Lactobacilli and Streptococci Activate NF-κB and STAT Signaling Pathways in Human Macrophages

This information is current as Minja Miettinen, Anne Lehtonen, Ilkka Julkunen and of October 3, 2021. Sampsa Matikainen J Immunol 2000; 164:3733-3740; ; doi: 10.4049/jimmunol.164.7.3733 http://www.jimmunol.org/content/164/7/3733 Downloaded from

References This article cites 62 articles, 25 of which you can access for free at: http://www.jimmunol.org/content/164/7/3733.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 3, 2021

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 © 2000 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. Lactobacilli and Streptococci Activate NF-␬B and STAT Signaling Pathways in Human Macrophages1

Minja Miettinen,2 Anne Lehtonen, Ilkka Julkunen, and Sampsa Matikainen

Gram-positive bacteria induce the production of several cytokines in human leukocytes. The molecular mechanisms involved in Gram-positive bacteria-induced cytokine production have been poorly characterized. In this work we demonstrate that both nonpathogenic Lactobacillus rhamnosus GG and pathogenic Streptococcus pyogenes (group A streptococci) induce NF-␬B and STAT DNA-binding activity in human primary macrophages as analyzed by EMSA. NF-␬B activation was rapid and was not inhibited by a protein synthesis inhibitor cycloheximide, suggesting that these bacteria could directly activate NF-␬B. STAT1, STAT3, and IFN regulatory factor-1 DNA binding was induced by both bacteria with delayed kinetics compared with NF-␬B. In addition, streptococci induced the formation of IFN-␣-specific transcription factor complex and IFN-stimulated gene factor-3 (ISGF3). STAT1 and STAT3 activation and ISGF3 complex formation were inhibited by cycloheximide or by neutralization with IFN-␣/␤-specific Abs. Streptococci were more potent than lactobacilli in inducing STAT1, ISGF3, and IFN regulatory factor-1 Downloaded from DNA binding. Accordingly, only streptococci induced IFN-␣ production. The activation of the IFN-␣ signaling pathway by streptococci could play a role in the pathogenesis of these bacteria. These results indicate that extracellular Gram-positive bacteria activate transcription factors involved in cytokine signaling by two mechanisms: directly, leading to NF-␬B activation, and indirectly via cytokines, leading to STAT activation. The Journal of Immunology, 2000, 164: 3733–3740.

acrophages have a central role in initiating the innate activation of NF-␬B is followed by the production of TNF-␣ and http://www.jimmunol.org/ immune response, which leads to activation of the further activation of NF-␬B (8, 9). STAT activation occurs via M adaptive response. Macrophages phagocytose infected tyrosine phosphorylation; binding of cytokines to their receptors cells, present Ags to T and B cells, and produce cytokines and results in autophosphorylation of receptor-associated JAK kinases chemokines that modulate immune responses (1). These cytokines that phosphorylate and activate STATs. Activated STATs form include TNF-␣ and IL-6, which are among the first produced dur- homo- or heterodimers that translocate into the nucleus and bind ing the innate immune response toward bacteria. TNF-␣ and IL-6 specific target elements in the promoters of cytokine-inducible have pleiotropic effects such as induction of the acute phase re- genes (10–12). STAT1 and STAT3 can be activated by several sponse and activation of macrophages (2, 3). An innate immune cytokines, including IFN-␣ and IL-6, and bind to the IFN-␥ acti- 3 response to viruses is characterized by rapid production of IFN-␣ vation site (GAS) element. IFN-␣ is the only known activator of by guest on October 3, 2021 in macrophages. In addition to its direct antiviral functions, IFN-␣ STAT2 that, together with STAT1 and p48, form the IFN-stimu- has a range of immunoregulatory functions, including NK cell ac- lated gene factor-3 (ISGF3) (3) complex that binds to the IFN tivation and enhancement of Th1-type immunity (4–7). Although response element (ISRE) (13, 14). the role of IFN-␣ in viral and intracellular bacterial infections is Lactobacilli are nonpathogenic Gram-positive inhabitants of hu- well established, in extracellular bacterial infections it remains man normal microflora (15); some of them have been postulated to poorly characterized. have health beneficial effects, such as stimulation of the immune Cytokines initiate signaling cascades through their receptors, system (16). Strain Lactobacillus rhamnosus GG has been exten- leading to activation of transcription factors and target gene ex- sively studied regarding safety and clinical effects (17). Strepto- pression. NF-␬B and STATs are both latent cytoplasmic transcrip- coccus pyogenes (group A streptococci) is a major Gram-positive tion factors activated by Ag or cytokine stimulation. They regulate human pathogen causing a wide range of infections. Since the transcription of genes encoding proteins involved in immune, 1980s, highly invasive strains associated with shock and organ failure acute phase, and inflammatory responses. In humans, activated have emerged (18). To analyze the mechanisms by which Gram-pos- NF-␬B dimers consist mainly of the Rel family proteins p50 and itive bacteria induce immune responses we have compared nonpatho- p65 subunits. NF-␬B binds to responsive ␬B sites in the promoters genic and pathogenic strains. We have previously shown that L. rh- and enhancers of target genes, including TNF-␣ and IL-6. TNF-␣ amnosus GG and S. pyogenes induce the production of several is known to initiate an autoregulatory feedback loop, where the cytokines in human PBMC (19). In this work we have compared and analyzed the role of live L. rhamnosus GG and S. pyogenes in acti- vation of NF-␬B and STATs in human primary macrophages. Department of Virology, National Public Health Institute, Helsinki, Finland Received for publication July 19, 1999. Accepted for publication January 21, 2000. Materials and Methods The costs of publication of this article were defrayed in part by the payment of page Bacterial strains charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Lactobacillus rhamnosus GG (American Type Culture Collection 53103) was obtained from Valio R&D (Helsinki, Finland), and Streptococcus pyo- 1 This work was supported by the University of Helsinki 350th Anniversary Fund, the genes serotype T1 (IH32030), isolated from a child with bacteremia, was Medical Research Council of the Academy Finland, the Sigrid Juselius Foundation, and the Jenny and Antti Wihuri Foundation. 2 Address correspondence and reprint requests to Dr. Minja Miettinen, Department of 3 Abbreviations used in this paper: GAS, IFN-␥ activation site; ISGF3, IFN-stimulated Virology, National Public Health Institute, Mannerheimintie 166, 00300 Helsinki, gene factor-3; ISRE, IFN response element; CHX, cycloheximide; IRF, IFN regulatory Finland. E-mail address: minja.miettinen@ktl.fi factor; TLR, Toll-like receptor; GAS, group A streptococci; SIE, c-sis-inducible element.

