ORIGINAL RESEARCH published: 05 February 2021 doi: 10.3389/fmicb.2021.621280 Evaluation of an O2-Substituted (1–3)-β-D-Glucan, Produced by Pediococcus parvulus 2.6, in ex vivo Models of Crohn’s Disease Sara Notararigo 1,2,3, Encarnación Varela 2,4, Anna Otal 2, Iván Cristobo 1, María Antolín 2, Francisco Guarner 2, Alicia Prieto 1* and Paloma López 1* 1 Department of Microbial: and Plant Biotechnology, Margarita Salas Biological Research Centre (CIB-Margarita Salas-CSIC), Madrid, Spain, 2 Department of Gastroenterology, Digestive System Research Unit, Institut de Recerca Vall d’Hebron (VHIR), University Hospital Vall d’Hebron, Universitat Autònoma de Barcelona, Barcelona, Spain, 3 Foundation Health Research Institute of Santiago de Compostela (FIDIS), Santiago de Compostela, Spain, 4 CIBERehd, Instituto de Salud Carlos III, Madrid, Spain 1,3-β-glucans are extracellular polysaccharides synthesized by microorganisms and Edited by: plants, with therapeutic potential. Among them, the O2-substituted-(1–3)-β-D-glucan, Jasna Novak, University of Zagreb, Croatia synthesized by some lactic acid bacteria (LAB), has a prebiotic effect on probiotic strains, Reviewed by: an immunomodulatory effect on monocyte-derived macrophages, and potentiates the Jesús Navas, ability of the producer strain to adhere to Caco-2 cells differentiated to enterocytes. In University of Cantabria, Spain Rodrigo Dias De Oliveira Carvalho, this work, the O2-substituted-(1–3)-β-D-glucan polymers produced by GTF Federal University of Minas Gerais, glycoyltransferase in the natural host Pediococcus parvulus 2.6 and in the recombinant Brazil strain Lactococcus lactis NZ9000[pNGTF] were tested. Their immunomodulatory activity *Correspondence: was investigated in an ex vivo model using human biopsies from patients affected by Paloma López plg@cib.csic.es Crohn’s disease (CD). Both polymers had an anti-inflammatory effect including, a reduction Alicia Prieto of Interleukine 8 both at the level of its gene expression and its secreted levels. The overall aliprieto@cib.csic.es data indicate that the O2-substituted-(1–3)-β-D-glucan have a potential role in ameliorating Specialty section: inflammation via the gut immune system cell modulation. This article was submitted to Food Microbiology, Keywords: bacterial exopolysaccharides, O2-substituted-(1-3)-β-D-glucan, lactic acid bacteria, a section of the journal Immunomodulation, Crohn´s disease anti-inflammatory effect Frontiers in Microbiology Received: 25 October 2020 Accepted: 11 January 2021 INTRODUCTION Published: 05 February 2021 Citation: The exopolysaccharides (EPS) are large, lineal, or branched, extracellular carbohydrate polymers, Notararigo S, Varela E, Otal A, produced by algae, plants, and bacteria. They are commonly used as food additive due to Cristobo I, Antolín M, Guarner F, their rheological properties (bio-thickener, gelling, or viscosifier agents) in the food industry. Prieto A and López P (2021) The demand for new polymers in the food industry is positioning EPS from lactic acid bacteria Evaluation of an O2-Substituted (LAB), as the new generation of food thickeners. Due to their Generally Regarded As Safe (1–3)-β-D-Glucan, Produced by Pediococcus parvulus 2.6, in ex vivo (GRAS) status, they are suitable for the production of fermented and/or functional products. Models of Crohn’s Disease. The EPS synthesized by LAB are divided into homopolysaccharides, if they contain only one Front. Microbiol. 12:621280. monosaccharide type or heteropolysaccharides, if they contain various monosaccharides types doi: 10.3389/fmicb.2021.621280 (Werning et al., 2012). These EPS are known to support bacterial growth, as well as participate Frontiers in Microbiology | www.frontiersin.org 1 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan in cellular recognition and interaction, surface adhesion, and TNF-α, INF-γ, IL-17, and IL12) into the gut lumen, alteration biofilm formation. Currently, EPS are gaining interest as of microbiota, and the intestinal mucosal barrier (Knights et al., prebiotics, as modulators of the host immune system (Schmid 2013). High inflammation is responsible for abdominal pain, et al., 2015), and as antiviral agents (Nácher-Vázquez et al., bloody stools, weight loss, diarrhea, etc. 2015). Their immunomodulatory properties are dependent on To evaluate therapeutic treatments, current chemically induced their ability to form a suitable tertiary structure. For instance, animal models of IBD, including sodium dextran-sulfate (1–3)-β-D-glucans (β-glucans), linear or branched in position treatment, resembles UC symptoms, but emulates only partially O-4 and O-6, are gaining interest as therapeutic targets; various the inflammatory process that occurs in human CD (Chassaing studies have shown their positive influence in reduction of et al., 2014). However, an ex vivo model, using CD mucosal human serum cholesterol levels and their stimulation of the tissue, allows the investigation of how to modulate the human immune system (Chan et al., 2009; Jin et al., 2018), inflammation at the gut mucosa level. Previous studies, using as well as their potential anti-carcinogenic properties (Chan this model, have established that certain probiotic bacteria are et al., 2009; Ali et al., 2015). capable of interacting with immunocompetent cells using the Recent findings documented that (1–3)-β-glucans are capable mucosal interface, and thus can modulate locally the production of regulating the inflammatory response, modulate immune of pro-inflammatory cytokines by inflamed tissue (Borruel system cell types (such as peripheral blood immune system et al., 2003; Llopis et al., 2009; Hidalgo-Cantabrana et al., 2015). cells, intestinal epithelium, and mammalian microglia), through In this context, recently, new therapeutic strategies involving their interaction with membrane receptors, including toll-like the use of probiotic strains have been developed to ameliorate receptors, Dectin-1, SIGNR1, complement receptor 3 (CR3), CD patient’s symptoms. Some of these include microorganisms, LaCer, and Scavenger, which are differentially expressed in such as Lacticaseibacillus rhamnosus GG, Limosilactobacillus these cells types (Li et al., 2019). Their interaction with a reuteri, Lactobacillus acidophilus, Bifidobacterium infantis, receptor leads to downstream events including the activation Saccharomyces boulardii, Escherichia coli Nissle 1917, and of immune system cells, both innate (e.g., macrophages, Clostridium butyricum MIYAIRI 588 (Tsai et al., 2019), which monocytes, or neutrophils) and adaptive (e.g., T cells or B contribute to reduction of inflammation, due to their ability cells). These interactions may result in differential cytokine to reduce pathogen adhesion to the intestinal epithelium (blocking production [such as tumor necrosis factor-α (TNF-α), IL-10, their binding site), and by production of antibacterial substances IL-8, or IL-12], through the modulation of the nuclear factor or EPS (Orel and Kamhi Trop, 2014; Oka and Sartor, 2020). kappa-light-chain-enhancer of activate B cells (NFκB) In addition, probiotic strains modify the release of cytokines transcription factor, belonging to the mitogen-activated protein in the intestinal epithelium and inhibit, in the immune system kinase (MAPK) pathway (Chan et al., 2009; Volman et al., 2010). cells, the production of the transcription factor NFkB, leading Moreover, it has been reported that the EPS are metabolized as a consequence to a reduction of intestinal inflammation in the gut due to the presence of the glycolytic enzyme pools (Basson et al., 2017; Oka and Sartor, 2020). In this general of the microbiota. The exact nature of this process depends on context, some probiotic bacteria produce a O2-substituted- the molecular weight and biochemical composition of the EPS (1–3)-β-D-glucan with prebiotic activity and that stimulate the (Bodera, 2008; Kau et al., 2011; Ballesteros Pomar and Gonzalez growth of probiotic LAB (Russo et al., 2011; Pérez-Ramos Arnaiz, 2018). Finally, this hydrolysis is beneficial to both the et al., 2017). Therefore, the producing bacteria or their EPS host and the microbiota itself, because it generates secondary have potential as adjuvants in the treatment of IBD. Such a metabolites with the potential to act as probiotic or postbiotic