Abstract

   Milk kefir is a fermented dairy product composed of a complex
   consortium of microorganisms, including lactic acid bacteria (LAB),
   yeasts, and acetic acid bacteria. Despite growing interest in its
   health-promoting properties, the synergistic interactions among the
   diverse microbial constituents of kefir remain insufficiently
   characterized. This study investigated the probiotic potential of milk
   kefir through a multi-omics approach. Two strains, Kluyveromyces
   marxianus SLAM 005Y and Lentilactobacillus kefiri SLAM 023B, were
   isolated from Korean household milk kefir. Both strains exhibited key
   probiotic properties, including antibacterial activity, adherence to
   intestinal epithelial cells, and resistance in simulated
   gastrointestinal conditions. Supplementation with SLAM 005Y, SLAM 023B,
   or their synergistic combination (SYN) significantly enhanced lifespan
   and immune function in Caenorhabditis elegans. Transcriptomic analysis
   further revealed upregulation of genes involved in longevity and
   antimicrobial defense, suggesting the activation of conserved
   stress-response pathways. In a dextran sulfate sodium (DSS)-induced
   murine colitis model, SYN administration led to marked improvement in
   clinical outcomes. SYN also restored the expression of inflammatory and
   anti-inflammatory cytokines and tight junction proteins, indicating
   improved intestinal barrier function and immune regulation.
   Importantly, multi-omics analysis revealed that SYN altered both gut
   microbial composition and fecal metabolite profiles, supporting a
   mechanistic link between microbiota modulation and host health. Taken
   together, milk kefir strains SLAM 005Y and SLAM 023B, particularly in
   combination, exhibit strong probiotic properties. Our findings
   highlight the therapeutic potential of milk kefir-originated probiotics
   as a functional food for improving intestinal health.

   Keywords: Kluyveromyces marxianus, Lentilactobacillus kefiri, Synergy
   effect, Gut microbiome, Inflammatory bowel disease, Multi-omics
   analysis

Graphical abstract

   Image 1
   [49]Open in a new tab

Highlights

     * •
       Kefir strains showed probiotic traits including antibacterial
       activity, intestinal adherence, and gastrointestinal tolerance.
     * •
       Kefir derived strains activated stress-response pathways in
       Caenorhabditis elegans surrogate host model.
     * •
       Synbiotic kefir strains improved clinical outcomes in a dextran
       sulfate sodium (DSS)-induced murine colitis model.
     * •
       SYN with Kefir derived strains significantly altered both gut
       microbial composition and fecal metabolite profiles.

1. Introduction

   Inflammatory bowel disease (IBD), encompassing ulcerative colitis (UC)
   and Crohn's disease, comprises a group of chronic inflammatory
   disorders of the gastrointestinal tract ([50]Podolsky, 2002). Until
   present, the prevalence of IBD has surpassed 6.8 million cases,
   affecting individuals across diverse racial, geographic, and age groups
   ([51]Xia et al., 2023). The causes of IBD are still incompletely
   resolved, but many hypotheses exist, including genetic, environmental,
   and immunological factors ([52]Wirtz and Neurath, 2007). Approximately
   95 % of the human microbiota resides in the gut, where microbial genes
   and metabolites play essential roles in maintaining host physiological
   homeostasis and have been implicated in disease pathogenesis ([53]Lee
   et al., 2021). Numerous studies have identified consistent alterations
   in the composition and function of the gut microbiota in individuals
   with IBD compared to healthy controls, highlighting a strong
   association between microbial dysbiosis and IBD development
   ([54]Metwaly et al., 2020; [55]Miele et al., 2009; [56]Rigottier-Gois,
   2013; [57]Tamboli et al., 2004). Given that microbial imbalance is
   critical to the pathophysiology of IBD, strategies aimed at restoring
   homeostasis of gut microbiota are increasingly viewed as promising
   therapeutic approaches. Probiotics are defined as live microorganisms
   that, when administered in adequate amounts, confer health benefits on
   the host. Probiotics agents may contribute to the restoration of gut
   microbial balance, repair inflammation-induced epithelial barrier
   dysfunction, and normalize dysregulated intestinal motility ([58]Di
   Stefano et al., 2023; [59]Fischbach, 2018). Beyond their application in
   gastrointestinal disorders, probiotics have also demonstrated
   therapeutic potential in the context of systemic inflammatory diseases,
   including diabetes, obesity, and certain forms of cancer ([60]Bedada et
   al., 2020; [61]Bock et al., 2021; [62]Vallianou et al., 2020).

   Kefir, a traditional fermented milk product originating from the
   Caucasus and Tibetan regions, is produced through the fermentation of
   milk with kefir grains containing a complex microbial consortium
   ([63]Kazou et al., 2021). This probiotic-rich beverage typically
   comprises a diverse of bacteria and yeast ([64]Guangsen et al., 2021;
   [65]Leite et al., 2015; [66]Shavit, 2008; [67]TAŞ et al., 2012). Kefir
   has been reported to exhibit multiple health-promoting properties,
   including antioxidant, antibacterial, antifungal, anti-inflammatory,
   antidiabetic, and anti-atherosclerotic effects, thereby positioning it
   as a promising functional food ([68]Tzavaras et al., 2022). Recent
   experimental studies have demonstrated that kefir administration
   alleviates colonic inflammation in murine models of dextran sulfate
   sodium (DSS)-induced colitis ([69]da Silva et al., 2023). While these
   findings highlight the therapeutic potential of kefir in managing
   intestinal inflammation, the specific contributions and synergistic
   interactions among its constituent microbial species remain largely
   unexplored.

   In this study, we aimed to evaluate the probiotic potential of K.
   marxianus SLAM 005Y and L. kefiri SLAM 023B, isolated from traditional
   Korean milk kefir. We specifically investigated their individual and
   synergistic effects using both Caenorhabditis elegans (C. elegans) and
   a DSS-induced colitis mouse model. Through comprehensive in vivo and
   transcriptomic analyses, we sought to elucidate the mechanisms by which
   these strains modulate gut inflammation and host–microbe homeostasis.
   Our findings demonstrate that the combination of SLAM 005Y and SLAM
   023B exerts synergistic effects in alleviating gut inflammation and
   maintaining microbial balance, highlighting their therapeutic potential
   as kefir-derived probiotics.

2. Materials and methods

2.1. Prepare of kefir grain and fermented kefir milk

   Three traditional kefir samples were collected from local households in
   Korea between September and December, including two from Seoul and one
   from Suwon. Kefir grains were carefully transported to the laboratory
   in pasteurized milk and immediately transferred into sterilized milk
   (Seoul Milk Co., Ltd., Seoul, Korea) for fermentation. To restore
   microbial activity, 2–3 cycles of subculturing were performed prior to
   the experiment. The grains were then cultivated in sterilized milk at
   25 °C for 72 h, following previously described methods ([70]Kim et al.,
   2016, [71]2020).

2.2. Illumina sequencing for bacteria and yeast in kefir milk and kefir grain

   Following the fermentation process, kefir grains were separated via an
   autoclavable stainless steel sieve from kefir milk. Isolated kefir
   grain was washed with sterilized milk (Seoul Milk Co. Ltd.) to remove
   kefir milk remaining on its surface. Total DNA extraction from kefir
   milk (100 μl) and kefir grain (25 mg) were extracted using the DNeasy
   Blood & Tissue Kit (Qiagen, Hilden, Germany) according to the
   manufacturer's protocol. The V3-V4 region for bacteria and ITS region
   for yeast were amplified and sequenced using the MiSeq system platform
   (Macrogen Inc., Seoul, South Korea) based on the standard Illumina
   sequencing protocols. And then, fastq files obtained from Illumina
   sequencing data were analyzed using Mothur software (v. 1.44.3)
   according to MiSeq SOP protocol
   ([72]https://mothur.org/wiki/miseq_sop/) ([73]Kang et al., 2025) and
   MicrobiomeAnalyst.

2.3. Isolation and identification of lactic acid bacteria and yeast

   Bacteria and yeast are separately isolated from kefir milk and kefir
   grains. Kefir grains were placed in sterile stomacher bags containing
   saline solution and homogenized at high speed for 10 min using a
   stomacher. Subsequently, the suspension subjected to serial dilution
   and was plated onto Man, Rogosa, and Sharpe (MRS) agar (BD Difco,
   Sparks, MD, USA), Potato Dextrose Agar (PDA; BD Difco), and Yeast
   Extract Peptone Dextrose (YPD) agar (BD Difco). The plates were then
   incubated individually for 72 h at 37 °C and 48 h at 25 °C. To recover
   a broader spectrum of kefir-associated microorganisms, including
   coexisting lactic acid bacteria and yeasts, antifungal (amphotericin B)
   or antibacterial (chloramphenicol) agents were not supplemented in the
   media. This non-selective approach was employed to avoid artificial
   suppression of any microbial group and to better reflect the natural
   microbial composition of kefir.

2.4. Evaluating the fundamental characteristics of probiotics

2.4.1. Antimicrobial activity test

2.4.1.1. Pathogenic microbial strains and culture condition

   Staphylococcus aureus Newman and Listeria monocytogenes EGD-e were
   grown at 37 °C for 24 h in brain heart infusion (BHI) broth medium (BD
   Biosciences). Salmonella enterica serovar Typhimurium SL1344 was grown
   at 37 °C for 24 h in nutrient broth medium (BD Biosciences).
   Escherichia coli O157:H7 EDL933 was grown at 37 °C for 24 h in Luria–
   Bertani (LB) broth medium (BD Biosciences). Lactobacillus strain were
   grown at 37 °C for 48 h in MRS broth medium and Kluyveromyces strains
   were grown at 25 °C for 24 h. For the assay, pathogenic strains were
   prepared by inoculating overnight cultures into 0.4 % soft agar
   containing the respective broth medium at a final concentration of 5 %
   (v/v).

