight loss or decreased AB+ cells after colitis reoccurrence compared withtheir vehicle-treated counterparts (EtOH, and, respectively). Instead, AB+ cells were increasedin CSD females compared with those in VD females following colitis reoccurrence (). The lack ofincrease in MPO activity () and Il-1α levels () in all females with colitis reoccurrencesupports the stronger impact of colitis reoccurrence in CSD males. As previously described, CSD facilitates an exacerbated response to acute colitis in mice., Asexpected, DNBS caused marked body weight loss in all CSD groups (Supl. Figure 1). Another parameterused to evaluate the severity of the colitis induced by DNBS was local damage, which was measured usingcolonic macroscopic score, and male CSD animals showed a tendency toward a higher macroscopic score(p = 0.05) compared to VD counterparts, while no difference was detected between the female groups(Supl. Figure 1). ### First days' gut barrier immune development and gut microbial network centrality are shaped bydelivery mode with low sex influence To explore potential early-life factors that could precipitate colitis recurrence observed in CSD malesduring adulthood, we evaluated inflammatory and barrier-related parameters during the first days of life.Specifically, we profiled local inflammation (lipocalin 2, cytokines, and immune cells) and barrier integrity(sCD14) parameters in first days' pups born by VD or CSD. In parallel, the microbial network structurewas analyzed to determine whether early microbial interactions differed according to delivery modeand sex. ### Page 7 Some inflammatory parameters that increased in CSD were significant only in CSD females(), specifically the increase in serum lipocalin-2 levels () and colonic IL-6 and IFN-γ levels (), after pairwise comparisons. However, using linear modeling, CSD animals hadincreased lipocalin-2 levels [26,534 ± 10,762 pg/mL, p < 0.05; F(4, 55) = 97, p < 0.001], with only a trendfor sex and delivery mode interaction (p = 0.07). Also, delivery mode had an impact on both cytokines[14.9 ± 5.6 pg/mL, p < 0.01; F(2, 37) = 6.40, p < 0.01 and 5.2 ± 1.9 pg/mL, p < 0.01; F(2, 37) = 8.11, p < 0.001,respectively], but again, there was no sex effect. CSD males had a potential impact on colonic IL-1α(, p = 0.05) after pairwise comparisons. IL-1α is usually released due to cellular damage caused bybacterial translocation, and CSD is associated with higher colonic IL-1α levels [14.9 ± 5.6 pg/mg, p < 0.01; F(8,110) = 7.58, p < 0.001] and male animals with low levels (–15.1 ± 5.6 pg/mg, p < 0.01; F(8,110) = 7.58,p < 0.001), but there is no delivery‒sex interaction. Even though they were impacted by CSD model duringthe first days of life, no alterations in serum soluble CD14 (sCD14;), an important marker ofbarrier disruption, or in the proportion of splenic CD3+CD4+ROR-γ+ Th17 cells () wereobserved when groups were divided by sex after pairwise comparisons, showcasing the low effect of sexin CSD in gut barrier development during their first days of life. We also profiled sex-dependent gut microbial network features in first days' mice born by VD or CSD.Without accounting for sex, Lactobacillus was the key taxa centrality network of VD pups, while ![The four-panel graph arranged in two rows of two shows colitis reoccurrence in adulthood. Panel a, a scatter dot plot, displays percentage of weight loss on the vertical axis from minus 30 to 15. The horizontal axis shows VD and CSD groups for both female and male mice, under plus EtOH and plus DNBS conditions. Under plus DNBS, CSD males show significantly greater weight loss compared to VD males. Panel b, a violin plot, shows colonic AB plus cells per total on the vertical axis from 0.0 to 1.0. Under plus DNBS, CSD males show significantly fewer AB plus cells compared to VD males. Panel c, a box plot, shows colonic MPO in U per milligram on the vertical axis from 0.000 to 0.015. Under plus DNBS, CSD males show significantly higher MPO activity compared to VD males. Panel d, a box plot, shows colonic IL-1 alpha in picograms per milligram on the vertical axis from 0 to 10. Under plus DNBS, CSD males show significantly higher IL-1 alpha concentrations compared to VD males.]() ### Page 8 ![The 9-panel figure displays bar graphs of serum/colonic markers, T cell populations, and NetMoss scores for bacteria. Panels a and b show increased levels in c-section females vs. vaginal delivery counterparts. Panel c shows increased levels in c-section males vs. vaginal delivery counterparts. No differences were observed in Panels d and e between sexes. Panel a: serum lipocalin 2., vaginal delivery females/males near 100000 pg/mL, c-section near 200000 pg/mL. Panel b: colonic IL6 and IFN-gamma, low for vaginal delivery, increasing for c-section. Panel c: colonic IL1-alpha, vaginal delivery females/males near 20 pg/mg, c-section near 40 pg/mg. Panel d: serum soluble CD14, vaginal delivery females/males near 200 pg/mL, c-section near 400 pg/mL. Panel e: splenic CD3+CD4+ROR-gamma T cell populations, vaginal delivery females/males near 5%, c-section near 10%. Panels f and g show NetMoss scores \(0.00-1.00\) for top 40 microbial taxa in females and males, respectively. Panels h and i are tables of butyrate-producing taxa with increased NetMoss scores in c-section females and males, respectively.]