role is plausible, taking into account that β-glucans play an immunomodulators, that ultimately promote and/or restore a important role in the modulation of both the innate immune healthy environment for the microbiota (Iweala and Nagler, 2019). response, through interaction with dendritic cells (DCs) and Laminarin, an O6-substituted-(1–3)-β-D-glucan isolated from macrophages, and also in the adaptive immune response, brown algae, causes a reduction in the expression of increasing proliferation of T- and natural killer cells, via cytokine pro-inflammatory cytokines, such as IL-6 and IL-1β in pig release (Bodera, 2008; Zhang et al., 2015). intestinal mucosa (Heim et al., 2014; Walsh et al., 2015) and Only Pediococcus, Lactobacilli, and Oenococcus strains isolated counteracts dysbiosis of the microbiota (Rattigan et al., 2020). from alcoholic beverages synthesize the O2-substituted-(1–3)- Moreover, studies performed with human fecal microbiota and β-D-glucan (reviewed in Llamas-Arriba et al., 2019). Moreover, commercial laminarin from Laminaria digitata, or crude studies performed with Pediococcus parvulus 2.6 isolated from polysaccharide-rich extracts from this algae, indicated that the cider and with the recombinant L. lactis NZ9000[pGTF] strain polysaccharide influences mucus and gut microbiota composition, carrying the pediococcal gtf gene, demonstrated that this gene which results in a potentially beneficial production of short-chain encodes the GTF glycosyltransferase, which catalyzes the fatty acids (such as butyrate; Devillé et al., 2007; Strain et al., 2020). synthesis of the O2-substituted-(1–3)-β-D-glucan in LAB Inflammatory bowel disease (IBD) has a multifactorial etiology (Werning et al., 2006, 2008, 2012). Moreover, P. parvulus 2.6 and includes Crohn’s disease (CD) and ulcerative colitis (UC; and L. lactis NZ9000[pGTF] synthesize, respectively the EPS Kamada et al., 2013; Morhardt et al., 2019). They are chronic P and the EPS L polymers having the same primary structure, gastrointestinal disorders, where the dysregulation of the immune which is different from those of the β-glucans isolated from system is responsible for immunological imbalance characterized fungi and yeasts, which have ramifications at positions O-4 by the production of pro-inflammatory cytokines (such as and O-6 (Thompson et al., 2010). Frontiers in Microbiology | www.frontiersin.org 2 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan This peculiarity of structure was the reason why we investigated The study was approved by the Ethics Committee (Comité the probiotic potential of the producing strain P. parvulus 2.6 Ético de Investigación Clínica, Hospital Vall d’Hebron, Barcelona). (Werning et al., 2012; Pérez-Ramos et al., 2017), and the Written informed consent was obtained from all patients [CEIC: prebiotic potential of its EPS P (Russo et al., 2011; Pérez- PR(AG)56/2010]. Ramos et al., 2017). Also, our previous work supports an anti-inflammatory effect of this O2-substituted-(1–3)-β-D-glucan. EPS P and EPS L Production and Comparison of the behavior of P. parvulus 2.6 with its isogenic Purification EPS P-non-producing strain, revealed that the presence of the Exponential cultures of P. parvulus 2.6 (Dueñas-Chasco et al., polymer decreased the pro-inflammatory effect exerted by the 1997; Pérez-Ramos et al., 2018) and L. lactis NZ9000[pGTF] LAB on human macrophages M1, indicating a possible activity (Werning et al., 2008) were used to produce EPS P and EPS L, of this EPS as an immunomodulator in the innate immune respectively. The EPS were produced and purified, as previously response (Fernández de Palencia et al., 2009). Moreover, in described (Notararigo et al., 2013). Briefly, after removal of an induced inflammation model, using the zebrafish transgenic bacterial cells by centrifugation, the EPS present in the culture line Tg(mpx:GFP)i114, the polymer inhibited neutrophil supernatants were precipitated with three volumes of ethanol. recruitment and proliferation in the larvae, confirming once Then, the EPS were further purified by dialysis and fractionation again its potential as an immunomodulator (Pérez-Ramos et al., by size exclusion chromatography after resuspension in a 0.3 M 2018). Furthermore, the metabolic rate of macrophages derived NaOH solution. Afterwards, the EPS alkaline solution was from human monocytes increased upon exposure to either dialyzed as above. Finally, the EPS was lyophilized and left at the EPS P and the EPS L synthesized by L. lactis NZ9000[pGTF] room temperature until use. After the first and second (Notararigo et al., 2014). In addition, these EPS activated lyophilization, the purity of the EPS was tested fluorometrically processes involved in M1 differentiation, migration, and cellular using specific fluorescent staining kits for DNA, RNA, and proliferation, as well as inhibited AKT and mTor pathways proteins as previously reported (Zarour et al., 2018). Solutions implicated in the inflammatory response (Notararigo et al., of the purified polymers were prepared at 1 mg ml−1. No 2014). Therefore, this current work aimed to study the contaminants were detected (Table 2). immunomodulatory effect of the O2-substituted-(1–3)-β-D- glucan, vs. the O6-substituted-(1–3)-β-D-glucan, (laminarin, isolated from Laminaria digitata), on ileocolonic biopsies of Organ Culture of Human Colonic Mucosa CD patients. Organ culture assays were performed as described (Borruel et al., 2002, 2003; Llopis et al., 2009). Full-thickness ileal wall specimens, including areas with macroscopic lesions, were MATERIALS AND METHODS collected at surgery. After rinsing and washing with sterile saline solution, the specimens were transferred to the laboratory Patients in sterile saline solution at 4°C. The intestinal mucosa was Samples of intestinal mucosa were obtained during surgery removed from the tissue, and cut into pieces of approximately from four patients with CD (one male and three females, age 25–35 mg weight each, making an equal distribution of the 46.7 ± 19.9, SEM), who underwent ileal resection for stricture macroscopic lesion. Each piece was placed on the insert of a unresponsive to conventional medical treatment (Table 1). 12-well cell culture plate (Netwell culture system, Costar), with Biopsies of the intestinal tissue were kept at 4°C, from collection the epithelial surface uppermost. Filters were placed into the until their later handling. None of the patients had been on wells and incubated with 1,500 μl RPMI 1640 culture medium anti-TNF treatment for at least 2 months before the intervention. (CanSera), without antibiotic (37°C, 95% O2, 5% CO2 -carbogene-). The diagnosis of CD was previously established by the Before usage, the culture medium was filtered through a 0.22 μm clinical routine, as well as radiological and endoscopic, criteria, membrane, warmed to 37°C, and gassed for 90 min with and afterwards was confirmed by histological evaluation of carbogen. Tissue without lesion (not inflamed) was used as the surgical specimen. All patients received the same preparation negative control. for colonic surgery including gut lavage with electrolyte- The tissues obtained from biopsies of CD patients were polyethylene glycol solution and broad-spectrum antibiotic exposed independently to EPS L, EPS P, or laminarin. A solution therapy. of the polymers in culture medium at 100 μg ml−1 was added by dripping onto the tissues. The cell culture plate was covered TABLE 1 | Characteristics of the patients included in the pilot study. TABLE 2 | Detection of contaminants in EPS preparations. Patients Age Sex Affected Debut date Treatment segment 1 lyophilization 2 lyophilization Contaminants 1 20 F Ileocecal 2 years Azathioprine + budesonide EPS P EPS L EPS P EPS L 2 42 F Ileocecal less than Amoxicillin + clavulanate 6 months DNA (%) 0 0 0 0 3 57 F Ileocecal 8 years Mesalazine + azathioprine RNA (%) 0 <0.1 0 0 4 68 M Ileocecal 28 years Azathioprine Proteins (%) <0.1 1.2 0 0 Frontiers in Microbiology | www.frontiersin.org 3 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan with a supported lid, allowing a correct exchange of gases, The analysis with the array was performed according to and subsequently placed in a bath at 37°C, inside a container the manufacturer’s instructions, as follows. The substrate used shielded