Deciphering the Role of Postbiotics Derived from Bacillus subtilis natto on LPS-Induced Endothelial Cell Dysfunction

Abstract

Background: This study aimed to assess the effects of postbiotic derived from Bacillus subtilis natto (Szendi2020) on endothelial responses under LPS-induced inflammatory stress. Methods: In human umbilical vein endothelial cells (HUVECs), inflammation was induced with 200 ng/mL LPS. Cell viability, apoptosis, and mitochondrial integrity were assessed using MTT assay, DiIC, and Sytox Green permeability assays. Intracellular ROS levels, heat shock proteins (HSPB1/Hsp27, HSPA1L/Hsp70), adhesion molecules (ICAM-1, VCAM-1), tight junction protein (Occludin), transcription regulators (NF-κB, TNFα), and proinflammatory cytokines (IL-1β, IL-6, IL-8) were quantified using qPCR and ELISA. Results: LPS exposure significantly induced apoptosis in HUVECs, as reflected by decreased metabolic activity, decreased mitochondrial membrane potential, and increased cell death (p < 0.05). Concurrent postbiotic administration completely abolished LPS-induced cytotoxicity in all assay platforms, demonstrating a potent cytoprotective effect. Postbiotic treatment significantly reduced LPS-induced ROS accumulation (p < 0.05). LPS significantly increased Hsp27 and Hsp70 mRNA expression. However, combined LPS and postbiotic exposure mitigated Hsp27 and Hsp70 mRNA expression compared with LPS treatment alone (p < 0.001, p < 0.005). Postbiotic treatment also decreased the upregulation of adhesion molecules induced by LPS. Although this effect decreased after 24 h (p < 0.001). LPS strongly increased NF-κB, IL-1β and TNFα mRNA levels and was suppressed by postbiotics at early time points but not maintained over 24 h. Importantly, postbiotics significantly reduced IL-6, and IL-8 expression at both the mRNA and protein levels, highlighting the attenuation of endothelial inflammatory features (p < 0.05, p < 0.005, p < 0.001). Conclusions: Our results are the first to demonstrate that postbiotics derived from Bacillus subtilis natto (Szendi2020) exert potent cytoprotective and anti-inflammatory effects in LPS-induced endothelial inflammation. By reducing ROS accumulation, preventing apoptosis, stabilizing mitochondrial and barrier integrity, modulating HSP, NF-κB, and cytokine responses. Postbiotics may be promising therapeutic candidates for alleviating endothelial inflammation and the resulting endothelial dysfunction.

1. Introduction

Cardiovascular diseases (CVD), such as stroke, hypertension, atherosclerosis, and ischemic heart disease account as leading causes of death worldwide [1]. The prevalence of CVD continues to rise due to factors such as aging populations, increasing rates of obesity, physical inactivity, and the widespread prevalence of metabolic disorders like diabetes and hyperlipidemia [2]. Hence, there is a need to find effective, complementary strategies to traditional therapies for treating CVDs. Atherosclerosis underlies the central pathophysiological mechanism underpinning CVD and its spectrum of vascular manifestations, including myocardial infarction and acute stroke [3]. A critical component in the pathogenesis of atherosclerosis is endothelial dysfunction (ED), which is often described as an early and reversible step in the progression of vascular diseases [4,5,6].

The endothelium, a dynamic monolayer of cells lining the blood vessels, makes direct contact with the blood and serves as a regulator of vascular homeostasis [7]. However, several stressors such as oxidative stress, hyperglycemia, hyperlipidemia, and exposure to environmental toxins disrupt vascular homeostasis leading to ED, when the endothelium loses its protective functions, setting the stage for vascular damage and disease [8]. Activated endothelial cells upregulate pro-inflammatory cytokines, chemokines, and adhesion molecules [9]. Expression of adhesion molecules appears to be an early event in ED pathogenesis, changes initiate the recruitment and adhesion of immune cells to the vascular wall, amplifying inflammation and further compromising endothelial function [10]. During atherosclerosis, the balance between oxidative processes and the concomitant antioxidant mechanisms is disrupted, leading to the generation of reactive oxygen species (ROS) [11,12].

These molecules trigger oxidative stress, which reduces the bioavailability of endothelial-derived nitric oxide (NO), thereby impairing vascular relaxation and enhancing vascular wall inflammation [13]. The reduced bioavailability of NO results an enhance permeability to macromolecules at small arterial bifurcations, which guides to subendothelial lipid deposition, oxidation and aggregation [14,15]. Adhesion molecules including VCAM-1 ICAM-1 and E-selectin secrete on the endothelial surface initiate recruitment of circulating nanoparticles [13,16,17]. Furthermore, pro-inflammatory cytokine production IL-1α, IL-1β, IL-6, IL-12, IL-15, IL-18, and tumor necrosis factor-α (TNF-α), or anti-inflammatory cytokine IL-10 of endothelial cells were also a hallmark of vascular diseases [4].

A growing body of evidence shows crucial crosstalk between gut microbiome and distal organs through the well-described gut–liver, gut–kidney, and gut–brain axes [18,19,20]. Furthermore, the gut microbiome and its metabolites also have a great impact on our vascular health. Gut microbiota releases several bioactive metabolites, such as short-chain fatty acids (SCFAs) that can be transported to systemic circulation. SCFAs are able to induce the activation of G-protein coupled receptor 43 (GPR43 or FFA2) and G-protein coupled receptor 41 (GPR41 or FFA3) expressed on endothelial cells as well as vascular smooth muscle cells [21,22]. These receptors can prevent the activation of nuclear factor kappa-light-chain-enhancer of activated B cell (NF-κB) and the consequent production of pro-inflammatory cytokines [23]. Furthermore, the activation of FFA3 may trigger the biosynthesis and the bioavailability of endothelial NO, through the contribution of the Ca²⁺ dependent eNOS. According to previous studies, indole, which is a bacterial metabolite, exerts antioxidant effects, modulating the Nrf2 pathway and reducing reactive oxygen species production [24].

These mechanisms raise the possibility that gut microbiota-derived metabolites may represent potential alternative strategies for the prevention or treatment of CVD, via anti-inflammatory and antioxidant effects.

Postbiotics are identified as inanimate microorganisms or their cellular structures or metabolites that confer health benefits to the host and have been noted as a promising alternative strategy for gut dysbiosis and vascular diseases through the gut–vascular axis. Postbiotics poses several beneficial properties, including antioxidant and anti-inflammatory activities, which contribute to promote vascular health and homeostasis [25].

Taken together, these findings suggest that postbiotic treatment may mitigate endothelial inflammation—an underlying mechanism of endothelial dysfunction—and thereby promote vascular health and homeostasis. In this study, postbiotics (PostB) from a Bacillus subtilis natto (Szendi2020)-inoculated fermentation product were investigated on lipopolysaccharide (LPS)-induced inflammation in human umbilical vein endothelial cells (HUVECs).

