Secretome-Mediated Interactions with Intestinal Epithelial Cells

Secretome-Mediated Interactions with Intestinal Epithelial Cells: A Role for Secretome Components from Lactobacillus rhamnosus R0011 in the Attenuation of This information is current as Salmonella enterica Serovar Typhimurium of September 25, 2021. Secretome and TNF- −α Induced Proinflammatory Responses Michael P. Jeffrey, Chad W. MacPherson, Olivier Mathieu, Thomas A. Tompkins and Julia M. Green-Johnson Downloaded from J Immunol published online 1 April 2020 http://www.jimmunol.org/content/early/2020/03/31/jimmun ol.1901440 http://www.jimmunol.org/

Supplementary http://www.jimmunol.org/content/suppl/2020/03/31/jimmunol.190144 Material 0.DCSupplemental

Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision by guest on September 25, 2021

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2020 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Published April 1, 2020, doi:10.4049/jimmunol.1901440 The Journal of Immunology

Secretome-Mediated Interactions with Intestinal Epithelial Cells: A Role for Secretome Components from Lactobacillus rhamnosus R0011 in the Attenuation of Salmonella enterica Serovar Typhimurium Secretome and TNF-a–Induced Proinﬂammatory Responses

Michael P. Jeffrey,* Chad W. MacPherson,† Olivier Mathieu,† Thomas A. Tompkins,† and Julia M. Green-Johnson*

Recent evidence suggests that lactic acid bacteria communicate with host cells via secretome components to influence immune Downloaded from responses but less is known about gut-pathogen secretomes, impact of lactic acid bacteria secretomes on host–pathogen interactions, and the mechanisms underlying these interactions. Genome-wide microarrays and cytokine profiling were used to interrogate the impact of the Lactobacillus rhamnosus R0011 secretome (LrS) on TNF-a and Salmonella enterica subsp. enterica serovar Typhimurium secretome (STS)–induced outcomes in human intestinal epithelial cells. The LrS attenuated both TNF-a– and STS- induced gene expression involved in NF-kB and MAPK activation, as well as expression of genes involved in other immune-related signaling pathways. Specifically, the LrS induced the expression of dual specificity phosphatase 1 (DUSP1), activating transcrip- http://www.jimmunol.org/ tion factor 3 (ATF3), and tribbles pseudokinase 3 (TRIB3), negative regulators of innate immune signaling, in HT-29 intestinal epithelial cells challenged with TNF-a or STS. TNF-a– and STS-induced acetylation of H3 and H4 histones was attenuated by the LrS, as was the production of TNF-a– and STS-induced proinflammatory cytokines and chemokines. Interestingly, the LrS induced production of macrophage migration inhibitory factor (MIF), a cytokine involved in host–microbe interactions at the gut interface. We propose that the LrS attenuates proinflammatory mediator expression through increased transcription of negative regulators of innate immune activity and changes in global H3 and H4 histone acetylation. To our knowledge, these findings provide novel insights into the complex multifaceted mechanisms of action behind secretome-mediated interdomain communication at the gut-mucosal interface. The Journal of Immunology, 2020, 204: 000–000. by guest on September 25, 2021 actic acid bacteria (LAB) have been shown to influence predominately at mucosal interfaces and in GALTs where bacteria the activity of different key cell types involved in both can exert immunomodulatory effects via direct interactions with L innate and adaptive immunity. These interactions occur immune cells or indirectly through the production and secretion of bioactive molecules (1). Although the precise nature of these host–microbe interactions remain largely unknown, in vitro stud- *Applied Bioscience Graduate Program and the Faculty of Science, Ontario Techni- ies have aimed to elucidate the cellular mechanisms behind the cal University, Oshawa, Ontario L1G 0C5, Canada; and †Rosell Institute for Micro- biome and Probiotics, Montreal, Quebec H4P 2R2, Canada actions of LAB, with many modulating key cell signaling path- ORCIDs: 0000-0002-2990-2265 (T.A.T.); 0000-0001-7207-2431 (J.M.G.-J.). ways in various cell types involved in innate immunity, including intestinal epithelial cells (IECs). One of the earliest defined Received for publication December 4, 2019. Accepted for publication February 26, 2020. mechanisms of action proposed for the immunomodulatory ac- This work was supported by Mitacs Accelerate Program Fellowship Grant IT15498 tivity of commensal gut bacteria involves inhibition of NF-kB through a partnership with the Rosell Institute for Microbiome and Probiotics and by transcription factor activity via inhibiting its nuclear translocation Natural Sciences and Engineering Research Council of Canada Grant RGPIN-2017- or by inhibiting the ubiquitination of the IkB-a inhibitory protein 05237. complex in IECs (2). Expanding on this work, others have dem- The microarray data presented in this article have been submitted to the National Center for Biotechnology Information’s Gene Expression Omnibus (https://www. onstrated varied means of LAB-mediated inhibition of NF-kB and ncbi.nlm.nih.gov/geo/) under accession number GSE145091. its associated signaling pathways (3, 4). For example, the anti- Address correspondence and reprint requests to Dr. Julia M. Green-Johnson, Ontario inflammatory activity of Lactobacillus crispatus M247 has been Technical Institute, 2000 Simcoe Street North, Oshawa, ON L1G 0C5, Canada. attributed to the upregulation of PPARg, a potent inhibitor of E-mail address: [email protected] NF-kB activation, whereas other species and strains of lactobacilli The online version of this article contains supplemental material. have been demonstrated to increase the expression of A20 Abbreviations used in this article: ATF3, activating transcription factor 3; DCFH-DA, (TNFAIP3), a transcription factor responsible for terminating NF- 29,79-dichlorodihydrofluorescein diacetate; DUSP1, dual specificity phosphatase 1; GADD45b, growth arrest and DNA-damage-inducible b; HDAC, histone deacety- kB signaling in IECs (5–7). More recently, novel mechanisms of lase; IEC, intestinal epithelial cell; IPA, Ingenuity Pathway Analysis; LAB, lactic action of LAB involving epigenetic and posttranslational effects on acid bacteria; LrS, L. rhamnosus R0011 secretome; MIF, macrophage migration inhibitory factor; NLR, nucleotide-binding oligomerization domain-like receptor; IEC gene expression have been identified (8, 9). Lactic acid, a major ROS, reactive oxygen species; RT-qPCR, reverse transcription quantitative PCR; metabolite of lactobacilli, has also been shown to have histone ST, Salmonella enterica subsp. enterica serovar Typhimurium; STS, ST secretome; deacetylase (HDAC) inhibitory activity (10, 11), suggesting another TRIB3, tribbles pseudokinase 3. mechanism through which some lactobacilli may exert effects on Copyright Ó 2020 by The American Association of Immunologists, Inc. 0022-1767/20/$37.50 gene expression by modulating histone deacetylation patterns.

www.jimmunol.org/cgi/doi/10.4049/jimmunol.1901440 2 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION

The TNF superfamily encompasses numerous effector mol- communication as mediated through the respective secretomes of ecules and protein receptors that have pleiotropic effects on cells LAB and ST. of the immune system (12). TNF-a, the most studied molecule of the TNF superfamily, is a potent proinflammatory cytokine Materials and Methods with diverse effects on inflammatory and cell survival signaling Bacterial culture pathways. Binding of TNF-a to membrane-bound TNFR1 Lyophilized L. rhamnosus R0011 was obtained from Rosell Institute for present on HT-29 human IECs (13) results in the activation of Microbiome and Probiotics (Montreal, QC). The LrS was prepared as NF-kB, JNK, and p38 MAPK signaling pathways through the previously described (21). Briefly, bacteria were grown in de Man, recruitment of a protein complex involving TNFR type 1– Rogosa, and Sharpe medium (Difco) at 37˚C for 17 h in a shaking in- associated death domain (TRADD), TNFR-associated factor cubator and then diluted in nonsupplemented RPMI 1640 medium and allowed to further propagate for an additional 23 h under the same (TRAF)2, and receptor-interacting serine/threonine kinase 1 conditions. The pH of the LrS was measured, and the pH of the con- (RIPK1) (14). Activation of these signaling pathways through trol was adjusted to that of the bacterial culture using L-lactic acid and prolonged exposure to TNF-a leads to excessive production of HCl. To determine the L-lactic acid concentration within the LrS, the inflammatory mediators and has been implicated in several Megazyme D-/L-lactic acid kit was used following manufacturer’s human disease pathologies, including intestinal infections with protocols. Both the bacterial culture and controls were centrifuged at 3000 3 g for 20 min and filtered through a 0.22-mmfilter(Progene)to Gram-negative enteropathogens such as those from the genus remove any bacteria. The filtered supernatant samples were also sub- Salmonella (15). Growing evidence suggests that LAB secrete jected to size fractionation using ,10-kDa Amicon Ultra-15 Centrifugal or shed bioactive molecules with differential immunomodula- Filter (MilliporeSigma). tory activity on TNF-a–mediated signaling pathways. Kim For preparation of the ST secretome (STS), bacteria were propagated Downloaded from overnight in tryptone soya broth (Oxoid) in a shaking incubator at 37˚C. et al. (16) have shown that lipoteichoic acid, a key component Overnight cultures were centrifuged at 3000 3 g for 20 min at 4˚C and of the Gram-positive bacterial cell wall, derived from L. planta- filtered through a 0.22-mm filter, and the secretome was stored at 280˚C. rum K8, can downregulate TNF-a–induced proinflammatory cytokine production from HT-29 human IECs, whereas soluble Cell culture and challenge assay and RNA extraction proteins derived from L. rhamnosus GG inhibit TNF-a–mediated The HT-29 human colorectal adenocarcinoma cell line was obtained from