with a wet cloth. Then, the container was connected was cDNA synthesized from RNA samples (obtained from to the carbogen gas outlet, gassed at high pressure for 10 s patient biopsies) with the “RT2 First Strand” kit (Qiagen). The at 1 h intervals, and incubated for 4 h. quantitative-RT-PCR (qPCR) was carried out using the SYBR Then, the pro-inflammatory (TNF-α and IL-8) and anti- GREEN technology “RT2 SYBR Green,” and run in a thermal inflammatory (IL-10) cytokine levels released into the media cycler iQ5 (BioRad). Once the reaction was complete, the data were quantified by ELISA. At the end of the experiment, corresponding to the Ct values of each gene were exported aliquots of the supernatant were collected from each well, prior to a data sheet for further analysis. The analysis of the differential to being stored at −80°C, while the tissue was kept immersed gene expression was carried out through the web application in 400 μl of rRNA stabilization solution “RNA later” (Ambion) developed by the company for this purpose based on the at 4°C overnight, and then stored at −80°C. method ΔΔCt and using three of the five constitutive genes for the standardization of the data (β-actine, β-2-microglobulin, Determination of Cytokine Levels by Elisa and the ribosomal P0 protein; Seeger et al., 2014). The cytokine protein profiles of tissue culture supernatants, in response to EPS treatments, were analyzed by OptEIA ELISA Bioinformatic Functional Analysis (BD Pharmingen) to detect IL-10 and TNF-α, as well as by DuoSet The functional analysis of the selected genes was performed ELISA (RD Systems) for IL-8, following the suppliers’ instructions. using the Database for Annotation, Visualization and Integrated The cytokine concentrations were extrapolated from the Discovery (DAVID 6.7; http://david.abcc.ncifcrf.gov), which regression line generated using the absorbance values of a integrates the information of over 1.5 million genes in more standard curve, which was produced in the same test with than 65.000 species, from different public sources of gene and known concentrations of commercial cytokines. protein notations (Huang et al., 2008). DAVID provides a functional classification of genes by extracting the information Gene Expression Profiling Under EPS from various databases. We selected terms from Gene Ontology Treatment (GO; http://geneontology.org/) and from molecular pathways In gene expression organ culture assays, CD samples were integrated in: the Biological Biochemistry Image Database (BBID, harvested in the presence of RNA later (Ambion) and stored http://bbid.irp.nia.nih.gov/), Biocarta (http://www.biocarta.com/ at −80°C. RNA extraction from the patients’ biopsies was genes/index.asp), and the Kyoto Encyclopedia of Genes and performed using the RNeasy mini (Qiagen), according to the Genomes (KEGG, http://www.genome.jp/kegg/). supplier’s instructions. RNA concentration and integrity were Gene ontology terms allows the unification of the analyzed and determined with the RNA 6000 Nano Chip in representation of the attributes of genes and their products a Bioanalyzer 2100 (Agilent Technologies). RNA quantity and among the species, favoring the functional interpretation of integrity were considered acceptable if the 28S/18S ribosomal experimental data by dividing the terms into three categories: fragment ratio was over 1.5, and the RNA Integrity Number cellular compartment (CC), molecular function (MF), and (RIN) ranged in values between 9 and 10. Selected gene biological process (BP). The use of molecular pathways databases expression was evaluated with real-time PCR, and 1 μg of facilitates the production of graphic information of how genes total RNA was used for the reverse transcription to synthesize and their products interact. first strand cDNA following the recommended protocol of the High-Capacity cDNA Reverse Transcription kit (Applied Statistical Analysis Biosystems). Quantitative PCR was then carried out with the Statistical analysis was carried out using the Prism 8 (GraphPad) Taqman Gene Expression Assay (Thermo fisher Scientific), using software. The normality of the data was tested by the Kolmorov- TaqMan Fast Universal PCR Master Mix (2X; Applied Biosystems). Smirnov or Shapiro-Wilk Test normality test. For parametric Relative quantification of gene expression of thymic stomal data, a two-tailed paired t-test was applied, while for lymphopietin (TSLP), IL-12p35, IL-10, and IL-8 was determined, nonparametric data, the Friedman Test for paired data was used. using a 7500 Fast Real time PCR System (Applied Biosystems). For qPCR, for parametric data, unpaired two-tailed Student Data were obtained as threshold cycle (Ct) values. Gene expression t-test was performed with Welch’s correction. Values with levels for each individual sample were normalized relative to p < 0.05 were considered significant. the PPIA gene as housekeeping/constitutive/endogenous gene, which encodes the Peptidylprolyl Isomerase A (also called Cyclophilin A). Each condition was run in triplicate. RESULTS AND DISCUSSION Transcription Factor Signaling Pathways Effect of O2-Substituted-(1–3)-β-D-Glucans NFκB RNA Array on Cytokine Production in ex vivo Models The RT2Profiler PCR array (SABiosciences-Qiagen) was used The EPS P and EPS L synthesized by P. parvulus 2.6 and to obtain the expression profile of 84 fundamental genes of the L. lactis NZ9000[pGTF], respectively, are identical O2- NFκB pathway (Supplementary Table S1; supplementary data). substituted-(1–3)-β-D-glucans with the same primary structure Frontiers in Microbiology | www.frontiersin.org 4 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan (Dueñas-Chasco et al., 1997; Werning et al., 2014). However, Tissues obtained from biopsies of four CD patients were the polymers have different molecular masses, 9.6 × 106 Da exposed independently to the polysaccharides (Figure 1B); and for EPS P and 6.6 × 106 Da for EPS L (Werning et al., 2014). subsequently the pro-inflammatory (TNF-α and IL-8) and anti- The molecular mass of polysaccharides plays a role in their inflammatory (IL-10) cytokine levels released into the media biological activity, consequently, there may be differences between were quantified (Figure 1C). The results revealed that the the immunomodulatory activities of each EPS, and therefore concentration of the pro-inflammatory IL-8 secreted by the both polymers were tested. Thus, in the present pilot study, tissues during treatment with either of the three β-glucans to evaluate EPS P and EPS L (Figure 1A), ex vivo tissue tested was significantly lower (p < 0.05) than that released by cultures have been used as a model of pathological intestinal the untreated control tissue. In addition, the highest effect inflammation (Figure 1B), in which the immune system plays was obtained with EPS P or laminarin. These results indicated an important role in the development, propagation, and a role of the O2-substituted-(1–3)-β-D-glucans on the reduction maintenance of inflammation and disease. We have shown of inflammation associated with the intestinal epithelial mucosa. that this model is suitable to investigate the effects of LAB However, the levels of TNF-α and IL-10 secreted by the tissues treatment on the intestinal immunity axis (Bäuerl et al., 2013). were not affected by the treatments. This lack of influence Laminarin was also tested as a positive control due to its could be due to the cytokine’s short lifespan and to the large ability to activate the immune system through its interaction number of proteases released by the tissue of CD patients with the Dectin-1 receptor (Xie et al., 2010; Smith et al., 2018). (Borruel et al., 2002). These hypotheses presumably did not A B C FIGURE 1 | Evaluation of (1–3)-β-D-glucans in ex vivo model of inflamed gut mucosa. (A) The primary structure of EPS L and EPS M is depicted. (B) Organ culture assay, the tissues were collected from Crohn’s disease (CD) patients and incubated in the culture medium for 4 h. Then, the media were used to perform ELISA Immunoblot. (C) Concentration of tumor necrosis factor-α (TNF-α), IL-10, and IL-8 secreted by the tissues represented in Barplot mean with SEM. The levels of cytokines were quantified by the ELISA immunoblot method. Data passed the Kolmorov-Smirnov normality test, and statistical significance