Anti-Inflammatory effects of Lactobacillus helveticus and Arthrospira platensis on colonic cells inflamed by Crohn’s disease-associated Escherichia coli

Abstract

Background

Adherent-invasive Escherichia coli (AIEC) is linked to intestinal inflammation in inflammatory bowel disease (IBD). Arthrospira platensis and Lactobacillus helveticus exhibit anti-inflammatory properties individually, yet their effects remain underexplored in IBD-associated inflammation. We aimed to investigate the anti-inflammatory potential of L. helveticus and the hydroalcoholic extract of A. platensis (HA-A. platensis) in Caco-2 cells inflamed by IBD-associated E. coli.

Methods

Caco-2 cells inflamed by a Crohn’s disease (CD)-associated E. coli strain (MOI 10) were treated with HA-A. platensis (2 mg/mL) and/or L. helveticus (MOI 50) in live (LBC), heat-killed (HKC), or cell-free supernatant (CFS) forms. The anti-invasion/adhesion properties of L. helveticus and/or HA-A. platensis were investigated by assessing the CD-associated E. coli invasion/adhesion rate (%). Signaling molecules (NF-κB, STAT3, NOD2) were analyzed via qPCR to capture pathway activation dynamics, while cytokines (TNF-α, IL-1β, IL-8, IL-10) were quantified by ELISA to assess secreted functional proteins.

Results

HA-A. platensis reduced E. coli adhesion by 68% (P < 0.001) and completely inhibited invasion. L. helveticus (live form) decreased adhesion by 88% and invasion by 90%. Combined treatment showed synergistic effects, reducing adhesion by 89% and fully blocking invasion. HA-A. platensis downregulated STAT3 expression by 0.4-fold (P < 0.01), while L. helveticus (heat-killed form) reduced NF-κB by 0.51-fold (P < 0.05) and increased NOD2 by 1.8-fold (P < 0.01). Cytokine analysis revealed that HA-A. platensis decreased IL-1β by 0.61-fold (P < 0.001), and L. helveticus (heat-killed) reduced TNF-α (0.51-fold) and IL-8 (0.23-fold) while elevating anti-inflammatory IL-10 (4.39-fold; P < 0.001).

Conclusions

L. helveticus and HA-A. platensis synergistically inhibit CD-associated E. coli pathogenicity and modulate inflammatory responses in vitro. These findings highlight their potential as adjunctive therapies for CD, warranting further preclinical validation.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12860-025-00545-9.

Introduction

Inflammatory bowel disease (IBD) refers to a cluster of long-term disorders that display an unrestrained and heightened immune response in the gastrointestinal (GI) tract. The development and progression of IBD involve multiple factors, including genetic susceptibility, dysregulated immune reactions, and environmental influences [1, 2]. For a long time, there has been a strong belief that an imbalance in the gut microbiota, known as dysbiosis, is an important factor in the onset and advancement of IBD [3]. The overrepresentation of gut pathobionts, a group of microorganisms believed to have a pathological impact when their interaction with the host is disrupted, has been postulated as a potential factor contributing to the inflammation observed in individuals diagnosed with IBD [4]. Notably, pathobionts have been discovered in the microbiota of healthy individuals without causing any observable diseases, implying that the gut environment plays a crucial role in determining their capacity for pathogenesis [4]. The adherent-invasive Escherichia coli (AIEC), predominantly classified in the B2 and D phylogenetic groups, is the leading pathobiont that has been extensively investigated in the context of IBD [5, 6].

Arthrospira platensis (commonly called spirulina), a nutrient-rich cyanobacterium, exhibits potent immunomodulatory effects. Its bioactive compounds—including phycocyanin, polysaccharides, and polyunsaturated fatty acids—attenuate inflammation via multiple pathways [7, 8]. For instance, phycocyanin suppresses NF-κB activation, reducing TNF-α and IL-6 production in murine colitis models [9]. Researchers have also found that the hydroalcoholic extract of A. platensis has the potential to regulate gut permeability in animal models of IBD induced by dextran sulfate sodium (DSS) [10]. Specifically, A. platensis upregulates tight junction proteins (occludin, ZO-1) and mitigates DSS-induced colitis by lowering myeloperoxidase (MPO) activity. Furthermore, A. platensis has been investigated as a possible means to modulate gut dysbiosis and promote the proliferation of beneficial microorganisms such as Lactobacillus, Bifidobacterium, Akkermansia, and Butyricimonas taxa, while reducing the relative abundance of harmful Proteobacteria and the Firmicutes/Bacteroidetes ratio [11, 12].

