The Bacterial Fermentation Product Butyrate Influences Epithelial Signaling via Reactive Oxygen Species-Mediated Changes in Cullin-1 Neddylation This information is current as of September 24, 2021. Amrita Kumar, Huixia Wu, Lauren S. Collier-Hyams, Young-Man Kwon, Jason M. Hanson and Andrew S. Neish J Immunol 2009; 182:538-546; ; doi: 10.4049/jimmunol.182.1.538 http://www.jimmunol.org/content/182/1/538 Downloaded from

References This article cites 75 articles, 30 of which you can access for free at:

http://www.jimmunol.org/content/182/1/538.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 by guest on September 24, 2021

*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 © 2009 by The American Association of Immunologists, Inc. All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

The Bacterial Fermentation Product Butyrate Influences Epithelial Signaling via Reactive Oxygen Species-Mediated Changes in Cullin-1 Neddylation1

Amrita Kumar,* Huixia Wu,* Lauren S. Collier-Hyams,* Young-Man Kwon,* Jason M. Hanson,† and Andrew S. Neish2*

The human enteric flora plays a significant role in intestinal health and disease. Populations of enteric bacteria can inhibit the NF-␬B pathway by blockade of I␬B-␣ ubiquitination, a process catalyzed by the E3-SCF␤-TrCP . The activity of this is regulated via covalent modification of the Cullin-1 subunit by the ubiquitin-like NEDD8. We previously reported that interaction of viable commensal bacteria with mammalian intestinal epithelial cells resulted in a rapid and reversible generation of reactive oxygen species (ROS) that modulated neddylation of Cullin-1 and Downloaded from resulted in suppressive effects on the NF-␬B pathway. Herein, we demonstrate that butyrate and other short chain fatty acids supplemented to model human intestinal epithelia in vitro and human tissue ex vivo results in loss of neddylated Cul-1 and show that physiological concentrations of butyrate modulate the ubiquitination and degradation of a target of the E3- SCF␤-TrCP ubiquitin ligase, the NF-␬B inhibitor I␬B-␣. Mechanistically, we show that physiological concentrations of bu- tyrate induces reactive oxygen species that transiently alters the intracellular redox balance and results in inactivation of the

NEDD8-conjugating Ubc12 in a manner similar to effects mediated by viable bacteria. Because the normal flora http://www.jimmunol.org/ produces significant amounts of butyrate and other short chain fatty acids, these data provide a functional link between a natural product of the intestinal normal flora and important epithelial inflammatory and proliferative signaling pathways. The Journal of Immunology, 2009, 182: 538–546.

irtually unique among mammalian cell types, the intes- ucts such as lactate (4, 5). These organic compounds are an im- tinal epithelium coexists in intimate contact with a nor- portant energy source for the colonic epithelium and may influence V mal flora of ϳ1012 prokaryotic organisms. The greatest various aspects of gut physiology. For example, butyrate and other numbers and diversity are in the cecum and ascending colon where SCFAs have well-known differentiating and growth-promoting ac- bacterial density can reach 1011 cells/g of contents, yielding a bio- tivities in vitro and in vivo, a biological effect ascribed to histone by guest on September 24, 2021 mass of Ͼ1 kg. This bacterial community of ϳ500–1000 species deacetylase activity (6). Additionally, SCFAs have been noted to has diverse beneficial roles including vitamin synthesis, bile salt have immunomodulatory effects on colonic inflammation, sup- metabolism, and degradation of complex carbohydrates (1). Fur- pressing inflammatory cytokine secretion in cultured epithelial thermore, through studies with germfree or gnotobiotic mice, it is cells, and ameliorating model colitis in mice, suggesting that these also now established that the enteric flora can fundamentally affect molecules contribute to the ability of the mucosa to tolerate the epithelial gene transcription, ultimately affecting enterocyte pro- presence of vast quantities of living microorganisms and associ- liferation and homeostasis (2, 3). ated microbial-associated molecular patterns (MAMPs) (6, 7). Fur- The normal flora thrives in a largely anaerobic environment, thermore, luminal instillation of butyrate has been shown to be a generating energy by the fermentation of luminal complex carbo- promising experimental therapy in ulcerative colitis and related hydrates. The end products of fermentation are a spectrum of or- inflammatory disorders (8, 9). How these naturally occurring bac- ganic acids, including short chain fatty acids (SCFAs)3 such as terial metabolic products influence epithelial homeostasis are un- butyrate, succinate, and propionate, as well as other terminal prod- known and are a topic of considerable research interest. A key pathway that modulates inflammatory and/or growth sur- vival in the intestine is the NF-␬B or Rel signaling pathway. The *Department of Pathology and Laboratory Medicine, Epithelial Pathobiology Unit, Emory University School of Medicine, Atlanta, GA 30322; and †Department of Pe- NF-␬B pathway is controlled by regulated degradation of a phys- diatrics, Division of Pulmonary, Asthma, Cystic Fibrosis and Sleep, Emory Univer- ically associated inhibitor molecule, I␬B-␣, which when phosphor- sity, School of Medicine, Atlanta, GA 30322 ylated is ubiquitinated by a specific ubiquitin ligase complex des- Received for publication May 27, 2008. Accepted for publication October 9, 2008. ignated E3-SCF␤-TrCP. Ubiquitinated I␬B-␣ is targeted for The costs of publication of this article were defrayed in part by the payment of page degradation by the (10, 11). E3-SCF␤-TrCP and other charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. E3-SCF complexes are themselves regulated by transient covalent 1 This work was supported in part by National Institutes of Health Grants DK-71604 modifications. The ubiquitin paralog NEDD8 must be conjugated and AI-64462 (to A.S.N.). to the Cullin-1 (Cul-1) subunit of the E3-SCF complex for ubiq- 2 Address correspondence and reprint requests to Dr. Andrew S. Neish, Department uitin ligase activity (12–15). NEDD8 modification of Cul-1 has of Pathology, Emory University School of Medicine, Room 105F, Whitehead Build- ing, 615 Michael Street, Atlanta, GA, 30322. E-mail address: [email protected] 3 Abbreviations used in this paper: SCFA, short chain fatty acid; Cul-1, cullin-1; tiplicity of infection; IAA, iodoacetamide; HMM, high molecular mass; Eh, redox MAMP, microbial-associated molecular pattern; ROS, reactive oxygen species; NAC, potential; Trx, Thioredoxin. N-acetyl cysteine; DPI, diphenyleneiodonium; DCF, 5-(and-6)-chloromethyl-2Ј,7Ј- dichlorodihydrofluorescein diacetate acetyl ester; DHE, dihydroethidium; MOI, mul- Copyright © 2008 by The American Association of Immunologists, Inc. 0022-1767/08/$2.00 www.jimmunol.org The Journal of Immunology 539 been demonstrated to be necessary for ubiquitination of I␬B-␣ and (Sigma-Aldrich) using the enhanced chemiluminescent protocol (ECL; p100/p105, and for the subsequent activation of NF-␬B in mam- Amersham) and a HRP-conjugated secondary Ab. Blots were exposed to malian cells (16–20). filmfor1sto30min. We have demonstrated that viable commensal bacteria can block I␬B-␣ ubiquitination by blockade of neddylation of the Immunofluorescence studies Cul-1 subunit of E3-SCF␤-TrCP, accounting for the attenuation of ␬ Immunofluorescent labeling of p65 in adherent HeLa cells grown on NF- B activation (21). We further reported that loss of Cul-1 ned- 12-mm glass coverslips was performed as follows: cells were fixed for 20 dylation was mediated by rapid and reversible oxidative inactiva- min in 3.7% paraformaldehyde in PBS, washed in PBS, permeabilized with tion of Ubc12, the ubiquitin-like conjugating enzyme responsible 0.1% Triton X-100 in PBS for 5 min, and washed again. Fixed samples for the neddylation of Cullin subunits (22). Herein, we demon- were incubated in blocking solution (5% normal goat serum in PBS) over- strate that butyrate and other SCFAs supplemented to model epi- night at 4°C. A 1-h incubation with each Ab diluted in blocking buffer followed: 1/500 rabbit anti-p65 (Rockland); 1/200 fluorescein (FITC)-con- thelia in vitro and human tissue ex vivo also cause loss of neddy- jugated goat anti-rabbit IgG (Jackson ImmunoResearch Laboratories). lated Cul-1 resulting in consequent inhibition of the NF-␬B Cells were washed three times between each Ab. The coverslips were pathway. We also show that de novo and transient reactive oxygen mounted on glass slides and stained cells were observed by laser confocal species (ROS) generation mediate these regulatory effects of bu- epifluorescence microscopy (Zeiss). tyrate. These results suggest that metabolic products from the nor- mal intestinal flora (and potentially other complex microbial com- Monitoring ROS generation munities) can influence mammalian signaling pathways by For monitoring ROS generation, IEC-6 cell cultures were plated onto 24- modulating the activation of a rate-limiting enzymatic step. Downloaded from well plates and treated with butyrate for the duration indicated. Following treatment, the cells were further incubated with 5 ␮M each of the fluores- Materials and Methods cent ROS indicators 5-(and-6)-chloromethyl-2Ј,7Ј-dichlorodihydrofluores- Reagents cein diacetate acetyl ester (DCF) (Molecular Probes) or dihydroethidium (DHE) for 15 min. After loading the ROS dyes, cultures were washed three Butyric, lactic, propionic, and succinic acid were obtained from Sigma- times with warm HBSSϩ and the fluorescence intensity was acquired us- Aldrich. MG-262 was obtained from Affinity. TNF-␣ was from R&D ing a confocal laser-scanning microscope (Zeiss LSM 510). For DCF im-