was tested with two tailed t-test between each treatment and control (Ctrl). IL-8 was decreased in inflamed tissue treated with EPS L p < 0.03 (*), EPS P p < 0.0068 (**), and laminarin p < 0.0097 (**) vs. untreated control. Frontiers in Microbiology | www.frontiersin.org 5 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan apply for the results obtained for IL-8, since the levels detected the cell type that is expressing it (epithelial cells or DC for this cytokine were 40 times higher than those observed cells; Biancheri et al., 2016; Park et al., 2017). Recently, it for the TNF-α and IL-10. Therefore, the alteration of IL-8 has been proposed that TSLP also participate in Treg cell concentration seems to reflect the biological status of the tissue. development and gut homeostasis perpetuation (Tahaghoghi- One of the CD patients, whose ileocolonic biopsy was analyzed Hajghorbani et al., 2019). Our results suggest that EPS L and previously treated with antibiotic (Table 2). Therefore, the and laminarin exhibit normal values of gene expression, results obtained could have a bias due to the inclusion of his leading to a balanced TSLP function, while EPS P slightly biopsy in the study. However, quantification of the cytokine decreases its expression, and probably has no effect on TSLP levels released by the biopsies of only the three other patients turn over. EPS P fold change reached 0.74, which means provided the same kind of pattern and behavior that the decrease of relative expression is not so dramatic (Supplementary Figure S1), validating the performed assay. as expected in CD patients with active inflammatory reactions (Tahaghoghi-Hajghorbani et al., 2019). Our results also revealed that the NFκB transcript was Effect of O2-Substituted-(1–3)-β-D – over-expressed in tissues exposed to EPS L or laminarin Glucans on Gene Expression Profiling in (Figure 2). NFκB is a transcription factor that regulates the ex vivo Model expression of pro-inflammatory genes, including cytokines, To understand the immunomodulatory effect of the bacterial chemokines, and other molecules. NFκB is a heterodimer β-glucan on gene expression in the ex vivo model, levels of formed by RelA/1 that activates the canonical signaling pathway transcription of the TSLP, IL-8, IL-12p35, and NFκB coding and Rel B/NFKB2 that activates the non-canonical pathway genes in the treated vs. untreated tissues were determined by (see details in Figure 3). The first is activated by TNF-α or means of qPCR (Figure 2). IL1, while the latter is activated by cluster differentiation 40 The analysis of gene expression in the treated vs. untreated (CD40), B cell activating factor, and other molecules, but not tissues revealed in general a similar pattern of response to by TNF-α. The heterodimer activation mechanism is complex; the exposure to EPS L, EPS P, or laminarin. Thus, the additionally, the presence of cofactor molecules in the nucleus, treatment with any of the three β-glucans resulted in a that bind to the transcription factor complex, may influence decrease of TSLP and IL-8 transcription. However, this effect the transcription of pro- or anti-inflammatory proteins coding was only statistically significant (p = 0.0367) for TSLP in genes (Lawrence, 2009). Its activity is regulated by the endogenous tissues treated with EPS P. TSLP is a protein, which belongs cytoplasmic inhibitors IκKs kinases complex that ensures the to the cytokine’s family and, that promotes T cell maturation activation/repression of the heterodimer, that in specific when antigen presenting cells are activated. In IBD patients, conditions is able to transmigrate to the nucleus and induce low levels of TSLP are associated with a permeability increase transcription (Jobin and Sartor, 2000). In IBD, the canonical of the gut barrier, or Th2 cell differentiation, depending on pathway is activated constitutively, thus conspicuous amounts FIGURE 2 | (1–3)-β-D-glucans modulate the relative gene expression of the biopsies of patients affected by CD. Barplot representation mean with SEM of TSLP, NFκB, IL-8 IL-12p35, and IL-10. Gene expression was analyzed by quantitative-RT-PCR (qPCR) analysis. Data passed the Shapiro-Wilk normality test, the unpaired two-tailed Student t-test was performed with Welch’s correction Only TSLP demonstrated a significance difference with a p < 0.03 vs. untreated control. Frontiers in Microbiology | www.frontiersin.org 6 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan A B FIGURE 3 | Effect of treatment with EPS L (A,B) or EPS P (B) on apoptotic signaling pathway. Adapted from Kyoto Encyclopedia of Genes and Genomes (KEGG; Kanehisa et al., 2014). of TNF-α are released by immune system cells in gut mucosa, reduce inflammation (Pérez-Ramos et al., 2018). Also, when as well as interferon-γ (IFN-γ) or IL-8 (Lawrence, 2009). EPS P was tested in an induced inflammation model using Our results indicate that EPS L and laminarin increased the zebrafish transgenic line Tg(mpx:GFP)i114, a significant the relative expression of NFκB (RelA subunit) 1.5-fold higher decrease of neutrophil recruitment to the inflammation focus than that of the control, while EPS P does not have any was observed (Pérez-Ramos et al., 2018). major effect. To comprehend the EPS effect on NFκB activation, Analysis of IL12p35 gene expression confirmed that the it would be necessary to study all genes involved in the treatments did not augment the pro-inflammatory response in NFκB signaling pathway and clarify EPS activity. We would the ex vivo model (Figure 2). The fold change values were expect that the observed increase of NFκB expression would close to the untreated control, henceforth, taking into account determine the increase of pro-inflammatory mediators, like that IL12p35 is the specific subunit of IL12 cytokine, EPS do IL-8 or TNF-α, instead we detected the reduction of the not activate mechanisms involved in IL12 signaling pathway. IL-8 transcript (Figure 2) and no significant difference in Overall, the results obtained indicated a trend toward reduction TNF-α detection in cultured media (Figure 1), or in gene of inflammation in CD biopsies treated with any of the three expression (Figure 2). β-glucans. IL-8 is a chemokine produced by macrophages and intestinal IL-10 is an anti-inflammatory cytokine, that is released in epithelial cells (IEC); it is considered a good marker to the gut by macrophages and IEC to maintain homeostasis determine disease activity (Mohammed Vashist et al., 2018). (Latorre et al., 2018; Morhardt et al., 2019). According to our It is also known as neutrophil chemotactic factor, because results, IL-10 mRNA levels did not show significant differences it induces chemotaxis in target cells such as neutrophils or in the expression profiling (Figure 2), confirming the results other cells that migrate toward the infection focus (Llopis obtained in cytokine production (Figure 1). This agrees with et al., 2009; Walana et al., 2018). Our results showed that our previous observation that macrophages derived from human the amount of the protein IL-8 (Figure 1) and the expression monocytes treated independently with either EPS did not of its coding gene (Figure 2) decreased during treatments, stimulate IL-10 production (Notararigo et al., 2014). Moreover, suggesting that neutrophil recruitment also decreased as a we observed the same behavior in a gnotobiotic zebrafish larvae consequence of the reduction of the inflammation rate. In model, where EPS P treatment did not result in an increase line with these results, we have previously detected that EPS of IL-10 expression (Pérez-Ramos et al., 2018). Taken together, P treatment decreased IL-8 and TNF-α expression in a these results support that bacterial β-glucans do not interact gnotobiotic zebrafish larvae model, indicating its ability to with the IL10R receptor (Shouval et al., 2014). Frontiers in Microbiology | www.frontiersin.org 7 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan Effect of EPS L and EPS P on the NFκB indicating a possible function in IEC, restoring the gut barrier Gene Expression Profile of CD Patient integrity by preventing a dysregulated cell death mechanism. Biopsy Moreover, the results seemed to support that the EPS selectively After detecting IL-8 reduction at transcriptional and protein activated NFκB