() ### Page 9 Mucispirillum for CSD pups. Here, NetMoss-based microbial network analysis, divided by sex,revealed an increased centrality network of butyrate producers in CSD animals, indicating a moreinterconnected microbial structure in their first days of life (). Among these, the butyrate-producing genera Odoribacter, Alistipes, Lachnospiraceae (UCG-001 and FCS020 group), [Eubacterium]siraeum group, and Ruminococcus were enriched in CSD and exhibited high NetMoss scores, particularlyAlistipes with the highest network centrality, independent of sex (). Sex-specific butyrate-producers' signatures were detected in male CSD in the first days of life. Theyshowed potential increased network connectivity for Roseburia, Lachnospiraceae UCG-006,Anaerotruncus, and Butyricicoccus compared to male VD counterparts (). Roseburia had thehighest score. During childhood, it seems that a low relative abundance of Roseburia was associated withrespiratory diseases and becomes more abundant only after weaning, with early food introductionpromoting its enrichment in piglets. These results suggest that nutritional and developmental cues, such asthose observed in CSD animals, can accelerate Roseburia establishment and abundance within the gutearlier than normal microbiota development. Also, a denser network centrality in butyrate-producing taxa () was observed in male CSD.Unfortunately, butyrate levels were undetected during this period of life, limiting our conclusionsregarding the functional production of butyrate. ### Males CSD had impaired colonic barrier function, low caecal butyrate levels, few remainingbutyrate-producer taxa network centrality, and sex-specific abundant taxa during weaning Three gut barrier parameters impacted by CSD were assessed to evaluate sex effects altering CSD impactat weaning: (1) colonic and ileal epithelial conductance (), (2) colonic and ileal permeabilitymeasured by local tetramethylrhodamine isothiocyanate (TRITC) levels (), and (3) colonictransepithelial electrical resistance (TEER;). Caecal short-chain fatty acids (SCFAs) levelsimpacted in CSD animals during weaning were assessed by dividing the groups by sex ().Next, NetMoss score and linear models for differential abundance (LinDA) analyses were conducted toverify how sex interacts with the delivery mode shaping microbiota network centrality () anddifferential abundance (), respectively. A trend toward a delivery mode-by-tissue interaction in epithelial conductance was noted [p = 0.17; F(3,30) = 11.4, p < 0.001]. No differences between groups after pairwise comparisons were found ().Paracellular permeability was strongly affected by sex (p < 0.001), with potential delivery-by-sex interac-tions (p = 0.17). An increase in paracellular permeability in CSD males compared to VD males was foundafter pairwise comparisons (). Finally, sex significantly influenced colonic TEER (p < 0.01; F(3,75) = 3.1, p < 0.05), with a trend toward an interaction between delivery mode and sex (p = 0.08). Whenpairwise comparisons were made, a reduction in TEER was observed only in CSD males (). Thetotal variation in resistance over 2 h was significantly reduced in CSD males [; F(3, 65) = 3.999,p < 0.01], which is consistent with impaired barrier maturation. Caecal butyrate levels were significantly reduced in CSD males compared to VD males, while noreduction was observed in females (). In contrast, acetate concentrations were higher in CSDfemales compared to VD females, whereas male groups showed no increase (). These findingsindicate that delivery mode alters SCFA profiles during weaning in a sex-dependent manner, with maleCSD animals displaying reduced butyrate availability. Gut microbial network connectivity was assessed using NetMoss score around weaning, and the strongcentral connectivity observed in CSD animals in their first days of life remained with fewer taxa only toCSD males (). Specifically, Lachnospiraceae NK4A136 and [Eubacterium] xylanophilum taxadisplayed greater centrality in CSD males, followed by the butyrate-producers Butyricicoccus andAnaerotruncus, which maintained centrality in CSD males' microbial network compared to their VDmale counterparts (). Conversely, the non-butyrate-producer Peptococcus exhibited greatercentrality in the CSD male networks. Curiously, the butyrate-producer Lachnospiraceae FCS020 showedgreater centrality in VD males' microbiota networks (), although in the first days of life, it hadbeen more central in CSD animals (). ### Page 10 ![13-panel graph: gut microbiota and barrier function by delivery mode and sex at weaning. a\) Colon electrical conductance \(mS/cm^2\): Vaginal delivery \(VD\) females and males between 20 and 30, C-section \(CSD\) females near 20 and males near 30. b\) Colon TRITC \(ng/mL\): VD females near 500 and males near 100, CSD females near 1000, CSD males near 500. c\) Colon TEER over time \(minutes\): VD females,males, and CSD females show increase, CSD males decrease. d\) Colon TEER Total: All groups near 0, with CSD males the lowest. e\) Netmoss scores for top 40 bacteria: Acetatifactor and Lachnospiraceae NK4A136 group near 0.9. f\) Ileum electrical conductance \(mS/cm^2\): VD females and males near 10, CSD, females and males near 15. g\) Ileum TRITC \(ng/mL\): All groups similarly between 500 and 1000. h\) Butyrate \(mM\): VD females and males near 10, CSD females near 10, CSD males near 5. i\) Acetate \(mM\): VD females and males between 30 and 40, CSD females near 40, and CSD males near 20. j\) Butyrate producers table: Lachnospiraceae FCS020 group Netmoss 0.5 was the onlyenriched in VD. k\) Ligilactobacillus abundance: CSD females higher than VD females and CSD males lower than VD males . l\) Candidatus Arthromitus abundance: CSD females higher than VD females and CSD males lower than VD males . m\) Harryflintia abundance: VD females higher than CSD females and VD males lower than CSD males ]() ### Page 11 Finally, from LinDA integrative analysis, we observed some clear sex-dependent alterations aroundweaning in the abundance of the Ligilactobacillus, Candidatus arthromitus, and Harryflintia taxa (). Specifically, Ligilactobacillus and Candidatus arthromitus counts were higher in CSD femalescompared to VD females, whereas it was found decreased in CSD males compared to VD males,respectively (). Analysis at species levels also confirmed this pattern (Supl. Fig. 13).Ligilactobacillus lactic acid bacteria are usually related to benefits to the host, being decreased inCSD males. This is also the case for Candidatus Arthromitus. More present in the ileum, this bacteriumis recognized by its protective immune function. The absence