Lactobacillus helveticus is a well-characterized probiotic strain with potent anti-inflammatory and gut-protective properties, mediated through multiple mechanisms. It modulates cytokine production by downregulating pro-inflammatory mediators (TNF-α, IL-8) via TLR2/4 inhibition while promoting the secretion of anti-inflammatory IL-10 in dendritic cells [13]. Additionally, it provides competitive exclusion of pathogens like AIEC through both physical blockade of epithelial adhesion sites and production of antimicrobial peptides, including bacteriocins such as helveticin J [14]. It reinforces epithelial barrier integrity in the gut by upregulating mucin-2 and β-defensins in intestinal cells and preserving tight junction proteins during infections [15]. These coordinated actions - immunomodulation, pathogen exclusion, and barrier reinforcement - operate synergistically to reduce intestinal inflammation, making L. helveticus a promising therapeutic candidate for IBD management.

Importantly, scientific research has revealed promising insights into the potential synergistic effects of combining A. platensis as a prebiotic with lactobacilli for the treatment of colitis. This combination has been found to exert modulatory effects on the gut immune response and barrier function by influencing the gut microbiota, fostering the proliferation of beneficial probiotic strains, mitigating inflammation, and augmenting antioxidant activity in animal models [16–18]. Despite individual efficacy, the synergy between A. platensis, as a prebiotic, and L. helveticus, as a probiotic, remains underexplored in AIEC-driven inflammation. A. platensis may enhance L. helveticus survival and function by providing growth substrates (e.g., polysaccharides), while the probiotic could amplify the immunomodulatory effects of A. platensis. This study investigates their combined potential to mitigate AIEC-induced inflammation in Caco-2 cells, offering insights into novel dietary adjuncts for IBD management.

Materials and methods

Preparation of HA-A. platensis and bacterial culture

In this investigation, a 40-gram batch of A. platensis powder was utilized and subjected to homogenization using a solvent mixture composed of ethanol and water in a 60:40 ratio (v/v), with a total volume of 200 mL. The homogenization procedure involved stirring the mixture for a duration of two hours. Following the homogenization process, the resulting supernatants were filtered and meticulously collected. The solvents were subsequently concentrated using a rotary vacuum evaporator and freeze-dried to yield the final freeze-dried HA-A. platensis. The freeze-dried samples were stored at a temperature of 4 °C for subsequent analysis [10].

The strain L. helveticus IBRC-M 11,312, graciously provided by Dr. Baradaran Ghavami from the Basic and Molecular Epidemiology of Gastrointestinal Disorders Research Center at the Research Institute, was cultured in Man, Rogosa, and Sharpe (MRS) broth at 37 °C for 48 h until it reached the late exponential phase with a density of 10⁷ colony-forming units (CFU)/mL. In this study, L. helveticus was employed in three different forms: live bacterial cells (LBC), heat-killed bacterial culture (HKC), and cell-free culture supernatant (CFS). The LBC was prepared from bacterial suspensions that were centrifuged at 10,000 × g for 10 min at 4 °C, washed three times with sterile PBS, and adjusted to a live bacterial concentration of 10⁷ CFU/mL. The HKC was generated by subjecting L. helveticus to heat treatment at 121 °C for 15 min after the cells were washed three times with sterile PBS and adjusted to a cell density of 10⁷/mL. The CFS was obtained by centrifugation of a comparable amount (~ 10⁷ CFU/mL) of L. helveticus cultivated in MRS broth, and then passing the supernatant through a sterile filter with a pore size of 0.22 μm.

A mucosa-associated E. coli strain belonging to phylogroup B2, characterized by its adhesive and invasive properties, was isolated from the ileal biopsy of a patient with active CD in our previous research [19], was propagated in Luria-Bertani (LB) broth (Merck, Germany).