heterodimers, EPS P RelB/NFKB2, and EPS L levels, as well as NFκB modulation, we decided to investigate RelA/NFKB1 (Figure 4A; Jobin and Sartor, 2000). the effect of β-glucans on NFκB signaling pathways, carrying We selectively studied the NFκB heterodimer complex formed out a gene expression “array” with the RNA from the CD patient by RelB/NFKB2 and RelA/NFKB1 because NFκB1 and NFκB2 that showed the greatest IL-8 relative expression reduction. act as inhibitors, blocking transcription factor translocation to Moreover, to have a representative analysis, the results obtained the nucleus until an external stimulus activates the proteolysis for EPS treated and untreated samples were normalized taking that converts them into p50 and p52, respectively. The in account the expression of three houskeeping genes (see details overexpression of both inhibitors, showed in Figure 4D, could in “Materials and Methods”), and the results are depicted in be the key for the immunomodulatory effect of bacterial β-glucan. Figure 4. The profiles in response to treatment with either EPS According to these results, EPS P seemed to exert its impact P or EPS L were, in general similar (Figures 4B–E). The expression mainly on the MyD88 independent signaling pathway, as shown of the cytokines IL1A and IL1B coding genes was reduced (fold by its influence on RelB, Interleukin 1 Receptor Associated change: IL1A 0.8, 0.7; IL1B 0.5, 0.8 for EPS L and EPS P Kinase 1 (IRAK1), RIPK1, Inhibitor of Nuclear Factor Kappa treatment, respectively) as well as the levels of transcripts encoding B Kinase Subunit Epsilon (IKBKE), Signal Transducer and the colony-stimulating factor 3 (CSF3) and the MyD88 adapter Activator of Transcription 1 (STAT1), and CD40 (Figures 4C,D). protein. These results point out that both EPS have an On the other hand, EPS L could modulate both the canonical immunomodulatory effect on the gut immune system. Receptor and non-canonical pathways (Figure 4A). The first, MyD88 Interacting Serine/Threonine Kinase 1 (RIPK1) was upregulated dependent, by activating Rel, Toll Like Receptor 6 (TLR6; data A B C D E FIGURE 4 | Effect of EPS P and EPS L on expression of genes involved in NFκB pathway. (A) NFκB heterodimers activate canonical and non-canonical pathway. (B), (C), (D), and (E) Expression profiling of representative genes (fold change), whose expression resulted affected by the treatment. Data of the patient who showed the lower qPCR for IL-8. Frontiers in Microbiology | www.frontiersin.org 8 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan not shown), and RIPK1, and repressing IL1A and IL1B gene EPS L could affect extrinsic apoptosis pathways with the expression. The latter, MyD88 independent, by activation IKBKE inhibition of Tumor Necrosis Factor Apoptosis-Inducing Ligand and STAT1 (Figure 4C). In line with these findings, our previous 10 (TNFSF10) and IL-1B, while also it could activate survival results demonstrated that EPS P downregulated the relative pathway (RIPK1), with a possible healing effect on gut barrier, expression of three genes involved in inflammation, MyD88 through the NFκB pathway and chemotaxis reduction upon among them, in a zebrafish larvae model, pointing out the CSF3 downregulation. EPS P could activate apoptosis pathway potential of the O2-substituted-(1–3)-β-D-glucan as an by altering gene expression of Tumor Necrosis Factor Receptor immunomodulator (Pérez-Ramos et al., 2018). Superfamily Member 10a (TNFRSF10A), and the cell death Although EPS P and EPS L could have a different effect regulator BCL2L1 while repression of CSF3 may points to a on transduction pathways activated by NFκB, this analysis does role in modulation of intestinal immunity, with a possible not fully clarify what mechanisms they may be modulating. decrease in the chemotaxis of the neutrophils to the foci of Both have been shown to reduce the expression of inflammation (Carol et al., 2006). pro-inflammatory cytokines, such as IL-1A, IL-1B, TNF, and To support this hypothesis, some of the results depicted in IL-8 (Figure 4B), and these results have been validated in Figure 4E should be discussed in the context of the apoptotic different models (Notararigo et al., 2014; Pérez-Ramos et al., signal pathways. BCL2A1 is a gene which belongs to the BCL-2 2018). The overall results were quite complex, considering that family, the protein encoded by this gene is able to reduce the many of the genes studied are involved in the signaling of release of pro-apoptotic cytochrome c from mitochondria and different receptors and are present in the majority of cell types block caspase activation. The gene showed a 1.7- or 0.9-fold isolated in the CD biopsy. For this reason, it was necessary change upon treatment with EPS L or EPS P. This gene is to use bioinformatics tools to group the genes whose function directly activated by NFκB in response to inflammatory mediators, may be related, and improve the understanding of the mechanisms such as granulocyte-macrophage colony-stimulating factor underlying the stimulus provided by the O2-substituted-(1–3)- (GM-CSF), CD40, and cytokine like TNF-α or IL-1 which β-D-glucan, as described below. suggest lymphocyte activation/proliferation and cell survival. Yet, both EPS downregulate TNF-α, IL-1A, and IL-1B, and Functional Analysis With David have no effect on GM-CSF (data not shown), while only EPS Bioinformatic Resources P upregulated CD40. Taken together, these results suggest that In order to interpret the results of the differential response EPS L had an anti-apoptotic effect, while EPS P had a possible caused by each EPS on the tissue of the CD patient, a pro-apoptotic effect. Indeed, both β-glucans showed an anti- bioinformatic clustering with DAVID was carry out. The aim inflammatory effect, because of reduction of the relative of this analysis was to arrange genes and derive their biological expression of cytokines, that might reflect in reduced lymphocyte meaning, by association with identical or related functions activation for EPS L and in cell death activation for EPS P. and/or pathways. We restricted gene analysis to fold changes The BCL2L1 gene product is a potent pro-apoptotic activator, between ≤5 and/or ≥1.5, to establish strong interaction between and this gene was upregulated by both EPS treatments with selected genes. a fold change of 1.6 for EPS P and 1.2 for EPS L. We postulate The gene clustering revealed a modulation of 14 or 19 that the O2-substituted-(1–3)-β-D-glucans might reestablish genes upon treatment with either EPS L or EPS P. Among normal condition of the intestinal immune system cells through the regulated genes, the presence of EPS L resulted in a an apoptosis activation mechanism (Figure 3B). reduction of expression of 4, whereas downregulation of only In addition, expression of the cytokine TNFSF10 was one gene was observed with the EPS P treatment. Also, 10 downregulated in the presence of both EPS (fold change: 0.5 or 19 where upregulated after exposure of the tissues to EPS and 0.4 for EPS P and EPS L treatments, respectively) while L or EPS P, respectively. Furthermore, as a response to both TNFRSF10A, that binds TNFSF10 was slightly upregulated treatments, the alteration of expression of five genes showed under EPS P treatment (fold change 1.3), and downregulated the same pattern: upregulation for IKBKE, PC4, and SFRS1 upon exposure to EPS L (fold change: 0.7). Moreover, Tumor interacting protein 1 (PSIP1), RIPK1 and STAT1, and Necrosis Factor Receptor Superfamily Member 10b (TNFRSF10B), downregulation for Colony-Stimulating Factor 3 (CSF3). and IL-1R receptors coding genes were downregulated (fold The functional analysis for molecular function of GO depicted change: 0.9 and 0.7, respectively) in EPS L treated tissues, in Table 3 showed that the enrichment in functional terms and upregulated (fold change: 1.4 and 1.7, respectively) in for genes affected by treatment with EPS P, was greater than tissues exposed to EPS P (Figures 4B,C). These results indicate that with EPS L in number, degree of enrichment, and number that EPS L and EPS P reduced the expression of the apoptotic of members in each category. In addition, EPS L regulon ligand TNFSF10, but had a different effect on the expression included genes grouped into two categories related to cytokine of receptor