2.4.1.2. Antimicrobial activity screening

   Antimicrobial activity was evaluated using a three-layer agar diffusion
   method, modified from [74]Hossain et al. (2022). Briefly, 5 μL of
   overnight cultures of Kluyveromyces strains were spotted onto the
   bottom layer of YPD agar, and Lactobacillus strains were spotted onto
   MRS agar. These plates were incubated for 48 h at either 25 °C or 37 °C
   to allow probiotic growth. After incubation, a second layer consisting
   of 0.4 % soft agar seeded with the pathogenic strains was poured over
   the bottom layer and allowed to solidify. Subsequently, a third top
   layer of plain soft agar was added to stabilize diffusion and prevent
   desiccation. The plates were then incubated at 37 °C for an additional
   15 h. Zones of inhibition were measured to determine antimicrobial
   activity as previously described method ([75]Pundir et al., 2010).

2.4.1.3. Acid and bile tolerance test

   The tolerance of Kluyveromyces strains and Lactobacillus strains to
   acid and bile salt was tested following a previous study with some
   modifications ([76]Cho et al., 2024; [77]Oh et al., 2018).
   Kluyveromyces strains were cultured for 24 h in YPD broth medium, and
   Lactobacillus strains were cultured for 48 h in MRS medium. To create
   an acidic environment simulating the stomach, the pH of YPD and MRS
   broth was adjusted to pH 2.5 using 6N HCl. After autoclaving, porcine
   gastric mucosa pepsin (Sigma-Aldrich, St. Louis, MO, USA) was added to
   achieve a concentration of 1000 units/mL and filtered through a 0.45-μm
   pore size syringe filter. In the acid tolerance test of Kluyveromyces
   strains and Lactobacillus strains, cultures were inoculated into acidic
   solution and incubated for 0 h and 3 h at 37 °C. The bile solution was
   prepared by suspending oxgall powder (Acumedia, Lansing, MI, USA) in
   YPD and MRS broth medium to a final concentration of 0.5 % (w/v). All
   cultures of Kluyveromyces strains and Lactobacillus strains were
   inoculated into 0.5 % bile solution and incubated for 24 h at 37 °C.
   The survival rate of each strain was compared to the initial viable
   counts on plates at 0 h and after 3 h and 24 h of incubation. Survival
   rate (%) = CFU of 3 h or 24 h/CFU of 0 h) X 100(%)

2.4.1.4. Adhesion ability using Caco-2 intestinal epithelial cell

   Human colon epithelial colorectal adenocarcinoma Caco-2 cells were
   obtained from the Korean Cell Line Bank (Korea). Cell maintenance was
   performed following a previous study ([78]Yoon et al., 2022). After
   activating the culture medium, cells were maintained in minimal
   essential medium (MEM) (Gibco BRL, USA), 10 % heat-inactivated fetal
   bovine serum (FBS) (WelGENE Inc., Korea), and penicillin-streptomycin
   (15140-122, Gibco). Caco-2 cells were inoculated in a 100 mm plate and
   incubated at 37 °C under 5 % CO[2] conditions. The medium was exchanged
   every two days, the cells were incubated until a monolayer formed.
   Caco-2 cells were maintained until the experiment. To remove the
   culture medium and non-attached cells before experiments, the cells
   were washed 3 times with PBS. Cell adhesion assays were performed using
   modified methods ([79]Heo et al., 2018). Yeast and bacterial strains
   were treated at 10^9 CFU/mL MEM medium without penicillin-streptomycin
   was incubated for 2 h at 37 °C under 5 % CO[2] conditions. After 2 h of
   incubation, the monolayers were washed 3 times with PBS to remove
   non-attached yeast and bacteria. Trypsin-EDTA was used to lyse the
   attached cells. After lysis, serial dilutions of the yeast and
   bacterial mixture were separately plated on YPD and MRS agar plates.
   Each plate was incubated for 48 h at 25 °C and 72 h at 37 °C. The
   Adhesion ability was evaluated by counting CFU/ml.

2.4.1.5. Functional experimental using C. elegans surrogate animal model

   Kuyveromyces marxianus SLAM 005Y and Saccharomyces boulardii SLAM
   010SSB were grown at 25 °C for 24 h in YPD broth medium.
   Lentilactobacillus kefiri SLAM 023B and Lacticaseibacillus rhamnosus GG
   (LGG) were grown at 37 °C for 48 h in MRS broth. Escherichia coli OP50
   (OP50) and E. coli O157:H7 were grown in LB broth medium. S.
   Typhimurium was grown at 37 °C for 24 h in nutrient broth medium. To
   prepare live microbial lawns for C. elegans feeding, each microbial
   pellet was collected by centrifugation at 6000 rpm for 10 min, washed
   twice with sterile M9 buffer. After washing, the bacterial pellet was
   suspended in M9 buffer and seeded on nematode growth medium (NGM)
   plates.

2.4.1.6. C. elegans lifespan, killing assay, and fluorescence analysis

   C. elegans were maintained on NGM agar at 15 °C using standard
   techniques until experiment ([80]Brenner, 1974). Various of C. elegans
   strains were used for experiment, including
   CF512(fer-15(b26)II;fem-1(hc17)IV) and AY102(acEx102
   [vha-6p::pmk-1::GFP + rol-6(su1006)]), CF512, also known as temperature
   sensitive, was used for lifespan and killing assay and AY102 was
   employed for evaluating innate immunity, respectively. After
   Synchronization, L1 worms were grown on NGM plates, seeded with OP50 at
   a restrictive temperature (25 °C), to obtain sterile L4/young adult
   worms.

   To evaluate the microbial effect of kefir on C. elegans lifespan and
   immune response against pathogenic bacteria, we established methods
   based on a previous study ([81]Lee et al., 2024). For the lifespan
   assay, young adult L4 stage CF512 worms were transferred to 35 mm NGM
   agar plates and seeded with 100 μL of concentrated SLAM 005Y, SLAM
   023B, SYN and OP50 using a platinum wire. For the killing assay, young
   adult L4 stage CF512 worms were transferred to 60 mm NGM agar plates
   seeded with 200 μL of concentrated SLAM 005Y, SLAM 023B, SYN, and OP50
   using a platinum wire. After 48 h, worms were transferred to NGM plate
   seeded with pathogenic bacteria, S. Typhimurium, using a platinum wire.
   OP50 was used as normal control and positive control. The assessment of
   alive or dead C. elegans was conducted by gently touching them with a
   platinum wire, and the assay was conducted until all C. elegans died.
   The data from the C. elegans lifespan and killing assays were analyzed
   using the Kaplan-Meier method. The significance of differences between
   survival curves was assessed using the log-rank test with STATA6
   (STATA, TX, USA). Other data were statistically analyzed and graphed
   using Prism 10.2.2 (Graphpad Software, USA). Statistical significance
   was set at p value < 0.05 (∗), <0.01 (∗∗), <0.001(∗∗∗), <0.0001 (∗∗∗∗).
   All data are expressed as means ± the standard error of the mean (SEM).

   In addition, C. elegans fluorescence analysis was conducted to examine
   immune response. To examine the induction of pmk-1::GFP (the key gene
   for the p38 type MAPK pathway) in the nematode intestine, AY102
   transgenic worms were exposed to targeted strasin for 48 h. Then,
   individual worms (n = 10 per group) were transferred onto glass slides
   with 2 % agarose pads. C. elegans were anesthetized using 20 μL of
   30 mM NaN[3] (Sigma-Aldrich), and the GFP expression rate was observed
   using fluorescence microscopy (IX53; Olympus, Tokyo, Japan). ImageJ
   (MacBiophotonics ImageJ version 1.42I + loci tool) was used to analyze
   the fluorescent intensity.

2.4.1.7. C. elegans total RNA extraction and transcriptomic analysis

   Transcriptomic analysis was performed based on a previous study
   ([82]Kang et al., 2022). Young adult L4 worms were exposed to plates
   seeded with 200 μL of concentrated OP50, SLAM 005Y, SLAM 023B, and SYN
   for 48 h. Total RNA was extracted using TRIZOL reagent (Invitrogen,
   Carlsbad, CA, USA) and a mini bead beater (Biospec, Bartlesville, OK),
   followed by purification with the RNeasy Mini Kit (Qiagen) according to
   the manufacturer's instructions. Quality and quantity checks were
   conducted by measuring absorbance (ratio = A260/280) on a SpectraMax
   ABS Plus. RNA sequencing and transcriptomic analysis were performed by
   Macrogen (Macrogen Inc., South Korea). The ClueGO plug-in in Cytoscape
   software (version 3.9.1) was used to analyze Biological Process,
   Cellular Component, and Molecular functions of differentially expressed
   genes.

2.4.1.8. Fermentation of Intestinal Microbiota Model (FIMM)

   The Fermenter for Intestine Microbiome Model (FIMM) is an in vitro
   batch fermentation system designed to simulate the human colonic
   environment under anaerobic conditions. This model was adapted from the
   Simulator of the Human Intestinal Microbial Ecosystem (SHIME) and
   further developed following previously reported study ([83]Choi et al.,
   2025; [84]Kang et al., 2022, [85]2024). In this study, fresh fecal
   samples from healthy human donors were pooled and homogenized under
   anaerobic conditions to prepare a 10 % (w/v) fecal slurry. The
   fermentation was carried out in modified Gifu Anaerobic Medium (mGAM;
   Himedia, Mumbai, India), with the pH adjusted to 7.0 to mimic the human
   colon. To evaluate the microbiota-modulating potential of kefir-derived
   probiotics, 1 % (v/v) of SLAM 005Y, SLAM 023B, or SYN was inoculated
   into the culture. The cultures were then incubated in anaerobic
   conditions for 48 h at 37 °C with shaking and subsequently stored at
   −80 °C until metagenome analysis. Metagenomic analysis was performed by
   described methods as above.