Copyright © 2000 by The American Association of Immunologists 0022-1767/00/$02.00 3734 NF-␬B AND STAT ACTIVATION BY GRAM-POSITIVE BACTERIA obtained from the collection of the National Public Health Institute (Hel- sinki, Finland). Bacteria were stored in skimmed milk at Ϫ70°C and pas- saged three times as previously described (20) before their use in stimu- lation experiments. Lactobacilli were grown in MRS medium (Difco, Detroit, MI) and streptococci in TY medium supplemented with 0.2% glu- cose (21). For stimulation experiments bacteria were grown to logarithmic growth phase, and the number of bacterial cells was determined by count- ing in a Petroff Hausser counting chamber. Cell culture Freshly collected leukocyte-rich buffy coats from healthy blood donors were supplied by the Finnish Red Cross Blood Transfusion Service (Hel- sinki, Finland). PBMC were isolated by a density gradient centrifugation over Ficoll-Paque gradient (Pharmacia, Uppsala, Sweden). After washing, the cells were resuspended in RPMI 1640 medium (Sigma, St. Louis, MO) supplemented with 0.6 ␮g/ml penicillin, 60 ␮g/ml streptomycin, 2 mM L-glutamine, and 20 mM HEPES. For monocyte differentiation, PBMC were allowed to adhere to plastic six-well plates (Falcon, Becton Dickin- son, Franklin Lakes, NJ) for1hat37°C in RPMI 1640 medium supple- mented with antibiotics, glutamine, and HEPES without FCS (10 ϫ 106 cells/well). After incubation nonadherent cells were removed, and the wells were washed twice with PBS (pH 7.4). Adherent cells were then grown for

7 days in macrophage/serum-free medium (Life Technologies, Grand Is- Downloaded from land, NY) supplemented with antibiotics and recombinant GM-CSF at 10 ng/ml (Leucomax, Schering-Plough, Innishannon, Ireland). More than 90% of the cultured cells were macrophages as determined by their morphology and CD14 expression (data not shown). Stimulation experiments

To minimize interindividual variation all experiments were performed with http://www.jimmunol.org/ cells obtained from four to six blood donors. Stimulation experiments were conducted in RPMI 1640 medium either with or without 10% heat-inacti- vated FCS (Integro, Zaandam, The Netherlands). Macrophages were stim- ulated with live bacteria at a 1:1 ratio. Cycloheximide (CHX; Sigma) was FIGURE 1. Activation of NF-␬B DNA binding by lactobacilli and used to inhibit protein synthesis at concentration of 10 ␮g/ml. It was added to cell culture medium 0.5 h after the beginning of stimulations. Neutral- streptococci. Macrophages were stimulated with live lactobacilli (LAB) or izing sheep anti-IFN-␣/␤ (22) and goat anti-IL-6 (R & D Systems, Abing- streptococci (GAS) for 1, 3, 6, 9, or 24 h, and nuclear extracts were pre- don, U.K.) Abs were used at concentrations of 2400/165 neutralizing IU/ml pared and analyzed by EMSA using the NF-␬B oligonucleotide. A, Kinet- and 1 ␮g/ml, respectively. Cell culture supernatants and cells were col- ics of NF-␬B activation by lactobacilli and streptococci. B, Supershift ex- lected at different times after stimulation and pooled. Cells were used for periment with anti-p50 and anti-p65 Abs. Results are representative of preparing nuclear extracts, isolation of total cellular RNA, or SDS-PAGE three separate experiments. by guest on October 3, 2021 sample preparation. Supernatants stored at Ϫ20°C were used for cytokine determinations.