Cell culture conditions

Caco-2 cell line (ATCC^® HTB-37TM), purchased from the Iranian Biological Resource Center, Tehran, Iran, was cultured in 25 cm² culture flasks until they reached confluent growth at a temperature of 37 °C. The culture medium used was Dulbecco’s Modified Eagle’s Medium (DMEM) (Gibco, Life Technologies, Carlsbad, CA, USA), supplemented with 10% inactivated fetal bovine serum (FBS) (Euroclone, Milan, Italy), 1% nonessential amino acid (NEAA), 1% L-glutamine, and 1% penicillin-streptomycin (Gibco Life Technologies, Paisley, UK). In order to maintain optimal conditions, all cultures were incubated at 37 °C in a humidified atmosphere consisting of 5% carbon dioxide (CO₂) and 95% air. When the cells achieved 80–90% confluence, they were dissociated for further passaging using a solution consisting of 0.05% trypsin and 0.02% ethylenediaminetetraacetic acid (EDTA) at a ratio of 1:3.

Cytotoxicity assay of HA-A. Platensis

The assessment of cytotoxicity of HA-A. platensis on Caco-2 cells was carried out through the MTT colorimetric assay. Initially, Caco-2 cells were seeded in a 96-well plate at a density of 62.5 × 10³ cells per well and were allowed to incubate overnight. The cells were then treated with extracts dissolved in a fresh medium for 24 h. HA-A. platensis concentrations (0.125–8 mg/mL) were selected based on published efficacious ranges for intestinal cells [20, 21] and confirmed via pilot viability assays. Subsequently, the cells were exposed to MTT solution at 0.5 mg/mL for 4 h. To solubilize formazan crystals produced by viable cells, the culture medium was replaced with 100 µL of dimethyl sulfoxide (DMSO). Absorbance readings were taken at 570 nm with a reference wavelength of 630 nm. Cell viability was calculated using the formula described previously [22]. For the negative control, cells were solely incubated in a culture medium.

Co-culture of CD-associated E. coli, L. helveticus, and HA-A. Platensis

To assess the potential anti-inflammatory effects of HA-A. platensis and L. helveticus on inflamed colonic cells induced with CD-associated E. coli, Caco-2 cells were cultured in 24-well plates at a concentration of 3 × 10⁶ cells per well and allowed to attach for at least 48 h before the addition of bacteria. Following the incubation period, cells were washed twice with PBS, and then fresh DMEM medium, HA-A. platensis in the concentration of 2 mg/mL or/and L. helveticus at the multiplicity of infection (MOI) 50 (bacteria: cells) were added and incubated at 37 °C in 5% CO₂ for two hours. Subsequently, CD-associated E. coli was added at MOI 10 (bacteria: cells), and incubated at 37 °C in 5% CO₂ for a further four hours. The cells were washed three times with PBS and then fresh DMEM medium was replaced followed by 48 h incubation at 37 °C in 5% CO₂. Moreover, cells treated with sterile MRS medium and untreated cells were designated as the negative control groups. Each trial was conducted in triplicate to guarantee technical reproducibility.

Effect of L. helveticus/HA-A. Platensis on CD-associated E. coli adhesion and invasion

Following treatment, a portion of the cells underwent a triple wash with PBS buffer to eliminate non-adherent bacteria, followed by trypsinization using trypsin–EDTA (0.25% trypsin, 1 mM EDTA). The resulting cell suspension was then subjected to serial tenfold dilutions and spread-plated on LB agar, followed by overnight incubation at 37 °C to enumerate bacterial colonies, specifically focusing on CD-associated E. coli colonies. The anti-adhesion properties of L. helveticus and/or HA-A. platensis were examined through the determination of the CD-associated E. coli adhesion rate (%). This rate was calculated as one minus the ratio of the number of adherent bacteria co-cultured with L. helveticus/HA-A. platensis to the number of adherent bacteria in the control group, multiplied by 100.

A subset of the treated cells was subjected to a triple wash with PBS buffer to eliminate non-adherent bacteria, followed by the addition of 1 mL of gentamicin (200 µg/mL) to eradicate non-invasive E. coli cells. Following this, culture plates were placed in a CO₂ incubator at 37 °C for an additional hour to eradicate adhered non-invasive E. coli cells from the Caco-2 cells. Subsequently, the mixed culture medium was aspirated, and the cells were rinsed twice with PBS buffer, then 1 mL of 1% Triton-X 100 was introduced for a 10-minute incubation period to lyse the Caco-2 cells. Serial dilutions of the lysed Caco-2 cells were spread-plated on LB agar and incubated overnight at 37 °C to enumerate bacterial colonies. The anti-invasion properties of L. helveticus and/or HA-A. platensis were investigated by assessing the CD-associated E. coli invasion rate (%). This rate was derived by subtracting the ratio of invasive bacteria co-cultured with L. helveticus/HA-A. platensis from one, and then multiplying the result by 100.