TNFRSF10B, an IL-1R. It is possible that EPS P activity, while for EPS P regulon more grouping terms were activates a cell death mechanism, to regulate lymphocyte found, highlighting the regulation of caspase activity and proliferation, while EPS L acts at the cell mediation level by cytokine activity. These results confirmed the data obtained reducing cytokine production. from functional grouping (data not shown), and the overall Therefore, EPS P seems to modulate apoptosis, as several results allowed us to postulate a possible mechanism of action results previously collected support the bioinformatic analysis. of either EPS on the apoptosis pathway represented in Figure 3. Some examples are, the lowering of IL-8 in an ex vivo model, Frontiers in Microbiology | www.frontiersin.org 9 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan TABLE 3 | Molecular function gene ontology (GO) terms for the gene differentially expressed upon EPS P or EPS L treatment. GO TERM ENRICHMENTa GENES EPS P GO:0043028~caspase regulator activity 47,12 TNFRSF10A, BCL2L1 GO:0005125~cytokine activity 10,51 CSF3, CCL2, and CSF1 GO:0004674~protein serine/threonine kinase activity 7,94 IRAK1, IKBKE, MAP3K1, RIPK1, and RAF1 GO:0004672~protein kinase activity 6,76 IRAK1, IKBKE, CCL2, MAP3K1, RIPK1, and RAF1 GO:0008134~transcription factor binding 5,33 TNFRSF10A, RELB, BCL3, and NFKB2 GO:0019899~enzyme binding 5,23 CSF3, IRAK1, MAP3K1, and CD40 GO:0042802~identical protein binding 4,27 IRAK1, MAP3K1, CSF1, and BCL2L1 GO:0003700~transcription factor activity 3,50 RELB, IRF1, BCL3, NFKB2, and STAT1 GO:0030528~transcription regulator activity 2,71 IRAK1, RELB, IRF1, BCL3, NFKB2, and STAT1 EPS L GO:0032813~tumor necrosis factor receptor superfamily binding 64,43 TNFSF10, RIPK1 GO:0005125~cytokine activity 20,49 CSF3, TNFSF10, IL1B, and CCL5 a Enrichment of functional terms is shown as a proportion. It represents the gene list input divided by all the genes annotated to the selected GO term. a zebra fish model and in an in vitro transwell model It is worth noting that the differential response of the tissues (unpublished data). The decreased level of IL-1B and IL12p35 in to EPS P and EPS L might be related to their structural LPMCs isolated from CD biopsies and treated with either differences. EPS P has a molecular mass of 9.6 × 106 Da, EPS, detected with Luminex (data not shown). IL-8 takes higher than that of EPS L (6.6 × 106 Da, Werning et al., part in the neutrophil recruitment, while IL-1B and IL-12p35 2014), that could increase the affinity of the natural are Th1 mediators that augment when Th1 cells are deregulated, O2-substituted-(1–3)-β-D-glucan (EPS P) affinity for its therefore indicating a possible implication of physiological receptor(s). It has been shown that molecular mass and tertiary modulation of apoptosis in these T cells subpopulations. It structure of β-glucans, play a critical role in receptor binding is well-established that IBD pathogeneses determines Th1/Th2 (Legentil et al., 2015). We have previously shown that EPS P imbalance toward Th1 (Neurath, 2014). Th1 proliferation is increases adhesion of P. parvulus 2.6 to human enterocytes in decontrolled because apoptosis regulation is imbalanced, vitro and increases in vivo the colonization capacity of the increasing inflammation and chemotaxis in the gut. Hence, bacterial strain in the zebrafish gut (Pérez-Ramos et al., 2018). Th1 cells overproduce pro-inflammatory mediators like: TNF-α, Therefore, it is tempting to assume that, after ingestion in a IL1A, IL1B, IFN-γ etc. In turn TNF-α, that has a pivotal role beverage, P. parvulus 2.6 could synthesize EPS P in the human in the mucosal inflammation, recruit, and over stimulate intestine and the action of glycosidases synthesized by the immune system cell like Th1. microbiota could reduce its molecular mass generating an EPS Our data suggest that apoptosis intrinsic pathway mechanism L-like polymer with its detected dual role. (Stevens and Oltean, 2019), or in ameliorating gut barrier inflammation (Lu et al., 2020), could be re-established after O2-substituted-(1–3)-β-D-glucan administration, due to the fact CONCLUSION that Bcl2 genes are overexpressed (Figure 4). According to that, Th1 proliferation could be regulated and therefore reducing In an ex vivo model of CD biopsies, treatment with EPS P proinflammatory cytokine production’s (Figures 1C, 2, 4E). and EPS L are concomitant with decreased levels of As IBD is Th1 shifted pathology, and EPS P is able to pro-inflammatory cytokine IL-8 in the supernatant of tissue ameliorate the rate of inflammation, it is plausible to postulate biopsies, and in relative gene expression. RNA array demonstrated that it could play a role in Th1 through apoptosis modulation. that O2-substituted-(1–3)-β-D-glucans modulate the NFkB Bioinformatic analysis revealed that many mediators of this pathway, activating both the canonical and non-canonical pathway have been modulated by the treatment (Table 3), and pathway. The differences found in the expression profiling we also observed a reduction of pro-inflammatory mediators indicate that the 9.6 × 106 Da EPS P may be able to restore (Figures 1C, 2). As the Th1 of CD patients cycles faster than the apoptosis mechanism in Th1 cells, whereas the 6.6 × 106 Da the healthy controls (Sturm et al., 2004), it is fair to believe EPS L might restore the intestinal barrier through activation that Th1 could play a role in the re-establishment of Th1 of RIPK1 in IEC. Further studies have to be performed to population to a normal concentration, and therefore lowering determine which receptor is activated by EPS P and EPS L inflammation rate. and, if a co-receptor is necessary for receptor recognition, and Regarding EPS L, gene clustering showed a dual behavior, if EPS have different effects on immune system cells. by the reduction of cell death ligands expression such as TNFSF10 and IL-1B, and over-expression of RIPK, a kinase involved in necroptosis and cell survival, by modulating DATA AVAILABILITY STATEMENT dysregulated apoptosis in IEC (Figure 3A; Dannappel et al., 2014). Thus, the EPS treatments affect mechanisms that restore The original contributions presented in the study are included gut burrier homeostasis due to its wound healing properties in the article/Supplementary Material, further inquiries can and reduce inflammation mediators. be directed to the corresponding authors. Frontiers in Microbiology | www.frontiersin.org 10 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan ETHICS STATEMENT FUNDING The studies involving human participants were reviewed This work was supported by the Spanish Ministry of Science, and approved by Comité Ético de Investigación Clínica, Innovation, and Universities (grant RTI2018-097114-B-I00), by Hospital Vall d’Hebron, Barcelona. The patients/participants the Fondo de Investigación Sanitaria, grant number PI12/00263, provided their written informed consent to participate in and by CIBERehd. this study. ACKNOWLEDGMENTS AUTHOR CONTRIBUTIONS We thank Dr. Maria José Feito for her advice on the interpretation of the β-glucan immunological effect. We also thank Dr. Sara SN performed all the experiments described in this article, Ballester and Dr. Stephen Elson for critical reading of the realized the statistical analysis, and wrote the manuscript. manuscript. We acknowledge support of the publication fee EV and AO helped with the ex vivo model during SN stay by the CSIC Open Access Publication Support Initiative through at Vall d’Hebron laboratory. IC realized the bioinformatic its Unit of Information Resources for Research (URICI). analysis and with the design of the Figures. MA supervised the ex vivo model and the RNA Array. FG reviewed the results of the manuscript. PL and AP masterminded the SUPPLEMENTARY MATERIAL direction and conceptualization of the study, the interpretation of the data obtained, and the final version of the manuscript. The Supplementary Material for this article can be found online All authors listed have read and approved the final version at: https://www.frontiersin.org/articles/10.3389/fmicb.2021.621280/ of the manuscript. full#supplementary-material REFERENCES Chassaing, B., Aitken, J. D., Malleshappa, M., and Vijay-Kumar, M. (2014). Dextran sulfate sodium (DSS)-induced colitis in mice. Curr. Protoc. Immunol. Ali, M. F., Driscoll, C. B., Walters, P. R., Limper, A. H., and Carmona, E. M. 104, 15.25.11–15.25.14. doi: 10.1002/0471142735.im1525s104 (2015). Beta-Glucan-activated human B lymphocytes participate in innate Dannappel, M., Vlantis, K., Kumari, S., Polykratis, A., Kim, C., Wachsmuth, L., immune responses by releasing proinflammatory cytokines and stimulating et al. (2014). RIPK1 maintains epithelial homeostasis by inhibiting apoptosis neutrophil chemotaxis. J. Immunol. 