EMSA 1 h at room temperature. Primary Abs used in immunoblotting were guinea pig anti-human IRF-1 (1/2000 dilution) (25) and guinea pig anti-human Nuclear extracts were prepared as previously described (23). Nuclear pro- ␮ MxA (1/2000 dilution) (26). Peroxidase-conjugated rabbit anti-guinea pig tein/DNA-binding reactions were performed in a volume of 20 l contain- Igs (Dako, Copenhagen, Denmark) were used as secondary Abs (1/2000 ing 5 ␮g of nuclear extract protein, 10 mM HEPES-KOH (pH 7.9), 100 ␮ dilution). The bands were visualized on Amersham HyperMax film using mM NaCl, 1 mM EDTA, 1 mM DTT, 10% glycerol, and 2 g poly(dI-dC) the ECL chemiluminescence system according to the manufacturer’s (Am- as a nonspecific competitor. NF-␬B(5Ј-AGTTGAGGGGACTTTC Ј Ј ersham) instructions. Protein concentrations in the samples were deter- CCAGG-3 ), IRF1-GAS (5 -AGCTTCAGCCTGATTTCCCCGAAAT mined with the Bio-Rad protein assay kit (Hercules, CA). GACGGCA-3Ј), SIE (5Ј-GATCTAGGGATTTCCGGGAAATGAAGCT- 3Ј), ISRE15 (5Ј-AGCTTGATCGGGAAAGGGAAACCGAAACTGA RNA isolation and analysis AGCCA-3Ј), and PRDI (5Ј-GATCAAGTGAAAGTGAAAGTGA-3Ј) oligonucleotides were synthesized with an IBI oligonucleotide synthesizer For isolation of total cellular RNA, stimulated cells were collected, washed (Foster City, CA) and purified on PAGE in the presence of 8 mol/l urea. once with cold PBS (pH 7.4), and lysed in guanidium isothiocyanate The probes for NF-␬B, IRF1-GAS, SIE, and ISRE15 were end labeled with (stored at Ϫ70°C) (27) followed by centrifugation through a CsCl cushion [␥-32P]dATP (3000 Ci/mol; Amersham, Aylesbury, U.K.) by T4 polynu- (28). RNA was quantitated photometrically, and samples containing equal cleotide kinase and the probe for PRDI with the Klenow fill-in method. The amounts (20 ␮g) of total cellular RNA were size-fractionated on 1% form- binding reaction was performed at room temperature for 30 min. For su- aldehyde-agarose gels, transferred to Hybond-N nylon membranes (Am- pershift assays, nuclear protein extracts were incubated with Abs against ersham), and hybridized with human IRF-1 (29) and MxA (30) cDNA NF-␬B p50 (sc-1190 x), NF-␬B p65 (sc-372 x), STAT1␣ p91 (sc-345 x), probes. To control for equal RNA loading, ethidium bromide staining or STAT5 recognizing both STAT5A and STAT5B (sc-835 x; all from Santa hybridization with ␤-actin cDNA probe was used. The probes were labeled Cruz Biotechnology, Santa Cruz, CA), and STAT3 (71-0900, Zymed, San with [␣-32P]dCTP (3000 Ci/mmol; Amersham) using a random primed Francisco, CA) for1honicebefore addition of the radiolabeled probe. DNA labeling kit (Roche, Indianapolis, IN). Hybridizations were per- Nondenaturing low ionic strength PAGE gels (6%) in 0.25 ϫ Tris-borate- formed in a solution containing 50% formamide, 5ϫ Denhardt’s solution, EDTA buffer were used. Gels were dried, and bands were visualized by 5ϫ SSPE, and 0.5% SDS at 42°C. Membranes were washed twice with 1ϫ autoradiography. SSC/0.1% SDS at 42°C for 30 min and once at 65°C for 30 min. The membranes were exposed to Kodak X-OMAT AR films (Eastman Kodak, Western blotting Rochester, NY) at Ϫ70°C with intensifying screens. SDS-PAGE was conducted by using the Laemmli buffer system (24) on Cytokine-specific ELISAs and biological assay for IFN-␣/␤ 10% polyacrylamide gels. Proteins separated on gels were transferred to Immobilon-P (polyvinylidene difluoride) membranes (Millipore, Bedford, TNF-␣ and IL-6 levels in cell culture supernatants were determined by MA) with an Isophor electrotransfer device (Hoefer Scientific Instruments, ELISA methods as described previously (19) with sensitivities of 20 pg/ml. San Francisco, CA) at 200 mA for 2 h. Binding of the primary and sec- The IFN-␣/␤ assay was performed as previously described (31) with a ondary Abs was performed in PBS (pH 7.4) containing 5% nonfat milk for detection limit of 3 IU/ml. Briefly, cell culture supernatants were harvested The Journal of Immunology 3735 Downloaded from http://www.jimmunol.org/

FIGURE 2. Activation of STAT DNA binding to IRF1-GAS element by lactobacilli and streptococci. Macrophages were stimulated with live lac- tobacilli (LAB) and streptococci (GAS) for 1, 3, 6, 9, or 24 h, and nuclear FIGURE 3. Activation of STAT DNA binding to the SIE element by extracts were prepared and analyzed by EMSA using IRF1-GAS oligonu- lactobacilli and streptococci. Macrophages were stimulated with live lac- cleotide. A, Kinetics of STAT activation by lactobacilli and streptococci. B, tobacilli (LAB) or streptococci (GAS) for 1, 3, 6, 9, or 24 h, and nuclear Supershift experiment with anti-STAT1, anti-STAT3, and anti-STAT5 extracts were prepared and analyzed by EMSA using SIE oligonucleotide. Abs. Results are representative of two separate experiments. A, Kinetics of STAT activation by lactobacilli and streptococci. B, Super-

shift experiment with anti-STAT1, anti-STAT3, and anti-STAT5. Results by guest on October 3, 2021 are representative of two separate experiments.

and dialyzed against acidic glycine buffer (pH 2) followed by two dialyses in PBS. IFN-␣ titers in samples were assayed by VSV plaque reduction in Hep2 cells. binding to IRF1-GAS was weakly detectable at 9 and 24 h after stimulation (Fig. 2A). Supershift experiments with anti-STAT1, Results anti-STAT3 and anti-STAT5 Abs showed that both bacteria acti- ␬ Activation of NF- B DNA binding in human macrophages by vated STAT1 binding (Fig. 2B). As analyzed using the SIE ele- Gram-positive bacteria ment, streptococci apparently activated the lowest of the three To study lactobacilli- and streptococci-induced activation of tran- STAT DNA binding complexes at 3 h after stimulation. Both lac- scription factors, differentiated macrophages were stimulated with tobacilli and streptococci activated the STAT DNA binding com- live bacteria. Nuclear extracts from stimulated macrophages were plex at 6 h after stimulation, with the intensity of the complexes prepared and analyzed by EMSA using NF-␬B oligonucleotide. increasing at least up to 24 h (Fig. 3A). The SIE binding complexes Both bacteria activated NF-␬B DNA binding rapidly at 1 h after activated by bacteria containing STAT1 and STAT3 homodimers stimulation, and this activation increased with time for up to 24 h (the lowest and the uppermost of the three bands, respectively) as (Fig. 1A). As shown by the supershift experiment, the NF-␬B DNA well as STAT1/3 heterodimers (the middle band). Also, STAT5 binding complex containing p50 and p65 was activated at 1 h after binding to the SIE element induced by lactobacilli was seen at 24 h bacterial stimulation and was most prominent at the 24 h point after stimulation (Fig. 3B). ISGF3 complex binding to the ISRE15 (Fig. 1B). element was detectable at 6 and 9 h after stimulation with strep- tococci, and the complex started to disappear thereafter (Fig. 4). Activation of STAT1, STAT3, and ISGF3 complex formation in human macrophages by Gram-positive bacteria Effect of protein synthesis inhibition on the activation of NF-␬B Both NF-␬B and STATs are involved in cytokine signaling during and STATs by Gram-positive bacteria immune responses. We therefore studied the possible activation of To analyze the requirement of ongoing protein synthesis for STATs induced by Gram-positive bacteria. STAT activation was NF-␬B and STAT activation, macrophages were stimulated with analyzed using nuclear extracts prepared from live bacteria-stim- live bacteria for 3 and6hinthepresence or the absence of CHX. ulated macrophages by EMSA with IRF1-GAS, SIE, and ISRE15 NF-␬B activation by lactobacilli or streptococci was not inhibited oligonucleotides. STAT binding to IRF1-GAS in response to strep- by CHX at either time point (Fig. 5A). Streptococci-induced tococci was detected at 3 h after stimulation, and the intensity of STAT1 DNA binding to IRF1-GAS or SIE elements was inhibited this complex increased up to 24 h. Lactobacilli-induced STAT by CHX (Fig. 5, B and C). Lactobacilli activated weak STAT1 and 3736 NF-␬B AND STAT ACTIVATION BY GRAM-POSITIVE BACTERIA