RNA extraction and quantitative real-time PCR (qPCR) analysis

For assessing the expression levels of NF-kB, STAT3, and NOD2, total RNA extraction from cultured cells was executed utilizing the Total RNA Purification Mini kit (Yekta Tajhiz Azma, Iran) as per the prescribed protocol. The quality of the extracted RNA was evaluated through nanodrop spectrophotometry (by analyzing the 260/280 nm and 260/230 nm ratios). Subsequent to this, reverse transcription (RT-PCR) was implemented employing the 2-step 2X RT-PCR Premix (Taq) kit (BioFact™, South Korea) following the manufacturer’s guidelines. Quantitative real-time polymerase chain reaction (qPCR) was conducted in a total volume of 20 µL utilizing the SYBR Green Master Mix (BioFact™, South Korea) on the LightCycler^® 96 System (Roche Applied Science, Germany). The qPCR cycling parameters included an initial denaturation at 95 °C for 600 s, followed by 40 cycles of two steps: denaturation at 95 °C for 15 s and amplification at 60 °C for 60 s, along with a single fluorescence measurement. For normalizing the expression levels, glyceraldehyde 3-phosphate dehydrogenase (GAPDH) was applied as a housekeeping reference gene in every qPCR run. The primer sequences used in the investigation are disclosed in Table 1. The relative mRNA levels were computed utilizing the 2^−ΔΔCt method (where ΔC_t = C_t target – C_t housekeeping). All samples were analyzed in triplicate to ensure statistical reliability.

Table 1.

Oligonucleotide primers used in this study

Gene	Oligonucleotide Primer Sequence (5’-3’)	Annealing Temperature (C°)	Product Size (bp)	Reference
NF-κB	F: GGATTTCGTTTCCGTTATGTAT R: TCCTTGGGTCCAGCAGTTA	58	233	This study.
NOD2	F: GGGGTTTCGTCAGCCAGTAT R: GAAGGAAGGCAGCCAATCCA	59	123	This study.
STAT3	F: TGGCCCCTTGGATTGAGAGT R: ATTGGCTTCTCAAGATACCTGCT	60	196	This study.
GAPDH	F: TGCCTCCTGCACCACCAAC R: CGGAGGGGCCATCCACAG	60	131	This study.

Quantification of cytokine concentrations

Upon stimulation, cell culture supernatants (conditioned media) were collected to determine the levels of TNF-α, IL-1β, IL-8, and IL-10, enzyme-linked immunosorbent assay (ELISA) colorimetric kits (Carmania Parsgen Co, Iran) were employed, following the guidelines provided by the manufacturer.

Statistics

All experiments were performed in triplicate (n = 3 biological replicates). Data are presented as mean ± standard deviation (SD). Normality was confirmed using Shapiro-Wilk tests. For comparisons between multiple groups, one-way or two-way ANOVA (for combination treatments) with Tukey’s post-hoc test was applied, as appropriate. Student’s t-test was used for pairwise comparisons when justified. Synergistic effects were analyzed via two-way ANOVA with interaction terms, comparing observed combination effects to expected additive effects. For qPCR data, the 2^−ΔΔCt method was used with GAPDH normalization. All statistical tests were two-tailed, with P < 0.05 considered significant. Analyses were performed using SPSS v22 (IBM) and GraphPad Prism v6.07.

Results

Influence of HA-A. platensis on the cell viability and proliferation

The impact of HA-A. platensis on Caco-2 cell viability was evaluated over a 48-hour duration using the MTT assay. No significant decline in cell proliferation rates was observed across concentrations ranging from 0.125 to 2 mg/mL of HA-A. platensis. Impressively, concentrations up to 4 mg/mL maintained cell viability above 90%. Nonetheless, noticeable reductions in cell proliferation rates were noted at concentrations of 4 and 8 mg/mL compared to the untreated control group, resulting in viable cell numbers decreasing to 73% and 66%, respectively. Consequently, for subsequent experiments, a concentration of 2 mg/mL HA-A. platensis was selected for use.