of CSD effects in females during weaningsuggests a sex-driven counterbalance in the gut microbiota composition that may mitigate CSD-associatedalterations in females, which are observed more clearly in males. Harryflintia was virtually absent in CSDfemales and VD males but significantly enriched in VD females and CSD males (). Owing tochallenges in isolation, little is known about its effects on the host. ### Key butyrate-producers from first days of life until young adulthood are sex- and delivery mode-dependent To explore microbial network centrality that might be related to the increased susceptibility to colitisreoccurrence observed in CSD males during adulthood, we examined whether the establishment andpersistence of butyrate-producing taxa trajectories across development differed according to delivery modeand sex. During the first days of life, CSD animals exhibited butyrate producers' network centrality, withmales having more butyrate producers on their network centrality (), whereas at weaning, thispattern was restricted to CSD males (). Around the same period, sex-dependent effectsinteracting with CSD were also detected in the differential abundance of some taxa (). Theseobservations suggest that delivery mode, sex, and age jointly influence the establishment of butyrateproducers and their connectivity during early development. To evaluate the key taxa network across earlydevelopment, i.e., between the first days of life until young adulthood, we used NetMoss to identify sex-dependent key taxa in males and females in VD and CSD animals (). In females, some taxa showed high NetMoss scores enriched between VD and CSD animals across ages(). Among them, one non-butyrate-producer (Anaeroplasma, score = 0.7) was distinguished as akey taxon enriched in VD. In contrast, the butyrate-producers UBA1819 (score = 1.0) and Ruminococcus(score = 0.6) were the key taxa enriched in CSD female animals (). In males, distinct taxa also separated VD from CSD animals (). Opposite to the pattern infemales, butyrate-producers were the key enriched taxa in VD, including [Eubacterium] siraeum group(score = 0.9), Lachnospiraceae UCG-001 (score = 0.7), and Roseburia (score = 1.0). Conversely, the non-butyrate-producers Anaeroplasma (score = 0.77), Tuzzerella (score = 0.64), Rikenellaceae RC9 gut group(score = 1.0), and A2 (score = 0.62) were the key taxa enriched in male CSD animals (). Microbialkey signatures in both females (AUC = 0.82) and males (AUC = 0.83) were robustly discriminated in bothCSD and VD groups (). These results highlight that CSD leads to age- and sex-dependent shifts in butyrate-producing taxa abundanceand their cooccurring patterns that may impact immunity imprinting, gut barrier feature establishment, andcolitis severity. The key taxa in CSD females across early development were butyrate producers () withthe opposite profile in VD animals but not in CSD males (), clearly exhibiting the complexity of thegut microbial network across development in CSD animals and the impact of sex. ### Differential microbial abundance analysis across life stages until adulthood reveals a transitionfrom stronger early delivery-mode to later sex associated effects in gut microbiota To dissect how the relative influence of delivery mode and sex evolved across development and potential cueslinking sex-dependent colitis reoccurrence susceptibility, we performed differential microbial abundance analyses ### Page 12 using LinDA, which is specifically designed for compositional microbiome data and correct the bias typicallyintroduced by traditional linear regression models that are not based on centered log-ratio (CLR)–transformedabundances. Overall, LinDA-based analyses revealed a clear developmental transition in the determinants ofmicrobial abundance composition from stronger early effects driven by delivery mode to later patternsincreasingly shaped by sex. Venn diagrams highlight this taxa transition between delivery-associated(corrected for sex) and sex-associated (corrected for delivery) taxa across the first days of life, weaning,postweaning, young adulthood, colitis remission, and colitis reoccurrence in adulthood (). ![The six panel figure shows gut microbiota network analysis. Panel a, a horizontal bar graph, displays NetMoss scores for 40 bacteria in females. The horizontal axis is NetMoss score from 0.00 to 1.00. The vertical axis lists bacteria names. Bars extend right, with some enriched in VD and others in CSD. To the right, a scatter plot shows Log2\(FC\) values from minus 2 to 2 for the same bacteria. Panel b, a horizontal bar graph, displays NetMoss scores for 40 bacteria in males, similar to panel a. Panel c, a table, lists taxon names, NetMoss score, p value, p adjusted, butyrate producer status, and pattern for Anaeroplasma, UBA1819, and Ruminococcus. Panel d, a table, lists taxon names, NetMoss score, p value, p adjusted, butyrate producer status, and pattern for seven bacteria including Eubacterium siraeum group and Lachnospiraceae UCG-001. Panel e, a line graph, shows a receiver operating characteristic curve for females. The horizontal axis is false positive fraction from 0.00 to 1.00. The vertical axis is true positive fraction from 0.00 to 1.00. The curve rises from the bottom left to the top right, with an AUC of 0.82. Panel f, a line graph, shows a receiver operating characteristic curve for males, similar to panel e, with an AUC of 0.83.]