FIGURE 4. Activation of ISGF3 DNA binding by lactobacilli and strep- tococci. Macrophages were stimulated with live lactobacilli (LAB) or streptococci (GAS), and the kinetics of ISGF3 complex formation and binding to ISRE15 element at 1, 3, 6, 9 or 24 h after stimulation were analyzed by EMSA. Results are representative of two separate experiments. Downloaded from

STAT3 binding that was also inhibited by CHX (Fig. 5C). Strep- tococci-induced ISGF3 complex formation was similarly blocked by CHX (Fig. 5D).

Production of proinflammatory cytokines and IFN-␣ in human http://www.jimmunol.org/ macrophages stimulated with Gram-positive bacteria As CHX experiments showed, ongoing protein synthesis was re- quired for bacteria-induced STAT activation. This suggested that activation of STATs is likely to be cytokine mediated. The cyto- FIGURE 6. Kinetics of cytokine production in macrophages stimulated kines involved in STAT1 and STAT3 activation include macro- with lactobacilli and streptococci. Macrophages were stimulated with live phage-derived IFN-␣ and IL-6, while TNF-␣ is the most promi- lactobacilli (LAB) and streptococci (GAS) for 1, 3, 6, 9, or 24 h; cell nent activator of NF-␬B. To quantitate TNF-␣ and IL-6 production culture supernatants were collected; and the amounts of secreted TNF-␣, we used ELISA, whereas IFN-␣␤ levels were measured by a bi- IL-6, and IFN-␣ were measured. Results are representative of several sep- by guest on October 3, 2021 ological assay. Both lactobacilli and streptococci induced the pro- arate experiments.

FIGURE 5. Effect of CHX treatment on NF-␬B and STAT DNA binding activated by lactobacilli and streptococci. Macrophages were stimulated with live lactobacilli (LAB) or streptococci (GAS), and CHX (10 ␮g/ml) was added at 0.5 h after the beginning of bacterial stimulation. Nuclear extracts were prepared and analyzed by EMSA using NF-␬B(A), IRF1-GAS (B), SIE (C), and ISRE15 (D) oligonucleotides. Results are representative of four separate experiments. The Journal of Immunology 3737 Downloaded from http://www.jimmunol.org/

FIGURE 8. Expression of IRF-1 and MxA in lactobacilli- and strepto- cocci-stimulated macrophages. Macrophages were stimulated with live lac- tobacilli (LAB) and streptococci (GAS) for 2, 6, 12, or 24 h; total cellular RNA was collected; and expression of IRF-1 and MxA mRNAs was an-

alyzed by Northern blotting (A). Macrophages were stimulated in a 1:1 or by guest on October 3, 2021 1:10 ratio with live streptococci (GAS) for 6, 12, or 24 h, and the expres- sion of IRF-1 and MxA proteins was analyzed by Western blotting (B).

FIGURE 7. Effect of anti-IFN-␣ and anti-IL-6 Abs on activation of STAT1, STAT3, and ISGF3 DNA binding by lactobacilli and streptococci. Induction of IRF-1 and MxA mRNA and protein expression, and Macrophages were stimulated with live lactobacilli (LAB) or streptococci activation of IRF-1 DNA binding by Gram-positive bacteria ␣ ␣ (GAS) or with IFN- (20 IU/ml) and/or IL-6 (5 ng/ml). Abs to IFN- To characterize in more detail bacteria-induced IFN-␣ production ␮ (2400 IU/ml) and/or IL-6 (1 g/ml) were added at 1 and 4 h after the and IFN-␣-mediated target gene activation, we studied bacteria- beginning of bacterial stimulation. Nuclear extracts were prepared and an- induced IRF-1 and MxA mRNA and protein expression as well as alyzed by EMSA following binding to IRF1-GAS (A), SIE (B), and ISRE15 (C)oligonucleotides. Results are representative of three separate activation of IRF-1 DNA binding. Both lactobacilli and strepto- experiments. cocci induced IRF-1 mRNA at 2 h and MxA mRNA at 6 h after stimulation (Fig. 8A). Streptococci induced IRF-1 and MxA mRNAs more strongly than lactobacilli. Expression of IRF-1 and duction of TNF-␣ at 3 h and IL-6 at 6 h after stimulation. Bacteria- MxA proteins by streptococci was first detectable at 6 h after stim- induced TNF-␣ and IL-6 production increased up to 24 h after ulation, continuing up to 24 h (Fig. 8B). Increasing the streptococ- stimulation. Detectable amounts of IFN-␣ were found only after cal dose to 10-fold slightly potentiated the expression of IRF-1, but 24 h of stimulation with streptococci (Fig. 6). decreased that of MxA. Both lactobacilli and streptococci induced IRF-1 DNA binding to PRDI oligonucleotide at 3 h after stimu- Anti-IFN-␣␤ Abs inhibit Gram-positive bacteria-induced STAT lation. However, streptococci induced IRF-1 DNA binding more activation efficiently than lactobacilli at all time points studied (Fig. 9). To study the role of IFN-␣ and IL-6 in bacteria-induced STAT activation, we used neutralizing Abs. Macrophages were stimu- Discussion lated with live bacteria for8hinthepresence or the absence of Transcriptional regulation of the genes involved in inflammatory neutralizing Abs to IFN-␣␤ or IL-6. STAT1 binding to IRF1-GAS responses such as TNF-␣ and IL-6 is accomplished by NF-␬B. We by both bacteria was reduced by anti-IFN-␣␤ Abs, while anti-IL-6 have analyzed and compared the ability of live nonpathogenic Lac- Abs had no detectable effect on STAT1 or STAT3 DNA binding tobacillus rhamnosus GG and pathogenic Streptococcus pyogenes (Fig. 7A). Similarly, bacteria-induced SIE DNA binding was re- to induce NF-␬B activation in human primary macrophages. Lac- duced, while ISGF3 DNA binding activity was completely blocked tobacilli and streptococci similarly induced NF-␬B DNA binding by anti-IFN-␣␤ Abs (Fig. 7, B and C). with fast kinetics, and the intensity of the DNA binding complex 3738 NF-␬B AND STAT ACTIVATION BY GRAM-POSITIVE BACTERIA