Based on these, our study pursued several objectives: (i) to determine the biological effective dose of postbiotics; (ii) to examine whether postbiotics treatment could potentially inhibit the LPS-induced ROS production of HUVECs; (iii) to investigate heat shock protein (Hsp70 and Hsp27) expression profiles in both LPS and postbiotics treatments as well as using them in combination; (iv) to test/elucidate how postbiotics treatments affect the LPS-induced inflammation of HUVECs through pro-inflammatory cytokines levels. To the best of our knowledge, this is the first scientific research where postbiotics were used to mitigate endothelial inflammation.

2. Materials and Methods

2.1. Bacterial Strains and Preparation of Postbiotics

Bacillus subtilis is a Gram-positive, catalase-positive bacterium commonly found in soil. In this study, Bacillus subtilis natto Szendi 2020 (Szendi 2020 NCAIM Bizt 614/2024) postbiotics bacterial strains were cultured at 37 °C for 24 h with TSB (Biolab ZRT). The postbiotics used in this study were cell-free fractions of a collected probiotic growth medium. Bacterial culture media was centrifuged at 5000× g for 10 min. Next, the supernatant was filtered through a 0.2 micrometer filter.

2.2. Extraction of Surfactin Produced by Bacillus subtilis natto Culture

Sample preparation and purification were carried out according to the method described by Beata Koim-Puchowska et al. [26]. For the qualitative and quantitative evaluation of surfactin isoforms, affinity chromatographic extraction was performed using a solid-phase extraction (SPE) system with Bond Elut LRC-C18 columns (Agilent Technologies, Santa Clara, CA, USA), which are suitable for the retention of non-polar compounds. After removing the bacterial biomass by centrifugation (2400× g, 15 min, 25 °C), surfactin was extracted from the culture supernatant. Prior to extraction, the SPE columns were conditioned with methanol and equilibrated with distilled water following the manufacturer’s instructions. Subsequently, 20 mL of the culture medium was applied to the column, which was then washed with 5% methanol to remove impurities. The surfactin fraction was eluted using HPLC-grade methanol, and the eluate was filtered through a 0.22 µm membrane filter prior to chromatographic analysis. Quantitative determination was performed using the external standard method (ESTD) with an ethanolic surfactin standard solution derived from Bacillus subtilis (Sigma-Aldrich, Darmstadt, Germany).

Surfactin analysis was performed by high-performance liquid chromatography (HPLC) using a Waters Alliance e2695 Separation Module (Waters, Milford, MA, USA) equipped with a Waters 2998 photodiode array (PDA) detector (Figure 1). Data acquisition and processing were carried out with Waters Empower 3 software. Chromatographic separation was achieved under isocratic conditions on an XSelect HSS C18 column (150 × 4.6 mm, 5 µm; Waters Corporation, Milford, Ireland). The mobile phase consisted of 80% acetonitrile (eluent A) and 20% 3.8 mM trifluoroacetic acid in water (eluent B), at a constant flow rate of 1 mL min⁻¹. The column temperature was maintained at 30 °C, and 20 µL of each sample—filtered through a 0.22 µm PTFE syringe filter (Filter-Bio, Nantong, China)—was injected for analysis. Detection was carried out at wavelengths of 205 and 210 nm using the PDA detector.

2.3. Determination of Surfactin Profile with UHPLC-MS Method

Surfactin analysis was assessed by the UHPLC system (DionexUltimate3000RS) coupled to a Thermo Q Exactive Orbitrap mass spectrometer (Thermo Fisher Scientific Inc., Waltham, Massachusetts, USA) equipped with an electrospray ionization source (ESI). The HPLC separation was performed on a Themo Accucore C18 column (100 mm 2.1 mm 2.6 m). Sampler and oven temperature were maintained at 25 °C; the flow rate was 200 µL min⁻¹. Eluent A was water containing 0.1% formic acid and eluent B was acetonitril containing 0.1% formic acid. The following gradient elution program was applied: 0 min, 95% A; 0–3 min, 0% A; 3–11 min, 95% A; 11–12 min, 95% A; 12–20 min. A total of 2 µL of the sample was injected. The Q Exactive hybrid quadrupole–Orbitrap mass spectrometer was operated under the following conditions: capillary temperature of 320 °C and a spray voltage of 4.0 kV in positive ionization mode. The MS1 resolution was set to 35,000, with a scanned mass range of 150–1500 m/z and a maximum injection time of 100 ms. For MS/MS (MS2) scans, the resolution was set to 17,500, and the normalized collision energy was 35. Sheath gas and auxiliary gas flow rates were set to 32 and 7 arbitrary units, respectively. Data acquisition and analysis were performed using Xcalibur 4.0 software (Thermo Fisher Scientific Inc., Waltham, MA, USA).

2.4. Determination of K₂-MKn Vitamin Profile with HPLC Method

Sample preparation was performed according to the procedure described by Sato et al. [27]. Five milliliters of PostB were measured, followed by the addition of 6 mL of 2-propanol and 12 mL of n-hexane. The mixtures were stirred for 10 min and subsequently subjected to ultrasonic treatment in a water bath for 10 min to improve the efficiency of solvent extraction. The sample was then centrifuged at 10,000 rpm for 5 min to promote phase separation. The upper hexane layer was collected and evaporated to dryness using a rotary vacuum evaporator. Prior to chromatographic analysis, the dried residues were reconstituted in acetonitrile, centrifuged again at 10,000 rpm for 5 min, and the resulting supernatants were transferred into 1.5 mL HPLC sample vials for subsequent analysis. The measurements were carried out using an Alliance 2695 HPLC system (Waters) equipped with a PDA 2998 detector (Waters), operating at a detection wavelength of 248 nm. Data acquisition and processing were conducted using the Empower software. Separation was achieved by gradient elution according with the following solvent program: 0 min, 80% A; 0–3.5 min, 80% A; 3.5–4 min, 100% A; 4–6.5 min, 80% A; 6.5–10 min, 100% A; 10–15 min. The mobile phase consisted of eluent A: acetonitrile and eluent B: methanol–water (50:50, v/v; adjusted to pH = 3 with phosphoric acid). The flow rate was set to 1.2 mL/min, and the total run time was 15 min. Chromatographic separation was performed on a Nucleosil C18 column (5 µm, 125 × 4 mm; Phenomenex, San Juan, PR, USA). The injection volume was 20 µL. Compound identification was based on retention times and comparison with literature data, while quantification was expressed as mg/L for the bacterial cultures.

2.5. Cell Culture Conditions

Immortalized human umbilical vein endothelial cells (HUVECs/TERT) obtained from ATCC (Manassas, VA, USA) were applied for postbiotic treatments. Cells were cultured according to the manufacturer’s instructions. HUVECs were maintained in M199 medium supplemented with 10% heat-inactivated fetal bovine serum (FBS), 1% penicillin–streptomycin, 1% amphotericin B, 2 mM glutamine, and Endothelial Cell Growth Medium-2 (EGM-2). Cultures were incubated at 37 °C in a humidified atmosphere with 5% CO₂. The culture medium was replaced every 48 h until cells reached 80–90% confluence. At confluence, cells were either subcultured or used for experiments. All experiments were performed using cells at passage 16. The complete medium described above served as the control condition. Prior to seeding, culture surfaces were coated with 0.1% gelatin to enhance cell adhesion. To establish the inflammatory model, lipopolysaccharide (LPS; eBioscience, San Diego, CA, USA) was added to M199 medium at a final concentration of 200 ng/mL.