IEC damage (17). the American Type Culture Collection (no. HTB-38) and was maintained http://www.jimmunol.org/ Pathogens at the gut-mucosal interface can activate proin- in RPMI 1640 medium supplemented with 10% bovine calf serum and 0.05 mg/ml gentamicin (Sigma-Aldrich) and were grown in 75-cm2 flammatory signaling pathways in host IECs through direct tissue culture flasks (Greiner Bio-One) at 37˚C, 5% CO2 in a humidified recognition of bacterial components via innate immune receptors incubator (Thermo Fisher Scientific), as described previously (21). HT- such as the TLRs or the nucleotide-binding oligomerization 29 IECs were enumerated and viability determined using trypan blue domain-like receptors (NLRs), activating well-characterized following subculturing. Cells were then resuspended in complete culture medium (RPMI 1640 medium supplemented with 10% bovine calf serum signaling pathways. Detection of pathogenic bacteria by TLRs and 0.05 mg/ml gentamicin), and 2.0 3 106 cells were seeded into and NLRs expressed on or within IECs usually results in the 25-cm2 cell tissue culture flasks (Greiner Bio-One). This cell concen- upregulation of proinflammatory mediators via NF-kB activa- tration has been shown to yield sufficient amounts of RNA for subse- tion. However, less is known about how secreted molecules from quent microarray analyses (4). Seeded HT-29 IECs were incubated for by guest on September 25, 2021 pathogens influence host immune cells, although recent findings 48 h at 37˚C, 5% CO2 in a humidified incubator to obtain confluent monolayers prior to challenge. HT-29 IEC cell culture medium was suggest that this is also an important route for pathogenicity. For aspirated and replaced with fresh nonsupplemented (no calf serum) example, Neisseria meningitidis secretes bioactive molecules RPMI 1640 medium containing the LrS, L-lactic acid controls, TNF-a that serve multiple roles in host–pathogen interactions (18). (50 ng/ml), STS (1% v/v), or a combination of the challenges. Total Probiotic LAB have been shown to antagonize gut-pathogen RNA was harvested after 3 h of exposure to the various challenges using the phenol-based TRIzol method of RNA extraction (24) following activity through direct inhibition (19) and through the secre- manufacturer’s protocols (Thermo Fisher Scientific). Briefly, 2 ml of tion of bioactive molecules (20), suggesting the potential for TRIzol reagent was added to each culture flask to lyse the IECs. Cell complex and multifaceted host–microbe interactions mediated culture homogenates were added to Phase Lock Gel-Heavy tubes for by soluble molecules. Many questions currently remain re- phase separation of total RNA. Total extracted RNA was then purified garding the roles of probiotic-derived secreted bioactive mole- using the RNeasy Plus Mini Kit (QIAGEN). The purity and quality of RNA was determined using both the ND100 NanoDrop and an Agilent cules in these host–pathogen interactions. 2000 Bioanalyzer, respectively. Only samples with an RNA integrity We previously reported that the L. rhamnosus R0011 secretome number .9.0 were used for microarray analysis. (LrS) downregulates the production of IL-8 from IECs challenged Reverse transcription of RNA and direct method of labeling with a wide array of innate immune stimulants including TNF-a, TLR, and NOD1 agonists (21). In the current study, we aimed to Control and experimental total RNA (15 mg) was reverse transcribed with further interrogate the signaling pathways modulated by the LrS in SuperScript IV (Invitrogen) and labeled with Cy3-dCTP and Cy5-dCTP (GE Healthcare) using the direct method of dye labeling as previously HT-29 IECs exposed to TNF-a to better understand the mecha- described (4). Briefly, 3 mg of oligo(dT)23 primers were added to the RNA nism(s) of action behind the observed bioactivity, as TNF-a in- and samples were heated to 70˚C for 30 min to reduce secondary structure duction is a common feature of many pattern recognition receptor formation. A cDNA Synthesis Master Mix containing 53 First-Strand (PRR)–mediated pathways. Furthermore, secretome components Buffer, 0.1 M DTT, dNTPs, 200 U of SuperScript IV, and either 1 mM from Salmonella enterica subsp. enterica serovar Typhimurium of Cy3 or Cy5 dye were added, and the samples were heated at 42˚C for 3 h to allow the reverse transcription to occur. Dye swaps between treated (ST), a known gut pathogen, were examined for the ability to and control RNA were done to eliminate bias of dye labeling. Purification influence IEC innate immune activity as was the impact of the LrS of the cDNA product was done using the QIAquick PCR Purification Kit in this context. TNF-a signaling is well defined in HT-29 human following manufacturer’s protocols (QIAGEN). Labeling efficiency was IECs (13), and their use as an IEC model to study the impacts of determined by calculating the dye incorporation rate to ensure consistency between experiments. both TNF-a and invasive pathogens at the gut-mucosal interface is well established (22, 23). Using genome-wide gene expression Microarray analysis microarrays and cytokine/chemokine production analysis, our Hybridization of the labeled cDNA to the microarray was done using pre- findings, to our knowledge, provide novel insights into the complex, viously established protocols (4). Following hybridization, the microarrays multifaceted role of bacterial secretome components in interdomain were scanned using the ScanArray 5000 instrument from PerkinElmer, and The Journal of Immunology 3 spot intensities were quantified using ImaGene version 9.0 (BioDiscovery). chemokine profiling from all cell challenges was done following Normalization was done using locally weighted scatterplot smoothing (25). manufacturer’s instructions (Bio-Rad Laboratories) with four biological Statistical analyses and two-dimensional hierarchical clustering analyses replicates. Quality controls were also included to ensure the validity of were performed with MultiExperiment Viewer (version 4.2). Genes with the concentrations that were obtained. The Bio-Plex Manager software statistically significant changes in expression levels were selected based on was used to determine the concentration of the analytes within each a t test yielding a p value , 0.05 and a 1.5-fold change gene expression sample using the generated standard curves and concentration was cutoff. Set Distiller from GeneDecks, Cytoscape, and Ingenuity Pathway expressed in picograms per milliliter (concentration in range). Statis- Analysis (IPA) pathway enrichment analyses were used to ascertain the tical analysis was done using GraphPad Prism (Version 8) ANOVA and pathways in which genes were significantly modified by the different Tukey multiple comparison test when the ANOVA indicated significant treatments. Set Distiller groups gene sets into statistically significant (p , differences were present. All data are shown as the mean picograms per 0.05) and biologically relevant pathways and descriptors (26), whereas milliliter 6 SEM. String v 11.0 analysis was also done to determine GeneMANIA in Cytoscape determines functional links between genes in functional links between each of the different cytokines/chemokines gene sets (27, 28). Gene networks were generated through the use of IPA measured (30). (https://www.qiagenbioinformatics.com/products/ingenuity-pathway-analysis; QIAGEN) (29). Measurement of intracellular reactive oxygen species generation Reverse transcription quantitative PCR Intracellular reactive oxygen species (ROS) production was determined by Reverse transcription quantitative PCR (RT-qPCR) to determine transcript using 29,79-dichlorodihydrofluorescein diacetate (DCFH-DA) (Sigma- abundance of differentially expressed genes was performed to validate Aldrich). HT-29 IECs were pretreated with 100 mM DCFH-DA for 30 microarray expression data as per the Minimum Information for Publication min at 37˚C, 5% CO2 to allow sufficient time for cellular uptake. Cells of Quantitative Real-Time PCR Experiments guidelines. DNase-treated were washed twice with fresh medium and challenged with the LrS,

RNA (1 mg) from controls and each challenge were reverse transcribed Downloaded from TNF-a (50 ng/ml), STS (1% v/v), L-lactic acid–matched controls, or a with SuperScript IV following manufacturer’s protocols as previously combination of the treatments for 45 min. Fluorescence of 29,79- described (3). Reverse-transcribed cDNA was diluted 1:4 prior to am- dichlorodihydrofluorescein because of the oxidation of DCFH-DA by plification, and 2.5 ml of diluted cDNA was used in RT-qPCR using gene- ROS was quantified using a Synergy HT Microplate Reader (BioTek specific primers (Supplemental Table I) and SsoAdvanced Universal Instruments) set to 485/20 excitation and a 528/20 emission filter pair SYBR Green Supermix (Bio-Rad Laboratories) per the manufacturer’s and a photomultiplier sensitivity of 55. instructions. An initial incubation of 5 min at 95˚C was performed, followed by 40 cycles consisting of template denaturation (15 s at 95˚C) and one-step annealing and elongation (30 s at 60˚C) with a Bio-Rad CFX Results http://www.jimmunol.org/ Connect instrument (Bio-Rad Laboratories). Four biological replicates were analyzed for each gene tested, and fold-change expression levels The LrS attenuates TNF-a– and STS-induced changes in were normalized to the expression levels of two reference genes (RPLPO global gene expression profiles in HT-29 IECs and B2M) and negative controls using Bio-Rad CFX Manager 3.1 Genome-wide transcriptional profiling of HT-29 IECs treated Software. with TNF-a, STS, LrS, L-LA, or a combination of the challenges Histone extraction and H3/H4 histone acetylation assay was performed to evaluate the global changes in the IEC tran- H3 and H4 global acetylation patterns in HT-29 IECs were determined using scriptome in response to these challenges. Two-dimensional hi- the EpiQuik Global Histone H4 (P-4009) and H3 (P-4008) Acetylation erarchical heat-map cluster analysis revealed that TNF-a and Assay Kits following manufacturer’s protocols (EpiGentek). Briefly, his- STS challenges do not cluster closely together, suggesting dif- by guest on September 25, 2021 tones were extracted from HT-29 IECs following 3 h of challenge, and ferences between these treatments. Interestingly, L-LA appears total protein concentrations in each sample were determined using the Coomassie Plus Protein Assay Reagent (Thermo Fisher Scientific). A total to have a distinct effect on gene expression profiles when com- of 2 mg of protein from both untreated and challenged IEC histone extracts bined with either proinflammatory challenge, with the cochal- were spotted into strip wells and assayed for the amount of acetylation lenges of L-LA and either TNF-a or STS clustering closely with using Abs specific for acetylated H3 or H4 histones. Standards with known the TNF-a and STS challenges, respectively. In contrast, the LrS concentrations of acetylated histones were included, and percentage of and cochallenge of the LrS with either TNF-a or the STS cluster histone acetylation was determined by comparing treated and untreated controls. Statistical analysis was done using GraphPad Prism (Version 8) together, indicating distinct and unique global gene expression one-way ANOVA and Tukey multiple comparison test when the ANOVA profiles (Fig. 1A). TNF-a challenge had a large impact on the indicated significant differences were present. All data are shown as the global IEC transcriptomic response, with a total of 1531 dif- mean percentage of change in histone acetylation 6 SEM. ferentially modulated genes (886 upregulated and 645 down- , Cytokine/chemokine/inflammation marker analysis regulated) (n =4;p 0.05). When combined with the LrS, this was reduced to a total of 820 genes, indicating that the LrS at- Cell culture supernatants were collected following 6 h of challenge to tenuated the impact of TNF-a on IEC global gene expression allow sufficient time for the production of key inflammatory cytokines , and chemokines. Cytokine and chemokine profiling was performed using (n =4;p 0.05). A similar effect was observed on STS-induced the Bio-Plex Pro 40-Plex Human Chemokine Panel (no. 171ak99mr2; changes in IEC global gene expression, with the LrS reducing Bio-Rad Laboratories) and the Bio-Plex Pro Human Inflammation Panel the total number of differentially expressed genes from 618 to 1, 37-Plex (no. 171AL001M; Bio-Rad Laboratories). All 40 chemokines 342 (n =4;p , 0.05) (Fig. 1B). (CCL21, BCA-1/CXCL13, CTACK/CCL27, ENA-78/CXCL5, eotaxin/CCL11, To interrogate the cellular pathways that were impacted by the eotaxin-2/CCL24, eotaxin-3/CCL26, fractalkine/CX3CL1, GCP-2/CXCL6, GM-CSF, Gro-a/CXCL1, Gro-b/CXCL2, I-309/CCL1, IFN-g,IL-1b,IL-2, different challenges, gene enrichment analysis was performed IL-4, IL-6, IL-8/CXCL8, IL-10, IL-16, IP-10/CXCL10, I-TAC/CXCL11, using GeneDecks Set Distiller analysis. This analysis revealed that MCP-1/CCL2, MCP-2/CCL8, MCP-3/CCL7, MCP-4/CCL13, MDC/CCL22, treatment of HT-29 IECs with TNF-a led to the increased tran- macrophage migration inhibitory factor [MIF], MIG/CXCL9, MIP-1a/CCL3, scription of genes involved in immune-related pathways, includ- MIP-1d/CCL15, MIP-3a/CCL20, MIP-3b/CCL19, MPIF-1/CCL23, SCYB16/CXCL16, SDF-1a + b/CXCL12, TARC/CCL17, TECK/CCL25, ing the innate immune response (88 genes), TNF-a (36 genes), and TNF-a) or 37 cytokines (APRIL/TNFSF13, BAFF/TNFSF13B, MAPK (43 genes), and NF-kB (31 genes) signaling pathways sCD30/TNFRSF8, sCD163, chitinase-3–like 1, gp130/sIL-6Rb,IFN-a2, (p , 0.001) (Fig. 1C). Further analysis revealed that TNF-a IFN-b,IFN-g,IL-2,sIL-6Ra, IL-8, IL-10, IL-11, IL-12 [p40], IL-12 [p70], challenge led to increased transcription of key proinflammatory IL-19, IL-20, IL-22, IL-26, IL-27 [p28], IL-28A/IFN-l2, IL-29/IFN-l1, mediators common to these immune pathways, such as CXCL1 IL-32, IL-34, IL-35, LIGHT/TNFSF14, MMP-1, MMP-2, MMP-3, osteocalcin, osteopontin, pentraxin-3, sTNF-R1, sTNF-R2, TSLP, (13.2-fold change), CXCL10 (29.6-fold change), CCL20 (35.9- and TWEAK/TNFSF12) were multiplexed on the same 96-well fold change), BIRC3 (15.4-fold change), PTGS2 (11.0-fold plate. Chemokine/cytokine standards were serially diluted, and change), CXCL8 (23.6-fold change), and NFкB1 (4.1-fold change) 4 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION Downloaded from http://www.jimmunol.org/ FIGURE 1. (A) Two-dimensional hierarchical clustering analysis of global gene expression data in human HT-29 IECs exposed to TNF-a, the STS, the LrS alone, a combination of the LrS and either TNF-a or the STS, L-lactic acid–matched controls, or a combination of the STS or TNF-a and L-lactic acid (n =4;p , 0.05; fold change difference . 1.5 versus untreated cells). (B) Total number of up- and downregulated genes in response to each different challenge. Gene enrichment analysis using GeneDecks Set Distiller to delineate the number of genes in (C) immune-related or (D) cellular/signaling-related pathways differentially modified in HT-29 IECs exposed to TNF-a, the STS, the LrS, a combination of both, or to the STS or TNF-a plus L-lactic acid control (n =4;p , 0.05).