esis plausible. The marked similarity in the ability of lactobacilli and streptococci regardless of their pathogenicity to induce NF-␬B activation could be explained by the similarity in the components and structure of the cell wall. Pepetidoglycan that is present in abundance in the cell walls of all Gram-positive bacteria and has been shown to induce NF-␬B activation through CD14 (37) and TLR2 (44, 45, 48) receptors is a potential component in the cell walls of lactobacilli and streptococci mediating NF-␬B activation. STATs are activated as a result of cytokine binding to their receptors and, like NF-␬B, are involved in activating immune re- sponses. Streptococci-induced STAT1 DNA binding was more pronounced compared with that induced by lactobacilli. In con- trast, STAT3 was more strongly activated by lactobacilli than by streptococci. Bacteria-induced activation of STAT1 and STAT3 appeared to be dependent on protein synthesis. This suggested that STAT activation by Gram-positive bacteria was cytokine medi- FIGURE 9. Activation of IRF-1 DNA binding by lactobacilli and strep- ated. The kinetics of STAT activation and IL-6 production induced tococci. Macrophages were stimulated with live lactobacilli (LAB) or by bacteria coincided, suggesting that IL-6 has a role in bacteria-

streptococci (GAS), and the kinetics of IFR-1 binding to PRDI oligonu- induced STAT activation. Interestingly, neutralization of IL-6 did Downloaded from cleotide at 1, 3, 6, 9, or 24 h after stimulation were analyzed by EMSA. not have any effect on bacteria-induced activation of these STATs. Results are representative of three separate experiments. Although the role of STAT1 in intracellular bacterial infection (listeria) in mice is well established (49), to our knowledge the increased with time. As shown by the supershift experiment, bac- only reported work on bacteria-induced STAT activation in human teria-induced NF-␬B consisted of at least of p50 and p65 proteins, cells concerns Listeria monocytogenes-induced IL-6-dependent which form the most common active dimer in human cells (8). Our activation of STAT3 in hepatocytes (50). It could be that neutral- http://www.jimmunol.org/ data are in line with previous studies where Streptococcus pneu- ization of IL-6 in our experimental setting was incomplete, and a moniae and its cell wall (32), Listeria monocytogenes and cell wall small amount of IL-6 was sufficient to exert an activating effect. lipoteichoic acid (33), and Lactobacillus crispatus (34) were Neutralization of IFN-␣ clearly reduced STAT activation. IFN-␣ shown to activate NF-␬B in several cell lines. Lactobacilli and thus appears to have a significant role in both STAT1 and STAT3 streptococci induced the production of TNF-␣ and IL-6 in macro- activation, even though we were unable to detect IFN-␣ production phages with similar kinetics. These cytokines were produced only in 8 h supernatants (data not shown). However, we were able to after the initial activation of NF-␬B DNA binding, and their pro- measure IFN-␣ in 24 h supernatants of streptococci-stimulated duction increased with time. The increase in the intensity of macrophages. by guest on October 3, 2021 NF-␬B DNA binding signal as a function of time could thus result IRF-1 is a transcriptional regulator participating in the up-reg- from the additive activating effect of TNF-␣ produced by macro- ulation of IFN-␣- and IFN-inducible genes by binding to their phages stimulated with bacteria. ISRE elements. The expression of IRF-1 is induced, e.g., by vi- The rapid activation of NF-␬B DNA binding without measur- ruses, IFNs, TNF-␣, and IL-6 (51). While acting as a positive able cytokine production suggested that lactobacilli and strepto- regulator of innate immune responses, IRF-1 also mediates Th1- ␬ ␬ cocci were able to directly activate NF- B in macrophages. NF- B type immune responses, thus being of importance in both viral and DNA binding was activated by both bacteria in the presence of bacterial infections (52, 53). Lactobacilli and streptococci induced protein synthesis inhibitor, CHX, supporting this observation. rapid expression of IRF-1 mRNA that coincided with TNF-␣ and ␬ Likewise, rapid NF- B activation after binding of pathogenic bac- IL-6 mRNA expression (data not shown). Based on our data it teria to their target cells has been observed with listeria, mycobac- remains open whether bacteria-induced TNF-␣ and IL-6 have a teria, Escherichia coli, and salmonella (33, 35, 36). The possible role in IRF-1 induction. Likewise, IFN-␣ production at early time direct activation of NF-␬B as a result of binding of lactobacilli and points after bacterial stimulation could not be detected. It is thus streptococci to macrophages raises the question of the nature of the possible that these bacteria directly induce the expression of IRF-1, receptor binding these bacteria and mediating the signal. CD14 as has been shown to be the case with LPS (54). Rapid induction receptor has been implicated in NF-␬B activation by Gram-posi- of IRF-1 production could also suggest the involvement of NF-␬B tive bacteria and their cell wall components (37). However, other in bacteria-induced IRF-1 activation. Since it is known that IRF-1 receptors have been suggested to mediate binding and signaling ␬ (38–40). Human Toll-like receptor 2 (TLR2) and TLR4 were first gene expression is regulated by both STAT1 and NF- B (55, 56), ␣ ␣ demonstrated to mediate signaling by LPS leading to NF-␬B ac- it is likely that IFN- , TNF- , and IL-6 produced at later time tivation, and this response was shown to be enhanced by CD14 points after bacterial stimulation may then contribute to activation (41–43). TLR2 and TLR4 were then shown to have a role in of IRF-1 gene expression. Both bacteria induced IRF-1 DNA bind- Gram-positive bacteria-mediated signaling, leading to NF-␬B ac- ing. However, streptococci activated IRF-1 DNA binding more tivation (44–46). Most recent data indicate that TLR2 is mainly efficiently than lactobacilli, which correlated to IFN-␣ production. involved in responses to Gram-positive bacteria, while TLR4 has Bacteria-induced IRF-1 activation was inhibited by CHX (data not a role in recognition and signaling in response to Gram-negative shown), suggesting that the activation of IRF-1 DNA binding is bacteria (47, 48). It is thus likely that in our experimental system not induced directly by these bacteria but, instead, could involve with human primary macrophages TLRs in addition to CD14 may bacteria-induced cytokines, IFN-␣, TNF-␣, or IL-6. play a role in mediating lactobacilli- and streptococci-induced Formation of ISGF3 complex is specifically induced by IFN-␣ NF-␬B activation. TLR2 and TLR4 mRNAs are expressed in hu- (57). Streptococci induced the formation of ISGF3 complex, which man primary macrophages (data not shown), making our hypoth- was inhibited by CHX and abolished with anti-IFN-␣ Ab, while The Journal of Immunology 3739