2.6. Cell Viability Measurements

Cell viability was assessed using the MTT mitochondrial assay. HUVECs were seeded into 96-well plates at a density of 2 × 10⁴ cells/well. After reaching approximately 90% confluence, cells were treated with various concentrations of postbiotic (PostB) (19, 3.8, 1.9, 0.19, and 0.0095 mg/mL), either alone or in combination with LPS (200 ng/mL), for 24 or 48 h. Following treatment, cells were incubated with MTT solution (0.5 mg/mL) for 3 h to allow the formation of formazan crystals proportional to cell viability. After that, the crystals were dissolved in 100 µL/well of solubilizing solution consisting of 81% (v/v) isopropyl alcohol (Serva, Heidelberg, Germany), 10% (v/v) Triton X-100 (Serva, Heidelberg, Germany), and 9% (v/v) 1 M hydrochloric acid (HCl; Serva, Heidelberg, Germany). Clariostar microplate reader (BMG Labtech, Ortenberg, Germany) was used to measure the absorbance at 465 nm. Data are expressed relative to the control group, which was defined as 100%.

2.7. Evaluation of Apoptotic Cell Death

To monitor early apoptotic events, mitochondrial membrane potential in HUVECs was assessed using DilC1(5) (1,1′,3,3,3′,3′-hexamethylindodicarbocyanine iodide). Cells were seeded in 96-well plates at a density of 2 × 10⁴ cells/well and allowed to reach approximately 90% confluence. They were then treated with various concentrations of postbiotic (PostB) (19, 3.8, 1.9, 0.19, and 0.0095 mg/mL), either alone or in combination with LPS (200 ng/mL), for 24 or 48 h. Following treatment, culture supernatants were removed, and cells were incubated with DilC1(5) working solution (50 µL/well) for 30 min. Afterward, cells were washed twice with PBS, and fluorescence was measured at 630 nm excitation and 670 nm emission using a Clariostar microplate reader (BMG Labtech, Ortenberg, Germany). Data were expressed as fluorescence intensity normalized to the control group.

2.8. Evaluation of Necrotic Cell Death

Necrotic cell death was analyzed using SYTOX Green staining. The dye selectively penetrates necrotic cells with disrupted membranes and binds to intracellular nucleic acids, whereas viable cells with intact membranes show minimal staining. HUVECs were seeded into 96-well plates at a density of 2 × 10⁴ cells/well and allowed to reach approximately 90% confluence. Cells were then treated with various concentrations of postbiotic (PostB) (19, 3.8, 1.9, 0.19, and 0.0095 mg/mL) for 24 or 48 h. Following treatment, the medium was removed and cells were incubated with SYTOX Green (1 µM in Dulbecco’s modified Eagle’s medium, 50 µL/well) for 30 min, then washed with PBS. Fluorescence was measured using a Clariostar microplate reader (BMG Labtech, Ortenberg, Germany) at 490 nm excitation and 520 nm emission. Results were expressed as fluorescence intensity normalized to the control group.

2.9. Measurement of Intracellular ROS (Reactive Oxygen Species) Production on HUVEC

Intracellular ROS productions were measured using CM-H2DCFDA (chloromethyl-2′,7′-dichlorodihydrofluorescein diacetate). Cells were pretreated in 96-well plates at a density of 2 × 10⁴ cells/well with different concentrations of postbiotics (PostB) (19, 3.8, 1.9, 0.19 and 0.0095 mg/mL) for 16 h. After treatment, the cells were treated with 100 µM CM-H2DCFDA for 30 min at 37 °C to label intracellular ROS. After incubation, cells were washed twice with PBS. Subsequently, the labeled cells were monitored at 0, 3, 10, 20, 30, 40, 60, 90, 120, 153 and 174 min using a microplate reader (excitation = 485 nm; emission = 530 nm) (Clariostar; BMG Labtech). The results represent fluorescence intensity.

2.10. ELISA

HUVECs were seeded into a 6-well plate (6 × 10⁵ cells/well). After reaching 90% confluence, cells were pretreated with PostB (0.0095 mg/mL) for 16 h. Then cells were washed twice with sterile PBS with Mg²⁺ and Ca²⁺ and treated with 0.0095 mg/mL PostB alone and in combination with LPS (200 ng/mL) for 24 h. Supernatants were collected, centrifuged for 10 min 10,000 r*min⁻¹, and the released amount of IL-6 and IL-8 was determined by using Human ELISA kit (Thermo Fisher Scientific, MA, USA) based on the manufacturer’s recommendation. The ELISA was measured using a microplate reader (Clariostar; BMG Labtech).

2.11. PCR

Quantitative PCR (qPCR) was assessed on a Roche LightCycler 480 System (Roche, Basel, Switzerland) using the 5′-nuclease assay. Total RNA was isolated using the VeZol-Pure Total RNA Isolation System (Vazyme, Nanjing, China) according to the manufacturer’s instructions. Complementary DNA (cDNA) was synthesized from 1 µg of total RNA using the High-Capacity cDNA Reverse Transcription Kit (Thermo Fisher Scientific, Waltham, Massachusetts, USA). The reaction was implemented using the TaqMan™ Fast Advanced Master Mix (Thermo Fisher Scientific, MA, USA) and TaqMan™ Gene Expression Assay (FAM) (Thermo Fisher Scientific, MA, USA). As an internal control, glyceraldehyde-3-phosphate dehydrogenase (GAPDH) (Thermo Fisher Scientific, MA, USA) was used. The amount of the transcripts was normalized to those of the housekeeping gene (GAPDH) using the ΔCT method. Finally, relative gene expressions were calculated with the comparative ΔΔCt method according to Livak’s formula. We had negative control (cell culture media).

The following TaqMan™ Gene Expression Assays were applied for analyses (FAM): IL-6 (Hs00174131_m1), TNF-α (Hs00174128_m1), GAPDH (Hs02786624_g1), VCAM-1 (Hs01003372_m1), ICAM-1 (Hs00164932_m1), E-selectin (Hs00174057_m1), IL-1 β (Hs01555410_m1), HSPA1L (heat shock protein family A (Hsp70)) (Hs05046051_s1), and heat shock protein family B (small) member 1 (Hs00356629_g1).

2.12. Statistics

Data analysis was performed using Microsoft Excel (Microsoft Corporation) and GraphPad Prism Version 8.0 (GraphPad Software Inc.) software. Statistical comparisons were conducted using one-way ANOVA followed by Dunnett’s post hoc test. The results were expressed as mean  ±  standard deviation. Differences were considered statistically significant at p < 0.05. Significance levels are indicated as follows: * p < 0.05, ** p < 0.01, and *** p < 0.001 compared to untreated control cells. Double crosses represent statistical significance compare with LPS 100 ng/mL and LPS 200 ng/mL cell treatments # p < 0.05; ## p < 0.005; ### p < 0.001.