2.4.2. DSS-induced colitis mice model

2.4.2.1. Animal administration

   All experimental protocols and animal procedures were approved by the
   Institutional Animal Care and Use Committee of Seoul National
   University (certificate SNU-230718-5). C57BL/6, 5-week-old male mice
   (n = 25) were obtained from SamTako Bio (Korea) and housed 5 mice per
   cage. All mice were acclimatized without any intervention for a week
   under standard laboratory conditions (23 °C ± 1 °C with 55 % ± 5 %
   humidity, 12 h light/dark cycle), provided with normal chow diet and
   sterile water. After an acclimation period for one week, mice were
   placed in one of the following groups: (1) no-DSS control group (CONT;
   mice administered PBS only for 3-wk; no DSS induction of colitis). (2)
   DSS-alone (DSS), (3) SLAM 005Y with DSS (SLAM 005Y), (4) SLAM 023B with
   DSS (SLAM 023B), or (5) synbiotics (SYN). These mice were administered,
   respectively, PBS (200 μL of suspension per mouse), SLAM 005Y
   (1.0 × 10^9 CFU/day), SLAM 023B (1.0 × 10^9 CFU/day) or SYN (5.0 × 10^8
   of SLAM 005Y + 5.0 × 10^8 of SLAM 023B CFU/day) every other day for
   3-wk. Suspensions were given by oral gavage (Fuchigami Kikai, Kyoto,
   Japan). To induce experimental acute colitis, mice in the 4 treatment
   groups (DSS-C, SLAM 005Y, SLAM 023B, SYN) were provided drinking water
   containing low-molecular-weight DSS (2.5 % DSS with a molecular weight
   of 36,000–50,000 Da, MP Biomedicals, Solon, OH) for an additional week
   after the 2-wk study period. During the 3-wk period, daily clinical
   assessment of DSS-induced colitis was performed, including measuring BW
   and food intake and noting rectal bleeding, stool conditions, and blood
   in stool using the disease activity index (DAI) scoring system
   ([86]Chassaing et al., 2014). Stool bleeding score was determined using
   the Hema-Screen Lab Pack for fecal occult blood (EKF Diagnostics, TX,
   USA). Stool bleeding score was determined using the Hema-Screen Lab
   Pack for fecal occult blood (EKF Diagnostics, TX, USA).

2.4.2.2. Morphological and biological analysis

   Spleen weight and colon length were measured after sacrifice. Spleen
   index was calculated by dividing the weight of the spleen (mg) by the
   weight of the mouse (g) at the time of dissection. Colon samples were
   washed with PBS and collected from the cecum to the rectum. After
   photographing the mouse colon tissue of each group with a ruler, the
   length of the colon was measured using ImageJ software version 1.8.0
   (W. Rasband, USA). Colon samples from each mouse were washed with
   sterilized PBS. Samples were fixed in 10 % formalin for dehydration and
   then embedded in paraffin for staining with hematoxylin and eosin (H&E)
   analyzed by imaging with a digital slide scanning system (Konfoong
   bioinformation tech co., LTD). Histological score was based on previous
   studies ([87]Kyoung et al., 2025; [88]Ye et al., 2023).

2.4.2.3. RNA isolation and RT-qPCR

   Total RNA from colon tissues were extraction by Trizol reagent and a
   RNeasy Mini kit (QIAGEN) according to the manufacturer's instructions.
   Quantity and purity were confirmed by measuring absorbance
   (ratio = A260/280 and A260/230) on a SpectraMax ABS Plus spectrometer
   ([89]Kwak et al., 2024). All RNA samples were diluted to 1 μg for
   RT-qPCR using iScript cDNA synthesis Kit (Bio-rad, CA, USA). After
   then, real-time PCR was performed using SsoAdvanced Universal SYBR
   Green Supermix with CFX96™ System (Bio-rad, CA, USA). Expression of
   each gene was normalized to the housekeeping gene, Gapdh (2^−ΔΔCt)
   ([90]Kwak et al., 2022). Primer sequences used in experiments are
   listed in [91]Supplementary Table S1. RNA sequencing data have been
   deposited in the NCBI SRA under BioProject accession number
   PRJNA1288232.

2.4.2.4. Fecal metagenomic and metabolomic analysis

   Fecal samples were collected and stored at −80 °C until metabolomic
   analysis. Metagenomic analysis was based on previous method ([92]Lee et
   al., 2022). gDNA was homogenized and extracted using the Dneasy
   PowerSoil Pro Kit (Qiagen, Hilden, Germany). Metagenomic analysis were
   performed by described methods as above. All raw sequence reads from
   the 16S rRNA gene analysis have been submitted to the NCBI SRA under
   BioProject accession number PRJNA1288167.

   For metabolomic analysis, GC-MS analysis was conducted using a Thermo
   Trace 1310 GC (Waltham, MA, USA) coupled with a Thermo ISQ LT Single
   Quadrupole Mass Spectrometer (Waltham, MA, USA). A DB-5MS column
   (Agilent, Santa Clara, CA, USA). The sample was injected at 300 °C with
   a split ratio of 1:60 and a helium split flow of 7.5 mL/min. Mass
   spectra were acquired at a scan range of 35–650 m/z with a scan rate of
   5 spectra per second. Ionization was performed in electron impact mode
   with the ion source temperature set at 270 °C. GC-MS spectra were
   deconvoluted and processed using the Automated Mass Spectral
   Deconvolution and Identification System (AMDIS) software ([93]Meyer et
   al., 2010) for peak detection and compound identification, and
   metabolites were identified by matching mass spectra and retention
   indices with the NIST Mass Spectral Search Tool (version 2.0,
   Gaithersburg, MD, USA) ([94]Lee et al., 2022). Further multivariate
   statistical analyses, including partial least squares-discriminant
   analysis (PLS-DA), were conducted using MetaboAnalyst 6.0 ([95]Pang et
   al., 2024) to evaluate group discrimination and determine key
   metabolites contributing to intergroup variation.

2.5. Statistics analysis

   All data were statistically analyzed and visualized using GraphPad
   Prism version 10.2.2 (GraphPad Software, San Diego, CA, USA). Multiple
   group comparisons were analyzed using one-way ANOVA or two-way ANOVA,
   as appropriate, followed by Tukey's post hoc test. Statistical
   significance was defined as p < 0.05 (∗), p < 0.01 (∗∗), p < 0.001
   (∗∗∗), and p < 0.0001 (∗∗∗∗). All data are presented as
   means ± standard error of the mean (SEM).

3. Result

3.1. Metagenomic analysis for myco- and microbiota in kefir milk and kefir
grain

   The relative abundance of fungal and bacterial communities in kefir
   milk and kefir grain samples from three individual samples is presented
   in [96]Fig. 1. Kefir-associated mycobiota and microbiota profiles were
   analyzed using next-generation sequencing (NGS). In the yeast community
   composition of kefir milk, KM1 was dominated by Kluyveromyces
   (55.65 %), followed by Issatchenkia (35.85 %) and Kazachstania
   (8.36 %). KM2 also exhibited high relative abundances of Kluyveromyces
   (42.89 %) and Issatchenkia (38.85 %), with lower levels of Kazachstania
   (2.53 %) and Pichia (0.79 %). Similarly, KM3 was composed predominantly
   of Kluyveromyces (42.88 %), followed by Issatchenkia (18.66 %),
   Kazachstania (0.37 %), and Pichia (1.19 %) ([97]Fig. 1A). In the
   corresponding kefir grains, distinct mycobiota patterns were observed.
   KG1 consisted almost exclusively of Kazachstania (99.28 %), with trace
   amounts of Kluyveromyces (0.39 %) and Issatchenkia (0.31 %). KG2
   contained Kazachstania (81.95 %), Kluyveromyces (16.49 %), and
   Issatchenkia (0.66 %). Notably, KG3 exhibited a shift in dominance,
   with Kluyveromyces (70.78 %) as the most abundant genus, followed by
   Kazachstania (23.32 %), Pichia (0.44 %), and Issatchenkia (0.37 %)
   ([98]Fig. 1C). Regarding bacterial composition, KM1 was largely
   dominated by Lactobacillus (89.23 %), along with Lentilactobacillus
   (9.57 %), and minor contributions from Acetobacter (0.38 %) and
   Brevibacillus (0.63 %). In contrast, KM2 had a higher proportion of
   Acetobacter (56.32 %), with Lactobacillus (35.81 %) and
   Lentilactobacillus (7.76 %). KM3 consisted of Lactobacillus (56.39 %),
   Acetobacter (41.55 %), and Lentilactobacillus (2.04 %) ([99]Fig. 1B).
   The bacterial profiles of the kefir grains also observed consistency in
   Lactobacillus dominance. KG1 contained Lactobacillus (99.13 %), with
   minor contributions from Lentilactobacillus (0.59 %) and Acetobacter
   (0.26 %). KG2 followed a similar pattern, with Lactobacillus (98.42 %),
   Lentilactobacillus (1.14 %), and Acetobacter (0.42 %). KG3 was composed
   of Lactobacillus (97.22 %), Lentilactobacillus (2.31 %), and
   Acetobacter (0.33 %). Overall, these results indicate that
   Kluyveromyces and Lactobacillus were the predominant fungal and
   bacterial genera, respectively, in both kefir milk and grains across
   all Korean household kefir samples analyzed ([100]Fig. 1D). Our data
   revealed a core kefir microbiota composed mainly of Lactobacillus and
   Acetobacter species, and a core mycobiota dominated by Kluyveromyces,
   consistent with previous studies on traditional kefir fermentation
   ([101]Dertli and Çon, 2017; [102]Kim et al., 2020; [103]Plessas et al.,
   2017). We also observed inter-sample variability in the relative
   abundances of these taxa, particularly in the proportions of
   Lactobacillus and Kluyveromyces, which varied depending on the
   household source.

Fig. 1.

   [104]Fig. 1
   [105]Open in a new tab

   Metagenomic and culturomic analysis of kefir milk and kefir grain. The
   microbial composition of Korean household kefir was analyzed at the
   genus level following 72 h of fermentation. Relative abundances of
   fungal and bacterial taxa were determined using Illumina sequencing of
   the ITS region for yeasts and the 16S rRNA gene for bacteria. (A)
   Genus-level composition of yeasts in three kefir milk (KM) samples. (B)
   Genus-level composition of bacteria in the same KM samples. (C)
   Genus-level composition of yeasts in the corresponding kefir grain (KG)
   samples. (D) Genus-level composition of bacteria in the same KG
   samples. Subsequently, kefir samples were serially diluted from KM and
   KG and cultured for microbial isolation. A total of 43 yeast strains
   (E) and 44 bacterial strains (F) were isolated and taxonomically
   identified using ITS and 16S rRNA gene sequencing, respectively.