(Table I). Other cellular pathways identiﬁed as being impacted by apoptosis (36 genes), PAK (36 genes), and TWEAK (19 genes)

TNF-a challenge involved apoptosis (65 genes), p21-activated (p , 0.001), an effect that was independent of L-lactic acid by guest on September 25, 2021 protein kinases (PAK) (77 genes), and TWEAK pathways (30 (Fig. 1C, 1D, Table I). genes) (p , 0.001) (Fig. 1D). Consistent with the changes ob- The STS also induced transcriptional changes, leading to the served in global gene expression proﬁles, cochallenge of IECs activation of genes involved in immune-related pathways, with with the LrS resulted in the attenuation of TNF-a–induced the greatest impact on those involved in regulating innate immune transcription of genes involved in the innate immune response activity (28 genes), NF-kB (16 genes), TLR signaling (21 genes), (38 genes), TNF-a (20 genes), MAPK (13 genes), NF-kB (11 genes), and NLR signaling (21 genes) (p , 0.001) (Fig. 1C). As was

Table I. Fold-change difference in expression of genes encoding proinﬂammatory mediators and negative regulators of innate immunity in HT-29 IECs exposed to TNF-a, STS, the LrS, L-LA, or a combination of the treatments

TNF-a STS TNF-a STS Gene Name TNF-a STS LrS L-LA plus LrS plus LrS plus L-LA plus L-LA Proinflammatory mediator CXCL1 13.2 7.85 21.62 — 13.72 — 10.35 — CXCL10 29.62 — — — — — 17.1 — CXCL11 13.2 — — — — — 4.17 — CCL20 35.93 6.39 21.54 — 8.77 — 6.75 14.38 IL-1b 2.134 — — — — — 2.48 — IL-8 23.61 5.97 — — 5.50 2.02 18.96 7.52 IL-17C 10.68 3.69 — — — — 11.31 5.79 IL-23A 4.30 — — — — — 2.02 — IL-32 5.28 — — — — — 8.32 — BIRC3 15.39 2.47 — — — — 4.46 6.65 PTGS2 10.99 1.67 — — — — 14.41 2.98 EGFR 3.80 — — — — — 1.64 — Negative regulator of innate immunity DUSP1 2.86 — 24.48 — 15.67 7.40 4.49 1.50 ATF3 2.47 — 23.39 — 12.65 6.37 5.66 — TRIB3 ——25.75 — 5.50 7.00 2.00 — HSPA6 ——26.69 — 14.95 4.11 — — GADD45b ——25.67 — 10.06 5.05 2.07 — The numbers represent the fold change increase or decrease in gene expression. The Journal of Immunology 5 seen with the TNF-a challenge, the STS also induced the ex- transcription factor 3 (ATF3)(23.4-fold change) and tribbles pseu- pression of CXCL1 (7.9-fold change), CCL20 (6.4-fold change), dokinase 3 (TRIB3)(25.8-fold change), negative regulators of innate BIRC3 (2.5-fold change), PTGS2 (1.7-fold change), and NFкB1 immunity (Fig. 3A–D). However, when HT-29 IECs were (1.8-fold change). Concurrent challenge of IECs with the LrS cochallenged with the LrS and TNF-a, there was a 15.7-fold and the STS resulted in the attenuation of all STS-induced change in DUSP1 expression, 12.7-fold change in ATF3 ex- transcription of genes involved in the aforementioned immune- pression, and 5.5-fold change in TRIB3 expression (Fig. 3B). related pathways (p , 0.001; Fig. 1C, Table I), suggesting that This pattern of upregulation of negative regulator expression the LrS attenuates both TNF-a– and STS-induced transcriptional was also seen in HT-29 IECs cochallenged with the LrS and changes in IECs. STS, in which a 7.4-fold change in DUSP1 expression, 6.37- fold change in ATF3 expression, and 7.0-fold change in TRIB3 The LrS induces negative regulators of innate immunity in expression was observed (Fig. 3C). Effects on expression of TNF-a and STS-challenged IECs these negative regulators were confirmed by RT-qPCR (Fig. 4). IPA was used to further interrogate the underlying mechanism(s) Predictive modeling of these negative regulators suggest that they of action of the LrS on TNF-a– and STS-induced transcriptional play key roles in the modulation of TNF-a– or STS-induced ac- responses in IECs. Using molecular activity prediction, path- tivation of the NF-kB, MAPK, and TNF-a signaling pathways way analysis revealed that treatment of TNF-a–challenged when HT-29 IECs are treated with the LrS (Fig. 3D), sug- IECs with the LrS results in the attenuation of components gesting a possible mechanism of action behind the observed of the TNF-aRcomplex(TNFR, FADD,andRIPP), the IKK immunoregulatory activity of the LrS. Moreover, a similar complex (IKKa, IKKb,andIKKg), as well as the NF-kBsig- trend was observed for heat shock protein family A (Hsp70) Downloaded from naling complex (NFkB1, NFkB2, RELA,andRELB)(Fig.2). member 6 (HSPA6) (15.0- and 4.1-fold change) and growth GeneMANIA gene enrichment analysis confirmed these find- arrest and DNA-damage-inducible b (GADD45b)(10.1-and ings, as the LrS significantly attenuated STS- and TNF-a–in- 5.1-fold change), key cellular players in mediating oxidative duced transcription of NFkB1, NFkB2, RELA,andRELB, and cellular stressors, in response to the combination of LrS TNFR2–TRAF signaling complex protein (BIRC3), CASP8 and and TNF-a or STS (Table I).

FADD–like apoptosis regulator (CFLAR), IRAK-2,andseveral http://www.jimmunol.org/ genes involved in the MAPK signaling pathway (Supplemental The LrS attenuates TNF-a– and STS-induced global Fig. 1). histone acetylation To elucidate how the LrS was attenuating the activation of these The ability of the LrS to modulate global histone acetylation cellular pathways in response to TNF-a challenge, HT-29 IECs patterns was examined to determine if the observed reduc- treated with the LrS were examined for differential activation of tion in proinﬂammatory gene expression correlated with de- negative regulators of the innate immune response. Interestingly, creased histone acetylation. To this end, analysis of the amount treatment of HT-29 IECs with the LrS alone resulted in decreased of acetylated H3 and H4 histones indicated that the LrS signif- expression of dual speciﬁcity phosphatase 1 (DUSP1)(24.5-fold icantly reduced TNF-a– and STS-induced H3 histone acetylation change), a key regulator of the MAPK pathway, and of activating (n =4;p , 0.05) (Fig. 5A). In addition, the LrS also attenuated by guest on September 25, 2021

FIGURE 2. Pathway analysis of TNF-a signaling pathways using IPA predictive modeling of the cellular outcomes of HT-29 IECs exposed to (A) TNF-a challenge or (B) TNF-a plus the LrS. 6 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION Downloaded from http://www.jimmunol.org/ FIGURE 3. Negative regulators of the innate immune response that were identiﬁed to be differentially transcribed in response to the different treatments. Gene interaction maps using CytoScape and GeneMANIA showing genes that have been reported to interact directly with each other in response to treatment with (A) LrS, (B) TNF-a,(C) STS, (D) TNF-a plus the LrS, or (E) STS plus the LrS. (F) IPA pathway analysis using predictive modeling to determine the interactions between the identiﬁed negative regulators and the NF-kB and MAPK signaling pathways.