Systems. Flagellin was purified as described previously (23). N-acetyl aging, images were captured with a 10ϫ objective using an excitation http://www.jimmunol.org/ cysteine (NAC) and diphenyleneiodonium (DPI) were purchased from wavelength of 488 nm and an emission wavelength of 513 nm. For DHE, Sigma-Aldrich. images were captured with a 40ϫ objective using an excitation wavelength of 490 nm and an emission wavelength of 610 nm. For determination of Cell culture ROS generation in mitochondria, we used the mitochondrial superoxide Confluent monolayers of T84 cells were grown in a polarized fashion on dye indicator MitoSOX Red (Molecular Probes). Cells were treated with permeable supports as previously described (24). HeLa, Caco-2, and IEC-6 butyrate for the times indicated, after which they were loaded with 5 ␮M MitoSOX Red in HBSSϩ for 10 min at 37°C. Cells were washed with epithelial cells were grown in standard tissue culture vessels and main- ϩ tained as recommended by the American Type Culture Collection. Human warm HBSS and fluorescent images were captured with a 63ϫ objective promyelocytic leukemia cells (HL-60) were maintained in RPMI 1640 me- using an excitation wavelength of 510 nm and an emission wavelength of 580 nm. dium supplemented with 10% (v/v) FBS and 100 U/ml penicillin and 100 by guest on September 24, 2021 U/ml streptomycin. The cultures were maintained in humidified 95% air, ϫ 5 5% CO2, at 37°C by passage of 2 10 cells/ml every other day. Differ- entiation was induced by treatment with PMA (20 ng/ml) for 48 h and the Kinetic measurements of ROS extent of differentiation was measured by attachment of cells to the ROS generation was measured with DCF as described by Wang et al. (25), substratum. with the following modification. Caco-2 cells grown to confluence on a Human biopsy material 24-well plate were incubated in DMEM supplemented with 0.4% serum and containing 1 mM DCF dye. After incubation for2hin5%CO2 at Healthy human colonic mucosa was obtained from the outer borders of 37°C, the cells were washed and treated with Lactobacillus rhamnosus GG fresh distal colonic specimens resected for neoplastic disease. Mucosa was (multiplicity of infection (MOI) ϭ 1) or with butyrate (10 mM) in Krebs- dissected free from the underlying gut wall and 3-mm punches were in- Ringer-HEPES buffer. An increase in fluorescence units was measured by cubated for1hinHBSSϩ supplemented with butyrate at the indicated using a fluorescence microplate reader (SpectraMax M2; Molecular De- concentration. Tissue was immediately lysed in denaturing SDS-PAGE vices) after various time intervals with excitation at 475 nm and emission buffer. at 525 nm. Transient transfections and luciferase assays HeLa cells were transiently transfected using Lipofectamine 2000 (Invitro- Examination of thioredoxin1 and thioredoxin 2 redox states gen) or FuGENE 6 (Roche) according to the manufacturer’s instructions. For Thioredoxin (Trx) 1 (BD Biosciences) analysis, Caco-2 cells treated Briefly, plasmid DNA was mixed with serum-free DMEM and transfection with butyrate were washed with cold 1ϫ PBS and whole cell lysates were reagent. After 15 min, the mixture was added in a dropwise fashion to the ϳ prepared in G-lysis buffer (6 M guanidine-HCl, 50 mM Tris (pH 8.3), 3 growth medium of 70–90% confluent adherent cells in 24- or 6-well mM EDTA, 50 mM iodoacetamide (IAA)) as described in (26). Briefly, plates or 100-mm dishes. The cells were allowed 16–24 h for transfection lysates were incubated for 30 min at 37°C, after which excess IAA was and expression of transfected genes. For luciferase reporter assays, cells removed using a G-25 spin column (GE Healthcare). The samples were were transfected with pNF-␬B-Luc plasmid (Stratagene), allowed 18 h for ␣ electrophoresed on a native gel, transferred to nitrocellulose membrane, expression, treated with TNF- (20 ng/ml) for 6 h, and lysed according to probed with mouse anti-human Trx1 (BD Biosciences), and detected with the manufacturer’s protocol. All transfections were balanced to contain 200 an Alexa Fluor 680 anti-mouse secondary Ab. Trx2 redox states were ng of DNA with empty vector. Luciferase activity was determined using analyzed as described in a previous report (26). Following treatment with the Dual Luciferase Reporter Assay System (Promega). butyrate, total cell protein was precipitated with 10% trichloroacetic acid. Western blot analysis Samples were centrifuged, washed with acetone, and resuspended in 20 mM Tris (pH 8.0) buffer containing 15 mM 4-acetamido-4Ј-maleimidyl- Following experimental treatment, epithelial cells were washed in cold stilbene-2,2Ј-disulfonic acid Molecular Probes)]. Samples were incubated HBSSϩ and whole cell extracts were prepared by rapid lysis in denaturing for 30 min at room temperature. Oxidized and reduced forms of Trx2 were SDS-PAGE buffer. Cell lysates were electrophoresed on SDS-polyacryl- separated by nonreducing SDS-PAGE. Trx2 was detected using rabbit anti- amide gels and transferred to nitrocellulose using standard protocols. Im- human Trx2 Ab and Alexa Fluor 680 anti-rabbit secondary Ab. Bands munoreactive were detected with Abs to I␬B-␣ (Santa Cruz Bio- corresponding to Trx1 and Trx2 were visualized using an Odyssey scanner technology), phospho-I␬B-␣ (Cell Signaling), Cul-1 (Zymed), NEDD8 (LI-COR) and band densometric values were applied to the Nernst equa- (Zymed), Ubc12 (Rockland), ␤-tubulin (Sigma-Aldrich), or ␤-actin tion as described elsewhere (27). 540 BUTYRATE EFFECTS ON EPITHELIAL SIGNALING