3.2. Isolation of potential probiotic yeast and bacteria strains

   Following the identification of Kluyveromyces and Lactobacillus as the
   predominant genera in both kefir milk (KM) and kefir grains (KG) of
   Korean household kefir samples using metagenomic analysis, serial
   dilutions of the KM and KG samples were plated on MRS agar, YPD agar,
   and PDA to isolate individual microbial colonies. Subsequent
   identification of isolates was performed using internal transcribed
   spacer (ITS) sequencing for yeasts and 16S rRNA gene sequencing for
   bacteria. A total of 43 yeast isolates and 44 bacterial isolates were
   obtained. Among the yeast isolates, six distinct strains of
   Kluyveromyces (SLAM 001Y, SLAM 002Y, SLAM 003Y, SLAM 004Y, SLAM 005Y,
   and SLAM 006Y) were identified ([106]Fig. 1E). For bacterial isolates,
   several functionally relevant lactic acid bacteria (LAB) species were
   identified. These included Leuconostoc mesenteroides SLAM 003B,
   Lactobacillus brevis SLAM 006B, Lacticaseibacillus paracasei SLAM 008B
   and SLAM 012B, Lactococcus lactis SLAM 010B, and Lentilactobacillus
   kefiri SLAM 023B ([107]Fig. 1F). These findings suggest that
   Kluyveromyces and various LAB strains represent dominant taxa in the
   kefir samples analyzed in this study, as confirmed by both metagenomic
   and culture-based approaches.

3.3. Evaluating the fundamental characteristics of probiotics

   The antimicrobial activity was assessed by observing the formation of
   inhibition zones around the colonies of kefir-derived isolates
   ([108]Supplementary Table S2). Among the yeast isolates, K. marxianus
   strains SLAM 001Y through SLAM 005Y did not exhibit detectable
   antimicrobial activity against any of the tested pathogens under the
   experimental conditions. In contrast, several bacterial isolates
   demonstrated notable antibacterial effects. Most prominently,
   kefir-derived strains exhibited strong inhibitory activity against E.
   coli O157:H7, with strain Lentilactobacillus kefiri SLAM 023B and
   Lacticaseibacillus paracasei SLAM 008B also showing pronounced activity
   against S. Typhimurium.

   Probiotic functionality requires that candidate strains exhibit
   resistance to both gastric acidity and bile salts, which are critical
   for survival and adhesion within the human gastrointestinal tract
   ([109]Fig. 2A and B). To assess these attributes, we evaluated the acid
   and bile tolerance of selected kefir-derived strains under simulating
   gastrointestinal conditions. Initially, among the yeast isolates, K.
   marxianus SLAM 005Y demonstrated a high level of acid tolerance, with a
   survival rate of 87.04 %, comparable to that of the probiotic yeast
   reference strain Saccharomyces boulardii (SSB). Similarly, L. kefiri
   SLAM 023B showed a strong acid resistance profile, achieving a survival
   rate of 92.66 %, nearly identical to that of the well-established
   probiotic LGG strain. Moreover, in the bile tolerance assay, all K.
   marxianus strains revealed bile resistance comparable to SSB. Among the
   bacterial isolates, all strains except SLAM 012B maintained survival
   rates above 80 %. Notably, SLAM 010B and SLAM 023B demonstrated high
   bile tolerance, with survival rates of 93.65 % and 91.72 %,
   respectively, closely matching that of LGG. These findings highlight
   the robust acid and bile tolerance of kefir-derived microorganisms,
   particularly K. marxianus SLAM 005Y and L. kefiri SLAM 023B, suggesting
   their strong potential as viable probiotic candidates for
   gastrointestinal survival and functionality.

Fig. 2.

   [110]Fig. 2
   [111]Open in a new tab

   Evaluation of probiotic properties of yeast and bacterial strains
   isolated from Korean household kefir. The probiotic potential of
   kefir-derived microorganisms was assessed by measuring acid and bile
   tolerance, as well as adhesion capacity. (A) Survival rate after 3 h of
   exposure to acidic conditions (pH 2.5). (B) Survival rate after 24 h of
   exposure to 0.5 % (w/v) bile salts. (C) Adhesion capacity to
   Caco-2 cells, expressed as the percentage of adherent cells relative to
   baseline (0 h). Saccharomyces boulardii (SSB) and Lacticaseibacillus
   rhamnosus GG (LGG) were used as positive controls in all assays. Data
   are expressed as mean ± SEM from three independent experiments.
   Statistical significance was determined by one-way ANOVA followed by
   post hoc analysis; ∗p < 0.05, ∗∗p < 0.01, ∗∗∗p < 0.001, and
   ∗∗∗∗p < 0.0001.

   Effective probiotic function requires not only gastrointestinal
   survivability but also the ability to adhere to the intestinal mucosal
   surface, which facilitates colonization, persistence, and exclusion of
   pathogenic microbes ([112]Pennacchia et al., 2006; [113]Xie et al.,
   2012). To evaluate the adhesion capacity of kefir-derived strains, we
   utilized the Caco-2 cell line as an in vitro model of the human
   intestinal epithelium. In this study, two established reference
   probiotics, L. rhamnosus GG (LGG), and Saccharomyces boulardii (SSB),
   commonly recognized for its superior mucosal adherence and
   functionality, served as positive controls ([114]Gu et al., 2022). As
   shown in [115]Fig. 2C, all tested kefir-derived microorganisms
   exhibited strong mucin-binding capacity, with adhesion rates exceeding
   70 %. Of particular note, L. kefiri SLAM 023B displayed an adhesion
   rate above 80 %, surpassing that of several other tested strains,
   indicating a robust mucosal attachment potential. These results
   indicate that kefir isolates, especially L. kefiri SLAM 023B, exhibit
   strong adhesion to intestinal epithelial cells, supporting their
   potential for stable gut colonization.

3.4. Evaluation of functional properties using C. elegans

   To assess the functional effects of kefir-derived microorganisms on
   host physiology, we first evaluated their influence on the lifespan of
   C. elegans ([116]Supplementary Fig. S1). As shown in [117]Fig. 3A,
   several strains, including SLAM 003Y, SLAM 005Y, SLAM 006Y, and SLAM
   003B, significantly extended the mean lifespan of C. elegans compared
   to the E. coil OP50 normal feeding group. These results suggest that
   specific kefir isolates possess enhanced longevity effects in the
   nematode model. To further examine their impact on host immune defense,
   we performed C. elegans pathogen killing assays ([118]Supplementary
   Fig. S2). After pre-exposure to kefir-derived strains for 48 h, worms
   were challenged with S. Typhimurium. Among the tested strains, SLAM
   005Y showed a notable protective effect, reducing pathogen-induced
   mortality relative to the E. coli OP50 control group (p = 0.0242;
   [119]Fig. 3B). Similarly, strains SLAM 006B and SLAM 023B also
   conferred enhanced resistance to S. Typhimurium, suggesting an
   immunomodulatory role against Gram-negative bacterial infection. These
   functional assays that combined with previous results demonstrate acid
   resistance, bile tolerance, and strong epithelial adhesion, highlight
   K. marxianus SLAM 005Y and L. kefiri SLAM 023B as particularly
   promising probiotic candidates. Based on their superior performance in
   promoting lifespan extension and immune defense in C. elegans,
   subsequent experiments were focused on these two strains and their
   synergistic combination for further characterization.

Fig. 3.

   [120]Fig. 3
   [121]Open in a new tab

   Enhanced longevity and host immune response of Caenorhabditis elegans
   fed kefir-derived probiotic strains. (A) Lifespan analysis of C.
   elegans fed with kefir-derived Kluyveromyces marxianus SLAM 005Y and
   Lentilactobacillus kefiri SLAM 023B compared to the control diet of
   Escherichia coli OP50. (B) Survival of worms in the killing assay
   following infection with S. Typhimurium, assessing the protective
   effects of kefir-derived strains relative to OP50. (C) Fluorescence
   microscopy of AY102 transgenic worms carrying the pmk-1::GFP reporter,
   following 48 h exposure to kefir microorganisms. (D) Expression levels
   of immune-related genes following SYN treatment, with significantly
   upregulated genes defined as those showing a fold change >3 and
   p < 0.05 relative to OP50-fed controls. (E) Gene ontology enrichment
   analysis highlights the molecular response pathways activated by SYN
   supplementation. Lifespan data were analyzed using the Kaplan–Meier
   method with significance assessed by the log-rank test and immune and
   gene expression data were analyzed by one-way ANOVA. Statistical
   significance indicated as follows: p < 0.05 (∗), p < 0.01 (∗∗),
   p < 0.001 (∗∗∗), and p < 0.0001 (∗∗∗∗). Data represent means ± SEM from
   three independent experiments.

   Moreover, the PMK-1 signaling cascade, a central component of the p38
   MAPK pathway, is recognized as a key regulator of innate immune
   responses in C. elegans ([122]Yang et al., 2019). Previous studies have
   demonstrated that certain probiotic strains enhance host immunity by
   up-regulating pmk-1 expression, thereby protecting the nematode from
   pathogenic infections. In the present study, kefir-derived microbial
   strains, particularly those in the SYN group—a synergistic combination
   of the two selected strains—elicited significantly higher expression
   levels of pmk-1 compared to the OP50-fed control group ([123]Fig. 3C
   and D). These findings suggest that kefir-derived microorganisms
   stimulate the host's innate immune system, which contributes to
   enhanced resistance against microbial pathogens.