() ### Page 13 In the first days of life, most differentially abundant taxa were associated with delivery mode, with nosex effect associations observed (). The results revealed that Streptococcus, Odoribacter, andAcetatifactor were more abundant in VD animals at distinct stages (first days, weaning, and youngadulthood, respectively). These genera are nonspore-forming butyrate producers, with Streptococcus andOdoribacter being more abundant than Acetatifactor in normalized counts. In contrast, the majority oftaxa more abundant in CSD animals during early life belonged to Firmicutes and Bacteroidota, particularlyOscillospiraceae (12 taxa), Lachnospirales (13), and Bacteroidales (6), including Oscillospiraceae (6),Ruminococcaceae (6), and Lachnospiraceae (13), at the family level. Most of these differentially abundanttaxa were butyrate producers (see Supl. Table 3 for further details), indicating that among the taxaenriched early in life, CSD animals harbored more and broader varieties of abundant butyrate producersthan VD animals. As the animals aged, this broader representation of butyrate producers with increased abundancy inCSD animals diminished. Some taxa initially enriched in CSD later became more abundant in VD animals,including Butyricicoccus, [Eubacterium] xylanophilum group, Mucispirillum, LachnospiraceaeNK4A136, Lachnospiraceae FCS020, and Ruminococcaceae UCG-005. Except for Mucispirillum, allare butyrate producers. By adulthood, CSD animals displayed a greater differential abundance in a variety of non–butyrate-producing taxa. Interestingly, taxa enriched in VD animals were life stage-specific and transient, whereassome taxa enriched in CSD animals persisted across stages. In particular, the non-butyrate-producerRikenellaceae RC9 gut group remained consistently enriched in CSD animals at multiple stages, exceptweaning. Giving support, LinDA analysis at species levels maintains Rikenellaceae RC9 gut group as themost enriched at multiple stages, except during weaning and the colitis remission period in adulthood(Supl. Table 4). When analyses were corrected for delivery mode, no sex-associated taxa were identified until youngadulthood (). This turning point coincides with the onset of sexual hormone elevation in mice.The number of sex-associated taxa with higher abundance increased further into adulthood, with femalesshowing a greater representation than males. Colitis reoccurrence reduced the number of CSD-corrected abundant taxa in males compared to colitisremission, leaving Prevotellaceae UCG-001 as the only persistent enrichment at the genus () andspecies levels (Supl. Table 4). Almost all enriched taxa in females shifted from colitis remission toreoccurrence, with the enrichment persistence of [Eubacterium] xylanophilum group and Biophilaonly. Curiously, in agreement with the network inferences, this shift was marked by more abundantbutyrate producers in colitis remission taxa to less in colitis reoccurrence in females compared to males inthe same period (). Worth mentioning, females had more abundant Lactobacillus johnsoniiduring colitis reoccurrence in adulthood, specifically. L. johnsonii is usually considered beneficial to thehost, such as decreasing colitis, which might be related to the lower colitis reoccurrence in females(). ## Discussion This study reveals that delivery mode and sex seem to interact to shape microbiota–gut barrier crosstalkacross life, with unique life stage-specific trajectories that reflect the dynamic co-development of the hostand microbiota. By integrating classical physiological, immunological, and microbial network analyses, weshow that cesarean delivery (CSD) might induce sex-dependent alterations in gut barrier function andmicrobiota composition that persist from early life into adulthood. These effects appear more pronouncedin males, who display increased susceptibility to colitis reoccurrence later in life, along with early-lifemicrobial and barrier alterations. Together, these observations highlight the potential necessity of consid-ering both delivery mode and sex when investigating early-life programming of host–microbiota interac-tions and their potential long-term consequences for inflammatory disease susceptibility. These observations may be relevant in light of human studies reporting an association between CSDand an increased risk of inflammatory bowel diseases. Although evidence remains heterogeneous,some studies have suggested sex-related differences., While these findings cannot fully establishcausality, they support the hypothesis that delivery mode influences early-life microbiota trajectories ### Page 14 ![The three panel figure shows microbial taxa abundance changes across five life stages. Panel a presents two Venn diagrams. The left diagram shows delivery effects with overlapping circles for First days of life, Weaning, Young adult, Colitis remission in adulthood, and Colitis reoccurrence in adulthood. The right diagram shows sex effects with similar overlapping circles. Panel b displays five volcano plots, one for each life stage, showing delivery effects. Each plot has a horizontal axis for log 2 fold change and a vertical axis for minus log 10 Padj. Insets show overall direction of change. A table below lists ASV ID, genus or family, butyrate producers, mean normalized counts, log 2 fold change, p value, adjusted p value, and life stage differences. Panel c displays five volcano plots, one for each life stage, showing sex effects. Each plot has a horizontal axis for log 2 fold change and a vertical axis for minus log 10 Padj. Insets show overall direction of change. A table below lists ASV ID, genus or family, butyrate producers, mean normalized counts, log 2 fold change, p value, adjusted p value, and life stage differences.]