A D Butyrate Butyrate 014 + (5 mM) (10 mM) HBSS [mM ] NEDD8-Cul-1

15’ 30’ 60’ 90’ 15’ 30’ 60’ 90’ 15’ 30’ 60’ 90’ Butyrate NEDD8-Cul-1 90 kDa Cul-1 Cul-1 85 kDa NEDD8-Cul-1 Lactate IB: Cul-1 Cul-1 NEDD8-Cul-1 Cul-1 Propionate NEDD8-Cul-1 90 kDa NEDD8-Cul-1 Succinate IB: NEDD8 Cul-1

7.5 IB: Cul-1 4.0 pH B Butyrate E HBSS+ (5 mM) 15’ 30’ 60’ 15’ 30’ 60’ NEDD8-Cul-1 [Butyrate (mM)] 90 kDa 0 0.61 1.4 1.5 2.7 3.8 5.5 14 Cul-1 85 kDa NEDD8-Cul-1 IB: Cul-1 Cul-1 7.1 6.6 6.3 5.9 5.6 5.0 4.8 4.4 pH Butyrate NEDD8-Cul-1 HBSS+ (5 mM) Cul-1 Buffered to pH 7.4 with NaOH 15’ 30’ 60’ 15’ 30’ 60’

NEDD8-Cul-1 90 kDa NEDD8-Cul-1 [HBSS+] Downloaded from Cul-1 85 kDa Cul-1 7.1 6.6 6.3 5.9 5.6 5.0 4.8 4.4 pH IB: Cul-1

C F WASH Butyrate (2.7 mM) 1.5 3.0 [Butyrate (mM)] NEDD8-Cul-1 control 90 kDa Cul-1 85 kDa http://www.jimmunol.org/ HBSS+ 30’ 60’ 60’ 30’ 60’ NEDD8-Cul-1 90 kDa IB: Cul-1 Cul-1 85 kDa IB: Cul-1 FIGURE 1. Bacterial fermentation products at physiological pH and concentration cause loss of Cul-1 neddylation. A, Immunoblot of Cul-1 and NEDD8 from T84 model epithelia treated for up to 90 min with physiological concentrations of butyrate. B, Immunoblot of Cul-1 from HeLa cell lines (top panel) and HL-60 differentiated into macrophage-like cells treated for up to 60 min with physiological concentrations of butyrate. C, Immunoblot of Cul-1 from T84 model epithelia treated for up to 1 h with an effective dose of butyrate, washed, and incubated for 30 min and 1 h longer in HBSSϩ. D, Immunoblot of Cul-1 from T84 epithelial cells treated with HBSSϩ supplemented with butyrate and other organic acids as indicated for 1 h. Decreasing acid

concentration is marked (open wedge; 0–14 mM) and has an inverse relationship to pH (solid wedge; pH 4.0–7.5). E, Immunoblot of Cul-1 from T84 model by guest on September 24, 2021 epithelia treated with HBSSϩ or HBSSϩ supplemented with butyrate for1hattheindicated concentrations. The effective concentration of HBSSϩ or HBSSϩ supplemented with butyrate was buffered to the pH as indicated below each panel. F, Immunoblot of Cul-1 from 3-mm human distal colonic mucosal punches treated for1hinHBSSϩ supplemented with butyrate.