lactobacilli-induced ISGF3 complex formation was variably de- 7. Sareneva, T., S. Matikainen, M. Kurimoto, and I. Julkunen. 1998. Influenza A ␣ ␤ ␥ tectable. These results together with those regarding bacteria-in- virus-induced IFN- / and IL-18 synergistically enhance IFN- gene expression in human T cells. J. Immunol. 160:6032. duced STAT1 and STAT3 activation, strongly suggest that despite 8. Bauerle, P. A., and T. Henkel. 1994. Function and activation of NF-␬Binthe the lack of measurable IFN-␣ production, bacteria-induced IFN-␣ immune system. Annu. Rev. Immunol. 12:141. 9. Ghosh, S., M. J. May, and E. B. Kopp. 1998. NF-␬B and Rel proteins: evolu- is involved in STAT activation. In some experiments we have been tionarily conserved mediators of immune responses. Annu. Rev. Immunol. 16: able to measure the production of 3–10 IU/ml IFN-␣ induced by 225. lactobacilli; however, streptococci was always a better inducer. In 10. Darnell, J. E. 1997. STATs and gene regulation. Science 277:1630. 11. Leonard, W. J., and J. J. O’Shea. 1998. JAKs and STATs: biological implica- previous studies in mice, it has been shown that nonpathogenic tions. Annu. Rev. Immunol. 16:293. lactobacilli are able to induce the production of IFN-␣ (58, 59). It 12. Hoey, T., and M. J. Grusby. 1999. STATs as mediators of cytokine-induced is known that as little as 0.3 IU/ml of IFN-␣ has biological effects responses. Adv. Immunol. 71:145. 13. Decker, T., P. Kovarik, and A. Meinke. 1997. GAS elements: a few nucleotides in human PBMC (60). Thus, it is likely that a small amount of with a major impact on cytokine-induced gene expression. J. Interferon Cytokine IFN-␣ is produced and used up by the macrophages at early time Res. 17:121. points after stimulation with both bacteria. The observation that 14. O’Shea, J. J. 1997. Jaks and STATs, cytokine signal transduction, and immuno- regulation: are we there yet? Immunity 7:1. both bacteria induced the expression of MxA, that is strictly under 15. Lidbeck, A., and C. E. Nord. 1993. Lactobacilli and the normal human anaerobic regulation of IFN-␣ (61, 62), further supports the idea that, in microflora. Clin. Infect. Dis. 16:S181. 16. Tannock, G. W. 1997. Probiotic properties of lactic-acid bacteria: plenty of scope addition to streptococci, lactobacilli induce the production of for fundamentalR&D.Trends Biotechnol. 15:270. small, yet biologically significant, amounts of IFN-␣. 17. Saxelin, M. 1997. Lactobacillus GG: a human probiotic strain with thorough Analysis of the activation of transcription factors involved in clinical documentation. Food Rev. Int. 13:293. 18. Stevens, D. L. 1999. The flesh-eating bacterium: what’s next? J. Infect. Dis. immune responses sheds light on the molecular mechanisms of 179:S366. Downloaded from interaction between bacteria and immune cells, especially those 19. Miettinen, M., S. Matikainen, J. Vuopio-Varkila, J. Pirhonen, K. Varkila, between Gram-positive bacteria and human leukocytes, which M. Kurimoto, and I. Julkunen. 1998. Lactobacilli and streptococci induce inter- leukin-12 (IL-12), IL-18, and ␥ interferon production in human peripheral blood have been poorly characterized. This work elucidates factors in- mononuclear cells. Infect. Immun. 66:6058. volved in cytokine signaling and regulation of immune responses 20. Miettinen, M., K. Varkila, and J. Vuopio-Varkila. 1996. Production of human ␣ by nonpathogenic and pathogenic Gram-positive bacteria. We tumor necrosis factor , interleukin-6, and interleukin-10 is induced by lactic acid bacteria. Infect. Immun. 64:5403. show that lactobacilli and streptococci similarly induce the acti- 21. Holm, S. E., and E. Falsen. 1967. An antigen-free medium for cultivation of vation of transcription factor NF-␬B in primary macrophages. We ␤-hemolytic streptococci. Acta Pathol. Microbiol. Scand. 69:264. http://www.jimmunol.org/ 22. Mogensen, K. E., R. Pyha¨la¨, and K. Cantell. 1975. Raising antibodies to human are also the first to report that both nonpathogenic and pathogenic leukocyte interferon. Acta Pathol. Microbiol. Scand. 83:443. extracellular bacteria can induce the activation of STATs. Our re- 23. Andrews, N. C., and D. V. Faller. 1991. A rapid micropreparation technique for sults suggest that NF-␬B activation is direct, while STAT activa- extraction of DNA-binding proteins from limited numbers of mammalian cells. ␣ Nucleic Acids Res. 29:2499. tion is mediated mainly by IFN- . Nonpathogenic and pathogenic 24. Laemmli, U. K. 1970. Cleavage of structural proteins during the assembly of the bacteria, or at least the studied strains, differ with regard to STAT head of bacteriophage T4. Nature 227:680. activation. Pathogenic streptococci potently induce IFN-␣ produc- 25. Matikainen, S., T. Ronni, M. Hurme, R. Pine, and I. Julkunen. 