Comparative Immunomodulatory Profiles of Bifidobacterium animalis subsp. lactis HOM2120 and Bifidobacterium longum subsp. longum HOM1190: Insights from Genomics, In Vitro, and In Vivo Studies

Abstract

Bifidobacterium species are widely used as probiotics; yet their strain-specific immunomodulatory mechanisms remain incompletely defined. In this study, we characterized the probiotic properties and immunomodulatory activities of two genetically distinct strains, B. lactis HOM2120, isolated from human milk, and B. longum HOM1190, isolated from infant feces. Genomic analyses revealed strain-specific features associated with environmental resilience and probiotic functionalities, including acid and bile tolerance, pathogen inhibition, and adhesion to intestinal epithelial cells, which were confirmed by in vitro assays. In RAW264.7 macrophages, HOM2120 preferentially enhanced phagocytosis and TNF-α production, whereas HOM1190 induced higher IL-6 secretion, reflecting distinct strain-specific immunomodulatory patterns. Transcriptomic profiling and qRT-PCR validation revealed distinct molecular signatures underlying these effects. HOM2120 preferentially activated NF-κB and STAT3 signaling pathways and induced higher Tnf expression, suggesting a bias toward cell-mediated immune responses. In contrast, HOM1190 more strongly modulated the PI3K–Akt signaling axis and upregulated Il6 and Cd80 expression, supporting enhanced antigen presentation and humoral immune–associated responses. These strain-specific immunological tendencies were further validated in mice. HOM2120 predominantly enhanced delayed-type hypersensitivity responses, indicative of cellular immunity, whereas HOM1190 more effectively increased hemolytic activity and antibody-producing cell numbers, reflecting preferential activation of humoral immunity. Collectively, these results suggest that HOM2120 and HOM1190 are functionally distinct yet synergistic probiotic candidates, supporting their potential application as functional food ingredients or therapeutic adjuncts to modulate host immunity.

Introduction

Immunity is one of the most important factors for the health and well-being of an individual, which was starkly demonstrated during the COVID-19 pandemic resulting from SARS-CoV-2 [1]. The immune system consists of a complex network that includes the less specific innate immunity and the more specific adaptive immunity, both of which help maintain homeostasis under normal physiological conditions. Innate immunity serves as the primary defense, comprising the skin, mucous membranes, immune cells like dendritic cells, monocytes, macrophages, neutrophils, and natural killer cells, as well as cytokine molecules. Adaptive immunity is a highly specific line of defense against pathogens, consisting of humoral immunity (driven by B cells) and cell-mediated immunity (driven by T cells) [2]. A strong immune system provides protection against diseases, promotes recovery, and contributes to overall well-being.

The commonly-agreed-upon definition of probiotics is: “live microorganisms that, when administered in adequate amounts, confer a health benefit on the host” [3]. As a key genus of probiotics, Bifidobacterium species are extensively utilized and possess a well-defined biological profile. Bifidobacterium represents one of the most abundant genera in a healthy gut during early life, especially in the intestine of healthy breastfed infants, and to interact with human immune cells and to modulate specific pathways, involving innate and adaptive immune processes [4]. Bifidobacterium animalis subsp. lactis and Bifidobacterium longum subsp. longum represent two widely used yet biologically distinct bifidobacterial taxa in commercial probiotic products. The beneficial properties of consuming B. longum and B. lactis on improving the host immune system have been demonstrated for treating a range of conditions and diseases, such as diarrhea [5, 6], infections [7-10], colds [11, 12], COVID-19 [13, 14], arthritis [15], infant colic [16, 17] and certain forms of cancer [18-20]. These health benefits are primarily mediated through immunomodulatory mechanisms, as B. lactis and B. longum can influence the host immune system by interacting with gut-associated lymphoid tissue (GALT), regulating cytokine production, and modulating the balance between innate and adaptive immunity. However, despite their widespread application and shared probiotic characteristics, direct head-to-head comparisons of the immunomodulatory properties of B. lactis and B. longum remain limited, particularly the underlying molecular mechanisms mediating their potentially complementary immune effects have not been systematically investigated. This gap becomes particularly pressing in light of the widespread use of multi-strain probiotics, whose strain combinations are typically selected on an empirical basis. Understanding these functional differences is critical for developing probiotic interventions tailored to specific immune-related diseases.

In the present study, we isolated and characterized a strain of B. lactis HOM2120 from human milk and a strain of B. longum HOM1190 from infant feces. To elucidate their functional differences, we performed a systematic comparison of their probiotic and immunomodulatory properties. This included whole-genome sequencing, a series of in vitro assays recommended by the FAO/WHO guidelines [21] such as tolerance to gastrointestinal conditions (gastric acid and bile), adherence to intestinal epithelial cells and/or mucus, and antimicrobial activity against potentially pathogenic bacteria, as well as immunomodulatory and transcriptomic analyses in immune cells. Key pathway-related genes were further validated by quantitative real-time PCR (qRT-PCR), and the strain-specific immune effects were ultimately confirmed in vivo using mouse models. This study aims to highlight the distinct immunological profiles and application potentials of each strain, providing a scientific basis for developing targeted probiotic formulations.