Influence of L. helveticus and HA-A. Platensis on the adhesion and invasion of CD-associated E. coli

Both L. helveticus and HA-A. platensis exhibited significant, form-dependent inhibition of CD-associated E. coli adhesion (Table 2). LBC showed the strongest anti-adhesion effect (87.9 ± 3.1% inhibition, P < 0.001 vs. control), followed by HKC (62.8 ± 5.3%, P < 0.001) and CFS (12.3 ± 3.7%, P = 0.023). HA-A. platensis alone demonstrated 68.2 ± 4.8% adhesion inhibition (P < 0.001). Combination therapies showed significant synergy over individual treatments (P < 0.05), particularly LBC + HA-A. platensis (94.8 ± 2.1% inhibition, synergy P < 0.01).

Table 2.

Anti-adhesion and anti-invasion effects of L. helveticus forms and HA-A. Platensis against CD-associated E. coli in Caco-2 cells

Treatment Group	Adhesion Inhibition (% ± SD)†	P-value (Adhesion)	Invasion Inhibition (% ± SD)†	P-value (Invasion)	Synergy P-value#
E. coli control	0 ± 0	-	0 ± 0	-	-
HA-A. platensis	68.2 ± 4.8	< 0.001‡	100 ± 0	< 0.001‡	-
L. helveticus (LBC)	87.9 ± 3.1	< 0.001‡	89.7 ± 3.8	< 0.001‡	-
L. helveticus (HKC)	62.8 ± 5.3	< 0.001‡	84.6 ± 4.2	< 0.001‡	-
L. helveticus (CFS)	12.3 ± 3.7	0.023‡	39.8 ± 5.1	0.002‡	-
LBC + HA-A.p	94.8 ± 2.1	< 0.001§	100 ± 0	< 0.001§	< 0.01
HKC + HA-A.p	81.7 ± 3.5	< 0.001§	100 ± 0	< 0.001§	< 0.05
CFS + HA-A.p	70.9 ± 4.2	< 0.001§	100 ± 0	< 0.001§	< 0.01

Abbreviations LBC: Live bacterial cells; HKC: Heat-killed culture; CFS: Cell-free supernatant; HA-A.p: Hydroalcoholic extract of A. platensis

Footnotes

^†Mean ± standard deviation of three independent experiments (n = 3)

^‡P-value versus control (one-way ANOVA with Tukey’s post-hoc)

^§P-value versus corresponding single treatment (two-way ANOVA)

^#Synergistic effect for both adhesion and invasion inhibition (two-way ANOVA interaction term)

For invasion assays, HA-A. platensis alone achieved complete inhibition (100%, P < 0.001), while L. helveticus forms showed differential efficacy: LBC (89.7 ± 3.8%, P < 0.001) > HKC (84.6 ± 4.2%, P < 0.001) > CFS (39.8 ± 5.1%, P = 0.002). All combinations with HA-A. platensis maintained complete invasion blockade (100%, P < 0.001 vs. control), demonstrating significant synergy versus individual components (P < 0.01 for LBC/CFS combinations; P < 0.05 for HKC combination) (Table 2).

Effect of L. helveticus and HA-A. Platensis on NF-κB, NOD2, and STAT3 expression levels in CD-associated E. coli-treated cells

To assess the efficacy of HA-A. platensis and various forms of L. helveticus in alleviating the inflammatory responses induced by CD-associated E. coli in Caco-2 cell models, the expression levels of NF-κB, NOD2, and STAT3 were analyzed. Our findings revealed that CD-associated E. coli amplified the mRNA expression levels of these markers in Caco-2 cells. Interestingly, while HA-A. platensis did not significantly influence the levels of NF-κB and NOD2 (refer to Fig. 1A and B); it demonstrated a notable reduction in STAT3 expression in Caco-2 cells stimulated by CD-associated E. coli (Fig. 1C). Treatment of CD-associated E. coli-stimulated Caco-2 cells with HKC, LBC, and CFS individually, as well as combinations of them with HA-A. platensis, led to decreased NF-κB expression and increased NOD2 expression. Furthermore, the combination of CFS and HA-A. platensis resulted in a reduced level of STAT3 in CD-associated E. coli-stimulated Caco2 cells.

Fig. 1.