1996. Retinoic acid activates interferon regulatory factor-1 gene expression in myeloid cells. tion, which is likely to contribute to the activation of STATs and Blood 88:114. IRF-1. It can be speculated that the efficient induction of IFN-␣ 26. Melen, K., P. Keskinen, T. Ronni, T. Sareneva, K. Lounatmaa, and I. Julkunen. by guest on October 3, 2021 production and transcription factor activation by streptococci 1996. Human MxA protein, an interferon-␣-inducible GTPase, contains a nuclear targeting signal and is localized in the heterochromatin region beneath the nuclear would lead to fast and effective immune responses and could play envelope. J. Biol. Chem. 38:23478. a role in the pathogenesis. The factors in lactobacilli and strepto- 27. Chirgwin, J. M., A. E. Przybyla, R. J. MacDonald, and W. J. Rutter. 1979. Isolation of biologically active ribonucleic acid from sources enriched in ribo- cocci mediating different interactions with macrophages remain nuclease. Biochemistry 18:5294. uncharacterized; however, it could be that streptococci have sur- 28. Glisin, V., R. Crkvenjakov, and C. Buys. 1974. Ribonucleic acid isolated by face components, such as lipoteichoic acids or proteins, that more cesium chloride centrifugation. Biochemistry 13:2633. 29. Maruyama, M., T. Fujita, and T. Taniguchi. 1989. Sequence of a cDNA coding avidly interact with macrophages. We have previously shown that for human IRF-1. Nucleic Acids Res. 17:3292. S. pyogenes is a better inducer of Th1 type cytokines IL-12 and 30. Ronni, T., T. Sareneva, J. Pirhonen, and I. Julkunen. 1995. Activation of IFN-␣, IFN-␥ than L. rhamnosus GG (19). The ability of streptococci to IFN-␥, MxA, and IFN regulatory factor 1 genes in influenza A virus-infected ␣ human peripheral blood mononuclear cells. J. Immunol. 154:2764. induce IFN- production better than lactobacilli could further con- 31. Cantell, K., S. Hirvonen, H.-L. Kauppinen, and N. Kalkkinen. 1991. Rapid pro- tribute to its ability to direct immune responses toward the duction of interferon-␥ in uninduced human leukocyte suspensions. J. Interferon Th1 type. Res. 11:231. 32. Spellerberg, B., C. Rosenow, W. Sha, and E. I. Tuomanen. 1996. Pneumococcal cell wall activates NF-␬B in human monocytes: aspects distinct from endotoxin. Microb. Pathog. 20:309. Acknowledgments 33. Hauf, N., W. Goebel, F. Fiedler, Z. Sokolovic, and M. Kuhn. 1997. Listeria ␬ We thank Dr. Tapani Hovi for critically reading the manuscript. We also monocytogenes infection of P388D1 macrophages results in a biphasic NF- B (RelA/p50) activation induced by lipoteichoic acid and bacterial phospholipases thank Katja Moilanen, Valma Ma¨kinen, Teija Westerlund, and Marika and mediated by I␬B␣ and I␬B␤ degradation. Proc. Natl. Acad. Sci. USA 94: Ylisela¨for their excellent technical assistance. 9394. 34. Klebanoff, S. J., D. H. Watts, C. Mehlin, and C. M. Headley. 1999. Lactobacilli and vaginal host defense: activation of the human immunodeficiency virus type References 1 long terminal repeat, cytokine production, and NF-␬B. J. Infect. Dis. 179:653. 35. Giri, D. K., R. T. Mehta, R. G. Kansal, and B. B. Aggarwal. 1998. Mycobacte- 1. Morrissette, N., E. Gold, and A. Alderem. 1999. The macrophage: a cell for all rium avium-intracellulare complex activates nuclear transcription factor-␬Bin seasons. Trends Cell Biol. 9:199. different cell types through reactive oxygen intermediates. J. Immunol. 161:4834. 2. Tracey, K. J., and A. Cerami. 1993. Tumor necrosis factor, other cytokines and 36. Eaves-Pyles, T., C. Szabo, and A. L. Salzman. 1999. Bacterial invasion is not disease. Annu. Rev. Cell Biol. 9:317. required for activation of NF-␬B in enterocytes. Infect. Immun. 67:800. 3. Taga, T., and T. Kishimoto. 1997. Gp130 and the interleukin-6 family of cyto- 37. Gupta, D., T. N. Kirkland, S. Viriyakosol, and R. Dziarski. 1996. CD14 is a kines. Annu. Rev. Immunol. 15:797. cell-activating receptor for bacterial peptidoglycan. J. Biol. Chem. 271:23310. 4. Belardelli, F. 1995. Role of interferons and other cytokines in the regulation of 38. Cauwels, A., E. Wan, M. Leishman, and E. Tuomanen. 1997. Coexistence of the immune response. APMIS 103:161. CD14-dependent and independent pathways for stimulation of human monocytes 5. Herberman, R. B. 1997. Effect of ␣-interferons on immune function. Semin. On- by Gram-positive bacteria. Infect. Immun. 65:3255. col. 24(Suppl. 9):78. 39. Medvedev, A. E., T. Flo, R. R. Ingalls, D. T. Golenbock, G. Teti, S. N. Vogel, 6. Rogge, L., L. Barberis-Maino, M. Biffi, N. Passini, D. H. Presky, U. Gubler, and and T. Espevik. 1998. Involvement of CD14 and complement receptors CR3 and F. Sinigaglia. 1997. Selective expression of an interleukin-12 receptor component CR4 in nuclear factor-␬B activation and TNF production induced by lipopoly- by human T helper 1 cells. J. Exp. Med. 185:825. saccharide and group B streptococcal cell walls. J. Immunol. 160:4535. 3740 NF-␬B AND STAT ACTIVATION BY GRAM-POSITIVE BACTERIA