Materials and Methods

The Bacteria Strains and Culture Conditions

The B. lactis HOM2120 and B. longum HOM1190 were isolated from human milk and infant feces, respectively, and have been deposited in the China General Microbiological Culture Collection Center (CGMCC) in Beijing, China, under accession numbers CGMCC No. 25681 and No. 25684. The reference strain B. lactis Bb12 and B. longum BB536, which were used for the comparison of phenotypic properties, were selected as they are widely recognized gold-standard probiotic strains with well-documented health benefits. Indicator strains were selected to cover clinically relevant enteric pathogens (Salmonella Typhimurium ATCC 14028, and Escherichia coli ATCC 8739), foodborne pathogens (Staphylococcus aureus ATCC 6538, and Listeria monocytogenes ATCC 19111), and opportunistic pathogens (Clostridium difficile ATCC 9689 and Pseudomonas aeruginosa ATCC 9027) to comprehensively evaluate the antimicrobial activity of the candidate strains. B. lactis, B. longum and C. difficile ATCC 9689 were grown respectively in de Man, Rogosa, and Sharpe (MRS) broth (OXOID, UK) plus 0.5 g/L L-cysteine hydrochloride (MRSL) and BHI medium (OXOID) at 37°C for 24 h anaerobically (80% N₂, 10% CO₂, 10% H₂). With the exception of ATCC 9689, all indicator bacteria were freshly cultured in Tryptone Soya Broth (OXOID) for 20 h at 37°C under aerobic conditions and subsequently used for antimicrobial activity evaluation. Lyophilized B. longum HOM1190 and B. lactis HOM2120 powder were used in the animal experiments. Initially, fermentation was carried out in a 30 L bioreactor at 37°C under anaerobic conditions, maintaining the pH at 5.5 and the stirring speed at 100 rpm for 24 h. After fermentation, the bacterial cells were collected by centrifugation at 10,000 ×g for 10 min at 4°C, rinsed twice with sterile saline, resuspended in a protective medium, freeze-dried for 48 h, and stored at -20°C until further use. The lyophilized HOM2120 and HOM1190 powder were freshly reconstituted in sterile saline each day prior to administration to animals.

Morphological Characterization

After 72 h of incubation on MRS agar, a single colony was initially characterized by Gram staining, and its morphology was examined using optical microscopy (Zeiss, Germany). After 24 h of incubation on MRSL medium, the bacterial culture was centrifuged at 10,000 ×g for 10 min. The pelleted bacteria were washed three times with phosphate buffer solution (Gibco, USA). The cells were subsequently fixed in 2.5% glutaraldehyde (Phygene, China) and maintained at 4°C for overnight preservation. Subsequently, the cells were dehydrated through an ethanol concentration gradient (50%, 70%, and then twice with 100%), freeze-dried for 1 h and coated with a thin layer of gold by sputtering, and then analyzed using scanning electron microscopy (Hitachi, Japan).

Whole Genome Analysis

Whole-genome sequencing of strains HOM2120 and HOM1190 was performed using the Illumina II and PacBio Sequel III platforms (Majorbio, China). The resulting reads were de novo assembled with Unicycler (version 0.4.8) [22]. Coding sequences (CDSs), as well as tRNA and rRNA genes, were predicted using Glimmer [23], tRNAscan-SE v2.0 [24] and rRNAmmer (v1.2) [25], respectively. Finally, a circular genome graph was generated using Circos (v0.69.6) [26]. Protein function analysis was performed using a database of clusters of orthologous groups (COG) [27]. The complete HOM2120 and HOM1190 genome have been submitted to GenBank (with GenBank Accession No. CP172403 and CP197406, respectively).

Simulated Gastrointestinal Transit

The survival of probiotics in the GIT was evaluated by using simulated gastrointestinal juices based on the methods of Huihui et al. [28], with some modifications. To simulate gastrointestinal transit, bacterial suspensions were sequentially introduced into simulated gastric fluid (SGF, pH 3.0; Yuan Ye, China) and simulated intestinal fluid (SIF, pH 6.8; Yuan Ye) containing 0.3% w/v bovine bile (USA). The bacterial suspension (1 ml) was added to 9 ml of SGF and anaerobically incubated at 37°C for 3 h. The cells were then collected, resuspended in SIF, and anaerobically incubated at 37°C for 24 h. Live cell counts were performed before and after each stage of gastrointestinal transit using MRS agar. The survival rate (%) was calculated as (log N_t / log N₀) × 100, where N₀ and N_t represent the initial and final viable counts (CFU/ml), respectively. Bb12 and BB536, well known for their good gastrointestinal tolerance, were used as reference strains.

Adhesive Determination of Caco-2 Cells

The ability of bacteria for adhering to Caco-2 cells was investigated as previously described [29, 30], with some modifications. Caco-2 cells (Cell Bank, Chinese Academy of Sciences, China) were grown in DMEM (Gibco) supplemented with 10% FBS (Corning, USA) and antibiotics (100 U/ml penicillin and 100 μg/ml streptomycin). The cells were maintained at 37°C in an environment with 5% CO₂ with constant humidity. The cells were plated in 12-well plates (Corning) at a density of 2 × 10⁵ cells/ml and cultured for 14 days until complete morphological and functional differentiation was achieved. The culture medium was changed every two days. A fresh strain suspension, at a density of 1 × 10⁸ CFU/ml, was prepared in DMEM supplemented with 10% FBS but without antibiotic-antimycotic solution. The suspension was applied to the cells and incubated at 37°C for 2 h to allow adhesion. After incubation, the cells were gently washed twice with DPBS to remove non-adherent bacteria. Subsequently, portions of the cells were immersed in 1.0 ml of 100% ethanol for 20 min. Following three washes with DPBS, Gram staining was performed, and the cells were examined under a 100× oil immersion lens (Zeiss). Other cell portions were subjected to lysis with 0.1% Triton X-100 (Jiancheng, China) and plated onto MRS agar (OXOID). The adhesion rate of bacteria was determined by comparing the number of bacterial colonies with the number of bacteria initially added. Bb12 and BB536, which are known for their good adhesion abilities, were used as references in this experiment.

Antimicrobial Activities

The antimicrobial activity of probiotics was assessed by the well-diffusion method [31] against six indicator strains: S. Typhimurium ATCC 14028, E. coli ATCC 8739, S. aureus ATCC 6538, L. monocytogenes ATCC 19111, C. difficile ATCC 9689, and P. aeruginosa ATCC 9027. Approximately 1 × 10⁷ CFU of the microbial inoculum were evenly spread over the agar surface. A 10 mm diameter well was then aseptically created using a sterile metal cylinder. A 0.2 ml volume of a 24 h cultured probiotic was introduced into the well and allowed to diffuse at 4°C for 2 h. The agar plates were then incubated at 37°C for 48 h. The antimicrobial effect was determined by recording the size (mm) of the zones of inhibition around each well.