Results loss in Cul-1 neddylation was observed within 15 min (Fig. 1A). Butyrate mediates loss of Cul-1 neddylation Typically Cul-1 appears as a doublet of 85 and 90 kDa, with the higher molecular band representing the neddylated Cul-1. When an We had previously shown that colonization of epithelial cells with identical blot was probed with anti-NEDD8 Ab, the upper band viable commensal bacterial strains in vitro and in vivo caused a reacted with the Ab, confirming the identity of this band as ned- rapid and reversible loss of epithelial Cul-1 neddylation (21). Dur- dylated Cul-1 (Fig. 1A, bottom panel). Similar kinetics of butyrate- ing the course of these experiments, we observed a reduction of mediated Cul-1 deneddylation was also observed in HeLa cells Cul-1 neddylation if bacteria were physically separated from the epithelial cells by use of Transwell inserts mounted with a perme- (Fig. 1B, top panel) and in human promyelocytic leukemia cells able membrane support (data not shown). This indicated that the (HL-60) differentiated to form macrophage-like cells (Fig. 1B, bot- bacterial effects, at least in part, were mediated by a small diffus- tom panel). Interestingly, deneddylation of Cul-1 observed upon ible molecule. To identify this soluble signal, we analyzed bacte- treatment with butyrate could be reversed if the butyrate was ϩ rially conditioned supernatants by gas chromatography/mass spec- washed off and the cells allowed to recover in HBSS for another troscopy, among other approaches. These studies revealed the 30–60 min (Fig. 1C). The kinetics of reversibility of Cul-1 ned- presence of significant amounts of lactate in culture supernatants. dylation was similar as we observed with viable bacteria (21). Consistently, the pH of supernatants after a minimum of 30 min of Furthermore, loss in Cul-1 neddylation within 1 h was also ob- coculture was 4.5–6.0, indicating the microbial production of served when T84 cells were cultured in HBSSϩ supplemented acidic compounds. It is well known that the fermentation products with at least 3 mM concentrations of other organic acids, such as of the intestinal flora include lactate, as well as SCFAs such as lactate, propionate, and succinate (Fig. 1D). In all of these in vitro butyrate (C4), succinate (C4), and propionate (C3) (4, 5). Further- conditions, the molarity of the organic acids was within the range more, butyrate mediates well-known effects on inflammation, in- reported for the physiological gut (8, 30). cluding repression of NF-␬B in vivo and in vitro (8, 9, 28, 29). In T84 cells, butyrate at a concentration of ϳ3–5 mM (resulting These observations suggested the possibility that the loss of ned- inapHofϳ4.5–5.0) was effective in causing loss of Cul-1 ned- dylated Cul-1, at least partially, resulted from the exposure of the dylation (Fig. 1E, top panel). Interestingly, when effective concen- cells to bacterial fermentation products. When fully polarized T84 trations of butyrate were buffered to neutral pH, the ability of this epithelial cells were treated with 5 mM concentrations of butyrate, agent to affect Cul-1 neddylation was lost (Fig. 1E, middle panel). The Journal of Immunology 541 Downloaded from http://www.jimmunol.org/

FIGURE 2. Loss of Cul-1 neddylation results in inhibition of TNF-␣-induced p65 nuclear translocation and I␬B-␣ ubiquitination and degradation. The effective concentration of TNF-␣ and purified flagellin used for these experiments are 10 and 100 ng/ml, respectively. A, Butyrate prevents TNF-␣-induced p65 nuclear translocation. Confocal image of cytoplasmic vs nuclear immunofluorescent staining of p65 (Rockland) in HeLa cells treated for 1 h with HBSSϩ supplemented with butyrate or lactate at the indicated concentrations and challenged with TNF-␣ for 20 min. B, Butyrate inhibits activity of a NF-␬B-luciferase reporter. Lysates from HeLa cells transfected with a NF-␬B-luciferase reporter plasmid and treated in the presence of butyrate with TNF-␣ for 5 h were assayed for luciferase activity. Data represent the mean of three independent assays and are shown as percent TNF-␣-induced NF-␬B-Luc. C, Butyrate does not attenuate polyubiquitination of proteins. The level of protein ubiquitination in cellular extracts of Caco-2 cells pretreated with 2.5 mM butyrate for 1 h before addition of MG-262 for additional time was determined by immunoblotting with anti-ubiquitin conjugate Ab (Affinity ␬ ␣ Research). D, Immunoblots of whole cell extracts with anti-total-I B- Ab from Caco-2 cells pretreated with the proteasomal inhibitor MG-262 (500 ng/ml) by guest on September 24, 2021 or HBSSϩ for 40 min before a 1-h incubation with 5 mM butyrate and subsequent TNF-␣ (⌻; 10 ng/ml) or flagellin (F; 100 ng/ml) challenge for 5 or 15 min. Ubiquitinated I␬B-␣ species are marked. Equal loading was confirmed by ␤-tubulin immunoblots. E, Immunoblots of whole cell extracts with anti-phospho-I␬B-␣ Ab from Caco-2 cells pretreated with the proteasomal inhibitor MG-262 (500 ng/ml) for 40 min before a 1-h incubation with 5 mM butyrate and subsequent TNF-␣ (⌻; 10 ng/ml) or flagellin (F; 100 ng/ml) challenge for 5 or 15 min.