() ### Page 15 and host responses in a sex-dependent manner, potentially contributing to differences in disease suscepti-bility later in life. Early-life microbial network butyrate-producing taxa were prominent in CSD animalsduring the first days of life but declined by weaning, suggesting dynamic reorganization of microbialinteractions across development. When microbial profiles from multiple developmental stages wereconsidered together, integrated LinDA differential abundance and NetMoss network centrality analysesrevealed that the microbial community structure evolved from delivery mode–driven differentiation inearly life toward increasingly sex-dependent patterns during maturation. This transition likely reflects theinfluence of sex-specific host regulation,, including immune, metabolic, and hormonal factors thatemerge during development. Further research is needed to better understand the developmental shift observed in microbial networkcentrality. In particular, it remains unclear how the key non–butyrate-producing taxa identified in maleCSD networks and the butyrate-producing taxa enriched in female CSD networks interact with themicrobial communities detected at earlier developmental stages. The differences observed between inte-grated network analyses and those performed at individual time points suggest that microbial interactionsevolve dynamically across development. Interestingly, when microbial data from multiple developmental stages were integrated, distinctmicrobial signatures robustly distinguished CSD from VD animals. This observation highlights thepotential value of integrative network approaches for identifying persistent microbial features associatedwith delivery mode and underscores the importance of examining microbial interactions across develop-mental time rather than focusing on isolated life stages. A particularly notable taxon emerging across analyses was the Rikenellaceae RC9 gut group, anon–butyrate-producer that was consistently enriched in CSD animals across most developmental stages,suggesting that this taxon may represent a persistent microbial feature associated with delivery mode. Inparallel, network analyses revealed that RC9 occupied a highly central position within microbial interac-tion networks, particularly in males, indicating that this taxon may play an important structural role withinthe CSD-associated microbial community. Previous studies portray RC9 as context-dependent withbeneficial effects in some settings,, yet detrimental in others. Notably, Rikenellaceae RC9 gutgroup has been reported to show negative correlation with butyrate levels and positive correlations withbutyrate derivatives, suggesting potential competition with butyrate producers for shared substrates.Additional evidence implicates RC9 in lipid metabolism., Higher Rikenellaceae RC9 gut groupabundance has been associated with increased body weight and fat mass, or increased subcutaneousadipose tissue. This study also acknowledges some limitations. First, the modest sample sizes and intragroup variabilityat certain timepoints and attritions (Sup. Tables 1 and 2) may have reduced our power to detect subtle sexdifferences. Second, the lack of estrous cycle monitoring. Fluctuations in ovarian hormones across thecycle can influence female physiological outcomes, and the absence of cycle monitoring by vaginal smearscould have introduced additional variability, thus masking sex-specific effects in females. Third, circulatingsex hormone levels were not measured, preventing us from directly correlating the observed sex differencesto hormonal status or from accurately defining the stage of puberty. Fourth, microbiota analyses werebased on taxa abundance and microbial network inference, which does not necessarily reflect theirmetabolic production, such as SCFA synthesis, and functional validation is needed to confirm theseassociations. Therefore, the functional implications of the identified taxa and network alterations remainpartially inferential. Additional functional validation would be needed to confirm these associations foreach stage of life, and experiments such as fecal microbiota transplantation (FMT) or cohousing studieswould be necessary to determine whether these microbial features causally contribute to the observedbarrier dysfunction and increased disease susceptibility later in life. Finally, all the experiments wereperformed in a single inbred mouse strain, limiting the generalizability of all the results. ### Page 16 Nevertheless, the consistent patterns across independent analyses and life stages underscore ourconclusions. Using new approaches to stratify data and assess sex effects, we reinforce the possibilitythat delivery mode programs sex-dependent trajectories of host–microbe interactions, with long-lastingconsequences for gut barrier fitness and colitis susceptibility more prominent in males. The integration ofclassical and compositional differential abundance analyses, LinDA linear modeling and NetMoss networkprofiling uncovered subtle interactions between sex and delivery mode that would be missed byabundance-based methods alone. These results are in line with the potential need to design early-lifemicrobiota-targeted strategies—potentially personalized by sex—to mitigate the long-term health effects ofC-section delivery. ## Disclosure of potential conflicts of interest The authors declare no competing interests. ## Acknowledgments We appreciate all the kind of technical assistance from @BRIDGe (Université Paris-Saclay, INRAE, AgroParisTech,GABI, 78350 Jouy-en-Josas, France). Animal experiments always have invaluable assistance from all the experiencedstaff of the INRAE Infectiology of Fishes and Rodents Facility (IERP-UE907, Jouy-en-Josas Research Center, from theNational Distributed Research Infrastructure for the Control of Animal and Zoonotic Emerging Infectious Diseasesthrough In Vivo Investigation France). The authors would also like to thank the INRAE MIGALE bioinformaticsfacility (MIGALE, INRAE, 2020). Migale bioinformatics Facility for providing the computational resources neededfor our bioinformatics analyses. ## Funding This work was supported by Biostime Institute for Nutrition and Care (BINC - Geneva), and the Institut National deRecherche pour l'Agriculture, l'Alimentation et l'Environnement (INRAE - France). # Author contributions CRediT: Bruna Maitan Santos: Formal analysis, Writing – original draft, Writing – review & editing; Jordi Estellé:Data curation, Formal analysis, Validation, Writing – review & editing; Yuliaxis Ramayo-Caldas: Data curation,Formal analysis, Writing – review & editing; Sead Chadi: Investigation, Methodology, Writing – review & editing;Monica Barone: Investigation, Writing – review & editing; Florian