Analysis of Lactic Acid and Short-Chain Fatty Acids

Gas chromatography-mass spectrometry (Agilent 8860, USA) was employed to analyze lactic acid and short-chain fatty acids (SCFAs) in the fermented broth. The samples were separated with the application of a DB-23 chromatographic column (60 m × 250 μm × 0.25 μm) (Agilent). Specific chromatographic conditions were applied, as previously described [32]. Lactic, formic, acetic, propionic, and butyric acids (Sigma-Aldrich, USA) were used as standards. The concentrations of lactic acid and each SCFAs were determined by referencing a standard curve created from seven different concentrations of the standards.

Macrophage Stimulation, Transcriptomic Analysis and qRT-PCR Validation

RAW 264.7 macrophage cells (Cell Bank, Chinese Academy of Sciences, China) were used to investigate the immunostimulatory activity of probiotics in vitro [33]. After proliferation, RAW 264.7 macrophages were seeded at 5 × 10⁵ cells/ml in 24-well plates (Corning) in triplicate and incubated for 4 h to allow cell adhesion. RAW 264.7 cells were treated with probiotic suspensions that were heat-inactivated by incubating at 70°C for 30 min, at final concentrations of 5 × 10⁷ and 5 × 10⁸ total forming units per milliliter (TFU/ml) in 1 ml of culture medium. The positive control was added with 1μg/ml lipopolysaccharide (LPS; Sigma-Aldrich). Cell viability was assessed by the MTS assay (Promega, USA) [34]. Phagocytosis was measured using the neutral red uptake assay [35] following the method reported previously. The levels of cytokines, namely TNF-α and IL-6 were measured using the respective ELISA kit (RayBiotech, USA), as stated by the manufacturer’s directions.

RAW 264.7 cells were seeded at 5.0 × 10⁵ cells/ml in 6-well plates (Corning) in triplicate and incubated for 4 h to allow cell adhesion. The culture medium was refreshed with complete DMEM containing 5 × 10⁸ TFU/ml of either HOM2120 or HOM1190, followed by incubation for an additional 24 h. Complete DMEM without probiotics served as the negative control. After treatment, the cells were collected and washed three times with DPBS. Total RNA was isolated using TRIzol reagent (TIANGEN, China), and the purified RNA samples were sent to Majorbio Bio-Pharm Biotechnology Co., Ltd. (China) for sequencing. RNA sequencing was performed on the Illumina NovaSeq X Plus platform using paired-end 150 bp reads and the corresponding reagent kit (USA). Low-quality reads (Q < 20), adaptor sequences, and reads containing poly-N were removed using fastp (v0.19.5) [36]. The clean reads were mapped to the Mus musculus reference genome in a strand-specific manner using HISAT2 [37], and transcript assemblies were generated with RSEM (version 1.3.3) [38]. Principal component analysis (PCA) was applied to normalized expression data to visualize sample variation and assess replicate consistency. Differential gene expression was analyzed with DESeq2 (v1.24.0) [39], and genes with an absolute log₂ fold change ≥ 1 and an adjusted p < 0.05 were considered significant differentially expressed genes (DEGs). Enrichment of KEGG pathways in DEGs was examined, and pathways with adjusted p < 0.05 were considered significantly enriched following HOM2120 or HOM1190 treatment.

To validate the RNA-seq results, especially the key pathways identified from the differential expression analysis, qRT-PCR was performed on selected genes. Total RNA (1 μg) was reverse-transcribed into cDNA using HiScript Q RT SuperMix for qPCR (+gDNA wiper) (Vazyme Biotech Co., Ltd., China) according to the manufacturer’s instructions. QRT-PCR was conducted using ChamQ SYBR Color qPCR Master Mix (2X) (Vazyme Biotech Co., Ltd., China) on a QuantStudio 5 384-well Real-Time PCR System (Applied Biosystems, USA) under the following conditions: an initial denaturation at 95°C for 5 min, followed by 40 cycles of 95°C for 5 sec and 55°C for 30 sec. The relative expression levels of target genes were calculated using the 2^{^-ΔΔCt} method [40] with Gapdh as the internal reference and all reactions were performed in triplicate. The forward and reverse primer sequences of the target genes are listed in Table S2.

Immunity Boosting Experiment in Healthy Mice

A total of 336 male, specific pathogen-free Kunming mice (8 weeks old, 18–20 g) were obtained from Beijing HFK Bioscience Co., Ltd. (China). Mice received a regular laboratory chow diet and were allowed unlimited access to distilled water. All procedures involving animals were assessed and authorized by the Laboratory Animal Center of Beijing Union University (China). Following a one-week acclimation period, a total of seven groups were established by randomization (n = 12): NC; three HOM2120 dose groups (LD/MD/HD); and three HOM1190 dose groups (LD/MD/HD). The NC group received saline via intragastric administration, while the other six groups were administered viable HOM2120 and HOM1190 bacteria, respectively, prepared by suspending lyophilized bacterial powder in saline. Numerous studies have demonstrated that daily gavage with Bifidobacterium at doses of 1 × 10⁸–1 × 10⁹ CFU per mouse can effectively enhance immune function in mice [41-43]. Considering the influence of body weight on dosage, mice in the LD, MD, and HD groups were administered Bifidobacterium at doses of 2.5 × 10⁹, 2.5 × 10¹⁰, and 5 × 10¹⁰ CFU/kg body weight, respectively. The gavage volume was uniformly set at 10 ml/kg body weight, once daily for 32 consecutive days. Mice in the control group were administered sterile saline by gavage at a volume of 10 ml/kg. The experiments were conducted in four parts, following previously established methods [33]. The first phase involved carbon clearance experiments; the second phase included measurements of half-hemolytic value (HC₅₀), number of antibody-producing cells (APCs), and delayed-type hypersensitivity (DTH) assays; the third phase focused on macrophage phagocytosis assays; and the fourth phase involved concanavalin A (ConA)-induced lymphocyte proliferation tests in mice.

Statistical Analysis

Data from a minimum of three replicates are shown as mean ± SD or mean ± SEM. One-way ANOVA with Tukey’s or Dunnett’s post-hoc testing was performed using GraphPad Prism 9.5, with significance set at p < 0.05.