40. Haziot, A., N. Hijiya, K. Schultz, F. Zhang, S. C. Gangloff, and S. M. Goyert. 52. Lohoff, M., D. Ferrick, H.-W. Mittrucker, G. S. Duncan, S. Bischof, M. Ro¨ll- 1999. CD14 plays no major role in shock induced by Staphylococcus aureus but inghoff, and T. W. Mak. 1997. Interferon regulatory factor-1 is required for a T down-regulates TNF-␣ production. J. Immunol. 162:4801. helper 1 immune response in vivo. Immunity 6:681. 41. Kirschning, C. J., H. Wesche, T. M. Ayres, and M. Rothe. 1998. Human Toll-like 53. Taki, S., T. Sato, K. Ogasawara, T. Fukuda, M. Sato, S. Hida, G. Suzuki, receptor 2 confers responsiveness to bacterial lipopolysaccharide. J. Exp. Med. M. Mitsuyama, E.-H. Shin, S. Kojima, et al. 1997. Multistage regulation of Th1- 188:2091. type immune responses by the transcription factor IRF-1. Immunity 6:673. 42. Yang, R.-B., M. R. Mark, A. Gray, A. Huang, M. H. Xie, M. Zhang, A. Goddard, 54. Barber, S. A., M. J. Fultz, C. A. Salkowski, and S. N. Vogel. 1995. Differential W. I. Wood, A. L. Gurney, and P. J. Godowski. 1998. Toll-like receptor-2 me- expression of interferon regulatory factor 1 (IRF-1), IRF-2, and interferon con- diates lipopolysaccharide-induced cellular signalling. Nature 395:284. sensus sequence binding protein genes in lipopolysaccharide (LPS)-responsive 43. Chow, J. C., D. W. Young, D. T. Golenbock, W. J. Christ, and F. Gusovsky. and LPS-hyporesponsive macrophages. Infect. Immun. 63:601. 1999. Toll-like receptor-4 mediated lipopolysaccharide-induced signal transduc- 55. Ohmori, Y., R. D. Schreiber, and T. A. Hamilton. 1997. Synergy between inter- tion. J. Biol. Chem. 274:10689. feron-␥ and tumor necrosis factor-␣ in transcriptional activation is mediated by 44. Schwandner, R., R. Dziarski, H. Wesche, M. Rothe, and C. J. Kirscning. 1999. cooperation between signal transducer and activator of transcription 1 and nuclear Peptidoglycan- and lipoteichoic acid-induced cell activation is mediated by Toll- factor ␬B. J. Biol. Chem. 272:14899. like receptor 2. J. Biol. Chem. 274:17406. 56. Pine, R. 1997. Convergence of TNF␣ and IFN␥ signalling pathways through 45. Yoshimura, A., E. Lien, R. R. Ingalls, E. Tuomanen, R. Dziarski, and synergistic induction of IRF-1/ISGF-2 is mediated by a composite GAS/␬B pro- D. Golenbock. 1999. Recognition of Gram-positive bacterial cell wall compo- moter element. Nucleic Acids Res. 25:4346. nents by the innate immune system occurs via Toll-like receptor 2. J. Immunol. 57. Schindler, C., X.-Y. Fu, T. Improta, R. Aebersold, and J. E. Darnell, Jr. 1992. 163:1. Proteins of transcription factor ISGF-3: one gene encodes the 91- and 84-kDa 46. Means, T. K., S. Wang, E. Lien, A. Yoshimura, D. T. Golenbock, and ISGF proteins that are activated by interferon ␣. Proc. Natl. Acad. Sci. USA M. J. Fenton. 1999. Human Toll-like receptors mediate cellular activation by 89:7836. Mycobacterium tuberculosis. J. Immunol. 163:3920. 58. Solis Pereyra, B., R. Falcoff, E. Falcoff, and D. Lemonnier. 1991. Interferon 47. Underhill, D. M., A. Ozinsky, A. M. Hajjar, A. Stevens, C. B. Wilson, induction by Lactobacillus bulgaricus and Streptococcus thermophilus in mice. M. Bassetti, and A. Aderem. 1999. The Toll-like receptor 2 is recruited to mac- Eur. Cytokine Network 2:229. rophage phagosomes and discriminates between pathogens. Nature 401:811. 59. Kitazawa, H., Y. Tomioka, K. Matsumura, H. Aso, M. Mizugaki, T. Itoh, and 48. Takeuchi, O., K. Hoshino, T. Kawai, H. Sanjo, H. Takada, T. Ogawa, K. Takeda, T. Yamaguchi. 1994. Expression of mRNA encoding IFN␣ in macrophages stim- and S. Akira. 1999. Differential roles of TLR2 and TLR4 in recognition of Gram- ulated with Lactobacillus gasseri. FEMS Microbiol. Lett. 120:315. Downloaded from negative and Gram-positive bacterial cell wall components. Immunity 11:443. 60. Ronni, T., K. Melen, A. Malygin, and I. Julkunen. 1993. Control of IFN-inducible 49. Meraz, M. A., J. M. White, K. C. F. Sheehan, E. A. Bach, S. J. Rodig, MxA gene expression in human cells. J. Immunol. 150:1715. A. S. Dighe, D. H. Kaplan, J. K. Riley, A. C. Greenlund, D. Campbell, et al. 1996. 61. Goetschy, J.-F., H. Zeller, J. Content, and M. A. Horisberger. 1989. Regulation Targeted disruption of the Stat1 gene in mice reveals unexpected physiologic of the interferon-inducible IFI-78K gene, the human equivalent of the murine Mx specificity in the JAK-STAT signaling pathway. Cell 84:431. gene, by interferons, double-stranded RNA, certain cytokines, and viruses. J. Vi- 50. Gregory, S. H., E. J. Wing, K. L. Danowski, N. Van Rooijen, K. F. Dyer, and rol. 63:2616. D. J. Tweardy. 1998. IL-6 produced by Kupffer cells induces STAT protein 62. Ronni, T., S. Matikainen, A. Lehtonen, J. Palvimo, J. Dellis, F. Van Eylen, J.-F.

activation in hepatocytes early during the course of systemic listerial infections. Goetschy, M. Horisberger, J. Content, and I. Julkunen. 1998. The proximal in- http://www.jimmunol.org/ J. Immunol. 160:6056. terferon-stimulated response elements are essential for interferon responsiveness: 51. Nguyen, H., J. Hiscott, and P. M. Pitha. 1997. The growing family of interferon a promoter analysis of the antiviral MxA gene. J. Interferon Cytokine Res. regulatory factors. Cytokine Growth Factor Rev. 8:293. 18:773. by guest on October 3, 2021