3.5. Visualized and analyzed gene network using C. elegans RNA-sequencing

   To further investigate the molecular mechanisms underlying these
   probiotic effects, transcriptomic analysis was performed, and
   Cytoscape-based pathway enrichment analysis was conducted to elucidate
   the host transcriptional pathways modulated by kefir-derived
   microorganisms, in comparison with C. elegans fed the OP50 control. The
   RNA-sequencing results indicated that K. marxianus SLAM 005Y induced
   gene expression associated with metabolic defense responses to
   Gram-positive bacteria, as well as genes involved in stress regulation,
   cellular protective mechanisms, neuropeptide signaling, and cell
   growth. In contrast, L. kefiri SLAM 023B was primarily linked to
   reactive oxygen species (ROS)-associated pathways, including oxidative
   stress response, uronic acid metabolism, and protein localization
   within cells ([124]Supplementary Table S3). Notably, the SYN treatment
   group combined SLAM 005Y and SLAM 023B, exhibited enrichment in
   pathways related to regulation of response to stimulus, defense against
   microbial organisms, and cellular stress regulation ([125]Fig. 3E). The
   results enabled identification of genes associated with the observed
   lifespan extension and immune enhancement in Caenorhabditis elegans
   following individual or combined administration of K. marxianus SLAM
   005Y and L. kefiri SLAM 023B isolated from kefir. These findings
   suggest that kefir-derived strains may confer beneficial effects on
   host physiology by modulating gene networks involved in immune
   responses and microbial defense or environmental stress adaptation.

3.6. Dynamics of microbial composition using FIMM assay

   To evaluate the modulatory effects of kefir-derived microorganisms on
   the intestinal microbial community, we conducted a Fermentation of
   Intestinal Microbiota Model (FIMM) experiment using a colon-mimicking
   culture system ([126]Fig. 4). Following 24 h of co-culture with fecal
   samples and selected kefir microbial strains, beta diversity analysis
   via UniFrac revealed distinct clustering patterns among treatment
   groups, suggesting microbial compositional shifts ([127]Fig. 4A–B).
   Alpha diversity analysis using the Shannon index demonstrated a
   significant increase in microbial diversity in the kefir-treated groups
   compared to the untreated control (CONT) (p < 0.0001) ([128]Fig. 4C).
   However, no significant differences were observed in species richness
   based on the Chao index (p = 0.1577) ([129]Fig. 4D). At the phylum
   level, administration of kefir strains resulted in a reduction of
   Proteobacteria, with L. kefiri SLAM 023B exhibiting the most pronounced
   decrease (∼4 %) ([130]Fig. 4E). At the family level, SLAM 023B notably
   increased the abundance of Bifidobacteriaceae, surpassing that of the
   positive control LGG, and also elevated Lactobacillaceae levels
   relative to the CONT group ([131]Fig. 4F). Genus-level analysis
   revealed a significant increase in Lactobacillus abundance and a marked
   reduction in the relative abundance of pathogenic Escherichia_Shigella
   in kefir-treated samples, particularly in the SLAM 023B group.
   Furthermore, the combination of SLAM 005Y and SLAM 023B exhibited the
   highest enrichment of Akkermansia, a genus associated with mucosal
   integrity and anti-inflammatory effects ([132]Fig. 4G). These results
   indicate that the administration of kefir-derived probiotics, either
   alone or in combination, promotes the growth of beneficial taxa such as
   Bifidobacteriaceae and Akkermansia, while suppressing pathogenic
   bacteria such as Escherichi_Shigella. These findings suggest that these
   strains may beneficially reshape the intestinal microbiota, promoting a
   more balanced and health-associated microbial composition. While the
   FIMM model does not capture host physiological responses such as
   epithelial interaction, immune modulation, or systemic metabolism, it
   provides a controlled and reproducible environment for assessing direct
   microbiota–microbe interactions. This makes it a suitable platform for
   preliminary screening of probiotic effects on microbial composition and
   metabolic output in a human-relevant colonic context.

Fig. 4.

   [133]Fig. 4
   [134]Open in a new tab

   Kefir-derived microbial supplementation modulates gut microbiota
   composition in an in vitro digesting system. To evaluate the influence
   of kefir derived microorganisms on gut microbial communities, fermentor
   for intestine microbiota model (FIMM) was established by co-culturing
   healthy human fecal samples with kefir-derived strains. Microbial
   composition was assessed through next-generation sequencing of the 16S
   rRNA gene. (A–B) Principal Coordinates Analysis (PCoA) plots based on
   weighted and unweighted UniFrac distances reveal group-level clustering
   patterns. Statistical significance was determined using PERMANOVA, with
   F-values, R^2, and p-values indicated. (C–D) Alpha diversity indices
   (Shannon and Chao1) were calculated to assess microbial richness and
   evenness. Data are presented as mean ± SEM. (E–G) Relative abundance of
   microbial taxa at the phylum, family, and genus levels illustrates
   compositional changes across treatment groups.

3.7. Effect of kefir-derived probiotic strains in DSS-induced colitis mouse

   Body weight was monitored as a general indicator of health status in
   mice ([135]Fig. 5A). After exposure to the DSS solution, the percent
   changes in body weight relative to baseline (pre-DSS exposure) were as
   follows: 103.42 % for the CONT group, 91.30 % for the DSS group,
   96.19 % for the SLAM 005Y group, 97.34 % for the SLAM 023B group, and
   95.45 % for the SYN group. Although the CONT group maintained the
   highest relative body weight compared to the DSS group, the difference
   did not reach statistical significance (p = 0.2051; 95 % CI: −8.98 to
   33.23; [136]Fig. 5B). These results suggest that kefir-derived strains,
   particularly SLAM 005Y and SLAM 023B, may attenuate DSS-induced weight
   loss, potentially reflecting a protective effect against
   colitis-associated physiological decline. In addition, Disease Activity
   Index (DAI) was used to quantitatively assess the progression and
   severity of colitis, incorporating weight loss, stool consistency, and
   the presence of rectal bleeding as evaluation parameters. Importantly,
   SLAM 005Y, SLAM 023B, and SYN treatment groups exhibited significantly
   reduced DAI scores compared to the DSS-only group (p < 0.0001,
   p = 0.0061, and p < 0.0001, respectively), while the CONT group showed
   no pathological signs and maintained the lowest score ([137]Fig. 5C).
   These results indicate that the kefir-derived strains SLAM 005Y and
   SLAM 023B, either individually or in combination, significantly
   attenuate DSS-induced weight loss and disease activity, suggesting
   their potential to alleviate colitis-associated clinical symptoms.

Fig. 5.

   [138]Fig. 5
   [139]Open in a new tab

   Effects of kefir-derived microorganisms on clinical and physiological
   gut health in a DSS-induced colitis mouse model. (A) Experimental
   schematic of DSS-induced acute colitis and kefir microbial
   intervention. Mice were orally administered 200 μL of treatment daily:
   the CONT group received autoclaved tap water with PBS; the DSS group
   received 2.5 % DSS in drinking water with PBS; SLAM 005Y group received
   Kluyveromyces marxianus SLAM 005Y with DSS; SLAM 023B group received
   Lactobacillus kefiri SLAM 023B with DSS; and SYN group received a
   combination of SLAM 005Y and SLAM 023B with DSS. (B) Disease Activity
   Index (DAI) scores were assessed based on weight loss, stool
   consistency, and rectal bleeding. (C) Spleen index was calculated to
   evaluate systemic inflammation.(D) Colon length was measured as a
   surrogate marker for colonic inflammation and tissue damage.
   Statistical significance was determined using one-way or two-way ANOVA.
   Asterisks indicate significance relative to the DSS group: p < 0.05
   (∗), p < 0.01 (∗∗), p < 0.001 (∗∗∗), and p < 0.0001 (∗∗∗∗).

   The spleen index, a well-established marker for evaluating systemic
   inflammatory responses, was used to assess the physiological impact of
   DSS-induced colitis. As expected, DSS exposure resulted in a notable
   increase in the spleen index relative to CONT group. While the SLAM
   005Y, SLAM 023B, and SYN combined groups also exhibited elevated spleen
   indices compared to CONT, treatment with kefir-derived microorganisms
   led to a significant reduction in the spleen index relative to the DSS
   group, particularly in the SLAM 023B (p = 0.0443) and SYN (p = 0.0064)
   groups ([140]Fig. 5D). These results suggest that kefir microbial
   treatments can attenuate systemic inflammation induced by DSS.
   Furthermore, colon length was measured as a direct indicator of colonic
   inflammation and mucosal damage, with shortening reflecting the
   severity of tissue injury. As expected, DSS exposure led to a marked
   reduction in colon length across all treated groups. However,
   probiotics treated groups showed a protective effect, as evidenced by
   longer colon lengths compared to the DSS group. This effect was most
   pronounced in the SYN group, which exhibited a statistically
   significant improvement (p = 0.0013) ([141]Fig. 5E). Taken together,
   these findings suggest that the co-administration of kefir-derived
   strains mitigates DSS-induced colonic tissue damage and systemic
   inflammation, supporting their potential role in alleviating
   colitis-related inflammatory responses.

3.8. Morphological and biological impacts of kefir-derived probiotics

   Hematoxylin and eosin (H&E) staining was employed to evaluate
   histopathological changes in colonic tissue following DSS-induced
   colitis. This widely utilized method enables detailed visualization of
   tissue architecture and inflammatory cell infiltration, allowing for
   assessment of mucosal integrity and inflammatory severity. As shown in
   [142]Fig. 6A, DSS treatment resulted in pronounced epithelial
   disruption, goblet cell depletion, and increased inflammatory
   infiltrates. In contrast, the kefir-treated groups including SLAM 005Y,
   SLAM 023B, and SYN, exhibited markedly improved tissue morphology, with
   reduced inflammatory cell infiltration and preservation of epithelial
   structure, similar to the non-treated CONT group. Quantitative
   histological scoring confirmed these observations ([143]Fig. 6B), with
   the DSS group displaying significantly elevated histopathological
   scores compared to all kefir strains-treated groups (p < 0.0001; 95 %
   CI: 2.042 to 3.958). These findings indicate that administration of
   SLAM 005Y and SLAM 023B, either individually or in combination, reduced
   histopathological signs of DSS-induced colonic damage, as evidenced by
   improved colon architecture and lower histological scores. Taken
   together, these results suggest that kefir-derived probiotics may exert
   beneficial effects in alleviating colonic tissue injury associated with
   DSS-induced colitis.

Fig. 6.