Results and Discussion

Morphological Analysis of HOM2120 and HOM1190

Fig. 1A illustrated the colony characteristics of HOM2120, which were round or slightly oval, measured approximately 1-2 mm in diameter, had a smooth and moist surface, white or off-white, regular margins, and a soft texture that was easy to pick. In contrast, Fig. 1B depicted the colonies of HOM1190, which were irregular or slightly flattened, approximately 2-3 mm in diameter, had a somewhat rough or slightly wrinkled surface, milky white to grayish-white coloration, slightly irregular margins, and a denser texture with slight stickiness. These observations indicated clear differences in colony morphology and texture between the two strains. Under light microscopy, the bacteria of HOM2120 (Fig. 1C) appeared relatively short and uniform, with cell lengths mainly ranging from 1-4 μm and few chain-like arrangements. In contrast, the bacteria of HOM1190 (Fig. 1D) displayed greater morphological diversity, with some elongated cells reaching 2-5 μm and frequent short chain formations. The scanning electron microscopy (SEM) images provided higher resolution confirmation. The SEM image of HOM2120 (Fig. 1E) showed smooth, rod-shaped cells of relatively uniform size (1 4 μm × 0.5 1 μm), with occasional Y-shaped branching. In contrast, the SEM image of HOM1190 (Fig. 1F) exhibited greater heterogeneity, with some cells appearing swollen, measuring 2-5 μm in length and 0.5-1 μm in width, and a notable presence of extracellular mucus-like material. Taken together, both strains HOM2120 and HOM1190 exhibited morphological characteristics consistent with previously reported features of Bifidobacterium species, including anaerobic growth, Gram-positive, non-spore forming and non-motile as well as occasional V-shaped or Y-shaped cellular arrangements [44].

Fig. 1. The morphological features of HOM2120 and HOM1190,

Colony morphology of HOM2120 (A) and HOM1190 (B) on agar plates, Cellular morphology of HOM2120 (C) and HOM1190 (D) observed by optical microscope using oil-immersion objective lens at 1,000 × magnification) and by scanning electron microscope at 10,000 × magnification for HOM2120 (E) and HOM1190 (F).

General Genome Features of HOM2120 and HOM1190

The whole nucleotide sequence of B. lactis HOM2120 was submitted to GenBank under the accession number CP172403, while B. longum HOM1190 was deposited under the accession number CP197406. The complete genome of B. lactis HOM2120 consists of one 1,946,900 bp circular chromosome with no plasmid,which is consistent with observations from other B. lactis genomes [45-47], The chromosome has a G+C percentage of 60.5% and contains1568 CDSs, 51 tRNA genes, and 9 rRNA operons (Fig. 2A). The complete genome sequence of the B. longum HOM1190 strain was determined to be 2,389,412 bp in size with a G+C % percentage of 60.1% (Fig. 2B). The genome contained 1,598 CDSs, 54 tRNA genes, and 12 rRNA operons. No plasmids were detected, which is consistent with values reported for other B. longum genomes [48].

Fig. 2. Genomic characteristics of HOM2120 and HOM1190.

The circular genome structures of HOM2120 (A) and HOM1190 (B) were visualized using Circos and comprise seven concentric rings. From the outermost to the innermost, the rings represent genome size, coding sequences (CDSs) on both DNA strands, prophage regions, ribosomal and transfer RNA genes, along with GC content and GC skew. The functional classification of coding genes in HOM2120 and HOM1190 (C) were annotated based on Clusters of Orthologous Groups (COG) categories.

COG annotation identified 1,419 genes in HOM2120, and 1,699 genes in HOM1190. As shown in Fig. 2C, functional classification revealed distinct differences between the two strains. HOM2120 exhibited higher proportions of genes associated with translation, ribosomal structure and biogenesis (12.19% vs 9.77%) and amino acid transport and metabolism (11.35% vs 10.06%), suggesting enhanced capacity for protein synthesis and amino acid utilization. Cell wall/membrane/envelope biogenesis was slightly higher in HOM2120 (6.84% vs 6.36%). Conversely, HOM1190 displayed higher proportions of genes in carbohydrate transport and metabolism (12.18% vs 10.15%), transcription (8.00% vs 6.13%), and Mobilome: prophages, transposons (2.06% vs 0.85%), indicating more active gene expression and increased genomic plasticity. Genes related to extracellular structures were minimal in both strains (0.21% vs 0.18%). Further genome analysis revealed that the genomes of HOM2120 and HOM1190 contain probiotic-associated genes encoding proteins involved in essential physiological processes, including stress tolerance (temperature: groEL, dnaK; oxygen: trxB, mutT; acid and bile: atpC, atpD), adhesion (tuf, adhE), lactate/acetate production (ldh, ackA), and immunity (srtA, slyA), supporting their environmental resilience and potential probiotic functionalities (Table S1) [49-51]. These genetic features suggest that HOM2120 and HOM1190 may possess robust environmental resilience during industrial processing and gastrointestinal transit, as well as potential probiotic functionality through adhesion to intestinal epithelial cells, modulation of host immune responses, and the generation of beneficial metabolic products, such as lactic acid and acetic acid.

Taken together, these findings suggest that HOM2120 and HOM1190 represent promising candidates for probiotic applications, particularly in infant-targeted formulations. Future studies focusing on transcriptomic profiling and in vivo functional validation of their unique genes and probiotic properties will be essential to further elucidate their mechanisms of action and assess their commercial potential.