3. Results

3.1. Chemical Profiling of Postbiotic Composition

3.1.1. Determination of the Concentration of Surfactin Produced by Bacillus subtilis natto Culture

The quantitative analysis of the sample revealed notable differences in the concentrations of individual surfactin isoforms. Among the detected isoforms, isoform E exhibited the highest concentration, reaching 33.10 mg·L⁻¹, which accounted for the major proportion of the total surfactin content. Isoform C also showed a relatively high level (12.55 mg·L⁻¹), followed by isoform F (9.81 mg·L⁻¹) and isoform B (7.16 mg·L⁻¹). In contrast, isoforms D and A were present in lower amounts, with concentrations of 6.28 mg·L⁻¹ and 3.36 mg·L⁻¹, respectively. The observed distribution pattern suggests a predominance of isoform E in the surfactin profile of the bacteria strain, indicating that this variant may be preferentially synthesized under the applied cultivation conditions.

3.1.2. Determination of the Concentration of Surfactin Produced by Bacillus subtilis natto Culture Using UHPLC-MS Technique

The presence of surfactin in the PostB was further corroborated by UHPLC–MS analysis, as illustrated in Figure 2, providing additional analytical evidence for its production in the fermentation system.

Furthermore, an in-depth compositional analysis of the PostB was conducted. This profiling revealed a diverse set of metabolites, including amino acids, short-chain carboxylic acids, vitamins, and cyclic dipeptides. The qualitative distribution of these components is summarized in

Table 1. Metabolite compounds identified by LC-MS in Postbioticum. RT: retention time (min), m/z: mass/charge number.

No.	Rt. (min)	Compound	Chemical Formula	Exact Mass (m/z)		Fragment Ions (Relative Abundance,%)
				(M + H)+	(M − H)−
1	1.52	Lysine	C6H14N2O2	147.11336		130.0865; 84.0813; 67.0548; 56.0502
2	1.89	Arginine	C6H14N4O2	175.11951		158.0811; 130.0865; 112.0760; 70.0657; 60.0563
3	1.93	Histidine	C6H9N3O2	156.07731		110.0716; 95.0608; 93.045; 83.0609
4	1.95	Proline	C5H9NO2	116.07116		70.0657
5	1.95	Valine	C5H11NO2	118.08681		72.0814; 57.0580; 55.0550
6	1.95	γ-Aminobutyric acid	C4H9NO2	104.07116		87.0446; 86.0606; 69.0341; 58.0658
7	1.97	Aspartic acid	C4H7NO4	134.04534		116.0345; 88.0398; 74.0243; 70.0294
8	1.97	Glutamic acid	C5H9NO4	148.06099		130.0501; 102.0554; 84.0449; 56.0502
9	1.98	Methionine sulfoxide	C5H11NO3S	166.05379		149.0267; 102.0554; 75.0269; 74.0242; 56.0502
10	2.08	Malic acid	C4H6O5		133.01370	115.0022; 89.0228; 87.0073; 72.9916; 71.0122
11	2.27	Pyridoxine	C8H11NO3	170.08172		152.0706; 134.0601; 124.0758; 96.0812
12	2.31	Methionine	C5H11NO2S	150.05888		133.0319; 104.0532; 102.0554; 61.0113; 56.0502
13	2.46	Nicotinic acid (Niacin)	C6H5NO2	124.03986		96.0449; 80.0500; 78.0342
14	2.76	Leucine or Isoleucine	C6H13NO2	132.10246		86.0969; 69.0705
15	2.83	Citric acid	C6H8O7		191.01918	173.008; 129.0181; 111.0073; 87.0072; 85.0279
16	2.90	Tyrosine	C9H11NO3	182.08172		165.0547; 147.0441; 136.0758; 123.0443;119.0493
17	4.14	Phenethylamine	C8H11N	122.09698		105.0702; 103.0546; 79.0547
18	4.15	Cyclo-glycyl-l-proline	C7H10N2O2	155.08206		127.0868; 99.0922; 82.0657; 70.0658
19	4.88	Aconitic acid	C6H6O6		173.00862	129.0181; 111.0073; 85.0279
20	4.94	Phenylalanine	C9H11NO2	166.08681		149.06; 131.0494; 120.0810; 103.0546; 93.0704
21	5.03	Methylsuccinic acid	C5H8O4		131.03444	113.0229; 87.0436; 71.0122; 69.0329; 59.0122
22	5.76	Leucyl-valine	C11H22N2O3	231.17087		118.0862; 86.0969; 72.0813
23	6.80	Glutaric acid	C5H8O4		131.03444	113.0229; 87.0436; 71.0122; 69.0329; 59.0122
24	7.14	Cyclo-l-alanyl-l-proline	C8H12N2O2	169.09771		152.071; 141.1023; 96.0813; 72.0450; 70.0657
25	9.81	Pantothenic acid	C9H17NO5	220.11850		202.1074; 184.0968; 174.1124; 116.0344; 90.0554
26	10.52	Tryptophan	C11H12N2O2	205.09771		188.0707; 159.0916; 146.0601; 132.0809; 118.0653
27	12.85	Cyclo-l-prolyl-l-proline	C10H14N2O2	195.11336		98.0605; 70.0657
28	15.11	Cyclo-l-tyrosyl-l-proline	C14H16N2O3	261.12392		233.1285; 155.0815; 136.0758; 107.0494; 70.0657
29	15.65	Cyclo-l-prolyl-l-valine	C10H16N2O2	197.12901		169.1334; 141.1386; 136.0758; 72.0813; 70.0657
30	18.36	Biotin	C10H16N2O3S	245.09599		227.0847; 209.0742; 199.0899; 184.0792; 166.0684
31	20.26	Riboflavin	C17H20N4O6	377.14611		359.1336; 243.0875; 200.0818; 172.0868; 69.0341
32	20.86	Indole-3-lactic acid	C11H11NO3		204.06607	186.0553; 158.0599; 142.0651; 130.0651; 116.0461
33	21.88	Cyclo-l-phenylalanyl-l-proline	C14H16N2O2	245.12901		217.1336; 172.1120; 154.0737; 120.0810; 70.0657

.

3.1.3. Determination of K₂-MKn Vitamin Content in PostB Using an HPLC Method

The chromatographic results revealed the presence of both vitamin K₂ (menaquinone) and its homolog K₂-MK₇ in the analyzed PostB (Figure 3). The K₂ content in the sample (2.831 mg/L) and the K₂-MK₇ concentration was 0.557 mg/L. The retention times of the detected compounds (2.38–2.57 min) were consistent with those reported in the literature for K₂ and K₂-MK₇, confirming the reliability of the identification.