To rule out artifacts caused by acidic pH conditions, T84 cells diators (i.e., IL-8) and NF-␬B activation in stimulated cultured were similarly treated with HBSSϩ supplemented with hydrochlo- intestinal epithelial cells (31). We thus undertook a mechanistic ric acid to pHs at which butyrate could induce loss of Cul-1 ned- analysis of NF-␬B signaling to determine whether SCFA effects on dylation (Fig. 1E, lower panel). These conditions had no effect on Cul-1 neddylation affected the NF-␬B pathway. As shown in Fig. Cul-1 neddylation, indicating that although the inhibitory effect of 2A, in cells pretreated with 3 mM butyrate at pH 5.6 or with lactate butyrate was pH dependent, it was not a function of pH alone, at pH 5.0, we observed inhibition of p65 translocation, as evalu- likely because butyrate at pHs approaching its pKa (4.5) becomes ated by immunofluorescence staining of the normal TNF-␣ in- protonated and is freely diffusible across cellular membranes. duced cytoplasmic to nuclear translocation, consistent with effects Thus, these data suggest that the butyrate effects observed likely do observed with intestinal epithelial cell colonization with live bac- not require a receptor or active transport. teria (32). Furthermore, as seen in Fig. 2A, this butyrate-mediated To determine whether the presence of SCFAs could influence block in p65 translocation was abrogated if butyrate was buffered Cul-1 neddylation in intact mucosa, we similarly treated human to pH 7.4 or with lactate at pH 6.0, similar to the observed effects colonic mucosal segments ex vivo in medium supplemented with of pH on neddylation of Cul-1 in Fig. 1D. This blockade was butyrate and assayed Cul-1 neddylation by immunoblot (Fig. 1F). functionally relevant because the same or lower concentration of Reduction of Cul-1 neddylation was observed at concentrations of butyrate inhibited activation of a NF-␬B- dependent reporter gene 1.5–3 mM butyrate. The lower concentrations required in this ex- in a transient transfection assay (Fig. 2B). Thus, butyrate pretreat- periment presumably reflect the differences between intact, fully ment of cultured cells inhibited activation on NF-␬B, consistent differentiated mucosa and cultured cells. Thus, by supplementing with previous reports of butyrate-mediated effects on inflammatory medium with SCFAs, we were able to recapitulate the effects of cytokine induction. viable bacteria on the neddylation of Cul-1. Activation of the NF-␬B pathway is tightly regulated by the phosphorylation, ubiquitination, and of its physically ␬ ␬ ␣ Butyrate represses NF- B activation by blockade of I B- associated inhibitor molecule, I␬B-␣. Yin et al. (29) reported that ubiquitination prolonged exposure (24 h) of HT29 cells to butyrate suppressed Butyrate is known to be a potent inflammatory modulator in vivo; the NF-␬B activation via attenuation of the cellular proteasome it has been shown to block up-regulation of proinflammatory me- activity. To determine whether short-term exposure (as used in our 542 BUTYRATE EFFECTS ON EPITHELIAL SIGNALING

FIGURE 3. Butyrate induces gen- eration of epithelial ROS and causes redox changes. IEC-6 cells were treated with HBSS with or without 10 mM butyrate, washed, and incubated with 5 ␮M of DCF (A), 10 ␮M DHE (B), and 10 ␮M Mito-Sox Red (C). Downloaded from Fluorescence was captured by con- focal laser-scanning microscopy (Zeiss). D, Kinetics of butyrate-in- duced ROS generation in epithelial cells. Confluent Caco-2 cells were preloaded with 100 ␮M DCF and then ROS production was measured http://www.jimmunol.org/ upon colonization with Lactobacillus (MOI ϭ 1) or treatment with 10 mM butyrate. Data represent the mean of three independent assays and are shown as percent induction of ROS upon untreated controls. E and F, Bu- tyrate cause oxidation of Trx1 and Trx2 pools. Confluent Caco2 cells

were treated with PBS with or with- by guest on September 24, 2021 out butyrate as indicated over a time course. Graphs show the redox poten- tial (Eh) for Trx1 and Trx2, respec- tively. Data are represented as means Ϯ SE.

study) of intestinal epithelial cells to butyrate inhibited cellular compare lanes 2–4 to 6–8, respectively). These data indicate that proteasome activity, we treated intestinal epithelial Caco-2 cells within the time frame used in this study butyrate does not suppress with butyrate and monitored cellular proteasome activity by ob- accumulation of cellular polyubiquitinated proteins. serving the accumulation of polyubiquitin-conjugated proteins. Next, to determine whether butyrate-mediated inhibition of Pretreatment of Caco-2 cells with 2.5 mM butyrate for 1 h had no NF-␬B was related to ubiquitination of I␬B-␣, a process controlled gross effects on accumulation of polyubiquitinated proteins (Fig. by Cul-1 neddylation, we undertook an analysis of I␬B-␣ ubiq- 2C, compare lanes 1 and 5). We also monitored the accumulation uitination. Pretreatment of cells with butyrate before stimulation of the ubiquitinated proteins in the presence of MG-262, an inhib- with proinflammatory inducers such as TNF-␣ and flagellin re- itor of the 26S proteasome. In cells pretreated with butyrate and sulted in stabilization of I␬B-␣ in contrast to near to complete further incubated with MG-262 for several hours, we saw no gross degradation observed with TNF-␣ and flagellin, respectively (Fig. differences in accumulation of polyubiquitinated proteins (Fig. 2C, 2D, top panel). To allow further study of phosphorylated and The Journal of Immunology 543

FIGURE 4. Butyrate-mediated ox- A B idation of Ubc12. A, Butyrate inhibits Butyrate Butyrate (10 mM) formation of the NEDD8ϳUbc12 thi- HBSS+ (5mM) olester bond of endogenous Ubc12. kDa 15’ 30’ 90’ 15’ 30’ 60’ 90’ 15’ 30’ 60’ 90’ Western blot analysis with Ubc12- 200 Butyrate (5 mM)

+ specific antisera of whole cell protein S 90 S DPI (µM) NAC (mM) extracted from HeLa cells treated B

H 0 20 40 60 0 20 40 60 with butyrate for1hattheindicated NEDD8-Cul-1 dose and lysed in SDS lysis buffer Cul-1 IB: Cul-1 without (Ϫ DTT) or with (ϩDTT) re- ducing agents. The asterisk marks the NEDD8~Ubc12 * 37 ϳ Ubc12 25 30-kDa DTT-sensitive NEDD8 IB: Ubc12; - DTT Ubc12 thiolester form and the arrow 200 indicates the oxidized forms of 90 Ubc12. B, NAC and DPI prevents bu- tyrate-mediated loss of Cul-1 neddy- lation. HeLa cells pretreated with NAC or DPI for 60 min and subse- quently treated with 5 mM butyrate 37 for 1 h were analyzed by immuno- Ubc12 25 blotting with anti-Cul-1 Ab. IB: Ubc12; + DTT Downloaded from