STS-induced acetylation of H4 histones (n =4;p , 0.05) translated into effects at the protein level, the impact on pro- (Fig. 5B) in IECs. duction of several cytokines and chemokines were measured. Of the 74 different proinflammatory and regulatory protein markers by guest on September 25, 2021 The LrS attenuates TNF-a– and ST-induced proinflammatory examined, TNF-a challenge resulted in significant production of cytokine and chemokine protein production 55 different cytokines/chemokines and markers of inflammation To determine whether the observed effects of the LrS on STS (n =4;p , 0.05) (Fig. 6A–D, Supplemental Figs. 2, 3). Consistent and TNF-a–induced proinflammatory mediator gene expression with the observed changes in TNF-a–induced transcriptional

FIGURE 4. RT-qPCR conﬁrmation of the expression of TRIB3, ATF3, DUSP1, and NFкB1. Data shown are the mean relative fold change 6 SEM (n =4). The Journal of Immunology 7

FIGURE 5. Changes in global histone (A)H3or(B) H4 acetylation patterns in challenged HT-29 IECs. Data shown are the mean change in percentage of acetylation compared with untreated controls 6 SEM (n = 4). Significant differences between treatments were determined by one-way ANOVA and Tukey post hoc test. *p , 0.05, **p , 0.01, ***p , 0.001. Downloaded from changes when IEC production of 48 of the observed TNF-a– to induce proinflammatory mediator production consistent with induced cytokines/chemokines and markers of inflammation was typical Salmonella spp.–induced cytokine and chemokine pro- significantly attenuated to constitutive levels by the addition of the files in IECs. Cochallenge with the LrS attenuated the production LrS (n =4;p , 0.05). of 17 STS-induced cytokines/chemokines and markers of in- Of the 74 different protein markers examined within this study, flammation, indicating that the immunomodulatory impact of the http://www.jimmunol.org/ String protein analysis identified IFN-g, GM-CSF, IL-1b, CXCL8, LrS on STS-induced transcriptional responses was correlated to and CXCL1 as being involved in responses to Salmonella spp. functional protein levels in IECs (Fig. 6A–D, Supplemental infection (Fig. 6E). STS challenge resulted in significant in- Figs. 2, 3). creases in production of these five cytokines/chemokines and 24 Again, L-lactic acid had marginal impacts on TNF-a–orSTS- other mediators of inflammation, providing further evidence of the induced proinflammatory mediator production from IECs, sug- ability of secretome components from this gut-associated pathogen gesting that the observed immunomodulatory activity of the LrS by guest on September 25, 2021

FIGURE 6. Cytokine and chemokine profiles from HT-29 IECs exposed to the different treatments for 6 h. (A) Total number of cytokines and chemokines showing significant levels of production relative to negative controls as determined by one-way ANOVA and Dunnett post hoc test (n =4;p , 0.05). (B–D) Cytokine and chemokine profiles from HT-29 IECs exposed to the different challenges for 6 h. Data shown are the mean cytokine/chemokine production (picograms per milliliter) 6 SEM (n = 4). (E) String v 11.0 analysis showing functional links between each of the different cytokines/chemokines measured. 8 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION is not simply due to the presence of L-lactic acid. Cochallenge of HT-29 IECs with L-lactic acid resulted in a significant reduction in the production of CXCL10, CXCL1, and CCL2 from TNF-a– challenged IECs (Fig. 6A–D, Supplemental Figs. 2, 3). Inter- estingly, treatment with the LrS alone or in combination with TNF-a did lead to the significant production of MIF (14,721.3 6 4,693.9 pg/ml) from constitutive levels (1934.4 6 142.8) (Fig. 7). This was the only instance in which challenge with the LrS induced the production of a cytokine/chemokine found within this panel. The LrS reduces ROS generation from HT-29 IECs IPA Upstream Regulator analytics identified NO and ROS as potential upstream regulators that may explain the observed gene expression changes when HT-29 IECs are exposed to the LrS and either TNF-a or STS (Fig. 8A). To this end, intracellular ROS production in response to the different treatments was examined using DHFA. HT-29 IECs treated with the LrS had significantly FIGURE 7. MIF production is induced in HT-29 IECs by the LrS. HT- less ROS production than the other treatments (Fig. 8B), sug- 29 IECs were exposed to the different challenges for 6 h, and the levels of Downloaded from gesting that intracellular ROS is not responsible for the observed MIF were quantified. Data shown are the mean cytokine/chemokine pro- changes in the transcript abundance of these key negative regu- duction (picograms per milliliter) 6 SEM (n = 4). Different letters (a and , lators of innate immunity. b) denote significance (p 0.05) between treatments as determined by one-way ANOVA and Tukey post hoc test. Discussion

Although many studies support direct contact-mediated routes for indicate that ST has strategies to exploit the host inflammatory http://www.jimmunol.org/ immunomodulatory activity of lactobacilli in their interactions with response, contributing further to its pathogenic activity. ST re- IECs (1, 31), evidence for alternate modes of microbe–host in- quires intestinal inflammation to circumvent colonization resis- teraction involving soluble mediators and secreted components tance provided by the normal gut microbiota (43, 44). ST also uses point toward multiple potential mechanisms of action (32, 33). tetrathionate generated by the host via the oxidation of free lu- Earlier reports of antimicrobial and immunomodulatory effects of minal thiosulfate by ROS produced by activated proinflammatory soluble mediators from conditioned media of lactobacilli sug- intestinal macrophages (45). Tetrathionate acts as a respiratory gesting the potential for secretome-mediated effects of these electron acceptor and facilitates the ability of ST to respire etha- bacteria on IECs (34–37) are supported by more recent studies nolamine released into the gut lumen during inflammation, pro- reflecting the impact of microbial metabolites on the immune viding a unique strategy for this intestinal pathogen to use the host by guest on September 25, 2021 system (38, 39). Although several mechanisms of action behind inflammatory response to gain a growth advantage over other the activity of LAB have been identified, few studies have in- bacteria in the gut (46, 47). For this reason, we examined the ability vestigated the effects of LAB secretome components on key sig- of the LrS to attenuate STS-induced expression of proinflammatory naling pathways involved in innate immune signaling in IECs. mediators as a potential mechanism to counteract this pathogenic Based on our earlier observations indicating that the LrS attenu- strategy. As was seen with TNF-a–challenged IECs, the LrS also ates TNF-a–induced IL-8 production by HT-29 IECs (21), we attenuated STS-induced expression of genes involved in NF-kB, examined the effects on TNF-a–induced gene expression to gain MAPK, TLR, and NLR signaling pathways, suggesting a novel insight into regulatory activity at the transcriptional level. route of LAB-mediated gut-pathogen antagonism at the IEC level. Cochallenge of TNF-a–challenged HT-29 IECs with the LrS at- Concurrent challenge of TNF-a– or STS-challenged HT-29 tenuated the activation of many genes involved in proinflammatory IECs with the LrS resulted in the upregulation of DUSP1, signaling activity, including those involved in NF-kB activation. TRIB3, and ATF3. DUSP1, a central regulator of the MAPK These findings are in keeping with previous transcriptomic analysis, pathway that acts to dephosphorylate p38a and Jun N-terminal which revealed that live L. rhamnosus R0011 modulates expression kinases (48), has been well characterized in its role as a negative of genes involved in TLR, NOD, MAPK, and cytokine–chemokine regulator of innate proinflammatory activity (49–53). TRIB3, a receptor interactions in HT-29 IECs under basal conditions (40), negative regulator of NF-kB–mediated signaling, acts as a pseu- indicating a potential role for soluble mediators and metabolites in dokinase to allosterically inhibit the phosphorylation of p65 by the immunomodulatory activity of live LAB. protein kinase A (54). TRIB3 transcription is activated by DDIT3/ Challenge of HT-29 IECs with the STS resulted in the upreg- CHOP, a member of the C/EBP family (55), and the LrS upregulation of several genes involved in proinflammatory activity and ulated expression of this transcription factor in TNF-a–orSTS- is, to our knowledge, the first report of the effects of secretome challenged HT-29 IECs. In contrast, LPS derived from Helicobacter derived from ST grown under normal conditions on IECs. ST pylori downregulates CHOP and TRIB3 expression in human gas- typically causes inflammation in the intestinal epithelium through tric epithelial cell lines (56), suggesting TRIB3 plays a complex the direct delivery of effector proteins via type III secretion systems role in IEC responses to gut bacteria. (41). However, the results presented in this study describe, to our ATF3 is a transcription factor also involved in the attenuation of knowledge, a potentially novel route of pathogenicity of this gut- NF-kB–mediated signaling (57). Its expression can be induced by associated pathogen via the shedding or secretion of bioactive ribosomal insults, which has been suggested as a mechanism molecules inducing proinflammatory signaling pathways. Recent through which NF-kB activation is regulated in IECs to prevent reports have suggested that effector proteins encoded by the ST excessive inflammation in response to commensal bacteria (36). pathogenicity island 1 are packaged into outer membrane vesicles Although there have been reports of differential modulation of for long-distance delivery to host cells (42), and several studies ATF3 in response to bacterial challenge (58, 59), to the best of our The Journal of Immunology 9 Downloaded from