3.2. In Vitro Bioassays

3.2.1. Postbiotic Treatment Was Able to Mitigate LPS-Induced Cell Death on HUVECs

Initially, the effect of postbiotics on the cell viability of human umbilical vein endothelial cells (HUVECs) were explored (Figure 4). We performed MTT cytotoxicity assays with different concentrations of postbiotic (19, 3.8, 1.9, 0.19 and 0.0095 mg/mL) alone and with combination of LPS (100 and 200 ng/mL) for 24 and 48 h. As shown in Figure 1A, 19 mg/mL doses of postbiotic (PostB) (7.26 ± 2.97) significantly reduced the cell viability compared to the control (100 ± 14.99) after 24 h treatment of HUVECs (Figure 4A). Nevertheless, postbiotic exposure at lower concentrations (3.8, 1.9, 0.19 and 0.0019 mg/mL) did not influence the cell viability within the 24 h time window.

In this study, LPS treatment was applied to induce endothelial inflammation and subsequent endothelial dysfunction in HUVECs. Therefore, we evaluated the LPS effects on endothelial cell viability. According to our results, LPS treatments were significantly reduced endothelial cell viability, at both 100 ng/mL (48.75 ± 10.89) and 200 ng/mL (53.94 ± 4.39) concentrations when compared with the control (100 ± 14.99) of 24 h cell treatments. While postbiotic treatment in combination with LPS 100 ng/mL demonstrated a dose-dependent protective effect, at 3.8 (73.85 ± 12.48), 1.9 (76.05 ± 20.31) and 0.19 (92.92 ± 15.14) mg/mL concentrations compared to LPS 100 ng/mL treatment alone (48.75 ± 10.89). Furthermore, 0.0095 ng/mL PostB (with LPS 100 ng/mL: 85.32 ± 23.01; with LPS 200 ng/mL: 75.93 ± 5.48) significantly attenuated the aforementioned LPS-induced cytotoxicity at both 100 ng/mL (48.75 ± 10.89) and 200 ng/mL (53.94 ± 4.39) concentrations.

A number of 48 h viability studies were also performed (Figure 4B). Similarly to the 24 h treatments, the application of 19 mg/mL (6.36 ± 2.83) concentration of PostB reduced cell viability relative to the control (100 ± 15.62). However, concentrations of 3.8 (70.55 ± 18.32) and 1.9 mg/mL (66 ± 13.75) also exhibited negative cytotoxic effects in relation to the control. When we used postbiotics in 0.0095 mg/mL (95 ± 14.75) concentration in combination with LPS 100 ng/mL (29.92 ± 8.34), no cytotoxic effects were observed. However, treatment with LPS alone at concentrations of 100 ng/mL (29.92 ± 8.34) and 200 ng/mL (63.52 ± 12.79) significantly reduced cell viability compared with the control group.

3.2.2. Postbiotics Potentially Inhibited the LPS-Induced Apoptosis and Necrosis

Considering that high concentrations of postbiotics exhibit cytotoxic effects, it is reasonable to question whether lower concentrations may also induce adverse cellular processes that remain undetected by MTT assays. Furthermore, the postbiotics exert these cytotoxic effects through promoting apoptosis or necrosis. To address these questions, we used DilC1(5) and SYTOX Green fluorescent labeling (Figure 5). For DilC1(5) measurements, statistical significance is indicated for treatments showing a significant reduction in fluorescence relative to the negative control, as loss of signal reflects mitochondrial membrane depolarization associated with apoptotic cell death. SYTOX Green fluorescence signals exceeding the negative control reflect biologically relevant necrotic cell death, whereas values at or below control levels do not indicate necrosis.

In our studies, we found that postbiotics at 19 mg/mL concentration induced apoptosis in endothelial cells during 24 h of treatment (Figure 5A). Thus, it can be said that the cytotoxic effect of the PostB observed with MTT assay was mediated by the activation of apoptotic processes.

Application of postbiotics at lower concentrations did not induce apoptotic processes. These results are consistent with those observed in cytotoxicity studies. Interestingly, the application of LPS in both 100 ng/mL and 200 ng/mL concentrations did not enhance apoptosis in cells in a 24 h time windows, which is inconsistent with our MTT results, where LPS decreases cell viability under similar treatment condition.

After 48 h treatments, LPS at 200 ng/mL (0.5 ± 0.07) and PostB at 19 mg/mL (0.265 ± 0.023) and 3.8 mg/mL (0.42 ± 0.01) induced apoptosis relative to the control (0.98 ± 0.04) (Figure 5B). Notably, PostB at lower concentrations, including 1.9 mg/mL, 0.19 mg/mL and 0.0095 mg/mL, did not induce apoptotic process in HUVECs. Moreover, in combination with LPS 200 ng/mL concentrations PostB at 1.9 mg/mL (0.85 ± 0.08), 0.19 mg/mL (1.274 ± 0.22) and 0.0095 mg/mL (1.43 ± 0.18) inhibited the observed LPS apoptotic effects.

We also investigated whether postbiotics and LPS alone and in combination induce necrosis in HUVECs (Figure 5C,D). During 24 h treatments, no necrotic cell death was observed with any of the treatment’s solutions (Figure 5C).

However, in 48 h treatments, LPS 100 ng/mL triggers necrosis compared to the control (Figure 5D). Application of PostB at 1.9 mg/mL, 0.19 mg/mL and 0.0095 mg/mL concentrations was able to mitigate the LPS-caused necrotic cell death.

According to the MTT, DILC, and Sytox results, 0.0095 mg/mL PostB was selected at the most optimal concentration and used in the following study.

3.2.3. Postbiotic Treatment Inhibited LPS-Induced Intracellular ROS Production as Well as Heat Shock Protein mRNA Increase in HUVECs

To further explore the role of postbiotics (0.0095 mg/mL) in the regulation of LPS (LPS 200 ng/mL) induced cell stress, we assessed the intracellular ROS levels and as well as HSPB1 (Hsp27) and HSPA1L (Hsp70) mRNA levels of HUVECs (Figure 6). For this purpose, HUVECs were treated with postbiotics (0.0095 mg/mL), LPS (200 ng/mL), or their combination for 2, 4, and 24 h.

After a 2 h treatment, we found that PostB (0.54 ± 0.07) and LPS (0.79 ± 0.14) treatments decreased the HSPB1 mRNA levels relative to the control (1) (Figure 6A). Investigating HSPA1L, we noticed that postbiotics (0.54 ± 0.06) mitigated HSPA1L mRNA levels in comparison with the control (Figure 6B).

We tested the effects of PostB alone and in combination with LPS in a 4 h treatment as well (Figure 6C,D). Our data demonstrated that LPS (2.10 ± 0.07) significantly increased the HSPB1 relative gene expression in comparison with the control (Figure 6C). Interestingly, PostB (1.35 ± 0.06) treatment alone was able to slightly enhance the HSPB1 mRNA expression; however, when used together with LPS (1.10 ± 0.02), mRNA expressions were found to be nearly equal to the control level. In the case of HSPA1L heat shock protein mRNA expression, a significant decrease was noticed with PostB (0.28 ± 0.16) treatment relative to the controls (Figure 6D). LPS notably increased HSPA1L (3.1 ± 0.24) mRNA levels compared with the control. Furthermore, combined treatment with postbiotics and LPS (0.37 ± 0.20) markedly downregulated HSPA1L expression compared with both the control (1.00) and LPS treatment alone (3.17 ± 0.24).