   [144]Fig. 6
   [145]Open in a new tab

   Histological and intestinal barrier fuunction analysis of colonic
   tissue following kefir-derived microbial supplementation in a
   DSS-induced colitis mouse model. (A) Representative images of
   hematoxylin and eosin (H&E)-stained colon sections at 40 × and
   200 × magnification showing mucosal architecture and inflammatory
   features. (B) Histopathological scoring was performed based on
   inflammatory cell infiltration, mucosal disruption, and glandular
   integrity. Quantitative real-time PCR (RT-qPCR) was used to evaluate
   gene expression of pro-inflammatory cytokines: (C) TNF-α, (D) IL-6, and
   (E) IL-1β; the anti-inflammatory cytokine (F) IL-10; tight junction
   proteins: (G) claudin-1, (H) occludin, and (I) ZO-1; and (J) mucin-2
   (MUC-2). Gene expression levels were normalized to GAPDH as
   housekeeping gene and calculated using the 2^-ΔΔCt method. Statistical
   analyses were performed using one-way or two-way ANOVA, with
   comparisons made to the DSS group. Significance levels are indicated as
   follows: p < 0.05 (∗), p < 0.01 (∗∗), p < 0.001 (∗∗∗), and p < 0.0001
   (∗∗∗∗).

3.9. Transcriptional regulation of inflammatory cytokine and tight junction
genes

   DSS is a well-established agent for inducing acute inflammation in the
   large intestine and serves as a reliable model for studying colitis. To
   elucidate the effects of kefir-derived microorganisms on inflammation
   and intestinal barrier function, we assessed the expression levels of
   pro- and anti-inflammatory cytokines, as well as tight junction-related
   genes, in colonic tissues using RT-qPCR.

   As expected, pro-inflammatory cytokines such as tumor necrosis
   factor-alpha (TNF-α), interleukin-6 (IL-6), and interleukin-1β (IL-1β),
   were markedly elevated in the DSS group compared to the CONT group,
   reflecting acute intestinal inflammation. Notably, mice treated with
   SLAM 005Y and SLAM 023B showed a modest attenuation of these
   inflammatory markers, while the SYN group exhibited a more substantial
   reduction in TNF-α, IL-6, and IL-1β expression relative to the DSS
   group ([146]Fig. 6C–E). In contrast, expression of the
   anti-inflammatory cytokine IL-10, which modulates immune responses and
   plays a pivotal role in suppressing inflammation, was significantly
   downregulated in the DSS group compared to the CONT group. However,
   IL-10 levels were restored in all kefir-treated groups, with the SYN
   group showing a statistically significant increase (p = 0.0277)
   compared to DSS alone ([147]Fig. 6F). These results suggest that
   kefir-derived strains, particularly the SYN combination, downregulate
   pro-inflammatory cytokine genes and upregulate anti-inflammatory genes,
   thereby contributing to the attenuation of inflammation in DSS-induced
   colonic tissues.

   To further investigate the protective effects of kefir-derived
   probiotics on intestinal barrier function, we analyzed the expression
   levels of key tight junction (TJ) proteins that are critical for
   maintaining colonic epithelial integrity. TJ proteins enhance
   intercellular adhesion and play a vital role in preventing the
   translocation of harmful luminal substances. Compared to the DSS group,
   expression of TJ-related genes was elevated in all kefir-treated
   groups. Notably, claudin-1 and ZO-1 expression was significantly
   upregulated in the SYN group (p < 0.0001), suggesting enhanced
   epithelial barrier function ([148]Fig. 6G–I). In addition, MUC-2, the
   principal component of intestinal mucus that contributes to mucosal
   protection, was significantly increased in both the SLAM 005Y and SLAM
   023B groups compared to DSS controls (p < 0.0001 and p = 0.0141,
   respectively) ([149]Fig. 6J). These findings indicate that
   kefir-derived strains promote the expression of tight junction-related
   genes, as well as mucin-producing gene, thereby contributing to the
   restoratif epithelial barrier integrity and enhancement of mucosal
   defense in DSS-induced colitis.

3.10. Multi-omics analysis with metagenomic and metabolomic approaches

   To evaluate the impact of kefir-derived microorganisms on gut
   microbiota composition in DSS-induced colitis, we performed metagenomic
   analyses using fecal samples collected on day 7 of DSS administration
   ([150]Fig. 7). Beta diversity, as visualized by PCoA based on both
   weighted and unweighted UniFrac distances, revealed distinct clustering
   patterns across treatment groups, indicating divergent microbial
   communities ([151]Fig. 7A–B). Alpha diversity was assessed using both
   the Shannon and Chao1 indices. The SYN group displayed the highest
   Shannon diversity, while the SLAM 023B group showed the lowest
   ([152]Fig. 7C); interestingly, the Chao1 index presented the inverse
   trend (p = 0.015 and p = 0.017, respectively) ([153]Fig. 7D). At the
   phylum level, DSS exposure was associated with a non-significant
   reduction in Firmicutes across all DSS-treated groups (p = 0.6581 for
   DSS, p = 0.9703 for SLAM 005Y, p = 0.8520 for SLAM 023B, and p = 0.9026
   for SYN) ([154]Fig. 7E). At the family level, the abundance of
   Muribaculaceae (formerly S24-7) significantly declined across all
   DSS-treated groups, with the greatest reduction observed in the SYN
   group (p = 0.0004) ([155]Fig. 7F). Bifidobacteriaceae levels increased
   in the SLAM 005Y and SLAM 023B groups compared to the DSS group
   (p = 0.9957 and p = 0.9967, respectively), while Lachnospiraceae and
   Ruminococcaceae were significantly enriched in the SLAM 005Y group
   (p < 0.0001 and p = 0.0044, respectively). At the genus level,
   Aerococcus and Corynebacterium were elevated in the DSS group relative
   to CONT (p = 0.9368 and p = 0.6846), while Bacteroides decreased in all
   DSS-treated groups (p = 0.1005 for DSS, p = 0.9537 for SLAM 005Y,
   p = 0.1439 for SLAM 023B, and p = 0.1797 for SYN). Notably,
   Oscillospira, Ruminococcus, and Bifidobacterium were elevated in the
   SLAM 005Y group (p = 0.6548, p = 0.9965, and p = 0.7527, respectively),
   whereas Corynebacterium was significantly reduced in both the SLAM 023B
   and SYN groups (p = 0.0063 and p = 0.0145, respectively) ([156]Fig.
   7G).

Fig. 7.

   [157]Fig. 7
   [158]Open in a new tab

   Multi-omics anaylsis with of fecal microbiota and fecal metabolites in
   DSS-induced colitis mice treated with kefir-derived microbial strains.
   (A–B) Principal Coordinate Analysis (PCoA) plots based on weighted and
   unweighted UniFrac distances revealed distinct clustering patterns,
   indicating significant differences in beta diversity between treatment
   groups. (C–D) Alpha diversity was assessed using Shannon and Chao1
   indices to evaluate microbial richness and evenness; data are presented
   as mean ± SEM. (E–G) Taxonomic composition of gut microbiota was
   analyzed at the phylum (E), family (F), and genus (G) levels based on
   16S rRNA gene sequencing. (H–J) Partial least squares discriminant
   analysis (PLS-DA) of fecal metabolomic profiles was performed on
   GC-MS–derived fecal metabolomic data using MetaboAnalyst 6.0 and
   revealed clear separation among treatment groups: (H) overall
   comparisons across all groups, (I) comparisons between DSS-treated
   groups, and (J) DSS vs. SYN groups, highlighting metabolic shifts
   following kefir microbial intervention. (K) Bar plot showing log[2]
   fold changes of significantly altered metabolites in the SYN group
   relative to the DSS group.

   To further explore metabolic shifts in gut environments, fecal
   metabolomic profiling was conducted using GC-MS. PLS-DA analysis
   demonstrated clear separation between groups, with distinct clustering
   of the CONT and DSS groups, while SLAM 005Y, SLAM 023B, and SYN formed
   a separate cluster ([159]Fig. 7H). When restricted to DSS-treated
   samples, SLAM 005Y and SLAM 023B clustered closely ([160]Fig. 7I),
   whereas SYN formed a distinct cluster from the DSS group ([161]Fig.
   7J). Comparison of the SYN group with DSS controls revealed
   upregulation of 19 metabolites, including L-valine, L-isoleucine,
   5-aminovaleric acid, aspartic acid, L-threonine, succinic acid, batyl
   alcohol, L-5-oxoproline, glycerol, coprostanol, and lactic acid
   ([162]Fig. 7K). These results suggest that kefir-derived microorganisms
   help restore the gut microbial balance disrupted by DSS-induced
   colitis, as evidenced by increased microbial diversity and enrichment
   of beneficial taxa such as Bifidobacterium, which are known to support
   intestinal health. Furthermore, metabolomic profiling revealed an
   increased abundance of metabolites, particularly branched-chain amino
   acids, in the SYN group, suggesting that kefir-derived microbes not
   only reshape the gut microbial community but also contribute to the
   restoration of metabolic homeostasis. These results collectively
   highlight the role of kefir-derived strains in regulating both
   microbial composition and metabolic profiles, supporting their
   potential as therapeutic candidates for gut dysbiosis and intestinal
   inflammation.