Gastrointestinal Transit Tolerance and Adhesive Capacity to Caco-2 cells

Gastric acid and bile play critical roles in defending the digestive tract against microorganisms and effectively killing and controlling many pathogens [52]. However, the same defense mechanism can destroy potentially beneficial microbes. The capacity to endure gastric acid and bile enables probiotics to survive gastrointestinal transit, establish in the gut, and exert long-term physiological effects. As shown in Fig. 3A, when exposed to gastric fluid at pH 3 for 3 h, B. lactis HOM2120 demonstrated tolerance comparable to that of the reference strain B. lactis Bb12 (p > 0.05), while B. longum HOM1190 displayed significantly higher tolerance than B. longum BB536 (p < 0.001). After gastric treatment, the samples were subsequently incubated in intestinal fluid (pH 6.8), where the tolerance patterns at both 3 h and 24 h were generally consistent with those observed in gastric fluid, except that B. lactis HOM2120 exhibited significantly higher tolerance than B. lactis Bb12 (p < 0.05 at 3 h; p < 0.001 at 24 h) (Fig. 3B and 3C). HOM2120 and HOM1190 showed strong resistance to simulated gastrointestinal conditions. HOM1190 tolerated gastric fluid better than the reference strain BB536, while HOM2120 exhibited superior intestinal fluid tolerance compared with Bb12, especially at 24 h (p < 0.001). These results suggest that both strains are promising probiotic candidates, with enhanced survival through the gastrointestinal tract and potential for improved gut persistence.

Fig. 3. Gastrointestinal transit tolerance and adhesion of HOM2120 and HOM1190 to Caco-2 cells.

(A, B, and C) Tolerance to simulated gastric and intestinal juices. (D) Quantification of Caco-2 cell adherence. Microscopic visualization of the adhesion of HOM2120 (E), Bb12 (F), HOM1190 (G) and BB536 (H) to Caco-2 cells. Data are presented as mean ± SD, n = 3. One-way ANOVA followed by Tukey’s post-hoc test was conducted using GraphPad Prism 9.5 (GraphPad Software, USA). *p < 0.05, **p < 0.01, ***p < 0.001 vs. control. SGF, simulated gastric fluid; SIF, simulated intestinal fluid.

The adhesion of HOM2120 to Caco-2 cells after 2 h was significantly higher than that of Bb12 (p < 0.001), whereas HOM1190 adhered significantly more than BB536 (p < 0.001). The adhesion capacity of HOM2120 was significantly higher than that of HOM1190 (Fig. 3D), which may be attributed to the presence of filamentous, reticular, or viscous extracellular materials surrounding HOM1190 (Fig. 1F). This phenomenon will be investigated in greater detail in future studies. Microscopic examination of bacterial adhesion further corroborated these quantitative results (Fig. 3E-3H). Adhesion capacity is widely recognized as a strain-specific property rather than a general feature of a given species. Several comparative studies have demonstrated substantial variation in adhesion rates among strains of the same species, underscoring the importance of evaluating probiotic candidates at the strain level [53].These results align with previous reports that adhesion capacity may facilitate strain-specific immunomodulatory activity and epithelial interaction [54, 55]. Collectively, these findings suggest that HOM1190 and HOM2120 possess enhanced survival and adherence capabilities compared with conventional probiotic strains, indicating their potential as promising candidates for oral administration with improved gut colonization and persistence.

Antimicrobial Activities, Lactic Acid, and SCFAs Production

Both B. lactis HOM2120 and Bb12 strains showed antagonistic activity against E. coli, P. aeruginosa, L. monocytogenes, and C. difficile using the agar well diffusion method (Table 1). When it comes to combating S. Typhimurium and S. aureus, B. lactis HOM2120 was capable of generating zones that were not only clearer but also boasted larger diameters in comparison with Bb12. This may be because B. lactis HOM2120 produced more lactic acid (p < 0.01) (Table 2). Similarly, B. longum HOM1190 exhibited stronger inhibitory effects against P. aeruginosa and L. monocytogenes than B. longum BB536, likely due to increased production of lactic and acetic acids (p < 0.01) Overall, B. longum HOM1190 displayed superior antimicrobial activity compared with B. lactis HOM2120, with the exception of S. aureus, owing to its ability to produce a broader spectrum of organic acids, including lactic, acetic, formic, propionic, butyric, and isobutyric acids (p < 0.01). These findings suggest that the antimicrobial activity of these strains is closely associated with their capacity to produce organic acids, which can lower the local pH and disrupt the growth of pathogenic bacteria. SCFAs, secreted by probiotics, function both as antimicrobial agents and as enhancers of the relative abundance of beneficial gut bacteria. SCFAs, such as formic and acetic acid, have antimicrobial properties and can disrupt the metabolism and cell membranes of pathogenic bacteria, thus contributing to the overall antimicrobial activity of probiotics in the gut or other environments. In this study, the broader spectrum of acids produced by HOM1190 may contribute to its enhanced inhibitory effects against a wider range of pathogens. These SCFAs not only directly inhibit the growth of pathogens by creating an acidic microenvironment but are also widely recognized for their critical role in immune regulation, such as promoting the anti-inflammatory functions of macrophages and dendritic cells, modulating cytokine secretion, and maintaining intestinal immune homeostasis [56]. Moreover, SCFAs enhance the intestinal epithelial barrier, helping to reduce pathogen invasion and lower the risk of immune system overactivation [57]. These results support the potential application of HOM2120 and HOM1190 as effective probiotics for modulating gut microbiota and preventing or controlling enteric infections.

Table 1.

Inhibition of pathogen growth by HOM2120 and HOM1190.

Strains	S. Typhimurium	E. coli	S. aureus	L. monocytogenes	C. difficile	P. aeruginosa
MRSL medium	-	-	-	-	-	-
Bb12	-	+	++	+	+	++
HOM2120	++	+	+++	+	+	++
HOM1190	++	++	++	+++	+	+++
BB536	++	++	++	++	+	++

Growth inhibition was scored according to the diameter of the inhibition zone (mm): (-) no visible inhibition or <10 mm; (+) 11–16 mm; (++) 17–23 mm; (+++) ≥23 mm

Table 2.

Lactic acid and SCFA production by HOM2120 and HOM1190.