ubiquitinated I␬B-␣ adducts, we pretreated our model epithelia as compared with untreated controls. An increase in ROS was also with MG-262. Under these conditions, degradation of I␬B-␣ in confirmed with the use of an alternative ROS sensitive dye, DHE ␣ response to TNF- and flagellin was attenuated (Fig. 2D, bottom (Fig. 3B). DHE is freely permeable and is oxidized to fluorescent http://www.jimmunol.org/ panel). Furthermore, very high molecular bands consistent with ethidium bromide by superoxides. We have previously shown that polyubiquitinated I␬B-␣ was observed in cells stimulated with commensal bacteria elicited a rapid and transient increase in ROS TNF-␣ and flagellin, as previously reported (33) (Fig. 2D, bottom within 5–30 min of contact with intestinal epithelial cells (22). In panel). However, these higher molecular mass species represent- comparison to treatment with L. rhamnosus at MOI ϭ 1, butyrate ing polyubiquitinated I␬B-␣ were absent in cell lysates derived stimulated ROS with a slower kinetics that occurred over a period from butyrate-treated cells (Fig. 2D, bottom panel). Since ubiq- of 30–60 min after contact with intestinal epithelial cells (Fig. uitination of I␬B-␣ is dependent upon phosphorylation of I␬B-␣ 3D). To evaluate the sources of butyrate-mediated ROS genera- on two serine residues (Ser32 and 36), we also examined the phos- tion, we used a mitochondria-specific ROS fluorescent dye, phorylation of these residues with a phospho-specific Ab. Again, MitoSOX Red, that specifically detects superoxide generation in by guest on September 24, 2021 when we pretreated the cells with MG-262 before induction with mitochondria. Using this dye, we observed mitochondrial ROS TNF-␣ or flagellin, we observed an increase in high-molecular generation within 15 min of treatment with butyrate, suggesting mass (HMM) ubiquitinated species that was attenuated in the pres- the mitochondria as a potential source of induced ROS (Fig. 3C). ence of butyrate (Fig. 2E). Importantly, phosphorylated I␬B-␣ (de- Accumulation of intracellular levels of ROS in cells causes noted with asterisks in Fig. 2E) was abundant in control and en- changes in the oxidation profiles of the antioxidants. To evaluate hanced in butyrate-treated cell lysates (Fig. 2E). These results are changes in such intracellular redox pools, we assayed the thiol/ consistent with the bacterial mediated inhibition of NF-␬B previ- disulfide redox states of thioredoxins, a major thiol-dependent an- ously described (21, 33). Collectively, these data indicate that bu- tioxidant system of cells. Steady-state Eh values for cytosolic Trx1 tyrate prevents I␬B-␣ degradation by blocking ubiquitination of (Fig. 3E) and mitochondrial Trx2 (Fig. 3F) were calculated using phospho-I␬B-␣. the Nernst equation and show that treatment of cells with butyrate resulted in oxidation of both Trx1 and Trx2 over a 1-h time course. Butyrate causes generation of ROS and induces redox changes For Trx1, the Eh increased from –283 mV in untreated cells to in cultured epithelial cells Ϫ265 mV within 30 min in cells treated with 10 mM butyrate. For ␤-TrCP Ϫ Ϫ Neddylation of Cul-1, required for activity of the E3-SCF Trx2, the Eh increased from 359 mV to 336 mV within 15 min ubiquitin ligase, is dependent on sequential transfer of thiolester with 10 mM butyrate treatment. These cumulative data indicated NEDD8 from the NEDD8-charging enzyme Uba3-APP/BP-1 to a that butyrate treatment induced ROS in cultured epithelial cells catalytically active cysteine residue of the NEDD8-conjugating en- predominantly from mitochondrial sources and induced compen- zyme Ubc12. Ubc12 then catalyzes the formation of an isopeptide satory changes in the intracellular redox balance, albeit with bond between the C-terminal glycine residue of NEDD8 and a slower kinetics and to a lesser degree as ROS induced by viable target (Lys720) residue present on Cul-1 (34). We have pre- commensal bacteria. viously demonstrated that viable commensal bacteria cause tran- sient oxidative inactivation of Ubc12 with resultant effects on ned- dylation of Cul-1 (22). Given that butyrate treatment resulted in Butyrate-induced Ubc12 oxidation potent Cul-1 deneddylation, we asked whether SCFAs were also We next sought to examine whether butyrate-mediated ROS sig- stimulating ROS generation with resultant oxidation of Ubc12. To nals were being transduced via transient oxidation of Ubc12. To monitor ROS generation, IEC-6 intestinal epithelial cells were cul- evaluate the effects of butyrate on endogenous Ubc12, HeLa cells tured with physiological concentrations of butyrate and monitored were treated with 5 and 10 mM butyrate and cell lysates were for changes in DCF fluorescence, a dye sensitive to ROS, espe- analyzed under reducing and nonreducing conditions (absence of cially H2O2. An increase in fluorescence of DCF was observed DTT) to allow visualization of thiol modifications as previously within 15 min of treatment with 10 mM butyrate (Fig. 3, A and D) described (22). When cells were treated with 5 and 10 mM 544 BUTYRATE EFFECTS ON EPITHELIAL SIGNALING butyrate, we observed a loss of the 30-kDa NEDD8ϳUbc12 thi- terial MAMPs, butyrate-mediated signaling occurs via largely un- olester form when cell lysates were prepared and analyzed under known processes. nonreducing conditions (Fig. 4A, top panel, denoted with an as- Recent years have seen the establishment of the paradigm stat- terisk). Significantly, this loss of NEDD8ϳUbc12 thiolester form ing that inducible and “deliberate” generation of ROS have a wide corresponded with the appearance of a marked dose-dependent variety of physiological signaling functions. Several eukaryotic shift in the electrophoretic mobility of Ubc12 to HMM-oxidized signaling proteins, including growth factors, hormones, and cyto- species (Fig. 4A, top panel, denoted with an arrow). We have pre- kines, are capable of stimulating ROS generation in various cell viously confirmed the identity of the NEDD8ϳUbc12 thiolester types including epithelial cells (45–49). Physiological generation form in a similar immunoblot probed with anti-NEDD8 Ab under of these molecules can derive from mitochondrial sources, dedi- nonreducing conditions (22). In cell lysates prepared and analyzed cated homologous to the phagocytic NADPH oxidase, or under standard reducing conditions (with DTT), the HMM-oxi- by 5Ј-lipoxygenase (50, 51). ROS are short-lived molecules and dized species were abolished (Fig. 4A, bottom panel), indicating can exhibit exquisite microspatial localization within cells, allow- that these forms of the enzymes are mixed disulfides between ing specific targeting of signaling effects (51). ROS signaling can Ubc12 and HMM proteins. Furthermore, in cell lysates analyzed be transduced by an increasingly recognized subset of enzymes under reducing conditions, we did not observe the 30-kDa that can be transiently inactivated by reversible oxidation of cat- NEDD8ϳUbc12 thiolester form, further confirming the identity of alytic cysteine residues within the active sites (52, 53). Such en- this form of Ubc12 as being neddylated. zymes include a variety of tyrosine phosphatases such as PTEN Finally, to link the observed induction of intracellular ROS with (54), the DUSP family of MAPK phosphatases (45), antioxidants deneddylation of Cul-1, HeLa cells were treated with butyrate in the such as thioredoxins and peroxiredoxins (55), and, as we and oth- Downloaded from presence of a ROS scavenger NAC and a flavoprotein inhibitor DPI. ers have shown, members of the Ubc family of proteins (22, 56). As shown in Fig. 4B, pretreatment of cells with NAC or DPI atten- In the gut, sources of ROS could be exogenous, deriving from uated butyrate-mediated deneddylation of Cul-1. Collectively, these the metabolic processes of certain bacteria themselves or from in- data demonstrate that butyrate-mediated ROS generation causes at- filtrating neutrophils during an acute inflammatory event. Alterna- tenuation of Ubc12 activity in the same manner as of live bacteria. tively, we have found ROS generated within epithelial cells stim-