The relative expression levels of the NF-κB (A), NOD2 (B), and STAT3 (C) genes in Caco-2 cells induced by CD-associated E. coli and treated with various formulations of HA-A. platensis and L. helveticus. Fold change was calculated by comparing gene expression in cells subjected to different treatments with that in cells inflamed by CD-associated E. coli, serving as the control group. GAPDH served as an internal control for gene expression normalization. Data analysis was conducted using Student’s t-test. Mean values are provided, with error bars representing the standard deviation (SD) of triplicate data. Significance levels are indicated as follows: *, **, ***, and NS denote P < 0.05, P ≤ 0.01, P ≤ 0.001, and not significant, respectively. Abbreviations CFS, cell-free supernatant; HA-A.p, hydroalcoholic extract of A. platensis; HKC, heat-killed culture; LBC, live bacterial cells; NS, not significant

L. helveticus and HA-A. Platensis modulate CD-associated E. coli-induced production of TNF-a, IL-1β, IL-8 and IL-10

To elucidate the anti-inflammatory properties of the L. helveticus and HA-A. platensis on CD-associated E. coli treated Caco-2 cells, the release of TNF-α, IL-1β, IL-8, and IL-10 was assessed using ELISA. As depicted in Fig. 2, untreated Caco-2 cells exhibit minimal production of TNF-α (10.61 ± 3.52 pg/mL), IL-1β (28.13 ± 4.46 pg/mL), IL-8 (11.17 ± 4.07 pg/mL), and IL-10 (3.45 ± 1.97 pg/mL). Upon a four-hour exposure to CD-associated E. coli at an MOI of 10, a notable increase was observed in the production of TNF-α, IL-1β, IL-8, and IL-10, yielding concentrations of 84.43 ± 14.08 pg/mL, 132.3 ± 13.65 pg/mL, 202.7 ± 16.5 pg/mL, and 11.62 ± 3.41 pg/mL, respectively. The concentration of IL-1β in CD-associated E. coli-stimulated cells treated with HA-A. platensis exhibited a significant decrease by 0.61-fold (P < 0.001) (Fig. 2). Nonetheless, no substantial alterations were noted in the levels of TNF-α, IL-8, and IL-10 following exposure to 2 mg/mL of HA-A. platensis in inflamed Caco-2 cells induced by CD-associated E. coli.

Fig. 2.

The release of TNF-α (A), IL-1β (B), IL-8 (C), and IL-10 (D) in Caco-2 cells induced by CD-associated E. coli and treated with various formulations of HA-A. platensis and L. helveticus, as evaluated using ELISA. Error bars represent the standard deviation (SD) of triplicate data. Student’s t-test was used for data analysis. *, **, ***, ****, and ns indicate P < 0.05, P < 0.01, P < 0.001, P < 0.0001, and not significant, respectively. Abbreviations CFS, cell-free supernatant; HA-A.p, hydroalcoholic extract of A. platensis; HKC, heat-killed culture; LBC, live bacterial cells; NS, not significant

Remarkably, both L. helveticus alone and in conjunction with HA-A. platensis, administered through diverse treatment modalities, significantly reduced the concentration of inflammatory cytokines TNF-α, IL-1β, and IL-8, as evidenced by the fold changes outlined in Supplementary Table S1. However, our findings indicate that HKCs of L. helveticus alone may yield a more significant effect in reducing IL-8 levels, with a reduction of 0.23-fold change. Conversely, when used in combination with HA-A. platensis, including HKC + HA-A. platensis (0.37-fold change reduction), LBC + HA-A. platensis (0.34-fold change reduction), and CFS + HA-A. platensis (0.5-fold change reduction), the reduction in IL-8 levels appears to be less pronounced. Finally, both HKCs of L. helveticus alone and in combination with HA-A. platensis resulted in a notable elevation in IL-10 concentration, as an anti-inflammatory cytokine. Figure 2 delineates the alterations in cytokine concentration subsequent to treatment with various modalities of HA-A. platensis and L. helveticus.

Discussion

Our research has yielded valuable insights into the anti-adhesion, anti-invasive, and antiiflamatory properties of L. helveticus and HA-A. platensis against CD-associated E. coli in Caco-2 cell monolayers. Both L. helveticus and HA-A. platensis exhibited significant reductions in the adhesion and invasion of CD-associated E. coli, with synergistic effects apparent when used together. Specifically, various formulations of L. helveticus led to notable decreases in colony counts and adhesion capabilities of CD-associated E. coli, while HA-A. platensis demonstrated pronounced anti-adhesion effects and complete suppression of invasion and growth of CD-associated E. coli within Caco-2 cells. The combination of HA-A. platensis with different forms of L. helveticus further potentiated these effects, emphasizing the therapeutic potential of probiotic blends in mitigating bacterial adhesion and invasiveness associated with CD.