Strains	Lactic acid (mg/l)	Acetic acid (mg/l)	Formic acid (mg/l)	Propionic acid (mg/l)	Butyric acid (mg/l)	Isobutyric acid (mg/l)
MRSL medium	543.6 ± 10.7E	2134.7 ± 11.4D	160.6 ± 1.5C	11.4 ± 0.2C	12.3 ± 0.3C	10.0 ± 0.2C
HOM2120	1759.4 ± 8.4C	4355.9 ± 91.5C	1053.4 ± 46.1A	21.6 ± 0.1B	16.3 ± 0.3B	12.0 ± 0.3B
Bb12	1258.3 ± 51.9D	4281.0 ± 55.7C	1017.0 ± 16.5A	20.9 ± 0.7B	15.9 ± 0.5B	11.9 ± 0.20B
HOM1190	5842.1 ± 12.8A	6392.8 ± 95.8A	863.0 ± 34.6B	48.8 ± 0.6A	44.5 ± 0.4A	27.2 ± 0.1A
BB536	5243.1 ± 31.3B	5875.3 ± 68.1B	860.4 ± 6.9B	47.7 ± 1.3A	42.6 ± 1.8A	27.0 ± 0.2A

Data are expressed as mean ± SD, n = 3. One-way analysis of variance (ANOVA) followed by Tukey's post-hoc test for multiple comparisons was performed using GraphPad Prism 9.5(GraphPad Software). Within the same column, values labeled with different uppercase letters indicate highly significant differences (p < 0.01).

Macrophage Stimulation and Transcriptomic Analysis of HOM2120 and HOM1190

To evaluate the in vitro immunomodulatory properties of B. lactis HOM2120 and B. longum HOM1190, RAW 264.7 cells were cultured in the presence of heat-inactivated bacteria for 24 h. Heat inactivation was used to avoid the adverse effects of bacterial proliferation on cell viability, and to focus specifically on immune-modulatory effects mediated by bacterial structural components and residual metabolites. Subsequently, cell viability, phagocytic activity, and the levels of TNF-α and IL-6 were measured (Fig. 4). LPS (1 μg/ml) was used as the positive control to stimulate RAW 264.7 cells. We used B. lactis Bb12 and B. longum BB536 as reference strains, since they are well-established probiotics extensively employed in the food industry and documented to enhance host immune responses [58, 59]. None of the tested strains exhibited cytotoxic effects, as cell viability remained above 100% across all concentrations, indicating good biocompatibility under the experimental conditions (Fig. 4A). All treatments significantly enhanced the phagocytic activity of RAW264.7 macrophages compared with the control group (p < 0.05) (Fig. 4B). HOM1190 and Bb12 showed a significant dose-dependent effect (p < 0.05), whereas no such trend was observed for HOM2120 and BB536. Notably, all high-dose groups except BB536 exhibited higher phagocytic activity than the LPS-treated group (p < 0.05). Pairwise comparisons revealed that low-dose HOM2120 induced greater phagocytic activity than HOM1190 (p < 0.05), and high-dose HOM1190 was more effective than BB536 (p < 0.05). All strains significantly enhanced TNF-α production by RAW264.7 macrophages compared with LPS (p < 0.05), except for low-dose BB536. TNF-α production was dose-dependent for all strains, with higher concentrations eliciting greater responses. Notably, HOM2120 induced the highest TNF-α levels, followed by HOM1190, Bb12, and BB536 (p < 0.05), reflecting strain-specific differences in pro-inflammatory potential (Fig. 4C). As expected, LPS strongly induced IL-6 production (Fig. 4D). Similarly, HOM2120 and HOM1190 significantly increased IL-6 production in a dose-dependent manner (p < 0.05), whereas Bb12 and BB536 showed no clear dose-response. High-dose HOM2120 induced more IL-6 than Bb12, HOM1190 produced higher IL-6 than BB536 at both doses, and high-dose HOM1190 elicited greater IL-6 than high-dose HOM2120 (p < 0.05), highlighting strain-specific differences in IL-6–stimulating capacity. Macrophages possess defensive functions against pathogens, such as microbes, and they also play a substantial role in the homeostatic maintenance of the body by disposing internal waste materials and facilitating tissue repair. Phagocytosis is an essential function of macrophages and plays a crucial role in clearing infections and remodeling tissues [60]. Macrophages release various cytokines and chemokines to recruit other immune cells to the infection site. This coordinated immune response helps mount a more effective defense against pathogens. Treatment of RAW 264.7 cells with heat-inactivated HOM2120 and HOM1190 resulted in elevated levels of pro-inflammatory cytokines, including TNF-α and IL-6 (Fig. 4C and 4D) in RAW 264.7 cells, which are among the first cytokines produced by phagocytic cells in response to encountering pathogenic bacteria and agents. TNF-α and IL-6 are crucial cytokines that perform significant functions in promoting inflammation and activating T and B cells, thereby enhancing adaptive immunity [61]. Notably, HOM2120 induced a stronger increase in TNF-α, accompanied by enhanced phagocytic activity, suggesting a preferential role in cell-mediated immunity, whereas HOM1190 elicited higher IL-6 production, indicative of a bias toward humoral immune modulation. Importantly, the moderate but coordinated upregulation of TNF-α and IL-6 by HOM2120 and HOM1190 suggests a capacity to induce effective yet controlled immune activation—sufficient to combat intracellular pathogens and tumors while minimizing the risk of excessive inflammatory damage. This is particularly relevant given the central role of macrophage-mediated phagocytosis in pathogen clearance, immune cell recruitment, and tissue repair [60]. The enhanced phagocytic activity observed with HOM2120 and HOM1190 further supports their potential application as functional probiotics in immunotherapy. Together, these properties provide a mechanistic basis for their immunomodulatory potential.

Fig. 4. Immunostimulatory effects of heat-inactivated HOM2120 and HOM1190 on RAW264.7 cells.

(A) Assessment of cell viability. (B) Phagocytic activity. (C) TNF-α levels. (D) IL-6 levels. Data are presented as mean ± SD, n = 3. L and H represent low (5 × 10⁷ TFU/ml) and high (5 × 10⁸ TFU/ml) bacterial concentrations, respectively. One-way ANOVA followed by Tukey’s post-hoc test was conducted using GraphPad Prism 9.5 (GraphPad Software). Values marked with different letters as superscripts indicate significant differences (p < 0.05).