ulated by exogenous agents such as commensal bacteria (22), http://www.jimmunol.org/ soluble fermentation products (this report), and sterile components Discussion of bacterial cell walls (N-K. Young and A. S. Neish, unpublished Herein, we have described an important eukaryotic regulatory data). This is likely a highly conserved process, since ROS gen- node, namely, ubiquitin-mediated degradation by E3-SCF ubiq- eration as a response to environmental stressors, including bacte- uitin , that is influenced by the product of a complex pro- ria, is present in plants and lower metazoans (57). ROS generation karyotic community. Active E3-SCF ubiquitin ligases are required by butyrate has been reported, although with far more delayed for the regulation of the NF-␬B, ␤-catenin, Snail, Twist, and kinetics. Stimulation of ROS generation has been observed upon Hedgehog pathways and presumably others (35). The activity of exposure of human tongue cancer cells to 8 and 16 mM butyrate the E3-SCF ubiquitin ligases is modulated by neddylation of its for a period of 4–24 h (58, 59). Of note, although in our hands by guest on September 24, 2021 Cul-1 subunit (17). Neddylation is a posttranslation modification ROS generation elicited by butyrate was at least as strong as ROS specific to substrates of the Cullin family in which NEDD8, a induced by known ROS-dependant agonists such as insulin, epi- small 8-kDa ubiquitin-like protein, is attached to the substrates and dermal growth factor, flagellin, or platelet-derived growth factor is emerging as a key regulatory event in cellular processes that are (A. Kumar, M.-K. Young, and A. S. Neish, unpublished results), controlled by ubiquitin-mediated degradation of proteins. In a va- butyrate-induced ROS was markedly less than the amounts de- riety of experimental models, including yeast (36), Arabidopsis tected after cells were stimulated with viable bacteria and occurred (37), Caenorhabditis elegans (38), Drosophila (39), mice (40), and with noticeably slower kinetics. As suggested by our data, butyrate human cells in vitro (41), mutations leading to loss-of-function of may induce ROS via the mitochondrial pathway rather than by Cul-1 neddylation (or deneddylation) status can have profound receptor-mediated events. functional consequences. However, only limited evidence is avail- The physiochemical environment of the intestinal lumen in im- able implicating environmental signals in the physiological regu- mediate contact with the epithelium is largely anaerobic and is lation of Cul-1 neddylation. For example, yeast Cullin was shown suffused with not just with a complex ecosystem of living bacteria, to be hyperneddylated in response to UV exposure (42). In Ara- but also a miasma of MAMPs, small bacterially produced com- bidopsis studies, wild-type plants showed marked Cul-1 deneddy- pounds, and a spectrum of metabolic by-products that varies mark- lation when reared in darkness, while mutant plants that fail to edly over the greater than 1 m length of the organ (4, 5, 8). The respond to light were shown to have an abnormal accumulation of ascending colon contains a highly fermentative, “saccharolytic” neddylated Cul-1 (43). Furthermore, our previous results demon- bacterial milieu which produce high concentrations (up to 140 mM strate that interactions of commensal bacteria with mammalian in- for butyrate) of organic acids including lactate, acetate, and SCFAs testinal epithelia modulate Cul-1 neddylation (21) along with ex- (succinate, propionate, butyrate), resulting in a physiological lu- ogenous H2O2 (22), adenosine (44), and, in this report, bacterial minal pH of 5.5–6.5, as measured by a series of radiotelemetry fermentation products. We found commensal bacteria were able to studies in healthy human volunteers (reviewed in Ref. 30). As the mediate effects on Cul-1 neddylation via soluble bacterial fermen- carbohydrate-fermentative substrates are consumed through ␤ ox- tation products. Such SCFAs, particularly butyrate, have long been idation and bicarbonate is secreted by the epithelia, the luminal known to provide a nutritive source for the mammalian gut. More- environment shows a progressive decline in SCFA concentration over, butyrate has well-known immunomodulatory and trophic ef- (to 40 mM for butyrate) and a corresponding rise in pH, which in fects on gut homeostasis (6). In this sense, these compounds pos- the descending colon is near neutral. SCFAs are weak organic sess the ability to act as effectors or signaling intermediates in the acids, with pKa values generally in the range of 4.5–5.0, not co- symbiotic cross-talk between the mammalian host and the normal incidentally overlapping the range where loss of Cul-1 neddylation flora. Unlike other intermediates in the dialog between host and was observed (60). At these pH’s, the protonated forms of the microbe, such as eukaryotic pattern recognition receptors and bac- acids are lipid soluble, membrane permeable, and exhibit maximal The Journal of Immunology 545 bioactivity (61, 62). Thus, different regions of the colon are ex- monella, Shigella, etc., do not reach biomass densities of 1011 posed to significantly different concentrations of bioavailable organisms/ml before executing their effects on host cells, al- SCFAs and no doubt other chemical parameters as well. Possibly, though pathogens that exist in microbe-dense biofilms could regions of the colon exposed to maximal bioavailable SCFA con- plausibly generate local microenvironments of higher metabo- centrations may have attenuated inflammatory potential. Intrigu- lite concentrations. ingly, this anatomical distribution of the gradient of SCFA con- The relationship of the normal flora to the intestinal epithelia is centrations is the precise inverse to the anatomical distribution of a prime example of a complex microbial community associated the inflammatory lesions of acute ulcerative colitis, which is in- with a higher (76). It may be an oversimplification to variably seen in the distal colon and progressively attenuates prox- view this system as a binary host-pathogen interaction, with ex- imally (63). changes of individual and discrete biochemical signals, such as Excessively high concentrations of SCFAs may have deleterious secreted effector proteins, toxins, or even small molecules with clinical consequences. Necrotizing enterocolitis is a spontaneous specificity toward eukaryotic cellular processes. The intestinal lu- complication of prematurity that has been suggested to be medi- minal flora produces a massive and diverse metabolic output that ated by excessive levels of SCFAs (64, 65). Empirically, the pHs varies according to anatomical location, nutrient composition, and of luminal contents in patients with this disorder have been mea- host health. It may be more illuminating to view the normal flora sured at 3.8–4.6. Other workers have induced intestinal injury in as a holistic community rather than as a collection of semi-inde- rodents with very high doses of butyrate (300 mM/pH 4.0 with pendent organisms mediating “beneficial” or “detrimental” effects. normal being 70–140 mM/pH 5.5 in the proximal colon) and other Our results suggest a novel mechanism by which this complex