FIGURE 8. (A) IPA Upstream Regulator analytics identified ROS as a potential transcriptional regulator explaining the observed gene expression changes http://www.jimmunol.org/ when HT-29 IECs are exposed to the LrS and either TNF-a or STS. (B) Intracellular ROS production by HT-29 IECs in response to the different treatments (n = 3). Data shown are the mean fluorescence intensity of oxidized 29,79-dichlorodihydrofluorescein following 45 min of challenge. Different letters (a–d) denote significance (p , 0.05) between treatment groups as determined by one-way ANOVA and Tukey post hoc test. knowledge, this is the first report of induction of this transcrip- localization. Moreover, the HSP70 family of proteins can inhibit tional repressor in response to secretome components from LAB. NF-kB activation by preventing the oligomerization of NEMO ATF3 recruits HDAC1 to deacetylate the promoter regions of proteins and subsequent IKK activation (67, 68). Overall, this NF-kB target genes, such as IL-6 and IL-12 (60), hindering the indicates context-dependent regulation of the production of by guest on September 25, 2021 activity of the p65 subunit of NF-kB (61), or it can deacetylate proinflammatory biomarkers by the LrS and may provide insights p65 directly (62). For this reason, the ability of the LrS to mod- into how some bacteria use soluble mediators to maintain ho- ulate global histone acetylation patterns was examined. Epigenetic meostasis at the gut-mucosal interface by counteracting effects of reprogramming through changes in histone acetylation patterns is potential extracellular stressors and proinflammatory stimuli. well documented and has been shown to be an important aspect of ATF3 and TRIB3 expression can be induced in response to host–microbe interactions in the gut (63, 64). Furthermore, L. endoplasmic reticulum stress and has been shown to be activated by rhamnosus MTCC 5897 and L. fermentum MTCC 5898 have been ROS generation (69, 70). As ROS plays a complex role in regu- shown to limit Escherichia coli–induced increases in Caco-2 IEC lating IEC responses to proinflammatory stimuli and other cell H3 and H4 histone acetylation (65). In keeping with these find- stressors, we examined the impact of the LrS on ROS production ings, the LrS reduced TNF-a– and STS-induced H3 and H4 his- by HT-29 IECs. LAB-induced ROS production has been reported tone acetylation, suggesting a possible mechanism of action as a potential immunomodulatory mechanism (71), and varied behind the observed immunomodulatory activity of the LrS on roles for ROS in microbe–host communication are further illus- TNF-a– and STS-induced proinflammatory gene transcription trated by the ability of L. johnsonii N6.2 to modulate immune through chromatin structural remodeling. However, it is currently activity through effects on IDO activity (72), and the impact of unknown if this reduction in histone acetylation is a direct result H2O2 produced by lactobacilli on Citrobacter rodentium patho- of increased ATF3 expression when IECs are cochallenged with genicity in rodents (73). ROS induction has been shown to be a the LrS, and this warrants further investigation. route through which certain LAB inhibit NF-kB activation in Interestingly, upregulation of DUSP1, TRIB3, and ATF3 by the activated IECs via the oxidation of ubiquitin-conjugating enzyme LrS was observed when HT-29 IECs were also responding to either 12 both in vitro and in vivo (74–76). Indeed, NO and ROS were TNF-a or the STS, but not to the LrS alone. Impact on GADD45b identified as potential upstream regulators of the activation of and HSPA6 (HSP70B9) expression followed a similar pattern and DDIT3/CHOP, ATF3, and DUSP1 when HT-29 IECs were was significantly upregulated by the LrS only in IECs exposed to cochallenged with the LrS and either proinflammatory challenge. these proinflammatory stimuli. GADD45b is a central regulator of However, excessive ROS production can lead to pathological in- many cellular functions in response to certain cellular stressors flammation through the activation of NF-kB signaling pathways and has been shown to play a protective role against TNF-a– via stabilization of NEMO proteins, resulting in the production of induced apoptosis by antagonizing JNK activation (66). HSP70B9 proinflammatory mediators (77, 78). Recent reports indicate that and other members of the HSP70 family of proteins also have HSP70 can inhibit mitochondrial ROS production in response to diverse intracellular functions within cells in response to stress by bacterial toxins in human lung microvascular endothelial cells acting as molecular chaperones to prevent denaturation of proteins (79), and others have demonstrated the ability of LAB to inhibit and refold and guide damaged proteins to their correct cellular ROS production and exert antioxidant activity (80, 81). In keeping 10 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION with these findings, intracellular ROS production was significantly propose, to our knowledge, a novel mechanism of LAB lower in HT-29 IECs treated with the LrS, indicating that the in- secretome-mediated effects on IEC signaling pathways through duction of these negative regulators was not due to an increase in the induction of ATF3, TRIB3, DUSP1, negative regulators of ROS production and that the LrS may reduce oxidative stress on the NF-kB and MAPK signaling pathways, and through changes IECs by induction of HSP70. in global histone acetylation patterns. The LrS also induced the The ability of the LrS to attenuate production of TNF-a–and production of MIF, a cytokine that may exert context-dependent STS-induced proinflammatory cytokines and chemokines is in regulation of gut homeostasis by influencing IEC barrier integ- keeping with the observed effects on negative regulator expres- rity, M cell–dependent Ag uptake, and transcriptional responses sion, and with our previous findings indicating downregulation to proinflammatory challenge. Taken together, these findings, to of PRR-induced chemokine production in both HT-29 IECs and our knowledge, provide novel insight into the complex mecha- in nontransformed IEC-6 cells (21). Similar results were ob- nisms of action behind secretome-mediated interdomain com- served with the T84 IEC line, in which the LrS also reduced munication at the gut-mucosal interface and suggest new TNF-a–induced IL-8 production (Supplemental Fig. 4), indi- directions for approaches to delineate the activities of gut- cating that the observed effects of the LrS on chemokine pro- associated bacteria in vivo. Future work to characterize the duction are not limited to HT-29 IECs. Although downregulation LrS and STS will be needed to determine which bioactive of IEC proinflammatory cytokine production has been shown molecule(s) are responsible for the observed impacts of these with other LAB (82–85), to our knowledge, this is the first report of bacteria on IECs, their roles in pathogenicity and host im- secretome components of LAB demonstrating this range of immu- mune responses, and their potential roles in gut microbe–host nomodulatory bioactivity on proinflammatory mediator production interactions. Downloaded from by IECs. Interestingly, challenge with the LrS increased MIF production, whereas TNF-a or STS challenge and lactic acid controls Acknowledgments did not, a finding consistent with previous reports, suggesting that We thank Dr. Janice L. Strap (Faculty of Science, Ontario Tech University) proinflammatory cytokines also do not induce MIF production by and Dr. Holly Jones Taggart (Faculty of Health Sciences, Ontario Tech Uni- IECs (86). This was the only instance in which the LrS induced versity) for helpful discussion. cytokine production from HT-29 IECs, and is, to our knowledge, the http://www.jimmunol.org/ first report of secretome components of LAB to stimulate MIF Disclosures production. MIF is a pleiotropic cytokine (87) and has been reported C.W.M., T.A.T., and O.M. are employees of Lallemand Health Solutions. to be integral in the maintenance of IEC barrier function through its The other authors have no financial conflicts of interest. effects on colonic tight and adherens epithelial junctions (88). It may also act to inhibit AP-1 activity (89), resulting in the downregulation of inflammatory gene expression. MIF’s role in response References to Gram-negative bacterial challenge is well established (90, 91), 1. Suez, J., N. Zmora, E. Segal, and E. Elinav. 2019. The pros, cons, and many and it has been associated with increased macrophage-dependent unknowns of probiotics. Nat. Med. 25: 716–729. 2. Neish, A. S., A. T. Gewirtz, H. Zeng, A. N. Young, M. E. Hobert, V. Karmali, by guest on September 25, 2021 pathogen clearance (92), indicating that it may exert context- A. S. Rao, and J. L. Madara. 2000. Prokaryotic regulation of epithelial re- dependent regulation of the inflammatory response to bacteria. sponses by inhibition of IkappaB-alpha ubiquitination. Science 289: 1560– Although the precise nature of regulation of transcription and se- 1563. 3. Macpherson, C., J. Audy, O. Mathieu, and T. A. Tompkins. 2014. Multistrain cretion of MIF remains to be fully elucidated, it has been shown that probiotic modulation of intestinal epithelial cells’ immune response to a double- MIFissecretedwhenNF-kB activation is inhibited (93), suggesting stranded RNA ligand, poly(i·c). [Published erratum appears in 2019 Appl. En- viron. Microbiol. 85.] Appl. Environ. Microbiol. 80: 1692–1700. a possible mechanism of action behind LrS-induced MIF secretion. 4. MacPherson, C. W., P. Shastri, O. Mathieu, T. A. Tompkins, and P. Burguie`re. Moreover, MIF has also been shown to increase M cell–mediated 2017. Genome-wide immune modulation of TLR3-mediated inflammation in transport of luminal Ags, indicating that it plays a key role in fa- intestinal epithelial cells differs between single and multi-strain probiotic combination. PLoS One 12: e0169847. cilitating the complex cross-talk between bacterial secretome 5. O’Callaghan, J., L. F. Butto´, J. MacSharry, K. Nally, and P. W. O’Toole. 2012. components and the underlying immune cell population at the gut- Influence of adhesion and bacteriocin production by Lactobacillus salivarius on mucosal interface (94). the intestinal epithelial cell transcriptional response. Appl. Environ. Microbiol. 78: 5196–5203. L-Lactic acid, a key metabolite produced by lactobacilli and 6. O’Flaherty, S., and T. R. Klaenhammer. 2012. Influence of exposure time on present in the LrS, had distinct effects on IEC signaling pathways. gene expression by human intestinal epithelial cells exposed to Lactobacillus acidophilus. Appl. Environ. Microbiol. 78: 5028–5032. Most notably, unique gene expression profiles were observed when 7. Tomosada, Y., J. Villena, K. Murata, E. Chiba, T. Shimazu, H. Aso, N. Iwabuchi, HT-29 IECs were challenged with TNF-a and treated with L-lactic J. Z. Xiao, T. Saito, and H. Kitazawa. 2013. Immunoregulatory effect of bifi- acid concentration–matched controls, indicating that L-lactic acid is dobacteria strains in porcine intestinal epithelial cells through modulation of ubiquitin-editing enzyme A20 expression. PLoS One 8: e59259. not responsible for the observed bioactivity of the LrS. Although 8. Ghadimi, D., U. Helwig, J. Schrezenmeir, K. J. Heller, and M. de Vrese. 2012. many of the cellular pathways involved in TNF-a signaling were Epigenetic imprinting by commensal probiotics inhibits the IL-23/IL-17 axis in an in vitro model of the intestinal mucosal immune system. J. Leukoc. Biol. 92: unaffected by L-lactic acid, it did attenuate the production of TNF- 895–911. a–induced CXCL10, CXCL1, and CCL2. Indeed, lactic acid has 9. Zhao, H., C. Zhao, Y. Dong, M. Zhang, Y. Wang, F. Li, X. Li, C. McClain, been shown to attenuate the production of some proinflammatory S. Yang, and W. Feng. 2015. Inhibition of miR122a by Lactobacillus rhamnosus GG culture supernatant increases intestinal occludin expression and protects cytokines by human monocytes and T cells exposed to challenge mice from alcoholic liver disease. Toxicol. Lett. 234: 194–200. through the transcriptional and posttranscriptional regulation of 10. Garrote, G. L., A. G. Abraham, and M. Rumbo. 2015. Is lactate an undervalued Front. Microbiol. enzymes involved in glycolysis (95, 96). Moreover, L-lactic acid has functional component of fermented food products? 6: 629. 11.Latham,T.,L.Mackay,D.Sproul,M.Karim,J.Culley,D.J.Harrison, been associated with HDAC inhibitory activity, resulting in changes L. Hayward, P. Langridge-Smith, N. Gilbert, and B. H. Ramsahoye. 2012. to gene expression profiles (11), making it a potential immuno- Lactate, a product of glycolytic metabolism, inhibits histone deacetylase ac- modulatory molecule of interest, although its actions on IEC were tivity and promotes changes in gene expression. Nucleic Acids Res. 40: 4794– 4803. distinct from those mediated by the LrS. 12. Vanamee, E´ . S., and D. L. Faustman. 2018. Structural principles of tumor ne- The results presented in this study provide new insights into the crosis factor superfamily signaling. Sci. Signal. 11: eaao4910. 13. Janes, K. A., S. Gaudet, J. G. Albeck, U. B. Nielsen, D. A. Lauffenburger, and role of soluble bioactive molecules derived from both pathogenic P. K. Sorger. 2006. The response of human epithelial cells to TNF involves an bacteria and lactobacilli in communication with host IECs. We inducible autocrine cascade. Cell 124: 1225–1239. The Journal of Immunology 11