Chain: Investigation, Methodology, Writing –review & editing; Camille Kropp: Investigation, Writing – review & editing; Patrizia Bridigi: Conceptualization,Writing – review & editing; Philippe Langella: Conceptualization, Writing – review & editing; Rebeca Martín:Conceptualization, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration,Supervision, Validation, Writing – review & editing. ## ORCID Florian Chain Philippe Langella ## Data availability statement This work re-analyzed pre-clinical parameters originally reported as differing between cesarean-delivered (CSD) andvaginally delivered (VD) animals, now stratified by sex. Raw microbial 16S rRNA gene sequencing data available inthe NCBI Sequence Read Archive under BioProject PRJNA876103 (accessions SAMN30638467–SAMN30638852).Other data are available from the corresponding author upon reasonable request. ## References - 1. Koceva A, Herman R, Janez A, Rakusa M, Jensterle M. Sex- and gender-related differences in obesity: frompathophysiological mechanisms to clinical implications. Int J Mol Sci.;25(13):7342. doi:. ### Page 17 - 2. Liu DS, Wang X, Zhong X, Cao H, Zhang F. Sexual dimorphism in the gut microbiota and sexual dimorphismin chronic diseases: association or causation? J Steroid Biochem Mol Biol.;237:106451. doi:. - 3. Clocchiatti A, Cora E, Zhang Y, Dotto GP. Sexual dimorphism in cancer. Nat Rev Cancer.;16(5):330–339.doi:. - 4. Goodman WA, Erkkila IP, Pizarro TT. Sex matters: impact on pathogenesis, presentation and treatment ofinflammatory bowel disease. Nat Rev Gastroenterol Hepatol.;17(12):740–754. doi:. - 5. Ober C, Loisel DA, Gilad Y. Sex-specific genetic architecture of human disease. Nat Rev Genet.;9(12):911–922. doi:. - 6. Gao A, Su J, Liu R, Zhao S, Li W, Xu X, Shi J, Gu B, Zhang J, Wang X, et al. Sexual dimorphism in glucosemetabolism is shaped by androgen-driven gut microbiome. Nat Commun.;12(1):7080. doi:. - 7. Borgo F, Garbossa S, Riva A, Severgnini M, Luigiano C, Benetti A, Pontiroli AE, Morace G, Borghi E. Bodymass index and sex affect diverse microbial niches within the gut. Front Microbiol.;9:213. doi:. - 8. Falony G, Joossens M, Vieira-Silva S, Wang J, Darzi Y, Faust K, Kurilshikov A, Bonder MJ, Valles-Colomer M,Vandeputte D, et al. Population-level analysis of gut microbiome variation. Science.;352(6285):560–564.doi:. - 9. Sisk-Hackworth L, Kelley ST, Thackray VG. Sex, puberty, and the gut microbiome. Reproduction.;165(2):R61–R74. doi:. - 10. Ma X, Ding J, Ren H, Xin Q, Li Z, Han L, Liu D, Zhuo Z. Distinguishable influence of the delivery mode,feeding pattern, and infant sex on dynamic alterations in the intestinal microbiota in the first year of life. MicEc.;86(3):1799–1813. doi:. - 11. Neu J, Rushing J. Cesarean versus vaginal delivery: long-term infant outcomes and the hygiene hypothesis. ClinPerinatol.;38(2):321–331. doi:. - 12. Rebelo F, Da Rocha CMM, Cortes TR, Dutra CL, Kac G. High cesarean prevalence in a national population-based study in Brazil: the role of private practice. Acta Obstet Gynecol Scand.;89(7):903–908. doi:. - 13. Althabe F, Sosa C, Belizán JM, Gibbons L, Jacquerioz F, Bergel E. Cesarean section rates and maternal andneonatal mortality in low-, medium-, and high-income countries: an ecological study. Birth.;33(4):270–277. doi:. - 14. Betran AP, Torloni MR, Zhang J, Ye J, Mikolajczyk R, Deneux-Tharaux C, Oladapo OT, Souza JP, Tunçalp Ö,Vogel JP, et al. What is the optimal rate of caesarean section at population level? A systematic review of ecologicstudies. Reprod Health.;12:57. doi:. - 15. Betran AP, Ye J, Moller A, Souza JP, Zhang J. Trends and projections of caesarean section rates: global andregional estimates. BMJ Glob Health.;6(6):e005671. doi:. - 16. Eggesbo M, Eggesbø M, Botten G, Stigum H, Nafstad P, Magnus P. Is delivery by cesarean section a risk factorfor food allergy? J Allergy Clin Immunol.;112(2):420–426. doi:. - 17. Laubereau B, Filipiak-Pittroff B, von Berg A, Grübl A, Reinhardt D, Wichmann E, Koletzko S, GINI StudyGroup. Caesarean section and gastrointestinal symptoms, atopic dermatitis, and sensitisation during the firstyear of life. Arch Dis Child.;89(11):993–997. doi:. - 18. Bager P, Wohlfahrt J, Westergaard T. Caesarean delivery and risk of atopy and allergic disesase: meta-analyses.Clin Exp Allergy.;38(4):634–642. doi:. - 19. Huh SY, Rifas-Shiman SL, Zera CA, Edwards JWR, Oken E, Weiss ST, Gillman MW. Delivery by caesareansection and risk of obesity in preschool age children: a prospective cohort study. Arch Dis Child.;97(7):610–616. doi:. - 20. Pei Z, Heinrich J, Fuertes E, Flexeder C, Hoffmann B, Lehmann I, Schaaf B, von Berg A, Koletzko S. Cesareandelivery and risk of childhood obesity. J Pediatr.;164(5):1068–1073e2. doi:. - 21. Blustein J, Attina T, Liu M, Ryan AM, Cox LM, Blaser MJ, Trasande L. Association of caesarean delivery withchild adiposity from age 6 weeks to 15 years. Int J Obes (Lond).;37(7):900–906. doi:. - 22. Li Y, Tian Y, Zhu W, Gong J, Gu L, Zhang W, Guo Z. Cesarean delivery and risk of inflammatory boweldisease: a systematic review and meta-analysis. Scand J Gastroenterol.;49(7):834–844. doi:. - 23. Bruce A, Black M, Bhattacharya S. Mode of delivery and risk of inflammatory bowel disease in the offspring:systematic review and meta-analysis of observational studies. Inflamm Bowel Dis.;20(7):1217–1226. doi:. - 24. Bager P, Simonsen J, Nielsen NM, Frisch M. Cesarean section and offspring's risk of inflammatory boweldisease: a national cohort study. Inflamm Bowel Dis.;18(5):857–862. doi:. - 25. Mueller NT, Whyatt R, Hoepner L, Oberfield S, Dominguez-Bello MG, Widen EM, Hassoun A, Perera F,Rundle A. Prenatal exposure to antibiotics, cesarean section and risk of childhood obesity. Int J Obes (Lond).;39(4):665–670. doi:. ### Page 18 - 26. Maaser C, Langholz E, Gordon H, Burisch J, Ellul P, Hernández Ramirez V, Karakan T, Katsanos KH, KrustinsE, Levine A, et al. European Crohn's and colitis organisation topical review on environmental factors in IBD. JCrohns