4. Discussion

   This study aimed to evaluate the probiotic potential of kefir-derived
   microorganisms, specifically Kluyveromyces marxianus SLAM 005Y and
   Lactobacillus kefiri SLAM 023B and to explore the synergistic effects
   of their combination (SYN). kefir is a fermented dairy product
   traditionally produced through the symbiotic activity of various lactic
   acid bacteria and yeasts ([163]Garrote et al., 2001). Among its
   dominant microbial residents are Lactobacillus kefiranofaciens, L.
   kefiri, K. marxianus, and Saccharomyces cerevisiae, which collectively
   contribute to its probiotic functionality and unique sensory properties
   ([164]Kim et al., 2020; [165]Youn et al., 2022). K. marxianus, which is
   known to dominate approximately 95 % of the yeast population in kefir,
   has been reported to possess probiotic potential due to its high
   survivability in the intestinal environment and strong adhesion
   capability to intestinal epithelial cells ([166]Maccaferri et al.,
   2012; [167]Shiby and Mishra, 2013; [168]Youn et al., 2022). K.
   marxianus has been reported to modulate the intestinal microbiota
   composition following oral administration in BALB/c mice ([169]Youn et
   al., 2023). Moreover, in a murine model of high-fat diet–induced
   obesity, K. marxianus administration significantly reshaped gut
   microbial communities and mitigated lipid metabolic dysregulation,
   suggesting its potential role in restoring gut–metabolic axis
   homeostasis ([170]Tang et al., 2025). L. kefiri, one of the most
   prominent lactobacilli isolated from kefir, has been shown to protect
   intestinal epithelial cells against Salmonella infection through its
   surface-layer proteins ([171]Carasi et al., 2014; [172]Golowczyc et
   al., 2007). In vivo studies further demonstrated that oral
   administration of L. kefiri stimulated gut mucosal immunity and
   modulated the composition of the intestinal microbiota in mice
   ([173]Carasi et al., 2015). Furthermore, the exopolysaccharide produced
   by L. kefiri has been reported to exert anticancer effects by inducing
   apoptosis in HT-29 human colorectal cancer cells ([174]Rajoka et al.,
   2019). Clinically, administration of L. kefiri was also associated with
   improvements in diarrhea symptoms among cancer patients ([175]Ghidini
   et al., 2021).

   The present investigation examined also investigated the synergistic
   effects between kefir-derived yeasts and lactic acid bacteria. In the
   kefir microbial consortium, the metabolic interplay between yeasts and
   lactic acid bacteria exemplifies mutualistic cooperation, wherein
   yeast-derived amino acids and peptides foster an environment conducive
   to LAB growth and activity ([176]Ponomarova et al., 2017; [177]Simova
   et al., 2006). Previous studies have reported that co-culturing
   kefir-derived L. kefiranofaciens and K. marxianus enhances
   extracellular polysaccharide (EPS) production, thereby improving
   microbial survivability in simulated digestion models
   ([178]González-Orozco et al., 2023). Similarly, co-cultivation of L.
   helveticus with K. marxianus has been shown to exhibit greater
   antioxidant activity compared to individual treatments ([179]Wang et
   al., 2023). In line with previous observations, the present study
   revealed that both K. marxianus SLAM 005Y and L. kefiri SLAM 023B, as
   well as their synergistic combination, significantly enhanced host
   immune responses in the C. elegans model and alleviated DSS-induced
   colitis by restoring gut microbial and metabolic homeostasis.

   In this study, the surrogate model organism C. elegans was employed.
   Due to its short lifespan, genetic tractability, and ease of
   cultivation, C. elegans serves as a valuable system for studying
   host–microbe interactions, aging, and innate immunity ([180]Amrit et
   al., 2014; [181]Anderson et al., 2003; [182]Choi et al., 2023; [183]Yoo
   et al., 2022). When compared to the normal feed E. coli OP50, exposure
   to SLAM 005Y, SLAM 006Y, and SLAM 003B significantly prolonged the
   lifespan of C. elegans, indicating that these kefir-derived strains may
   promote longevity and enhance host resilience. In addition to lifespan
   extension, the ability of kefir strains to modulate early immune
   responses was also considered. C. elegans mounts rapid innate immune
   defenses to combat microbial infections, and intestinal colonization by
   pathogens can negatively impact host survival ([184]Martineau et al.,
   2021; [185]Scott et al., 2019). Prior research has demonstrated that
   the probiotic yeast Saccharomyces boulardii can colonize the C. elegans
   gut and exert antibacterial activity against enteric pathogens
   ([186]Chelliah et al., 2021). However, studies specifically
   investigating the probiotic efficacy of other yeast strains in C.
   elegans remain limited, underscoring the need for further functional
   validation ([187]Kunyeit et al., 2021). Similarly, various
   Lactobacillus species have been shown to exert protective effects in
   vivo by enhancing host immune responses and suppressing pathogenic
   colonization ([188]Lee et al., 2011; [189]Zhou et al., 2014).
   Consistent with these findings, our results suggest that both SLAM 005Y
   and SLAM 023B confer health-promoting effects in C. elegans, supporting
   their candidacy as novel probiotic strains. These data further
   highlight the utility of C. elegans as a preliminary in vivo platform
   for screening probiotic functionality prior to mammalian validation.

   Building upon our findings in C. elegans, which demonstrated the
   anti-aging and immunomodulatory properties of kefir-derived
   microorganisms, we next sought to evaluate their protective effects in
   a mouse animal model of intestinal inflammation. Specifically, we
   examined whether K. marxianus SLAM 005Y, L. kefiri SLAM 023B, and their
   combination (SYN) could ameliorate colonic damage in a DSS-induced
   colitis mouse model. This chemically induced model is widely accepted
   as a preclinical platform for investigating inflammatory bowel disease
   (IBD), as it recapitulates many clinical and histopathological features
   of human ulcerative colitis, including weight loss, diarrhea, rectal
   bleeding, mucosal barrier disruption, and colonic inflammation
   ([190]Goyal et al., 2014; [191]Sann et al., 2013). While kefir
   microorganism treatment attenuated DSS-induced weight loss, no
   statistically significant differences were observed compared to the DSS
   group. This could be due to the relatively short duration of DSS
   exposure and the treatment period with kefir strains, which may not
   have been sufficient to observe significant physiological recovery. A
   longer pre-treatment or extended recovery phase may enhance the
   manifestation of kefir's protective effects against colitis-induced
   symptoms. Colon length, a commonly used marker of inflammation and
   mucosal injury, was significantly preserved in the SYN group compared
   to DSS-only controls, indicating a protective effect against
   colitis-associated tissue damage ([192]Chen et al., 2017; [193]Duary et
   al., 2012). Histological analysis further supported these findings.
   Consistent with previous reports of DSS-induced IBD, the colons of
   untreated DSS mice exhibited epithelial distortion, crypt loss, and
   extensive infiltration of inflammatory cells ([194]Xu et al., 2020). In
   contrast, kefir-treated groups showed substantially less
   histopathological damage and inflammatory cell infiltration.
   Pro-inflammatory cytokines including TNF-α, IL-1β, and IL-6, were
   significantly elevated in the DSS group, in line with prior studies
   ([195]Lacruz-Guzman et al., 2013). Treatment with SLAM 005Y, SLAM 023B,
   and SYN led to marked reductions in these cytokines. Moreover,
   expression of the anti-inflammatory cytokine IL-10, which was
   suppressed by DSS exposure, was restored in all kefir-treated groups,
   with the SYN group showing the greatest increase ([196]Barbara et al.,
   2000). These findings suggest that kefir-derived strains modulate
   colonic inflammation by downregulating proinflammatory cytokines, while
   concurrently enhancing the expression of anti-inflammatory mediators.
   This modulation may contribute to the restoration of immune homeostasis
   and epithelial inflammation in the DSS-induced colitis model.
   Collectively, these results support the therapeutic potential of SLAM
   005Y, SLAM 023B, and their combination in attenuating intestinal
   inflammation and enhancing epithelial repair in inflammatory bowel
   disease.

   Although the precise etiology of IBD remains incompletely understood,
   disruption of the intestinal epithelial barrier is recognized as a
   hallmark feature of disease pathology ([197]Pope et al., 2014). TJ
   proteins play a crucial role in maintaining intestinal barrier
   integrity and regulating paracellular permeability ([198]Lee, 2015). In
   DSS-induced colitis models, expression of these proteins is typically
   downregulated, leading to increased intestinal permeability and
   inflammation ([199]Liu et al., 2023). In the present study, kefir
   microbial treatment significantly restored TJ protein expression levels
   suppressed by DSS exposure. Notably, Claudin-1 and ZO-1 were most
   strongly upregulated in the SYN-treated group compared to the DSS
   group, consistent with previous findings highlighting their loss in
   colitis. However, it should be acknowledged that the expression
   patterns and roles of specific TJ proteins may vary depending on the
   type and severity of intestinal disease ([200]Sawada, 2013). In
   addition to tight junction proteins, the mucus layer provides a
   critical line of defense against pathogenic infiltration. Among mucin
   glycoproteins, MUC2, a secretory mucin produced by goblet cells, is the
   principal component of this layer and is essential for mucosal
   protection in both humans and mice ([201]Van der Sluis et al., 2006).
   MUC2 expression is typically reduced in DSS-induced colitis,
   contributing to barrier dysfunction and increased susceptibility to
   infection ([202]Bankole et al., 2021). Our findings demonstrated that
   kefir microbial administration significantly enhanced colonic MUC2 gene
   expression, as determined by qPCR analysis. This elevation in MUC2
   levels suggests enhanced goblet cell activity and mucin production,
   which are critical for forming a protective mucus barrier. This
   increase was especially pronounced in the SLAM 005Y group and was
   associated with increased goblet cell numbers, consistent with previous
   studies showing similar outcomes following probiotic administration
   ([203]Morampudi et al., 2016). Collectively, these results suggest that
   SLAM 005Y, SLAM 023B, and their synergistic combination promote
   epithelial barrier restoration through upregulation of both tight
   junction and mucin genes. These effects likely contribute to improved
   intestinal structural integrity and a stabilized mucosal environment in
   the context of colitis. Such findings support the therapeutic potential
   of kefir-derived probiotics in managing IBD by enhancing mucosal
   defense mechanisms. Transcriptomic analysis in C. elegans revealed that
   kefir-derived strains modulate the expression of genes associated with
   bacterial defense, external stimulus response, and immune activation
   including PMK-1. More importantly, oral administration of these strains
   in mice resulted in upregulated expression of MUC-2 and Il-10,
   indicating enhanced mucosal protection and anti-inflammatory responses.
   These findings suggest that the observed probiotic properties—such as
   mucosal adhesion and antimicrobial activity—may contribute to the
   therapeutic effects. However, the precise molecular pathways mediating
   these effects remain to be elucidated and warrant further mechanistic
   investigations.