195, 5318–5326. doi: 10.4049/ and necroptosis. Nature 513, 90–94. doi: 10.1038/nature13608 jimmunol.1500559 Devillé, C., Gharbi, M., Dandrifosse, G., and Peulen, O. (2007). Study on the Ballesteros Pomar, M. D., and Gonzalez Arnaiz, E. (2018). Role of prebiotics effects of laminarin, a polysaccharide from seaweed, on gut characteristics. and probiotics in the functionality of the microbiota in the patients receiving J. Sci. Food Agric. 87, 1717–1725. doi: 10.1002/jsfa.2901 enteral nutrition. Nutr. Hosp. 35, 18–26. doi: 10.20960/nh.1956 Dueñas-Chasco, M., Rodríguez-Carvajal, M. A., Tejero-Mateo, P., Basson, A. R., Lam, M., and Cominelli, F. (2017). Complementary and alternative Franco-Rodríguez, G., Espartero, J. L., Irastorza-Iribas, A., et al. (1997). medicine strategies for therapeutic gut microbiota modulation in inflammatory Structural analysis of the exopolysaccharide produced by Pediococcus damnosus bowel disease and their next-generation approaches. Gastroenterol. Clin. N. 2.6. Carbohydr. Res. 303, 453–458. doi: 10.1016/s0008-6215(97)00192-4 Am. 46, 689–729. doi: 10.1016/j.gtc.2017.08.002 Fernández de Palencia, P., Werning, M. L., Sierra-Filardi, E., Dueñas-Chasco, M., Bäuerl, C., Llopis, M., Antolín, M., Monedero, V., Mata, M., Zúñiga, M., et al. Irastorza, A., Corbí, A. L., et al. (2009). Probiotic properties of the 2-substituted (2013). Lactobacillus paracasei and Lactobacillus plantarum strains downregulate (1-3)-β-D-glucan-producing bacterium Pediococcus parvulus 2.6. Appl. Environ. proinflammatory genes in an ex vivo system of cultured human colonic Microbiol. 75, 4887–4891. doi: 10.1128/AEM.00394-09 mucosa. Genes Nutr. 8, 165–180. doi: 10.1007/s12263-012-0301-y Heim, G., Walsh, A. M., Sweeney, T., Doyle, D. N., O’Shea, C. J., Ryan, M. T., Biancheri, P., Di Sabatino, A., Rescigno, M., Giuffrida, P., Fornasa, G., Tsilingiri, K., et al. (2014). Effect of seaweed-derived laminarin and fucoidan and zinc et al. (2016). Abnormal thymic stromal lymphopoietin expression in the oxide on gut morphology, nutrient transporters, nutrient digestibility, growth duodenal mucosa of patients with coeliac disease. Gut 65, 1670–1680. doi: performance and selected microbial populations in weaned pigs. Br. J. Nutr. 10.1136/gutjnl-2014-308876 111, 1577–1585. doi: 10.1017/s0007114513004224 Bodera, P. (2008). Influence of prebiotics on the human immune system (GALT). Hidalgo-Cantabrana, C., Sánchez, B., Álvarez-Martín, P., López, P., Recent Pat. Inflamm. Allergy Drug Discov. 2, 149–153. doi: 10.2174/18722130 Martínez-Álvarez, N., Delley, M., et al. (2015). A single mutation in the 8784543656 gene responsible for the mucoid phenotype of Bifidobacterium animalis subsp. Borruel, N., Carol, M., Casellas, F., Antolín, M., de Lara, F., Espín, E., et al. lactis confers surface and functional characteristics. Appl. Environ. Microbiol. (2002). Increased mucosal tumour necrosis factor α production in Crohn’s 81, 7960–7968. doi: 10.1128/AEM.02095-15. disease can be downregulated ex vivo by probiotic bacteria. Gut 51, 659–664. Huang, D. W., Sherman, B. T., and Lempicki, R. A. (2008). Systematic and doi: 10.1136/gut.51.5.659 integrative analysis of large gene lists using DAVID bioinformatics resources. Borruel, N., Casellas, F., Antolı́n, M. A., Llopis, M., Carol, M., Espı́in, E., Nat. Protoc. 4, 44–57. doi: 10.1038/nprot.2008.211 et al. (2003). Effects of nonpathogenic bacteria on cytokine secretion by Iweala, O. I., and Nagler, C. R. (2019). The microbiome and food allergy. human intestinal mucosa. Am. J. Gastroenterol. 98, 865–870. doi: 10.1111/j. Annu. Rev. Immunol. 37, 377–403. doi: 10.1146/annurev-immunol-042718- 1572-0241.2003.07384.x 041621 Carol, M., Borruel, N., Antolin, M., Llopis, M., Casellas, F., Guarner, F., et al. Jin, Y., Li, P., and Wang, F. (2018). Beta-glucans as potential immunoadjuvants: (2006). Modulation of apoptosis in intestinal lymphocytes by a probiotic a review on the adjuvanticity, structure-activity relationship and receptor bacteria in Crohn’s disease. J. Leukoc. Biol. 79, 917–922. doi: 10.1189/ recognition properties. Vaccine 36, 5235–5244. doi: 10.1016/j.vaccine. jlb.0405188 2018.07.038 Chan, G. C., Chan, W. K., and Sze, D. M. (2009). The effects of beta-glucan Jobin, C., and Sartor, R. B. (2000). The IκB/NF-κB system: a key determinant on human immune and cancer cells. J. Hematol. Oncol. 2:25. doi: 10.1186/ of mucosal inflammation and protection. Am. J. Phys. Cell Phys. 278, 1756-8722-2-25 C451–C462. doi: 10.1152/ajpcell.2000.278.3.C451 Frontiers in Microbiology | www.frontiersin.org 11 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan Kamada, N., Seo, S. U., Chen, G. Y., and Nunez, G. (2013). Role of the gut Pediococcus parvulus 2.6: technological aspects and prebiotic potential. Int. microbiota in immunity and inflammatory disease. Nat. Rev. Immunol. 13, J. Mol. Sci. 18:1588. doi: 10.3390/ijms18071588 321–335. doi: 10.1038/nri3430 Pérez-Ramos, A., Mohedano, M. L., Pardo, M. Á., and López, P. (2018). Kanehisa, M., Goto, S., Sato, Y., Kawashima, M., Furumichi, M., and Tanabe, M. β-Glucan-producing Pediococcus parvulus 2.6: test of probiotic and (2014). Data, information, knowledge and principle: back to metabolism in immunomodulatory properties in zebrafish models. Front. Microbiol. 9:1684. KEGG. Nucleic Acids Res. 42, D199–D205. doi: 10.1093/nar/gkt1076 doi: 10.3389/fmicb.2018.01684 Kau, A. L., Ahern, P. P., Griffin, N. W., Goodman, A. L., and Gordon, J. I. Rattigan, R., O’Doherty, J. V., Vigors, S., Ryan, M. T., Sebastiano, R. S., (2011). Human nutrition, the gut microbiome and the immune system. Callanan, J. J., et al. (2020). The effects of the marine-derived polysaccharides Nature 474, 327–336. doi: 10.1038/nature10213 laminarin and chitosan on aspects of colonic health in pigs challenged Knights, D., Lassen, K., and Xavier, R. (2013). Advances in inflammatory bowel with dextran sodium sulphate. Mar. Drugs 18:262. doi: 10.3390/md18 disease pathogenesis: linking host genetics and the microbiome. Gut 62, 050262 1505–1510. doi: 10.1136/gutjnl-2012-303954 Russo, P., López, P., Capozzi, V., Fernández de Palencia, P., Dueñas, M. A. T., Latorre, E., Layunta, E., Grasa, L., Pardo, J., García, S., Alcalde, A. I., et al. Spano, G., et al. (2011). Beta-glucans improve growth, viability and colonization (2018). Toll-like receptors 2 and 4 modulate intestinal IL-10 differently in of probiotic microorganisms. Int. J. Mol. Sci. 13, 6026–6039. doi: 10.3390/ ileum and colon. United European Gastroenterol J 6, 446–453. doi: 10.1177/ ijms13056026 2050640617727180 Schmid, J., Sieber, V., and Rehm, B. (2015). Bacterial exopolysaccharides: Lawrence, T. (2009). The nuclear factor NF-kappaB pathway in inflammation. biosynthesis pathways and engineering strategies. Front. Microbiol. 6:496. Cold Spring Harb. Perspect. Biol. 1:a001651. doi: 10.1101/cshperspect.a001651 doi: 10.3389/fmicb.2015.00496 Legentil, L., Paris, F., Ballet, C., Trouvelot, S., Daire, X., Vetvicka, V., et al. Seeger, P., Bosisio, D., Parolini, S., Badolato, R., Gismondi, A., Santoni, A., (2015). Molecular interactions of beta-(1-->3)-glucans with their receptors. et al. (2014). Activin A as a mediator of NK–dendritic cell functional Molecules 20, 9745–9766. doi: 10.3390/molecules20069745 interactions. J. Immunol. 192, 1241–1248. doi: 10.4049/jimmunol.1301487 Li, X., Luo, H., Ye, Y., Chen, X., Zou, Y., Duan, J., et al. (2019). Betaglucan, Shouval, D. S., Ouahed, J., Biswas, A., Goettel, J. A., Horwitz, B. H., Klein, C., a dectin1 ligand, promotes macrophage M1 polarization via NFkappaB/ et al. (2014). Interleukin 10 receptor signaling: master regulator of intestinal autophagy pathway. Int. J. Oncol. 54, 271–282. doi: 10.3892/ijo.2018.4630 mucosal homeostasis in mice and humans. Adv. Immunol. 