14. Leppkes, M., M. Roulis, M. F. Neurath, G. Kollias, and C. Becker. 2014. lipopolysaccharides, Gram-negative bacteria or potentially probiotic microbes. Pleiotropic functions of TNF-a in the regulation of the intestinal epithelial re- Benef. Microbes 3: 273–286. sponse to inflammation. Int. Immunol. 26: 509–515. 41. Keestra-Gounder, A. M., R. M. Tsolis, and A. J. Baümler. 2015. Now you see 15. Hotamisligil, G. S. 2017. Inflammation, metaflammation and immunometabolic me, now you don’t: the interaction of Salmonella with innate immune receptors. disorders. Nature 542: 177–185. Nat. Rev. Microbiol. 13: 206–216. 16. Kim, H., B. J. Jung, J. H. Jung, J. Y. Kim, S. K. Chung, and D. K. Chung. 2012. 42. Kim, S. I., S. Kim, E. Kim, S. Y. Hwang, and H. Yoon. 2018. Secretion of Lactobacillus plantarum lipoteichoic acid alleviates TNF-a-induced inflamma- Salmonella pathogenicity island 1-encoded type III secretion system effectors by tion in the HT-29 intestinal epithelial cell line. Mol. Cells 33: 479–486. outer membrane vesicles in Salmonella enterica serovar typhimurium. [Pub- 17. Yan, F., H. Cao, T. L. Cover, R. Whitehead, M. K. Washington, and D. B. Polk. lished erratum appears in 2019 Front. Microbiol. 10: 411.] Front. Microbiol. 9: 2007. Soluble proteins produced by probiotic bacteria regulate intestinal epi- 2810. thelial cell survival and growth. Gastroenterology 132: 562–575. 43. Barman, M., D. Unold, K. Shifley, E. Amir, K. Hung, N. Bos, and N. Salzman. 18. Tommassen, J., and J. Arenas. 2017. Biological functions of the secretome of 2008. Enteric salmonellosis disrupts the microbial ecology of the murine gas- Neisseria meningitidis. Front. Cell. Infect. Microbiol. 7: 256. trointestinal tract. Infect. Immun. 76: 907–915. 19. Sherman, P. M., K. C. Johnson-Henry, H. P. Yeung, P. S. Ngo, J. Goulet, and 44. Stecher, B., R. Robbiani, A. W. Walker, A. M. Westendorf, M. Barthel, T. A. Tompkins. 2005. Probiotics reduce enterohemorrhagic Escherichia coli M. Kremer, S. Chaffron, A. J. Macpherson, J. Buer, J. Parkhill, et al. 2007. O157:H7- and enteropathogenic E. coli O127:H6-induced changes in polarized Salmonella enterica serovar typhimurium exploits inflammation to compete with T84 epithelial cell monolayers by reducing bacterial adhesion and cytoskeletal the intestinal microbiota. PLoS Biol. 5: 2177–2189. rearrangements. Infect. Immun. 73: 5183–5188. 45. Winter, S. E., P. Thiennimitr, M. G. Winter, B. P. Butler, D. L. Huseby, 20. Fayol-Messaoudi, D., C. N. Berger, M. H. Coconnier-Polter, V. Lie´vin-Le Moal, R. W. Crawford, J. M. Russell, C. L. Bevins, L. G. Adams, R. M. Tsolis, et al. and A. L. Servin. 2005. pH-, lactic acid-, and non-lactic acid-dependent activities 2010. Gut inflammation provides a respiratory electron acceptor for Salmonella. of probiotic Lactobacilli against Salmonella enterica Serovar Typhimurium. Nature 467: 426–429. Appl. Environ. Microbiol. 71: 6008–6013. 46. LaRock, D. L., A. Chaudhary, and S. I. Miller. 2015. Salmonellae interactions 21. Jeffrey, M. P., J. L. Strap, H. Jones Taggart, and J. M. Green-Johnson. 2018. with host processes. Nat. Rev. Microbiol. 13: 191–205. Suppression of intestinal epithelial cell chemokine production by Lactobacillus 47. Thiennimitr, P., S. E. Winter, M. G. Winter, M. N. Xavier, V. Tolstikov, rhamnosus R0011 and Lactobacillus helveticus R0389 is mediated by secreted D. L. Huseby, T. Sterzenbach, R. M. Tsolis, J. R. Roth, and A. J. Baümler. 2011. bioactive molecules. Front. Immunol. 9: 2639. Intestinal inflammation allows Salmonella to use ethanolamine to compete with Downloaded from 22. Jung, H. C., L. Eckmann, S. K. Yang, A. Panja, J. Fierer, E. Morzycka- the microbiota. Proc. Natl. Acad. Sci. USA 108: 17480–17485. Wroblewska, and M. F. Kagnoff. 1995. A distinct array of proinflammatory 48. Arthur, J. S., and S. C. Ley. 2013. Mitogen-activated protein kinases in innate cytokines is expressed in human colon epithelial cells in response to bacterial immunity. Nat. Rev. Immunol. 13: 679–692. invasion. J. Clin. Invest. 95: 55–65. 49. Chi, H., S. P. Barry, R. J. Roth, J. J. Wu, E. A. Jones, A. M. Bennett, and 23. Rousset, M. 1986. The human colon carcinoma cell lines HT-29 and Caco-2: two R. A. Flavell. 2006. Dynamic regulation of pro- and anti-inflammatory cytokines in vitro models for the study of intestinal differentiation. Biochimie 68: 1035– by MAPK phosphatase 1 (MKP-1) in innate immune responses. Proc. Natl. 1040. Acad. Sci. USA 103: 2274–2279.

24. Chomczynski, P. 1993. A reagent for the single-step simultaneous isolation of 50. Hammer, M., J. Mages, H. Dietrich, A. Servatius, N. Howells, A. C. Cato, and http://www.jimmunol.org/ RNA, DNA and proteins from cell and tissue samples. Biotechniques 15: 532– R. Lang. 2006. Dual specificity phosphatase 1 (DUSP1) regulates a subset of 534, 536–537. LPS-induced genes and protects mice from lethal endotoxin shock. J. Exp. Med. 25. Berger, J. A., S. Hautaniemi, A. K. Ja¨rvinen, H. Edgren, S. K. Mitra, and 203: 15–20. J. Astola. 2004. Optimized LOWESS normalization parameter selection for 51. Ko, H. M., S. H. Oh, H. S. Bang, N. I. Kang, B. H. Cho, S. Y. Im, and H. K. Lee. DNA microarray data. BMC Bioinformatics 5: 194. 2009. Glutamine protects mice from lethal endotoxic shock via a rapid induction 26. Stelzer, G., A. Inger, T. Olender, T. Iny-Stein, I. Dalah, A. Harel, M. Safran, and of MAPK phosphatase-1. J. Immunol. 182: 7957–7962. D. Lancet. 2009. GeneDecks: paralog hunting and gene-set distillation with 52. Lang, R., M. Hammer, and J. Mages. 2006. DUSP meet immunology: GeneCards annotation. OMICS 13: 477–487. dual specificity MAPK phosphatases in control of the inflammatory response. 27. Shannon, P., A. Markiel, O. Ozier, N. S. Baliga, J. T. Wang, D. Ramage, J. Immunol. 177: 7497–7504. N. Amin, B. Schwikowski, and T. Ideker. 2003. Cytoscape: a software envi- 53. Salojin, K. V., I. B. Owusu, K. A. Millerchip, M. Potter, K. A. Platt, and ronment for integrated models of biomolecular interaction networks. Genome T. Oravecz. 2006. Essential role of MAPK phosphatase-1 in the negative control