   The gut microbiome is increasingly recognized as a key factor in the
   pathogenesis and progression of IBD, playing a dual role in either
   promoting or attenuating intestinal inflammation ([204]Manichanh et
   al., 2012). In murine models, including DSS-induced colitis, microbial
   dysbiosis and a marked reduction in gut microbial diversity are
   commonly observed ([205]Munyaka et al., 2016). Similar patterns of
   decreased microbial richness and compositional shifts have been
   documented in both human IBD patients and animal models ([206]Samanta
   et al., 2012). In the present study, the administration of
   kefir-derived microorganisms, particularly SLAM 005Y and SYN, resulted
   in increased alpha diversity compared to the DSS group, suggesting a
   partial restoration of microbial diversity disrupted by colitis.
   Furthermore, beta diversity analysis revealed distinct clustering
   between the SYN and DSS groups, indicating significant shifts in
   microbial community composition following kefir microbial treatment.
   The dominant bacterial phyla in the mammalian gut microbiota include
   Firmicutes, Bacteroidetes, Actinobacteria, and Proteobacteria ([207]Jo
   et al., 2024; [208]Tlaskalova-Hogenova et al., 2011; [209]Yu et al.,
   2022). A consistent hallmark of IBD-associated dysbiosis is the
   expansion of Proteobacteria, which is thought to reflect a destabilized
   gut ecosystem and heightened inflammatory tone ([210]Ni et al., 2017).
   In line with previous findings, the DSS group in this study exhibited
   an increased relative abundance of Proteobacteria. However, kefir
   microbial treatment led to a notable reduction in Proteobacteria
   abundance. In contrast, the abundance of S24-7, a commensal bacterial
   group typically enriched in healthy mice, increased significantly in
   the SLAM 023B and SYN groups. The observed decrease in Proteobacteria
   following kefir treatment further supports its anti-inflammatory
   potential, consistent with earlier studies linking Proteobacterial
   overgrowth to intestinal inflammation and colitis ([211]Mukhopadhya et
   al., 2012; [212]Rizzatti et al., 2017; [213]Shin et al., 2015).
   Moreover, genera such as Bifidobacterium and Parabacteroides, known for
   their probiotic activity and considered next-generation probiotic
   candidates ([214]Gomes and Malcata, 1999; [215]Menberu et al., 2022;
   [216]Sun et al., 2023), were increased in the SLAM 023B and SYN groups.
   These changes suggest that kefir microorganisms contribute to the
   restoration of gut microbial balance following DSS-induced dysbiosis,
   as evidenced by increased microbial diversity and the enrichment of
   beneficial taxa. Taken together, these findings support the potential
   of SLAM 005Y, SLAM 023B, and their combination to modulate gut
   microbial communities in colitis, thereby promoting microbial and
   metabolic homeostasis. Further mechanistic investigations are required
   to delineate the specific host–microbe interactions through which
   kefir-induced alterations in the gut microbiota mediate
   immunomodulatory effects and facilitate epithelial regeneration in the
   context of intestinal inflammation.

   In conjunction with our metagenomic findings, metabolomic profiling was
   conducted to further explore host responses to kefir microbial
   supplementation. Comparative analysis of fecal metabolites between the
   SYN and DSS groups revealed several distinct changes, particularly in
   amino acid profiles. Amino acids are central regulators of numerous
   metabolic pathways, and our results showed that BCAAs including
   L-valine and L-isoleucine, as well as aspartic acid and L-threonine,
   were significantly upregulated in the kefir-treated group. Prior
   studies have demonstrated that BCAAs play key roles in modulating
   inflammation, cellular stress responses, energy production, muscle
   metabolism, and immune function ([217]Xie et al., 2021; [218]Yan et
   al., 2023). Previous studies have demonstrated that dietary
   supplementation with BCAAs enhances intestinal development and
   epithelial cell proliferation by increasing the intestinal absorption
   of amino acids and glucose ([219]Zhou et al., 2018). In mice, BCAAs
   restriction has been shown to impair immune function and increase
   susceptibility to pathogenic infections ([220]Jose and Good, 1973).
   Amino acids such as L-valine and L-isoleucine are known to serve as
   substrates for immune cells and influence protein synthesis, thereby
   contributing to immune competence ([221]Calder, 2006). Moreover, BCAAs
   have been reported to regulate immune responses through activation of
   the mTOR signaling pathway, further underscoring their role in
   maintaining mucosal immunity ([222]Kimball and Jefferson, 2006).
   Notably, aspartic acid has been implicated in the development of
   visceral pain associated with intestinal inflammation, suggesting a
   possible link between elevated levels and inflammatory signaling in the
   gut ([223]Zhou and Nicholas Verne, 2008). Among the differentially
   regulated metabolites, arachidonic acid that a lipid mediator known to
   support mucosal repair, was also elevated following kefir microbial
   intervention ([224]Xu et al., 2021). These findings indicate that kefir
   microorganisms may contribute to the restoration of metabolic
   homeostasis, possibly through the modulation of key metabolites
   involved in intestinal regeneration and energy metabolism. The
   upregulation of BCAAs and other bioactive metabolites suggests that
   kefir supplementation could counteract the metabolic dysregulation
   associated with DSS-induced colitis. While these metabolomic
   alterations provide important insights into the potential mechanisms
   underlying the protective effects of kefir microbes, several
   limitations must be acknowledged. The precise microbial–host
   interactions responsible for these metabolic shifts remain to be
   clarified, and the causal links between specific metabolites and
   disease outcomes require further investigation. Future studies should
   focus on elucidating the microbial enzymatic pathways involved in
   metabolite production and their downstream effects on host physiology
   and immune regulation. A deeper mechanistic understanding—particularly
   of the microbial metabolic pathways and their downstream effects on
   epithelial repair and immune modulation—will be critical for developing
   targeted probiotic strategies to treat IBD using kefir-derived strains.

5. Conclusion

   Our study highlights the promising potential of K. marxianus SLAM 005Y,
   L. kefiri SLAM 023B, and their synergistic combination as
   multifunctional probiotic candidates isolated from milk kefir capable
   of enhancing host immune responses, extending lifespan, and
   ameliorating intestinal inflammation ([225]Fig. 8). Through an
   integrative approach combining gene expression profiling, microbiota
   analysis, and metabolomic evaluation, we provide mechanistic insight
   into how these kefir-derived microorganisms exert their beneficial
   effects. This study provides a foundational framework for future
   research investigating the specific mechanistic pathways through which
   kefir-derived microorganisms interact with the gut microbiota to
   regulate host immune responses and promote gastrointestinal health.
   Importantly, while the current study demonstrates efficacy in C.
   elegans and murine models of colitis, translational validation through
   human clinical trials is essential to determine their therapeutic
   relevance in the prevention and management of IBD. Further exploration
   of the interactions between kefir microbes and the host gut microbiome
   will be critical for understanding their broader impact on intestinal
   homeostasis and systemic health. Our work contributes to the growing
   body of evidence supporting the use of probiotic yeasts and bacteria as
   functional agents with therapeutic potential. By demonstrating
   consistent benefits across multiple biological models, this study
   supports the advancement of kefir-based probiotics as novel
   interventions for gut health and inflammation-related disorders.

Fig. 8.

   [226]Fig. 8
   [227]Open in a new tab

   Schematic representation of the protective effects of the synergistic
   combination of K. marxianus SLAM 005Y and L. kefiri SLAM 023B in a
   DSS-induced colitis mouse model. The left panel illustrates the
   pathological features associated with DSS-induced colitis, including
   dysbiosis, increased pro-inflammatory cytokines, disrupted intestinal
   barrier integrity, and reduced mucin and tight junction (TJ) protein
   expression. In contrast, the right panel depicts the therapeutic
   effects of SYN administration. SYN modulates the gut microbiota
   composition, leading to elevated production of beneficial metabolites
   such as branched-chain amino acids, amino acids, and fatty acids. These
   changes enhance mucin secretion and upregulate the expression of TJ
   proteins (claudin-1, occludin, ZO-1), thereby reinforcing epithelial
   barrier integrity and limiting microbial translocation. Furthermore,
   SYN suppresses pro-inflammatory cytokines (TNF-α, IL-6, IL-1β) while
   enhancing anti-inflammatory cytokine IL-10 expression, collectively
   contributing to reduced intestinal inflammation. This schematic
   highlights the role of milk kefir-derived microbial supplementation in
   restoring gut homeostasis and mitigating DSS-induced colitis via
   microbiome, metabolite, and immune modulation.

CRediT authorship contribution statement

   Se Hyun Lim: Writing – original draft, Methodology, Investigation,
   Formal analysis, Conceptualization. Visualization, Software, Writing –
   review & editing. Daye Mun: Writing – original draft, Methodology,
   Investigation, Formal analysis, Conceptualization. Visualization,
   Software, Writing – review & editing. Sangdon Ryu: Writing – original
   draft, Methodology, Investigation, Formal analysis, Conceptualization.
   Min-Geun Kang: Writing – review & editing, Visualization, Software,
   Methodology. Anna Kang: Writing – review & editing, Visualization,
   Software, Methodology. Daniel Junpyo Lee: Writing – review & editing,
   Visualization, Software, Methodology, Data curation. Eunsol Seo:
   Writing – review & editing, Visualization, Software, Methodology.
   Seon-hui Son: Writing – review & editing, Visualization, Software,
   Methodology. Soyeon Kim: Writing – review & editing, Visualization,
   Software, Methodology. Ki Beom Jang: Writing – review & editing,
   Visualization, Software, Methodology. Bum-Keun Kim: Visualization,
   Software, Methodology. Dong-Jun Park: Visualization, Software,
   Methodology. Sangnam Oh: Writing – review & editing, Supervision,
   Project administration, Investigation, Conceptualization. Younghoon
   Kim: Writing – review & editing, Supervision, Project administration,
   Investigation, Funding acquisition, Conceptualization.

Ethics approval and consent to participate

   Not applicable.

Consent for publication

   Not applicable.

Data availability

   Additional data is available from the authors upon request.

Declaration of competing interest

   The authors declare that they have no known competing financial
   interests or personal relationships that could have appeared to
   influence the work reported in this paper.

Acknowledgments