122, 177–210. Llamas-Arriba, M. G., Hernández-Alcántara, A. M., Yépez, A., Aznar, R., doi: 10.1016/B978-0-12-800267-4.00005-5 Dueñas, M. T., and López, P. (2019). “12—functional and nutritious beverages Smith, A. J., Graves, B., Child, R., Rice, P. J., Ma, Z., Lowman, D. W., et al. produced by lactic acid bacteria” in Nutrients in beverages. eds. A. M. Grumezescu (2018). Immunoregulatory activity of the natural product laminarin varies and A. M. Holban (Duxford: Academic Press), 419–465. widely as a result of its physical properties. J. Immunol. 200, 788–799. doi: Llopis, M., Antolin, M., Carol, M., Borruel, N., Casellas, F., Martinez, C., et al. 10.4049/jimmunol.1701258 (2009). Lactobacillus casei downregulates commensals’ inflammatory signals Stevens, M., and Oltean, S. (2019). Modulation of the apoptosis gene Bcl-x in Crohn’s disease mucosa. Inflamm. Bowel Dis. 15, 275–283. doi: 10.1002/ function through alternative splicing. Front. Genet. 10:804. doi: 10.3389/ ibd.20736 fgene.2019.00804 Lu, H., Li, H., Fan, C., Qi, Q., Yan, Y., Wu, Y., et al. (2020). RIPK1 inhibitor Strain, C. R., Collins, K. C., Naughton, V., McSorley, E. M., Stanton, C., ameliorates colitis by directly maintaining intestinal barrier homeostasis and Smyth, T. J., et al. (2020). Effects of a polysaccharide-rich extract derived regulating following IECs-immuno crosstalk. Biochem. Pharmacol. 172:113751. from Irish-sourced Laminaria digitata on the composition and metabolic doi: 10.1016/j.bcp.2019.113751 activity of the human gut microbiota using an in vitro colonic model. Eur. Mohammed Vashist, N., Samaan, M., Mosli, M. H., Parker, C. E., MacDonald, J. K., J. Nutr. 59, 309–325. doi: 10.1007/s00394-019-01909-6 Nelson, S. A., et al. (2018). Endoscopic scoring indices for evaluation of Sturm, A., Leite, A. Z., Danese, S., Krivacic, K. A., West, G. A., Mohr, S., disease activity in ulcerative colitis. Cochrane Database Syst. Rev. 1:Cd011450. et al. (2004). Divergent cell cycle kinetics underlie the distinct functional doi: 10.1002/14651858.CD011450.pub2 capacity of mucosal T cells in Crohn’s disease and ulcerative colitis. Gut Morhardt, T. L., Hayashi, A., Ochi, T., Quirós, M., Kitamoto, S., Nagao-Kitamoto, H., 53, 1624–1631. doi: 10.1136/gut.2003.033613 et al. (2019). IL-10 produced by macrophages regulates epithelial integrity Tahaghoghi-Hajghorbani, S., Ajami, A., Ghorbanalipoor, S., Hosseini-Khah, Z., in the small intestine. Sci. Rep. 9:1223. doi: 10.1038/s41598-018-38125-x Taghiloo, S., Khaje-Enayati, P., et al. (2019). Protective effect of TSLP and Nácher-Vázquez, M., Ballesteros, N., Canales, A., Rodríguez Saint-Jean, S., IL-33 cytokines in ulcerative colitis. Auto Immun. Highlights 10:1. doi: 10.1186/ Pérez-Prieto, S. I., Prieto, A., et al. (2015). Dextrans produced by lactic s13317-019-0110-z acid bacteria exhibit antiviral and immunomodulatory activity against salmonid Thompson, I. J., Oyston, P. C., and Williamson, D. E. (2010). Potential of the viruses. Carbohydr. Polym. 124, 292–301. doi: 10.1016/j.carbpol.2015.02.020 β-glucans to enhance innate resistance to biological agents. Expert Rev. Neurath, M. F. (2014). Cytokines in inflammatory bowel disease. Nat. Rev. Anti-Infect. Ther. 8, 339–352. doi: 10.1586/eri.10.10 Immunol. 14, 329–342. doi: 10.1038/nri3661 Tsai, Y. L., Lin, T. L., Chang, C. J., Wu, T. R., Lai, W. F., Lu, C. C., et al. Notararigo, S., de Las Casas-Engel, M., Fernández de Palencia, P., Corbí, A. L., (2019). Probiotics, prebiotics and amelioration of diseases. J. Biomed. Sci. and López, P. (2014). Immunomodulation of human macrophages and myeloid 26:3. doi: 10.1186/s12929-018-0493-6 cells by 2-substituted (1-3)-beta-D-glucan from P. parvulus 2.6. Carbohydr. Volman, J. J., Mensink, R. P., Ramakers, J. D., de Winther, M. P., Carlsen, H., Polym. 112, 109–113. doi: 10.1016/j.carbpol.2014.05.073 Blomhoff, R., et al. (2010). Dietary (1-->3), (1-->4)-beta-D-glucans from Notararigo, S., Nácher-Vázquez, M., Ibarburu, I., Werning, M. L., oat activate nuclear factor-kappaB in intestinal leukocytes and enterocytes Fernández de Palencia, P., Dueñas, M. T., et al. (2013). Comparative analysis from mice. Nutr. Res. 30, 40–48. doi: 10.1016/j.nutres.2009.10.023 of production and purification of homo- and hetero-polysaccharides produced Walana, W., Ye, Y., Li, M., Wang, J., Wang, B., Cheng, J. -W., et al. (2018). by lactic acid bacteria. Carbohydr. Polym. 93, 57–64. doi: 10.1016/j.carbpol.2012. IL-8 antagonist, CXCL8(3-72)K11R/G31P coupled with probiotic exhibit 05.016 variably enhanced therapeutic potential in ameliorating ulcerative colitis. Oka, A., and Sartor, R. B. (2020). Microbial-based and microbial-targeted Biomed. Pharmacother. 103, 253–261. doi: 10.1016/j.biopha.2018.04.008 therapies for inflammatory bowel diseases. Dig. Dis. Sci. 65, 757–788. doi: Walsh, A. M., Sweeney, T., O’Shea, C. J., Doyle, N. D., and O’Doherty, J. V. 10.1007/s10620-020-06090-z (2015). Effects of supplementing dietary laminarin and fucoidan on intestinal Orel, R., and Kamhi Trop, T. (2014). Intestinal microbiota, probiotics and morphology and the immune gene expression in the weaned pig. J. Anim. prebiotics in inflammatory bowel disease. World J. Gastroenterol. 20, 11505–11524. Sci. 90, 284–286. doi: 10.2527/jas.53949 doi: 10.3748/wjg.v20.i33.11505 Werning, M. L., Corrales, M. A., Prieto, A., Fernández de Palencia, P., Navas, J., Park, J. H., Jeong, D. Y., Peyrin-Biroulet, L., Eisenhut, M., and Shin, J. I. and López, P. (2008). Heterologous expression of a position 2-substituted (2017). Insight into the role of TSLP in inflammatory bowel diseases. (1-3)-β-D-glucan in Lactococcus lactis. Appl. Environ. Microbiol. 74, 5259–5262. Autoimmun. Rev. 16, 55–63. doi: 10.1016/j.autrev.2016.09.014 doi: 10.1128/AEM.00463-08 Pérez-Ramos, A., Mohedano, M. L., López, P., Spano, G., Fiocco, D., Russo, P., Werning, M. L., Ibarburu, I., Dueñas, M. T., Irastorza, A., Navas, J., and et al. (2017). In situ β-Glucan fortification of cereal-based matrices by López, P. (2006). Pediococcus parvulus gtf gene encoding the GTF Frontiers in Microbiology | www.frontiersin.org 12 February 2021 | Volume 12 | Article 621280 Notararigo et al. Immunomodulatory Effect of Bacterial β-Glucan glycosyltransferase and its application for specific PCR detection o fβ-D- isolated from dairy and non-dairy products. J. Funct. Foods 49, 351–361. glucan-producing bacteria in foods and beverages. J. Food Prot. 69, 161–169. doi: 10.1016/j.jff.2018.09.001 doi: 10.4315/0362-028x-69.1.161 Zhang, H., Sparks, J. B., Karyala, S. V., Settlage, R., and Luo, X. M. (2015). Werning, M. L., Notararigo, S., Nácher, M., Fernández de Palencia, P., Aznar, R., Host adaptive immunity alters gut microbiota. ISME J. 9, 770–781. doi: and López, P. (2012). “Biosynthesis, purification and biotechnological use 10.1038/ismej.2014.165 of exopolysaccharides produced by lactic acid bacteria” in Food additive. ed. Y. El-Samragy (London: Intechopen). Conflict of Interest: The authors declare that the research was conducted in Werning, M. L., Pérez-Rámos, A., Fernández de Palencia, P., Mohedano, M. L., the absence of any commercial or financial relationships that could be construed Dueñas, M. T., Prieto, A., et al. (2014). A specific immunological method as a potential conflict of interest. to detect and quantify bacterial 2-substituted (1,3)-beta-D-glucan. Carbohydr. Polym. 113, 39–45. doi: 10.1016/j.carbpol.2014.06.072 Copyright © 2021 Notararigo, Varela, Otal, Cristobo, Antolín, Guarner, Prieto and Xie, J., Guo, L., Ruan, Y., Zhu, H., Wang, L., Zhou, L., et al. (2010). Laminarin- López. This is an open-access article distributed under the terms of the Creative mediated targeting to Dectin-1 enhances antigen-specific immune responses. Commons Attribution License (CC BY). The use, distribution or reproduction in Biochem. Biophys. Res. Commun. 391, 958–692. doi: 10.1016/j.bbrc.2009. other forums is permitted, provided the original author(s) and the copyright owner(s) 11.173 are credited and that the original publication in this journal is cited, in accordance Zarour, K., Prieto, A., Pérez-Ramos, A., Kihal, M., and López, P. (2018). Analysis with accepted academic practice. No use, distribution or reproduction is permitted of technological and probiotic properties of Algerian L. mesenteroides strains which does not comply with these terms. Frontiers in Microbiology | www.frontiersin.org 13 February 2021 | Volume 12 | Article 621280