Res. 13: 2498–2504. of innate immune responses. J. Immunol. 176: 1899–1907. by guest on September 25, 2021 28. Warde-Farley, D., S. L. Donaldson, O. Comes, K. Zuberi, R. Badrawi, P. Chao, 54. Wu, M., L. G. Xu, Z. Zhai, and H. B. Shu. 2003. SINK is a p65-interacting M. Franz, C. Grouios, F. Kazi, C. T. Lopes, et al. 2010. The geneMANIA pre- negative regulator of NF-kappaB-dependent transcription. J. Biol. Chem. 278: diction server: biological network integration for gene prioritization and pre- 27072–27079. dicting gene function. Nucleic Acids Res. 38: W214–W220. 55. Ohoka, N., S. Yoshii, T. Hattori, K. Onozaki, and H. Hayashi. 2005. TRB3, a 29. Kra¨mer, A., J. Green, J. Pollard, Jr., and S. Tugendreich. 2014. Causal analysis novel ER stress-inducible gene, is induced via ATF4-CHOP pathway and is approaches in ingenuity pathway analysis. Bioinformatics 30: 523–530. involved in cell death. EMBO J. 24: 1243–1255. 30. Szklarczyk, D., A. L. Gable, D. Lyon, A. Junge, S. Wyder, J. Huerta-Cepas, 56. Smith, S. M., A. P. Moran, S. P. Duggan, S. E. Ahmed, A. S. Mohamed, M. Simonovic, N. T. Doncheva, J. H. Morris, P. Bork, et al. 2019. STRING v11: H. J. Windle, L. A. O’Neill, and D. P. Kelleher. 2011. Tribbles 3: a novel reg- protein-protein association networks with increased coverage, supporting func- ulator of TLR2-mediated signaling in response to Helicobacter pylori lipo- tional discovery in genome-wide experimental datasets. Nucleic Acids Res. 47: polysaccharide. J. Immunol. 186: 2462–2471. D607–D613. 57. Whitmore, M. M., A. Iparraguirre, L. Kubelka, W. Weninger, T. Hai, and 31. Thomas, C. M., and J. Versalovic. 2010. Probiotics-host communication: mod- B. R. Williams. 2007. Negative regulation of TLR-signaling pathways by acti- ulation of signaling pathways in the intestine. Gut Microbes 1: 148–163. vating transcription factor-3. J. Immunol. 179: 3622–3630. 32. Lebeer, S., J. Vanderleyden, and S. C. De Keersmaecker. 2008. Genes and 58. Lee, S., G. L. Kim, N. Y. Kim, S. J. Kim, P. Ghosh, and D. K. Rhee. 2018. ATF3 molecules of lactobacilli supporting probiotic action. Microbiol. Mol. Biol. Rev. stimulates IL-17a by regulating intracellular Ca2+/ROS-dependent IL-1b acti- 72: 728–764. vation during Streptococcus pneumoniae infection. Front. Immunol. 9: 1954. 33. Lebeer, S., J. Vanderleyden, and S. C. De Keersmaecker. 2010. Host interactions 59. Nguyen, C. T., T. T. Luong, S. Lee, G. L. Kim, S. Pyo, and D. K. Rhee. 2016. of probiotic bacterial surface molecules: comparison with commensals and ATF3 provides protection from Staphylococcus aureusandListeria mono- pathogens. Nat. Rev. Microbiol. 8: 171–184. cytogenes infections. [Published erratum appears in 2016 FEMS Microbiol. Lett. 34. Park, S. H., K. H. Do, H. J. Choi, J. Kim, K. H. Kim, J. Park, C. G. Oh, and 363: pii fnw090.] FEMS Microbiol. Lett. 363: pii fnw062. Y. Moon. 2013. Novel regulatory action of ribosomal inactivation on epithelial 60. Li, H. F., C. F. Cheng, W. J. Liao, H. Lin, and R. B. Yang. 2010. ATF3-mediated Nod2-linked proinflammatory signals in two convergent ATF3-associated path- epigenetic regulation protects against acute kidney injury. J. Am. Soc. Nephrol. ways. J. Immunol. 191: 5170–5181. 21: 1003–1013. 35. Broekaert, I. J., N. N. Nanthakumar, and W. A. Walker. 2007. Secreted probiotic 61. Gilchrist, M., V. Thorsson, B. Li, A. G. Rust, M. Korb, J. C. Roach, K. Kennedy, factors ameliorate enteropathogenic infection in zinc-deficient human Caco-2 T. Hai, H. Bolouri, and A. Aderem. 2006. Systems biology approaches identify and T84 cell lines. Pediatr. Res. 62: 139–144. ATF3 as a negative regulator of toll-like receptor 4. [Published erratum appears 36. Ganguli, K., D. Meng, S. Rautava, L. Lu, W. A. Walker, and N. Nanthakumar. in 2008 Nature 451: 1022.] Nature 441: 173–178. 2013. Probiotics prevent necrotizing enterocolitis by modulating enterocyte 62. Kwon, J. W., H. K. Kwon, H. J. Shin, Y. M. Choi, M. A. Anwar, and S. Choi. genes that regulate innate immune-mediated inflammation. Am. J. Physiol. 2015. Activating transcription factor 3 represses inflammatory responses by Gastrointest. Liver Physiol. 304: G132–G141. binding to the p65 subunit of NF-kB. Sci. Rep. 5: 14470. 37. Tao, Y., K. A. Drabik, T. S. Waypa, M. W. Musch, J. C. Alverdy, O. Schneewind, 63. Miro-Blanch, J., and O. Yanes. 2019. Epigenetic regulation at the interplay E. B. Chang, and E. O. Petrof. 2006. Soluble factors from Lactobacillus GG between gut microbiota and host metabolism. Front. Genet. 10: 638. activate MAPKs and induce cytoprotective heat shock proteins in intestinal 64. Qin, Y., and P. A. Wade. 2018. Crosstalk between the microbiome and epige- epithelial cells. [Published erratum appears in 2006 Am. J. Physiol. Cell Physiol. nome: messages from bugs. J. Biochem. 163: 105–112. 291: C194.] Am. J. Physiol. Cell Physiol. 290: C1018–C1030. 65. Bhat, M. I., A. Kumari, S. Kapila, and R. Kapila. 2019. Probiotic lactobacilli 38. Kim, C. H. 2018. Immune regulation by microbiome metabolites. Immunology mediated changes in global epigenetic signatures of human intestinal epithelial 154: 220–229. cells during Escherichia coli challenge. Ann. Microbiol. 69: 603–612. 39. Postler, T. S., and S. Ghosh. 2017. Understanding the holobiont: how microbial me- 66. Papa, S., F. Zazzeroni, C. Bubici, S. Jayawardena, K. Alvarez, S. Matsuda, tabolites affect human health and shape the immune system. Cell Metab. 26: 110–130. D. U. Nguyen, C. G. Pham, A. H. Nelsbach, T. Melis, et al. 2004. Gadd45 beta 40. Audy, J., O. Mathieu, J. Belvis, and T. A. Tompkins. 2012. Transcriptomic re- mediates the NF-kappa B suppression of JNK signalling by targeting MKK7/ sponse of immune signalling pathways in intestinal epithelial cells exposed to JNKK2. Nat. Cell Biol. 6: 146–153. 12 LACTOBACILLI SECRETOME–MEDIATED NEGATIVE REGULATOR INDUCTION

67. Ran, R., A. Lu, L. Zhang, Y. Tang, H. Zhu, H. Xu, Y. Feng, C. Han, G. Zhou, 81. Vong, L., R. J. Lorentz, A. Assa, M. Glogauer, and P. M. Sherman. 2014. Pro- A. C. Rigby, and F. R. Sharp. 2004. Hsp70 promotes TNF-mediated apoptosis by biotic Lactobacillus rhamnosus inhibits the formation of neutrophil extracellular binding IKK gamma and impairing NF-kappa B survival signaling. Genes Dev. traps. J. Immunol. 192: 1870–1877. 18: 1466–1481. 82. Bai, A. P., Q. Ouyang, W. Zhang, C. H. Wang, and S. F. Li. 2004. Probiotics 68. Salminen, A., T. Paimela, T. Suuronen, and K. Kaarniranta. 2008. Innate im- inhibit TNF-alpha-induced interleukin-8 secretion of HT29 cells. World J. munity meets with cellular stress at the IKK complex: regulation of the IKK Gastroenterol. 10: 455–457. complex by HSP70 and HSP90. Immunol. Lett. 117: 9–15. 83. Gao, K., C. Wang, L. Liu, X. Dou, J. Liu, L. Yuan, W. Zhang, and H. Wang. 69. Hoetzenecker, W., B. Echtenacher, E. Guenova, K. Hoetzenecker, F. Woelbing, 2017. Immunomodulation and signaling mechanism of Lactobacillus rhamnosus J. Bru¨ck, A. Teske, N. Valtcheva, K. Fuchs, M. Kneilling, et al. 2011. ROS- GG and its components on porcine intestinal epithelial cells stimulated by li- induced ATF3 causes susceptibility to secondary infections during sepsis- popolysaccharide. J. Microbiol. Immunol. Infect. 50: 700–713. associated immunosuppression. Nat. Med. 18: 128–134. 84. Tuo, Y., X. Song, Y. Song, W. Liu, Y. Tang, Y. Gao, S. Jiang, F. Qian, and G. Mu. 70. Kanwar, Y. S. 2010. TRB3: an oxidant stress-induced pseudokinase with a po- 2018. Screening probiotics from Lactobacillus strains according to their abilities tential to negatively modulate MCP-1 cytokine in diabetic nephropathy. Am. J. to inhibit pathogen adhesion and induction of pro-inflammatory cytokine IL-8. Physiol. Renal Physiol. 299: F963–F964. J. Dairy Sci. 101: 4822–4829. 71. Lin, Q. Y., O. Mathieu, T. A. Tompkins, N. D. Buckley, and J. M. Green- 85. Wan, M. L., Z. Chen, N. P. Shah, and H. El-Nezami. 2018. Effects of Lacto- Johnson. 2016. Modulation of the TNFa-induced gene expression profile of bacillus rhamnosus GG and Escherichia coli nissle 1917 cell-free supernatants intestinal epithelial cells by soy fermented with lactic acid bacteria. J. Funct. on modulation of mucin and cytokine secretion on human intestinal epithelial Foods 23: 400–411. HT29-MTX cells. J. Food Sci. 83: 1999–2007. 72.Valladares,R.B.,C.Graves,K.Wright,C.L.Gardner,G.L.Lorca,and 86. Maaser, C., L. Eckmann, G. Paesold, H. S. Kim, and M. F. Kagnoff. 2002. C. F. Gonzalez. 2015. H2O2 production rate in Lactobacillus johnsonii Ubiquitous production of macrophage migration inhibitory factor by human is modulated via the interplay of a heterodimeric flavin oxidoreductase gastric and intestinal epithelium. Gastroenterology 122: 667–680. with a soluble 28 Kd PAS domain containing protein. Front. Microbiol. 6: 87. Harris, J., S. VanPatten, N. S. Deen, Y. Al-Abed, and E. F. Morand. 2019. 716. Rediscovering MIF: new tricks for an old cytokine. Trends Immunol. 40: 447–462. 73. Pircalabioru, G., G. Aviello, M. Kubica, A. Zhdanov, M. H. Paclet, L. Brennan, 88. Vujicic, M., T. Saksida, S. Despotovic, S. S. Bajic, I. Lalic´,I.Koprivica,D.Gajic, R. Hertzberger, D. Papkovsky, B. Bourke, and U. G. Knaus. 2016. Defensive N. Golic, M. Tolinacki, and I. Stojanovic. 2018. The role of macrophage migration mutualism rescues NADPH oxidase inactivation in gut infection. Cell Host inhibitory factor in the function of intestinal barrier. Sci. Rep. 8: 6337. Downloaded from Microbe 19: 651–663. 89. Kleemann, R., A. Hausser, G. Geiger, R. Mischke, A. Burger-Kentischer, 74. Jones, R. M., L. Luo, C. S. Ardita, A. N. Richardson, Y. M. Kwon, O. Flieger, F. J. Johannes, T. Roger, T. Calandra, A. Kapurniotu, et al. 2000. J. W. Mercante, A. Alam, C. L. Gates, H. Wu, P. A. Swanson, et al. 2013. Intracellular action of the cytokine MIF to modulate AP-1 activity and the cell Symbiotic lactobacilli stimulate gut epithelial proliferation via Nox-mediated cycle through Jab1. Nature 408: 211–216. generation of reactive oxygen species. EMBO J. 32: 3017–3028. 90. Das, R., M. I. LaRose, C. B. Hergott, L. Leng, R. Bucala, and J. N. Weiser. 2014. 75. Kumar, A., H. Wu, L. S. Collier-Hyams, J. M. Hansen, T. Li, K. Yamoah, Macrophage migration inhibitory factor promotes clearance of pneumococcal Z. Q. Pan, D. P. Jones, and A. S. Neish. 2007. Commensal bacteria modulate colonization. J. Immunol. 193: 764–772.