In 24 h treatment time window, we found that LPS application to the cells remarkably increased both HSPB1 (1.73 ± 0.34) and HSPA1L (1.73 ± 0.34) levels in relation to the control (Figure 6E,F). Interestingly, PostB treatments with LPS were able to mitigate the above-mentioned HSPB1 (1.07 ± 0.05) and HSPA1L (1.07 ± 0.04) mRNA increase.

To investigate the effects of PostB on early ROS production induced by LPS in HUVECs, we assessed the time course of intracellular ROS generation evoked by LPS in cells (Figure 6G,H). According to our results, exposure of endothelial cells to 200 ng/mL LPS resulted in a significantly rapid dose-dependent increase in DCFDA fluorescence related to negative control (Figure 6G). Notably, intracellular ROS production remained significantly elevated in LPS-treated samples relative to the control throughout the entire experiment (0–174 min), with p-values of <0.05, <0.005, and <0.001 at different time points. However, when PostB was used in combination with LPS, intracellular ROS production was markedly reduced compared with the LPS-treated group throughout the experiment (p < 0.001). It is also interesting that a moderate increase was also observed during the application of PostB treatment. Furthermore, LPS (526.3 ± 70) induced intracellular ROS production was significantly higher relative to the control (304.3 ± 26). When LPS was used in combination with PostB (349 ± 47), the intracellular ROS production did not increase relative to the control and LPS treatment alone (Figure 6H).

3.2.4. Postbiotic Treatment Inhibited Pro-Inflammatory Cytokine Production in LPS-Stimulated Endothelial Cells

We aimed to evaluate whether postbiotics are able to promote possible positive effects on inflammatory stress induced by LPS. To answer this question, HUVECs were treated with PostB 0.0095 mg/mL concentration alone and in combination with LPS (200 ng/mL) for 4 and 24 h (Figure 7). We observed that LPS treatment induced notable morphological changes on HUVECs (Figure 7A). However, this morphological phenomenon was not observed if LPS treatment was applied in combination with PostB. Under control conditions, endothelial cells exhibited the characteristic elongated, spindle-shaped morphology and formed a confluent monolayer with preserved cell–cell contacts. In contrast, LPS treatment induced pronounced morphological changes, characterized by cell rounding, loss of elongation, and partial disruption of intercellular junctions. Notably, co-treatment with postbiotics and LPS largely preserved normal endothelial morphology, as cells maintained their elongated shape and the LPS-induced rounding phenotype was not observed (Figure 7A).

First, we examined the mRNA expression of NF-κB and the pro-inflammatory cytokines IL-1β, TNF-α, and IL-6, whose expression is known to be specifically and significantly upregulated during inflammatory responses. (Figure 7C–I). As shown in Figure 4B, NF-_KB (179.8 ± 1.442) relative gene expression was increased by LPS relative to the control after 4 h treatment. However, using PostB in combination of LPS mitigated the NF-_KB (1.46 ± 0.01) relative gene expression, compared to the control and LPS treatment alone. Interestingly, after 24 h the NF-_KB mRNA level was significantly higher when we used PostB treatment with LPS in comparison to negative control (Figure 7F).

As expected, the relative gene expression of IL-1β (4 h: 3.75 ± 1.2; 24 h: 18.3 ± 4.11), TNFα (4 h: 15.05 ± 0.83; 24 h: 2 ± 0.06) and IL-6 (4 h: 9.03 ± 1; 24 h: 3.61 ± 0.42) significantly increased as a result of LPS compared to the control in both 4 and 24 h treatments. Exposure to PostB (0.009 mg/mL) had no significant effect on pro-inflammatory cytokine mRNA expression at 4 h as well as 24 h. To assess the possible positive effects of PostB, HUVECs were exposed to 0.0095 mg/mL postbiotic with 200 ng/mL LPS for 4 and 24 h. Interestingly, similar gene expression changes were detectable for pro-inflammatory cytokine (IL-1β, TNFα and IL-6) at both 4 and 24 h treatments. Furthermore, PostB treatment was successfully applied to alleviate LPS-induced IL-1β (4 h: 2.22 ± 0.19; 24 h: 0.73 ± 0.17), TNFα (4 h: 2.41 ± 0.09; 24 h: 1.02 ± 0.02) and IL-6 (4 h: 0.93 ± 0.79; 24 h: 0.99 ± 0.01) pro-inflammatory cytokine mRNA expression compared with the LPS treatment alone.

To confirm the successful activation of endothelial cells and the production of inflammatory cytokines, the protein expression levels of IL-8 and IL-6 were also determined with ELISA from cell culture supernatants after 24 h treatment (Figure 7J,K). Endothelial cells exposed to 200 ng/mL LPS for 24 h remarkably increased IL-8 (3314 ± 727.9) as well as IL-6 (557 ± 149.5) protein expression relative to control (IL-8 = 139.1 ± 28.45; IL-6 = 26.99 ± 8.54). Postbiotic treatment was able to mitigate the LPS-induced protein expression of IL-8 (1547 ± 123.7) and IL-6 (557.7 ± 149.5) pro-inflammatory cytokines as well.

3.2.5. Postbiotic Treatment Improved Endothelial Barrier Functions

We tested the effect of PostB on endothelial cells integrity under inflammatory stress (Figure 8). Therefore, we incubated HUVECs with PostB (0.0095 mg/mL) in the absence or presence of LPS (200 ng/mL) for 4 or 24 h to assess adhesion molecules and tight junction (TJ) protein levels.

Quantitative real-time PCR reactions were performed to evaluate the mRNA expression levels of adhesion molecules, including E-selectin, vascular cell adhesion molecules-1 (VCAM-1) and intracellular adhesion molecule-1 (ICAM-1).

After 4 h of treatment, we observed that LPS (132 ± 4.9) significantly increased E-selectin expression level compared to the negative control (1 ± 0) (Figure 8A). PostB treatment was able to mitigate the LPS-induced E-selectin (343.3 ± 4.13) expression in our treatment conditions.

Prolonged exposure to LPS (24 h) resulted in greater induction of E-selectin (66.3 ± 3.09) mRNA expression compared to the control (Figure 8). However, PostB treatment was not successful in mitigating the robust E-selectin (60.41 ± 0.64) expression induced by LPS after 24 h treatment.

ICAM-1 mRNA expression was significantly increased after LPS (200 ng/mL) (3.27 ± 0.35) treatment for 4 h relative to the control (Figure 8C). PostB (0.42 ± 0.18) treatment exacerbated positive effects by inhibiting LPS (3.27 ± 0.35) caused ICAM-1 increase. After 24 h incubation, ICAM-1 mRNA expression was increased in response to both LPS and postbiotic treatment compared with the control (Figure 8D).

According to our results, LPS significantly enhanced VCAM-1 expression level after 4 and 24 h treatments (Figure 8E,F). Postbiotic treatment of endothelial cells reduced LPS-induced VCAM-1 expression at 4 h, but this effect was not observed at 24 h.