Colitis.:jjw223. doi:. - 27. Hviid A, Svanstrom H, Frisch M. Antibiotic use and inflammatory bowel diseases in childhood. Gut.;60(1):49–54. doi:. - 28. Zhong Z, Chen M, Dai S, Wang Y, Yao J, Shentu H, Huang J, Yu C, Zhang H, Ren W. Association of cesareansection with asthma in children/adolescents: a systematic review and meta-analysis based on cohort studies.BMC Pediatr.;23(1):571. doi:. - 29. Liang Y, Zhang J, Bai S, Du S, Yang X, Wang Z. Short-term and long-term effects of cesarean section on asthmaand wheezing: a cohort study and meta-analysis. Respir Med.;215:107300. doi:. - 30. Maynard CL, Elson CO, Hatton RD, Weaver CT. Reciprocal interactions of the intestinal microbiota andimmune system. Nature.;489(7415):231–241. doi:. - 31. Bokulich NA, Chung J, Battaglia T, Henderson N, Jay M, Li H, Lieber A, Wu F, Perez-Perez GI, Chen Y, et al.Antibiotics, birth mode, and diet shape microbiome maturation during early life. Sci Transl Med.;8(343):343–382. doi:. - 32. Yoon DY, Mansukhani NA, Stubbs VC, Helenowski IB, Woodruff TK, Kibbe MR. Sex bias exists in basicscience and translational surgical research. Surgery.;156(3):508–516. doi:. - 33. Melloni C, Berger JS, Wang TY, Gunes F, Stebbins A, Pieper KS, Dolor RJ, Douglas PS, Mark DB, Newby LK.Representation of women in randomized clinical trials of cardiovascular disease prevention. Circ CardiovascQual Outcomes.;3(2):135–142. doi:. - 34. Mazure CM, Jones DP. Twenty years and still counting: including women as participants and studying sex andgender in biomedical research. BMC Womens Health.;15:94. doi:. - 35. Solopov P, Colunga Biancatelli RML, Dimitropoulou C, Catravas JD. Sex-related differences in murine modelsof chemically induced pulmonary fibrosis. Int J Mol Sci.;22(11):5909. doi:. - 36. Orschell CM, Wu T, Patterson AM. Impact of age, sex, and genetic diversity in murine models of thehematopoietic acute radiation syndrome (H-ARS) and the delayed effects of acute radiation exposure(DEARE). Curr Stem Cell Rep.;8(3):139–149. doi:. - 37. Vichaya EG, Ford BG, Moltenkine JM, Taniguchi CM, Phillip West A, Dantzer R. Sex differences in thebehavioral and immune responses of mice to tumor growth and cancer therapy. Brain Behav Immun.;98:161–172. doi:. - 38. Bernardi JR, Pinheiro TV, Mueller NT, Sueno Goldani HA, Pereira Gutierrez MR, Bettiol H, Moura da SilvaAA, Antônio Barbieri M, Zubaran Goldani M. Cesarean delivery and metabolic risk factors in young adults: aBrazilian birth cohort study. AJCN.;102(2):295–301. doi:. - 39. Hansen S, Halldorsson TI, Olsen SF, Rytter D, Bech BH, Granström C, Henriksen TB, Chavarro JE. Birth bycesarean section in relation to adult offspring overweight and biomarkers of cardiometabolic risk. Int J Obes(Lond).;42(1):15–19. doi:. - 40. Mínguez-Alarcón L, Rifas-Shiman SL, Mitchell C, Sordillo J, Aris IM, Hivert M, Oken E, Chavarro JE. Cesareandelivery and metabolic health and inflammation biomarkers during mid-childhood and early adolescence.Pediatr Res.;91(3):672–680. doi:. - 41. Bridgman SL, Penfold S, Field CJ, Haqq AM, Mandhane PJ, Moraes TJ, Turvey SE, Simons E, Subbarao P,Kozyrskyj AL. Pre-labor and post-labor cesarean delivery and early childhood adiposity in the Canadian healthyinfant longitudinal development (CHILD) cohort study. Int J Obes (Lond).;48(5):717–724. doi:. - 42. Mueller NT, Zhang M, Rifas-Shiman SL, Oken E, Hivert M, Chavarro J. Mode of delivery, type of labor, andmeasures of adiposity from childhood to teenage: project viva. Int J Obes (Lond).;45(1):36–44. doi:. - 43. Zhang S, Zhou J, Yang M, Tao X, Huang K. Sex-specific association between elective cesarean section andgrowth trajectories in preschool children: a prospective birth cohort study. Front Public Health.;10:985851. doi:. - 44. Yoshida T, Matsumura K, Hatakeyama T, Inadera H, Kamijima M, Yamazaki S, Ohya Y, Kishi R, Yaegashi N,Hashimoto K, et al. Association between cesarean section and neurodevelopmental disorders in a Japanese birthcohort: the Japan environment and Children's study. BMC Pediatr.;23(1):306. doi:. - 45. Malmborg P, Bahmanyar S, Grahnquist L, Hildebrand H, Montgomery S. Cesarean section and the risk ofpediatric Crohn's disease. Inflamm Bowel Dis.;18(4):703–708. doi:. - 46. Andersen V, Erichsen R, Frøslev T, Sørensen HT, Ehrenstein V. Differential risk of ulcerative colitis andCrohn's disease among boys and girls after cesarean delivery. Inflamm Bowel Dis.;19(1):E8–E10. doi:. - 47. Barone M, Ramayo-Caldas Y, Estellé J, Tambosco K, Chadi S, Maillard F, Gallopin M, Planchais J, Chain F,Kropp C, et al. Gut barrier-microbiota imbalances in early life lead to higher sensitivity to inflammation in amurine model of C-section delivery. Microbiome.;11(1):140. doi:. ### Page 19 - 48. Barone M, Chain F, Sokol H, Brigidi P, Bermúdez-Humarán LG, Langella P, Martín R. A versatile new model ofchemically induced chronic colitis using an outbred murine strain. Front Microbiol.;9:565. doi:. - 49. Martin R, Martín R, Chamignon C, Mhedbi-Hajri N, Chain F, Derrien M, Escribano-Vázquez U, Garault P,Cotillard A, Pham HP, et al. The potential probiotic Lactobacillus rhamnosus CNCM I-3690 strain protects theintestinal barrier by stimulating both mucus production and cytoprotective response. Sci Rep.;9(1):5398.doi:. - 50. Perrier C CB. Gut permeability and food allergies. Clin Exp Allergy.;41(1):20–28. doi:. - 51. Wrzosek L, Miquel S, Noordine M, Bouet S, Chevalier-Curt MJ, Robert V, Philippe C, Bridonneau C, CherbuyC, Robbe-Masselot C, et al. Bacteroides thetaiotaomicron and Faecalibacterium prausnitzii influence theproduction of mucus glycans and the development of goblet cells in the colonic epithelium of a gnotobioticmodel rodent. BMC Biol.;11:61. doi:. - 52. Callahan BJ, McMurdie PJ, Rosen MJ, Han AW, Johnson AJA, Holmes SP. DADA2: high-resolution sampleinference from illumina amplicon data. Nat Methods.;13(7):581–583. doi:. - 53. McMurdie PJ, Holmes S. Phyloseq: an R package for reproducible interactive analysis and graphics ofmicrobiome census