To clarify the molecular basis of the immunoregulatory activities of HOM2120 and HOM1190, we carried out RNA-seq to assess transcriptomic alterations in RAW264.7 cells. Principal component analysis (PCA) revealed clear segregation among the control, HOM2120-, and HOM1190-treated groups, indicating distinct global transcriptional responses induced by each strain (Fig. 5A). HOM2120 treatment resulted in 3,287 upregulated and 2,248 downregulated genes (Fig. 5B), while HOM1190 treatment led to 4,193 upregulated and 3,082 downregulated genes (Fig. 5C). Both treatments significantly altered gene expression profiles, suggesting strong immunomodulatory activity. Clusters of DEGs were presented using heatmap visualization (Fig. 5D) among the three groups, confirming the reproducibility of transcriptomic differences. The heatmap suggests that the two strains modulate overlapping yet distinct sets of genes. To further investigate the biological functions associated with the transcriptional changes induced by B. lactis HOM2120 and B. longum HOM1190, KEGG pathway enrichment analysis was performed based on DEGs identified in each treatment group. As shown in Fig. 5E and 5F, KEGG pathway enrichment analysis revealed that both HOM2120 and HOM1190 significantly upregulated multiple immune-related signaling pathways. Their shared pathways mainly included pattern recognition receptors (Toll-like, NOD-like, and C-type lectin receptors), key signal transduction hubs (NF-κB, MAPK, and JAK–STAT), downstream cytokine and chemokine signaling (TNF and cytokine–cytokine receptor interactions), and immune effector pathways, including B-cell receptor signaling, FcγR-mediated phagocytosis, complement and coagulation cascades, and neutrophil extracellular trap formation. In addition to the shared pathways, HOM2120 uniquely or preferentially enriched pathways related to T-cell–mediated immunity, including Th1 and Th2 cell differentiation, Th17 cell differentiation, and the intestinal immune network for IgA production. These findings suggest that HOM2120 not only promotes T-cell polarization but also facilitates T-cell–dependent mucosal humoral responses. Specifically, Th17 cells are known to play a key role in mucosal immune responses and can be induced by specific microbial signals in the gut, contributing to intestinal barrier defense [62] Moreover, the intestinal immune network for IgA production involves coordinated T-cell–B-cell interactions that drive class switching to IgA in gut-associated lymphoid tissues [63]. In contrast, HOM1190 uniquely enriched the PI3K–Akt signaling pathway, which is centrally involved in B-cell proliferation, survival, and differentiation toward antibody-secreting cells, highlighting its potential role in modulating humoral immune modulation (Fig. 5F) [64]. In summary, the in vitro and pathway analyses suggest that HOM2120 may preferentially modulate T-cell–dependent immunity, whereas HOM1190 shows a trend toward humoral immune responses.

Fig. 5. Differential mRNA expression profiles and KEGG-based functional enrichment of DEGs in RAW264.7 cells treated with heatinactivated HOM2120 and HOM1190.

(A) PCA analysis (B) Volcano plot of DEGs between the HOM2120-treated groups and control. (C) Volcano plot of DEGs between the HOM1190 groups and control. (D) Heatmap of DEGs in control, HOM2120-, and HOM1190-treated groups (E) Scatter plot showing KEGG-enriched DEGs (p < 0.05) in the HOM2120 treatment compared with the control group. (F) Scatter plot showing KEGG-enriched DEGs (p < 0.05) in the HOM1190 treatment compared with the control group. Each dot represents a pathway, with color showing the adjusted p-value and size indicating the number of DEGs.

To validate the predicted strain-specific immune modulation, we performed qPCR on eight key genes corresponding to the enriched pathways: RelA and Nfkb1 (NF-κB), Stat3 (T-helper cell differentiation), Akt1 and Pik3ca (PI3K–Akt) and Tnf, Il6, and Cd80 (cytokine/humoral immune pathways). As shown in Fig. 6A–6D, treatment with both HOM2120 and HOM1190 significantly upregulated RelA, Nfkb1, Stat3, and Tnf compared with the control (p < 0.001), with HOM2120 inducing significantly greater increases than HOM1190 (p < 0.001). Additionally, both HOM2120 and HOM1190 treatments significantly upregulated Akt1, Pik3ca, Il6, and Cd80 compared with the control (p < 0.001) (Fig. 6E–6H). For these genes, HOM1190 induced significantly higher increases than HOM2120 (p < 0.001 for Akt1 and Pik3ca; p < 0.01 for Il6 and p < 0.05for Cd80). The qRT-PCR results, consistent with the transcriptomic analysis, demonstrate that HOM2120 and HOM1190 exert distinct immunomodulatory effects in RAW264.7 macrophages through differential regulation of key immune signaling pathways. Both strains significantly upregulated RelA and Nfkb1, core components of the NF-κB signaling pathway, accompanied by increased expression of the downstream pro-inflammatory cytokine Tnf. Notably, these effects were more pronounced in HOM2120-treated macrophages, indicating that HOM2120 preferentially activates NF-κB–mediated inflammatory signaling, a central regulatory pathway in innate immune activation and a key driver of downstream cellular immune responses[65]. In addition to NF-κB activation, Stat3 expression was significantly upregulated by both strains, with a stronger induction observed following HOM2120 treatment. STAT3 is a key transcription factor in the JAK–STAT signaling pathway and plays a central role in immune regulation, including macrophage activation and T-helper cell differentiation [66]. Enhanced Stat3 expression has been shown to modulate macrophage functions and to shape adaptive immune responses, suggesting that HOM2120 may further promote macrophage-mediated regulation of adaptive immunity and reinforce its bias toward cellular immune responses. In contrast, the two strains exhibited distinct regulatory patterns within the PI3K–Akt signaling pathway. Both strains significantly upregulated Akt1 and Pik3ca expression in macrophages; however, HOM1190 elicited a stronger induction of these genes compared with HOM2120. PI3K–Akt signaling has been shown to play a central role in regulating macrophage activation and polarization, with differential modulation of this axis affecting cytokine secretion and immune responses [67]. The stronger modulation of this pathway by HOM1190 implies a role in fine-tuning immune activation rather than driving robust pro-inflammatory responses. Consistent with this notion, HOM1190 also induced higher expression of Il6 and the costimulatory molecule Cd80, both of which are involved in antigen presentation and the activation of adaptive immune responses. Costimulatory molecules such as CD80 provide critical second signals for T-cell activation through interaction with CD28, thereby facilitating lymphocyte activation and differentiation[68]. Elevated CD80 expression indicates enhanced macrophage capacity to provide costimulatory signals to lymphocytes, while increased IL6 has been implicated in shaping downstream immune responses and supporting immune cell communication. These observations together suggest that HOM1190 preferentially supports humoral immunity–associated pathways, whereas HOM2120 predominantly enhances NF-κB- and STAT3-mediated cellular immune responses.