The levels of TJ protein of Occludin after a 24 h treatment, in endothelial cell cultures were determined in the presence of PostB alone and in combination with LPS by ELISA (Figure 8G).

Our data demonstrates that LPS treatments alone exerted decrease in protein expression levels of Occludin compared with the control. The protein level of Occludin significantly increased after induction with LPS in combination with PostB compared with LPS exposure alone.

4. Discussion

The aim of the present study was to investigate how PostB derived from Bacillus subtilis natto (Szendi2020) influences the inflammatory and stress responses of endothelial cells triggered by LPS. Given that endothelial inflammation and related molecular processes play a key role in the development of acute and chronic inflammatory conditions and cardiovascular diseases, it is particularly important to identify biological modulators that can mitigate these harmful processes. Postbiotics—as non-living bioactive molecules of microbial origin—are receiving increasing attention for their potential anti-inflammatory, antioxidant, and barrier-protective properties, but their effects in endothelial cells remain poorly understood [23,25,28].

Bacillus subtilis natto (Szendi2020) contains a potent fibrinolytic enzyme called nattokinase (NK), which is a serine protease and has been recognized for its cardiovascular beneficial effects due to its thrombolytic and anticoagulant activities [29]. According to previous study, NK treatment exacerbated notable anti-inflammatory activity with reduced intracellular ROS production of macrophages [29].

Motivated by this, we tracked the molecular effects of postbiotics derived from Bacillus subtilis natto (Szendi2020) in an endothelial inflammatory stress model. We utilized LPS to induce inflammation in HUVECs.

Here, we demonstrated that exposure of HUVECs to LPS at concentrations of 100 and 200 ng/mL significantly induced apoptosis, as reflected by decreased metabolic activity, loss of mitochondrial membrane potential, and increased cell death marker positivity. These findings are consistent with previous reports describing the proapoptotic effect of LPS on vascular endothelial cells, mediated by the activation of inflammatory and oxidative stress pathways that converge on mitochondrial dysfunction and caspase-dependent signaling [30]. Importantly, our data provide novel evidence that the concomitant administration of postbiotics completely abolished LPS-induced cell death in HUVECs. This protective effect was consistently observed across different methodological approaches, including MTT viability assay, DiIC measurement of mitochondrial membrane integrity, and Sytox Green staining of plasma membrane permeability, thereby strengthening the robustness of our findings. The precise mechanisms underlying the protective action of postbiotics remain to be elucidated, but several possibilities can be considered. Postbiotics may counteract LPS-induced oxidative stress and modulate inflammatory signaling pathways. Previous studies have shown that certain postbiotic components can enhance endothelial resilience by reducing reactive oxygen species generation, stabilizing mitochondrial function, and upregulating cytoprotective molecules such as heat shock proteins and antioxidant enzymes. Thus, the prevention of LPS-triggered apoptosis observed in our study may reflect a multifaceted protective response initiated by these bioactive compounds.

The examination of oxidative stress further supported these conclusions. According to LPS induced rapid, dose-dependent ROS accumulation, suggesting early inflammatory activation of HUVECs. In contrast, co-treatment with PostB inhibited this ROS increase (p < 0.05), indicating that the postbiotic exerts an antioxidant or ROS-modulating effect on endothelial cells. Overall, our results show that 0.0095 mg/mL PostB effectively modulated the LPS-induced oxidative and molecular stress response, reduced the activation of inflammatory signaling pathways, and thus had a potential cytoprotective and anti-inflammatory effect in endothelial cells. This concentration therefore proved to be a biologically favorable and relevant choice for further functional studies.

Since our results clearly demonstrated that postbiotic treatment can moderate the accumulation of ROS induced by LPS in endothelial cells, the next step was to examine how this effect on redox homeostasis is reflected at the molecular level of the cells’ stress response. The increase in ROS levels is a widely known trigger for the activation of cellular proteostatic and cytoprotective mechanisms, among which heat shock proteins (HSPs) play a prominent role [31]. These molecules stabilize damaged proteins through their basic chaperone function, promote their refolding, and protect against structural and functional disorders caused by oxidative stress. HSPB1 (Hsp27) and HSPA1L (Hsp70) are particularly sensitive to oxidative and inflammatory stimuli affecting endothelial cells, and their expression is therefore a reliable indicator of cellular stress and adaptive response. Thus, measuring changes in Hsp27 and Hsp70 helps clarify how the ROS-reducing effects of the postbiotic influence cellular protective and survival pathways [27,28].

In concordance with previous findings, LPS treatment notably increased Hsp27 as well as Hsp 70 mRNA expression levels after 4- and 24 h exposures (p < 0.05, p < 0.005, p < 0.001) [32]. When LPS was combined with postbiotic treatment, neither Hsp27 nor Hsp70 expression was induced at 2, 4, or 24 h (p < 0.05, p < 0.005, p < 0.001). Moreover, except for the 2 h Hsp27 result, postbiotic treatment significantly decreased the gene expression of both Hsp27 and Hsp70. This phenomenon may suggest that PostB is capable of modulating LPS-activated stress signaling pathways and potentially interferes with the regulation of the endothelial inflammatory response. The 24 h treatments further reinforced this observation: although LPS persistently increased Hsp27 and Hsp70 mRNA levels, co-administration of PostB attenuated the excessive induction of both genes. This suggests that the postbiotic is able to effectively modulate inflammatory signaling pathways (and their downstream molecular effectors) not only during the early, acute cellular stress response, but also during prolonged inflammatory activation.

Adhesion molecules and tight junction (TJ) proteins are key regulators of endothelial barrier integrity and vascular homeostasis. During endothelial inflammation, molecules such as ICAM-1 and VCAM-1 are upregulated, promoting leukocyte adhesion and transmigration into tissues. ROS molecules induce the upregulation of adhesion molecules. At the same time, the disruption of tight junction proteins such as claudins and occludins damages the endothelial barrier, leading to increased vascular permeability [33,34]. Therefore, our aim was to investigate how postbiotics influence adhesion molecule expression and tight junction (TJ) protein signatures during LPS-induced inflammation in endothelial cells.

During the 4 h treatments, LPS significantly increased the expression of E-selectin, ICAM-1, and VCAM-1, which are early markers of inflammatory activation. However, the presence of PostB attenuated LPS-induced E-selectin and VCAM-1 induction and inhibited ICAM-1 elevation, suggesting that at this early stage, the postbiotic is able to attenuate endothelial cell activation and excessive expression of adhesion molecules. During longer, 24 h exposure, LPS further increased the expression of all three adhesion molecules, reflecting the sustained inflammatory activation of endothelial cells. At this point, the anti-inflammatory effect of PostB was only partially observable: while ICAM-1 expression continued to increase in the presence of LPS. PostB was no longer able to significantly reduce the LPS-induced increase in E-selectin and VCAM-1. This may suggest that the effect of the postbiotic primarily extends to the early phase of the inflammatory response rather than the sustained phase. According to previous results, LPS treatment significantly decreases the expression of Occludin, thus increasing the cell permeability, which is in concordance with our results [35]. Examination of the integrity of tight junction structures further confirmed the protective effect of PostB. While LPS reduced Occludin protein levels within 24 h, indicating damage to the endothelial barrier, the combined use of PostB significantly restored Occludin expression compared to LPS treatment. This suggests that the postbiotic effectively mitigates LPS-induced endothelial barrier damage and may contribute to the structural and functional maintenance of tight junction structures. Overall, our results indicate that PostB is able to effectively modulate the early inflammatory activation of endothelial cells, reduce the overexpression of adhesion molecules, and partially restore the levels of cell barrier components (Occludin).