cullin-dependent signaling via generation of reactive oxygen species. EMBO J. 91. Roger, T., C. Froidevaux, C. Martin, and T. Calandra. 2003. Macrophage mi- http://www.jimmunol.org/ 26: 4457–4466. gration inhibitory factor (MIF) regulates host responses to endotoxin through 76. Wentworth, C. C., A. Alam, R. M. Jones, A. Nusrat, and A. S. Neish. 2011. modulation of toll-like receptor 4 (TLR4). J. Endotoxin Res. 9: 119–123. Enteric commensal bacteria induce extracellular signal-regulated kinase pathway 92. Roger, T., J. Delaloye, A. L. Chanson, M. Giddey, D. Le Roy, and T. Calandra. signaling via formyl peptide receptor-dependent redox modulation of dual spe- 2013. Macrophage migration inhibitory factor deficiency is associated with cific phosphatase 3. J. Biol. Chem. 286: 38448–38455. impaired killing of gram-negative bacteria by macrophages and increased sus- 77. Herb, M., A. Gluschko, K. Wiegmann, A. Farid, A. Wolf, O. Utermo¨hlen, ceptibility to Klebsiella pneumoniae sepsis. J. Infect. Dis. 207: 331–339. O. Krut, M. Kro¨nke, and M. Schramm. 2019. Mitochondrial reactive oxygen 93. Cho, M. L., Y. M. Moon, Y. J. Heo, Y. J. Woo, J. H. Ju, K. S. Park, S. I. Kim, species enable proinflammatory signaling through disulfide linkage of NEMO. S. H. Park, H. Y. Kim, and J. K. Min. 2009. NF-kappaB inhibition leads to increased Sci. Signal. 12: eaar5926. synthesis and secretion of MIF in human CD4+ T cells. Immunol. Lett. 123: 21–30. 78. Reuter, S., S. C. Gupta, M. M. Chaturvedi, and B. B. Aggarwal. 2010. Oxidative 94. Man, A. L., F. Lodi, E. Bertelli, M. Regoli, C. Pin, F. Mulholland, A. R. Satoskar, stress, inflammation, and cancer: how are they linked? Free Radic. Biol. Med. M. J. Taussig, and C. Nicoletti. 2008. Macrophage migration inhibitory factor

49: 1603–1616. plays a role in the regulation of microfold (M) cell-mediated transport in the gut. by guest on September 25, 2021 79.Li,X.,Y.Yu,B.Gorshkov,S.Haigh,Z.Bordan,D.Weintraub,R.D.Rudic, J. Immunol. 181: 5673–5680. T. Chakraborty, S. A. Barman, A. D. Verin, et al. 2018. Hsp70 suppresses 95. Dietl, K., K. Renner, K. Dettmer, B. Timischl, K. Eberhart, C. Dorn, mitochondrial reactive oxygen species and preserves pulmonary microvascu- C. Hellerbrand, M. Kastenberger, L. A. Kunz-Schughart, P. J. Oefner, et al. 2010. lar barrier integrity following exposure to bacterial toxins. Front. Immunol. 9: Lactic acid and acidification inhibit TNF secretion and glycolysis of human 1309. monocytes. J. Immunol. 184: 1200–1209. 80. Gaisawat, M. B., M. M. Iskandar, C. W. MacPherson, T. A. Tompkins, and 96. Haas, R., J. Smith, V. Rocher-Ros, S. Nadkarni, T. Montero-Melendez, S. Kubow. 2019. Probiotic supplementation is associated with increased an- F. D’Acquisto, E. J. Bland, M. Bombardieri, C. Pitzalis, M. Perretti, et al. 2015. tioxidant capacity and copper chelation in C. difficile-infected fecal water. Lactate regulates metabolic and pro-inflammatory circuits in control of T cell Nutrients 11: 2007. migration and effector functions. PLoS Biol. 13: e1002202.

Supplemental Figure S1. Gene-interaction maps of genes involved in the NF-κB signaling pathways in HT-29 IECs that were differentially expressed relative to untreated cells as determined by the GeneDecks gene enrichment analysis and GeneMania (p < 0.05) by A. LrS B. TNF-α C. STS D. L- lactic acid controls E. TNF-α + the LrS F. TNF-α + L-lactic acid controls G. STS + the LrS or H. STS and L-lactic acid controls. 1

Supplemental Figure S2. String protein-interaction maps of cytokines and chemokines induced by the different challenges involved in inflammatory responses, TNF-α signaling, NF-κB signaling or Salmonella infection in HT-29 IECs by A. TNF-α B. STS C. TNF-α + < LrS D. TNF-α + L-lactic acid controls E. STS + the LrS or H. STS and L-lactic acid controls. 2

)

L L 20000 500 L 5000 b

10000 L b b 2500

m m

b b m

/ /

/ b

g g

g b b

p p

p 400 4000

(

( (

8000 p b 2000

15000

(

i i

o 300 3000

t t

6000 i 1500

c c

10000 c

d d

d 200 2000

o o

o 4000 1000

o r r a

r a

P P

P a

5000

a a P

8 a a

0 1 0 100 1000 a a a

2000 2 500

3 C

0 I 0 0

X C

0 C 0 C C A rS - rS A C A S  S A C A S  S A C M L F L C A S  S A L r - r L L r - r L C A S  S A - L L - L r - r L M - L F L - M - L F L - L r - r L L N + L M - L F L - L N + L L N + L M - L F L - T  + L N + L T  + T  + L N + L -  T  + -  -  T  + F - -  F - F - -  N F F - N F N F F - T N N F T N T N N F T T N T T T N

T T

)

L L

800 600 b b 150 L 800 b b

/ /

b b 20000 /

p p

b b p (

( b

(

600 600

n n

400 15000 n 100

c c

400 c 400

u u

b u

d d

10000 d

r r

200 r r

r 50

200 200

a a a a a a a a

B 2

5000 a 6 L

a L

X C

0 0 X 0 0

C C

F 0 C A S - S A C A S - S A C A S  S A C A S  S A L r r L L r r L N L r - r L L r - r L M - L F L - M - L F L -  M - L F L - M - L F L - L L L L T C A S S A N + N + L r - r L N + L L N + L T  + T  + M - L F L -L T  + T  + -  -  L N L -  -  F - F - T + F - F - N F N F - + N F N F T N T N F - T N T N T T N F T T T N

)

L L 10000 400 50 b 3000 L

L 1500 m b b

b b m

m m

/ b b

p p

8000 ( 40

(

300

(

n c 2000

o o

i 1000

t o 6000 t

30 t

u u

200 c

d d

4000 o a

20 o

o r

r 1000

r r c o a a

P 500

P P a

a a a

a a

100 P

8 2000 10

3 L

X 0

I X

0 0 0 C

X 0

C C

C A S S S A C C A S S S A C A S S S A C A S S S r T r r T r r T r r r A M -L L L -L C A S S S A M -L L L -L M -L L L -L M -L L T L -L L S L L r T r L L S L L S L L S L + M - L S L - + + + S + L + L S + S + S + T S + T S T S T S T S T S T S S T S S T S S T

S S S S

)

L ) L b b b L b b

400 ) 600

L m

m b b

b b 500 m

600 /

/ 3000

(

300 400

(

n 400

i i

400 t 2000

i c

i 300 c

t a c t c

200 u

u c

u a a c

o d

o 200

r o

r 200

a o P

a r

P 200 1000

100 a a

a P a

6 100

C S

0 X 0 0

X C

F 0

0 C C C A rS S rS A N C A S S S A C A rS S S A M L T L L r T r M L T r L - L L - T C A S S S A L - L L - L S L C A rS S S A r T r M - L S L - S + L T r L M -L L L -L L + L L + L + M - L S L - L S L + S L + L + S + S T S + + T S T S S T S S S T S T S T S T S S S S T S T S S

Supplemental Figure S3. Cytokine profiles from HT-29 IECs exposed to the different challenges for 6 hours. Data shown is the mean cytokine production (pg/mL) ± SEM (n = 4). Different letters denote significance between treatment groups as determined by the one-way ANOVA and Tukey’s post-hoc test (P < 0.05). 3

✱ ✱ p = 0.0163 p = 0.0312

1500

)

p (

1000

u d

o 500

L I 0 l o - rS rS rS tr F L L L n N o T % % % C 0 0 0 3 2 1 m + + + iu - - - d F F F e N N N M T T T

Supplemental Figure S4. IL-8 production from T84 human IECs challenged with TNF-α and different concentrations of the LrS. Data shown is the mean IL-8 production (pg/mL) ± SEM (n = 3). * denote significant differences between treatments as determined by Dunnett’s one-way ANOVA.

Table SI. List of primers used for relative RT-qPCR for microarray validation

GenBank Accession Amplicon Gene Primer Sequence Source Number Length (bp) F: GTGCTCGCGCTACTCTCTC Dydensborg et al., B2M NM_004048 150 R: GTCAACTTCAATGTCGGAT 2006 F: GCAATGTTGCCAGTGTCTG Dydensborg et al., RPLPO NM_001002 142 R: GCCTTGACCTTTTCAGCAA 2006 F: TGGTACCCAGCTCCTCTACG TRIB3 NM_021158.4 184 Jousse et al., 2007 R: GACAAAGCGACACAGCTTGA F: AAGAACGAGAAGCAGCATTTGAT Bottone et al., ATF3 NM_001206484.3 71 R: TTCTGAGCCCGGACAATACAC 2005 F: GGCCCCGAGAACAGACAAA DUSP1 NM_004417 80 Locati et al., 2002 R: GTGCCCACTTCCATGACCAT F: GCAGCACTACTTCTTGACCACC MacPherson et al., NFкB1 NM_003998.3 130 R: TCTGCTCCTGAGCATTGACGTC 2014