Our investigation reveals that the exposure of Caco-2 cells to CD-associated E. coli elicited a notable elevation in levels of NF-κB, NOD2, and STAT3. Treating E. coli-stimulated cells using HKC, LBC, and CFS of L. helveticus, individually and in combination with HA-A. platensis, resulted in a reduction in NF-κB levels and an increase in NOD2 levels. The NF-κB signaling pathways play a pivotal role in the initiation and progression of intestinal inflammation [23]. Furthermore, research suggests that NOD2 activation triggers heightened expression of interferon regulatory factor 4 (IRF4), facilitating its interaction with tumor necrosis factor receptor-associated factor 6 (TRAF6) and receptor-interacting serine-threonine kinase (RICK). This interaction leads to IRF4-mediated suppression of Lys63-linked polyubiquitination of TRAF6 and RICK, consequently dampening NF-κB activation. Ultimately, NOD2 contributes to mitigating colonic inflammation by virtue of IRF4-mediated inhibition of K63-linked polyubiquitination of RICK and TRAF6 [24].

Following treatment with HA-A. platensis, a discernible reduction in STAT3 levels was observed in E. coli-stimulated cells. The JAK/STAT pathway assumes diverse roles in cytokine modulation, yet delineating the precise alterations in STAT functions during IBD proves intricate [25]. For example, while STAT3 is implicated in the signaling of certain anti-inflammatory cytokines such as IL-21, its upregulation may confer anti-inflammatory effects, thereby potentially ameliorating IBD symptoms [26]. Conversely, there exists literature suggesting that activated STAT3 could incite inflammation, potentially precipitating the onset of CD and UC [27]. Consequently, specific components of the STAT pathway, including STAT3, exhibit associations with a spectrum of cytokines and may harbor dual roles in regulating inflammation through cytokine modulation.

Notably, the differential STAT3 modulation observed between HA-A. platensis alone and combination treatments can be attributed to various interrelated mechanisms. Phycocyanin in the hydroalcoholic extract of phycocyanin could directly inhibit STAT3 phosphorylation via JAK2 kinase blockade, as well as upregulation of SOCS3 [28]. While this effect predominates in HA-A. platensis-only treatments, it may be partially attenuated in combination therapies due to competing probiotic signaling pathways. Additionally, probiotics and their metabolites could stimulate compensatory IL-6 production, reactivating STAT3 via JAK-STAT signaling [29]. Furthermore, probiotic-derived metabolites might shift signaling bias, further diminishing the overall STAT3 modulation.

IL-1β and TNF-α stand out as crucial immunoregulatory cytokines, pivotal in amplifying the inflammatory response and mediating tissue damage in IBD [30, 31]. This activation prompts the induction of epithelial cell damage, edema, neutrophil infiltration, and stimulates the proliferation of intestinal smooth muscle cells and fibroblasts [30, 31]. Correspondingly, the observed reduction in IL-1β levels following treatment with HA-A. platensis may exert a favorable influence on controlling inflammation induced by CD-associated E. coli in Caco-2 cells. Yu et al. [12] have highlighted the significant reduction in intestinal inflammation attributed to A. platensis, evidenced by diminished expression of myeloid differentiation factor 88 (MyD88), toll-like receptor 4 (TLR4), NF-κB, and inflammatory cytokines. Moreover, A. platensis has demonstrated efficacy in enhancing gut health by ameliorating chronic inflammation, rectifying gut dysbiosis, and modulating gut permeability in rats afflicted with gut dysbiosis induced by a high-fat diet [12]. Jahani-Sherafat et al. [32] have underscored the efficacy of A. platensis in mitigating inflammation triggered by gliadin in gliadin-stimulated Caco-2 cells, evidenced by reduced expression of IL-6 and NF-κB. Furthermore, research by Zhu et al. [33] has elucidated the therapeutic potential of selenium-containing phycocyanin derived from selenium-enriched Arthrospira in attenuating dextran sodium sulfate-induced IBD colitis in mice. This treatment regimen led to decreased levels of pro-inflammatory cytokines such as IL-6, TNF-α, and monocyte chemotactic protein-1 (MCP-1), coupled with an augmentation in anti-inflammatory cytokines like IL-10.