The Bacillus genus is widely utilized in industry, particularly in biotechnological processes, for the production of biologically active compounds such as high-value enzymes and a broad range of lipopeptides. Lipopeptides are secondary metabolites synthesized primarily by Bacillus and other microbial genera, and they have numerous potential applications. In veterinary medicine, they are employed as antimicrobial agents, adjuvants, and drug delivery systems. Surfactin (SF) is a biosurfactant lipopeptide produced by members of the Bacillus subtilis group. It consists of a cyclic heptapeptide linked to a β-hydroxy fatty acid chain of variable length (typically 13–15 carbon atoms). SF is biodegradable, exhibits lower toxicity compared with chemical surfactants, and has been reported to possess antiviral, antibacterial, antitumor, and anticoagulant activities. Several in vitro studies have shown that Bacillus-derived SF exerts dose-dependent anti-inflammatory effects [36]. Based on these findings, we hypothesize that postbiotic treatments could also mitigate the LPS-induced inflammatory cytokine production as well.

Human endothelium is able to secrete and respond to cytokines. Cytokines including IL-1β, IL-6, IL-8, TNFα, MCP-1, granulocyte colony-stimulating factor (GCSF) and granulocyte–macrophage colony-stimulating factor (GM-CSF) orchestrate acute inflammatory responses and alter endothelial cell responses, leading to increased vascular permeability and augmented leukocyte adhesion to the endothelial surface [37].

Previous findings have shown significantly elevated TNFα levels in patients with acute coronary syndrome, myocardial infarction, and heart failure. One mechanism by which TNFα contributes to endothelial dysfunction is through activation of the NF-κB transcription factor [35,38]. Therefore, the evaluation of cytokine signatures is crucial for understanding the potential beneficial effects of postbiotic treatment on endothelial inflammatory stress and for guiding therapeutic interventions. In this study, we assess both the NF-κB and TNFα mRNA expression levels on HUVECs, treated with LPS alone and in combination with PostB. In consistence with previous findings, we found that LPS treatment significantly elevated the NF-κB and TNFα mRNA expression levels (p < 0.001). However, when we used postbiotic treatment in combination with LPS, the aforementioned NF-κB and TNFα expression was abolished (p < 0.001). However, at 24 h, NF-κB expression was higher in the PostB + LPS treatment group than in the control group, suggesting that the effect of the postbiotic on NF-κB dynamics is time-dependent and that different regulatory patterns may develop during longer exposure. The expression patterns of IL-1β, TNFα, and IL-6 confirmed the proinflammatory properties of LPS: their mRNA levels were significantly increased after both 4 and 24 h. Although PostB exerted a clear anti-inflammatory effect, as it significantly reduced LPS-induced cytokine expression. This suggests that PostB is able to intervene at multiple points in the inflammatory response of HUVECs and moderate the LPS-activated transcriptional program. Protein-level studies further reinforced these observations. LPS significantly increased IL-8 and IL-6 secretion, confirming the inflammatory activation of endothelial cells. However, treatment with PostB reduced LPS-induced IL-8 and IL-6 protein expression, supporting its anti-inflammatory effect at the functional level.

In our study, we observed that LPS-induced oxidative stress, as measured by ROS accumulation, occurred rapidly within 90 min of treatment. Furthermore, ROS level steadily increased, indicating that ROS production is an early event in endothelial activation. A remarkable upregulation of Hsp27 and Hsp70 was followed by increased intracellular ROS production. Subsequent inflammatory responses, including upregulation of NF-κB, TNFα, IL-1β, IL-6 at mRNA levels after 4 h were observed. At later time points (8 h), downstream activation of translational pathways was reflected by significantly increased IL-6 and IL-8 protein levels following LPS treatment. Cell viability assays (MTT, DilC1(5), SYTOX) demonstrated increased cell death caused LPS after 24 h, suggesting that oxidative stress and inflammatory signaling precede detectable apoptotic or necrotic cell death. These results highlight a temporal hierarchy in LPS-induced endothelial dysfunction, where early ROS generation triggers inflammatory signaling that may eventually lead to cell death if exposure is prolonged.

Our study has several limitations. First, to induce endothelial inflammation we used only LPS; however, it would be interesting to test whether inflammation triggered by other agents—such as TNF-α or hypoxia-induced endothelial dysfunction—would also be mitigated by postbiotic treatment. Furthermore, we showed that it inhibited the LPS-induced inflammation at multiple levels, such as ROS and pro-inflammatory cytokine production; the exact mechanism which promotes the positive effects of the postbiotic remains unclear. We are planning to investigate postbiotic-induced changes in gene expression using high-throughput transcriptomic analyses. Moreover, NF-κB pathway activation was not performed by direct measurement of NF-κB phosphorylation. Instead, NF-κB activity was evaluated indirectly through the expression of well-established NF-κB-dependent downstream inflammatory markers, including pro-inflammatory cytokines and adhesion molecules. Nevertheless, direct analysis of phosphorylated NF-κB by Western blotting would provide complementary mechanistic insight into pathway activation and should be considered in future studies. Furthermore, LPS-induced endothelial activation is known to involve metabolic reprogramming toward enhanced glycolysis, we did not directly assess glycolytic activity or measure glycolytic intermediates. Future studies incorporating detailed metabolic profiling will be necessary to determine whether the beneficial effects of postbiotic treatment on endothelial inflammatory responses are mediated, at least in part, through the modulation of glycolytic pathways.

5. Conclusions

In summary, our results indicate that PostB may have a potentially beneficial effect on the inflammatory and stress response of endothelial cells. At the right concentration, the postbiotic can help keep cells alive and reduce the cytotoxic effects caused by LPS. Studies suggest that PostB may protect cells through mechanisms that inhibit apoptosis and reduce necrosis. Based on molecular stress responses, it can be assumed that the postbiotic modulates early and late stress-signaling pathways, including the expression of heat shock proteins. In addition, the reduction in ROS production may indicate that PostB also has an antioxidant or oxidative stress-regulating effect. Analysis of inflammatory cytokine responses indicates that the postbiotic may reduce the increased expression of LPS-induced inflammatory mediators. The time-dependent modulation of NF-κB activity suggests that PostB may selectively influence the dynamics of inflammatory signaling. As a strength of our study, it is important to note that despite the extensive body of research on postbiotics, we are the first to describe the potential beneficial effects of Bacillus subtilis natto-derived postbiotics on endothelial dysfunction cell models.