Fig. 6. qRT-PCR validation of selected immune-related genes in RAW264.7 macrophages treated with heat-inactivated HOM2120 and HOM1190.

The relative mRNA expression levels of (A) RelA, (B) Nfkb1, (C) Stat3, (D) Tnf, (E) Akt1, (F) Pik3ca, (G) Il6 and (H) Cd80 were determined by qRT-PCR. Data are presented as mean ± SD, n = 3. One-way ANOVA followed by Tukey’s post-hoc test was conducted using GraphPad Prism 9.5 (GraphPad Software). *p < 0.05, **p < 0.01, ***p < 0.001 vs. control.

Overall, these results highlight clear strain-specific immunomodulatory profiles: HOM2120 biases macrophage responses toward NF-κB activation and T-helper cell–related signaling, while HOM1190 modulates PI3K-Akt signaling and costimulatory pathways, contributing to differential regulation of innate and adaptive immunity. These mechanistic differences provide a molecular basis for the distinct immunological effects observed between the two strains.

Immunity Boosting Experiment in Healthy Mice

To validate the distinct immunomodulatory effects observed in vitro, the immune-enhancing activities of HOM2120 and HOM1190 were further examined in vivo. Fig. 7 presents the impact of HOM2120 and HOM1190 on the immunological parameters of healthy mice. Innate immunity was evaluated by carbon clearance and macrophage phagocytosis (A and B), cellular immunity by DTH response and lymphocyte proliferation (C and D), and humoral immunity by HC₅₀ and the number of antibody-producing cells (E and F). Low, medium, and high doses of HOM2120, as well as the low dose of HOM1190, significantly enhanced the carbon clearance capacity in mice compared with the negative control group (p < 0.05; Fig. 7A). Both strains at all tested doses also significantly increased macrophage phagocytosis of chicken red blood cells (p < 0.05; Fig. 7B). HOM2120 significantly enhanced toe swelling induced by sheep red blood cells at all tested doses compared with the normal control group (p < 0.05) (Fig. 7C). In contrast, HOM1190 only showed a slight enhancement of DTH responses, which did not reach statistical significance (p > 0.05). Lymphocyte proliferation was significantly enhanced by both strains in a clear dose-dependent manner (p < 0.05; Fig. 7D). At medium and high doses, HOM1190 significantly increased hemolytic activity (p < 0.05 and p < 0.001, respectively; Fig. 7E), whereas HOM2120 showed an upward trend that did not reach statistical significance (p > 0.05). Similarly, mice receiving the high dose of HOM1190 exhibited a significant increase in APCs (p < 0.05; Fig. 7F), while HOM2120 displayed a non-significant increase (p > 0.05).

Fig. 7. Effect of HOM2120 and HOM1190 on immunological indices in healthy mice.

Innate immunity: (A) Carbon clearance phagocytic index and (B) Macrophage phagocytic index. Cellular immunity (C) DTH-toe swelling thickness (mm) and (D) Lymphocyte proliferation ability (ΔOD). Humoral immunity: (E) Half hemolytic value (HC₅₀) and (F) Number of APCs (×10³/spleen cells). APCs, antibody-producing cells; DTH, delayed-type hypersensitivity. Data are presented as mean ± SEM, n = 12. One-way ANOVA followed by Dunnett’s post-hoc test was conducted using GraphPad Prism 9.5, *p < 0.05, **p < 0.01, ***p < 0.001 vs. control.

Our in vivo results demonstrate that HOM2120 and HOM1190 exert strain-specific immunomodulatory effects in healthy mice. Both HOM2120 and HOM1190 significantly enhanced carbon clearance and macrophage phagocytosis, indicating that both strains possess innate immunomodulatory activity. This aligns with previous reports showing that ingestion of B. lactis Bi07 enhanced phagocytic activity of monocytes and granulocytes, indicating improved immune defense and reduced infection risk [69]. In addition, HOM2120 showed stronger enhancement of DTH responses, indicative of a bias toward cellular immunity. This is consistent with a systematic review and meta-analysis demonstrated that short-term consumption of probiotics enhances cellular immune function, which is responsible for the first line of defense against pathogenic microorganisms in healthy older adults [70]. In contrast, HOM1190 more effectively increased hemolytic activity and APC numbers, reflecting a preference for humoral immunity. This is consistent with previous reports showing that supplementation with Bifidobacterium bifidum OLB6378 simultaneously enhances systemic and mucosal humoral immunity in low-birth-weight infants, as evidenced by increased serum IgG and secretory IgA levels[71].

Taken together, these results indicate that HOM2120 preferentially promotes cellular immunity, whereas HOM1190 favors humoral immune responses, consistent with both in vitro functional assays. Future studies in immunocompromised or disease models may provide further insight into their therapeutic potential, particularly regarding the combined administration of HOM2120 and HOM1190, which may offer complementary immunomodulatory benefits.

Conclusion

In conclusion, B. lactis HOM2120 and B. longum HOM1190 represent two genetically distinct probiotic strains with robust survival capacity and functionally differentiated immunomodulatory properties. Across integrated in vitro, transcriptomic, and in vivo analyses, HOM2120 was consistently biased toward cell-mediated immune activation, whereas HOM1190 preferentially supported humoral immune–associated responses. These complementary immunological profiles provide a mechanistic rationale for their targeted or combined application as functional probiotics. Further studies in disease-relevant models and human clinical trials will be essential to validate efficacy, clarify mechanisms, and support their development for functional food or therapeutic use.

Supplemental Materials

Supplementary data for this paper are available on-line only at http://jmb.or.kr.