Furthermore, the application of HKC, LBC, and CFS of L. helveticus, either independently or in conjunction with HA-A. platensis, led to a decrease in IL-8, IL-1β, and TNF-α levels, coupled with an increase in IL-10 levels. Notably, HKC alone and the combination of HKC with HA-A. platensis exhibited superior efficacy compared to other combinations, manifesting more pronounced effects on cytokine levels. This outcome resonates with established probiotic advantages, wherein probiotic bacteria can augment the innate immunity of the host and regulate the production of inflammatory cytokines like IL-8 and TNF-α. IL-10, an anti-inflammatory cytokine, assumes a critical role, as heightened IL-10 levels (alongside IL-6) can inhibit TNF-α production [34]. A separate investigation pinpointed L. helveticus NS8 as a beneficial strain, showcasing its ability to promote the production of the anti-inflammatory cytokine IL-10 [35]. This strain exhibited resilience within the gut environment, modulated the immune response, and conferred protective benefits against colitis in mouse models. Additionally, research on L. helveticus MIMLh5 and its surface-layer protein (SlpA) revealed their anti-inflammatory properties by reducing NF-κB activation [36]. These findings suggest potential therapeutic applications for these probiotic strains in managing inflammatory conditions.

The differential effects on cytokine modulation versus pathogen inhibition can be explained by three interrelated mechanisms. First, HA-A. platensis achieves near-complete E. coli eradication (100% invasion inhibition), creating a biological ceiling that limits further cytokine reduction. Second, the agents target distinct pathways—L. helveticus primarily could modulate TLR/NF-κB (reducing TNF-α/IL-8), while HA-A. platensis may act via STAT3/IL-10, leading to compensatory rather than additive effects when combined. Notably, HKC showed superior cytokine modulation due to preserved cell wall components (e.g., lipoteichoic acids) and lack of metabolic competition with HA-A. platensis compounds. These findings suggest a sequential therapeutic approach, including initial pathogen clearance by HA-A. platensis, followed by immunomodulation with L. helveticus, may optimize anti-inflammatory outcomes.

Our study has some limitations that warrant consideration. The monoculture system lacks microbiome-immune interactions critical to IBD pathogenesis. Moreover, our use of undifferentiated Caco-2 cells (48–72 h culture) may not fully replicate the complex barrier functions of mature enterocytes. While this model is validated for bacterial adhesion studies, findings should be confirmed in fully differentiated monolayers (18–21 days), co-culture models with immune cells (e.g., THP-1 macrophages), or patient-derived organoids. Furthermore, the study focused on one E. coli strain (phylogroup B2). While clinically relevant, results may not generalize to other IBD-associated pathobionts (e.g., Klebsiella). These findings thus require validation in complex models incorporating immune components and diverse microbial communities.

In summary, our study indicates that the combined treatment of Caco-2 cells with the HKC form of L. helveticus and HA-A. platensis shows promise in alleviating inflammation triggered by CD-associated E. coli. This combined approach suggests a supplementary therapeutic strategy for managing inflammation in IBD by targeting key inflammatory markers and modulating cytokine production. However, the study’s reliance on in vitro assays presents a limitation. Further research using animal models and human trials is essential to evaluate the safety and elucidate the underlying mechanisms for clinical application.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

H.H. and L.P. contributed significantly to the conception and design of the study, which involved project development and securing funding. H.H. and L.P. also played pivotal roles in reviewing and refining the manuscript for critical intellectual content and have provided approval for the final version for submission. S.A. and H.H. were involved in developing the experimental procedures and methodology. N.A. and M.R.N. contributed to data analysis, interpretation of findings, and drafting the article. All authors have diligently reviewed and endorsed the final manuscript version and consent to being listed as authors.

Funding

Financial support for this study was provided by the Celiac Disease and Gluten-Related Disorders Research Center, the Research Institute for Gastroenterology and Liver Diseases, affiliated with Shahid Beheshti University of Medical Sciences in Tehran, Iran, under grant number 43008075.

Data availability

All data generated or analyzed during this study are included in this published article.

Declarations

Ethics approval and consent to participate

This study was reviewed and approved by the Institutional Ethical Review Committee of the Research Institute for Gastroenterology and Liver Diseases at Shahid Beheshti University of Medical Sciences (No IR.SBMU.RETECH.REC.1402.645).

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.