data. PLoS One.;8(4):e61217. doi:. - 54. Li H, Sheng D, Jin C, Zhao G, Zhang L. Identifying and ranking causal microbial biomarkers for colorectalcancer at different cancer subsites and stages: a mendelian randomization study. Front Oncol.;13:1224705.doi:. - 55. Scardoni G, Tosadori G, Faizan M, Spoto F, Fabbri F, Laudanna C. Biological network analysis with CentiScaPe:centralities and experimental dataset integration. F1000Res.;3:139. doi:. - 56. Zhou H, He K, Chen J, Zhang X. LinDA: linear models for differential abundance analysis of microbiomecompositional data. Genome Biol.;23(1):95. doi:. - 57. Xia Y, Liu C, Li R, Zheng M, Feng B, Gao J, Long X, Zuo X. Lactobacillus-derived indole-3-lactic acid amelioratescolitis in cesarean-born offspring via activation of aryl hydrocarbon receptor. iSci.;26(11):108279. doi:. - 58. Xiao L, Zhang F, Zhao F. Large-scale microbiome data integration enables robust biomarker identification. NatComput Sci.;2(5):307–316. doi:. - 59. Parker BJ, Wearsch PA, Veloo ACM, Rodriguez-Palacios A. The Genus Alistipes: Gut Bacteria WithEmerging Implications to Inflammation, Cancer, and Mental Health. Front Immunol.;11:906. doi:. - 60. Alcazar CG, Paes VM, Shao Y, Oesser C, Miltz A, Lawley TD, Brocklehurst P, Rodger A, Field N. Theassociation between early-life gut microbiota and childhood respiratory diseases: a systematic review. LancetMicrobe.;3(11):e867–e880. doi:. - 61. Choudhury R, Middelkoop A, Boekhorst J, Gerrits WJJ, Kemp B, Bolhuis JE, Kleerebezem M. Early life feedingaccelerates gut microbiome maturation and suppresses acute post-weaning stress in piglets. Environ Microbiol.;23(11):7201–7213. doi:. - 62. Yang Y, Song X, Wang G, Xia Y, Xiong Z, Ai L. Understanding Ligilactobacillus salivarius from ProbioticProperties to Omics Technology: A Review. Foods.;13(6):895. doi:. - 63. Bolotin A, de Wouters T, Schnupf P, Bouchier C, Loux V, Rhimi M, Jamet A, Dervyn R, Boudebbouze S,Blottière HM, et al. Genome sequence of "Candidatus Arthromitus" sp. Strain SFB-Mouse-NL, a commensalbacterium with a key role in postnatal maturation of gut immune functions. Genome Announc.;2(4),. - 64. Lécuyer E, Rakotobe S, Lengliné-Garnier H, Lebreton C, Picard M, Juste C, Fritzen R, Eberl G, McCoy KD,Macpherson AJ, et al. Segmented filamentous bacterium uses secondary and tertiary lymphoid tissues to induce gutIgA and specific T helper 17 cell responses. Immunity.;40(4):608–620. doi:. - 65. Petzoldt H. Harryflintia acetispora gen. nov., sp. nov., isolated from chicken caecum. Int J Syst Evol Microbiol .;66:4099–4104. doi: . - 66. Bell MR. Comparing postnatal development of gonadal hormones and associated social behaviors in rats, mice,and humans. Endocrinology.;159(7):2596–2613. doi:. - 67. Liu HY, Yuan P, Li S, Ogamune KJ, Shi X, Zhu C, Ennab W, Hu P, Ahmed AA, Zhang Y, et al. Lactobacillusjohnsonii alleviates experimental colitis by restoring intestinal barrier function and reducing NET-mediatedgut-liver inflammation. Commun Biol.;8(1):1222. doi:. - 68. Sullivan O, Sie C, Ng KM, Cotton S, Rosete C, Hamden JE, Singh AP, Lee K, Choudhary J, Kim J, et al. Early-life gut inflammation drives sex-dependent shifts in the microbiome-endocrine-brain axis. Brain Behav Immun.;125:117–139. doi:. - 69. Jašarević E, Morrison KE, Bale TL. Sex differences in the gut microbiome-brain axis across the lifespan. PhilosTrans R Soc Lond B Biol Sci.;371(1688):20150122. doi:. - 70. Gryaznova M, Dvoretskaya Y, Burakova I, Syromyatnikov M, Popov E, Kokina A, Mikhaylov E. Dynamics ofchanges in the gut microbiota of healthy mice fed with lactic acid bacteria and bifidobacteria. Microorganisms.;10(5):1020. doi:. ### Page 20 - 71. Xu X, Ocansey DKW, Pei B, Zhang Y, Wang N, Mao F. Resveratrol alleviates DSS-induced IBD in mice byregulating the intestinal microbiota-macrophage-arginine metabolism axis. Eur J Med Res.;28(1):319. doi:. - 72. Wang JL, Han X, Li J, Shi R, Liu L, Liao Y, Jiang H, Zhang Y, Hu J. Differential analysis of intestinal microbiotaand metabolites in mice with dextran sulfate sodium-induced colitis. World J Gastroenterol.;28(43):6109–6130. doi:. - 73. Li H, Wen J, Zhang X, Dai Z, Liu M, Lei R, Luo P. Large-scale genetic correlation studies explore the causalrelationship and potential mechanism between gut microbiota and COVID-19-associated risks. BMC Microbiol.;24(1):292. doi:. - 74. Zheng P, Zeng B, Liu M, Chen J, Pan J, Han Y, Cheng K, Zhou C, Wang H, Gui S, et al. The gut microbiomefrom patients with schizophrenia modulates the glutamate-glutamine-gaba cycle and schizophrenia-relevantbehaviors in mice. Sci Adv.;5(2):eaau8317. doi:. - 75. Luo C, Tian B, Zhou Y, Shang Q, Yu S, Dai M, Li Y, Chen H, Fröde TS. Deciphering the interplay amonginflammatory bowel disease, gut microbiota, and inflammatory biomarkers in the risk of colorectal cancer.Mediators Inflamm.;2025:4967641. doi:. - 76. Uddin MK, Mahmud MR, Hasan S, Peltoniemi O, Oliviero C. Dietary micro-fibrillated cellulose improvesgrowth, reduces diarrhea, modulates gut microbiota, and increases butyrate production in post-weaning piglets.Sci Rep.;13(1):6194. doi:. - 77. Zhou L, Xiao X, Zhang Q, Zheng J, Li M, Yu M, Wang X, Deng M, Zhai X. Improved glucose and lipidmetabolism in the early life of female offspring by maternal dietary genistein is associated with alterations in thegut microbiota. Front Endocrinol (Lausanne).;9:516. doi:. - 78. Sun L, Jia H, Li J, Yu M, Yang Y, Tian D, Zhang H, Zou Z. Cecal gut microbiota and metabolites mightcontribute to the severity of acute myocardial ischemia by impacting the intestinal permeability, oxidativestress, and energy metabolism. Front Microbiol.;10:1745. doi:. - 79. Cao F, Pan F, Gong X, Wang W, Xu Y. Causal relationship between gut microbiota with subcutaneous and visceraladipose tissue: a bidirectional two-sample mendelian randomization study. Front Microbiol.;14:1285982. doi:. - 80. Migale bioinformatics core facility dataverse.. Available from: https://entrepot.recherche.data.gouv.fr/dataverse/migale.