SCFAs (66). Interestingly, it was also shown that butyrate anion at community can collectively influence key epithelial signaling Downloaded from neutral or alkaline pH had no effect nor did low pH alone, con- pathways. Metabolic products/small molecules produced at the eu- sistent with our observations in vitro (67). The transient nature of karyotic/prokaryotic interface may account for some of the widely Cul-1 deneddylation indicates that enterocytes can compensate and known effects of the bacterial flora on normal intestinal function adapt to dynamic changes; however, extreme conditions may limit (3) and may influence a range of eukaryotic regulatory processes. the ability of enterocytes to adapt. Given the wide variety of cel- lular processes other than NF-␬B and the Wnt signaling pathway Acknowledgments http://www.jimmunol.org/ activation that are regulated by SCF-mediated protein-degradative Human Biopsy Samples were provided by Dr. Shanti Srinivasan. We thank processes, it is not surprising that excessive exposure to very high Kirsten Gerner-Smidt for help with the confocal microscope. levels of compounds that cause loss of Cul-1 neddylation could inhibit many enzymatic processes and be rapidly toxic to the cell. Disclosures Luminal physiochemical properties are dynamic and may be The authors have no financial conflict of interest. manipulated in a therapeutic fashion by high-starch dietary sup- plements (prebiotics) or oral supplementation with live bacteria References (probiotics) (68, 69). Interestingly, one group of microbes with 1. Falk, P. G., L. V. Hooper, T. Midtvedt, and J. I. Gordon. 1998. Creating and

maintaining the gastrointestinal ecosystem: what we know and need to know by guest on September 24, 2021 therapeutic value are the lactic acid bacteria, well-known Gram- from gnotobiology. Microbiol. Mol. Biol. Rev. 62: 1157–1170. positive enteric commensals that include the genera Lactobacillus, 2. Hooper, L. V., M. H. Wong, A. Thelin, L. Hansson, P. G. Falk, and J. I. Gordon. Enterococcus, and Bifidobacteria, among others, and are defined 2001. Molecular analysis of commensal host-microbial relationships in the in- testine. Science 291: 881–884. by a common ability to produce lactic acid under microaerophilic 3. Hooper, L. V., and J. I. Gordon. 2001. Commensal host-bacterial relationships in to anaerobic conditions (70). As part of the human enteric flora, the gut. Science 292: 1115–1118. 4. Hooper, L. V., T. Midtvedt, and J. I. Gordon. 2002. How host-microbial inter- they contribute to the physiological acidic pH of the intraluminal actions shape the nutrient environment of the mammalian intestine. Annu. Rev. colonic environment (68) and lactate produced by this subset of Nutr. 22: 283–307. organisms may also serve as substrate for butyrate production by 5. Topping, D. L., and P. M. Clifton. 2001. Short-chain fatty acids and human colonic function: roles of resistant starch and nonstarch polysaccharides. Physiol. others, a process termed metabolic cross-feeding (71). More re- Rev. 81: 1031–1064. cently, these organisms have been recognized as possessing 6. Hamer, H. M., D. Jonkers, K. Venema, S. Vanhoutvin, F. J. Troost, and marked probiotic activity both clinically (reviewed in Ref. 69) and R. J. Brummer. 2008. Review article: the role of butyrate on colonic function. Aliment Pharmacol. Ther. 27: 104–119. in models (72). Conditioned medium from L. rhamnosis 7. Wong, J. M., R. de Souza, C. W. Kendall, A. Emam, and D. J. Jenkins. 2006. Colonic GG has been shown to repress proinflammatory gene expression in health: fermentation and short chain fatty acids. J. Clin. Gastroenterol. 40: 235–243. 8. Wachtershauser, A., and J. Stein. 2000. Rationale for the luminal provision of cultured epithelial cells (73). It has also been reported that condi- butyrate in intestinal diseases. Eur. J. Nutr. 39: 164–171. tioned medium from two of these bacteria, Bifidobacterium breve 9. Vernia, P., V. Annese, G. Bresci, G. d’Albasio, R. D’Inca, S. Giaccari, and Streptococcus thermophilus, exhibited inhibitory effects on M. Ingrosso, C. Mansi, G. Riegler, D. Valpiani, and R. Caprilli. 2003. Topical ␬ butyrate improves efficacy of 5-ASA in refractory distal ulcerative colitis: results NF- B activation in vitro (74). Furthermore, this activity was at- of a multicentre trial. Eur. J. Clin. Invest. 33: 244–248. tributed to a small (Ͻ3 kDa), soluble metabolite. These workers 10. Winston, J. T., P. Strack, P. Beer-Romero, C. Y. Chu, S. J. Elledge, and ␤ quantified lactate in the conditioned medium at 4.5–9 mM, but J. W. Harper. 1999. The SCF -TRCP-ubiquitin ligase complex associates spe- cifically with phosphorylated destruction motifs in I␬B␣ and ␤-catenin and stim- were unable to show NF-␬B inhibition at these concentrations. ulates I␬B␣ ubiquitination in vitro. Genes Dev. 13: 270–283. Finally, the bacterial metabolites butyrate, propionate, and acetate 11. Spencer, E., J. Jiang, and Z. J. Chen. 1999. Signal-induced ubiquitination of I␬B␣ by the F-box protein Slimb/␤-TrCP. Genes Dev. 13: 284–294. have been shown to have inhibitory activity on proinflammatory 12. Pan, Z. Q., A. Kentsis, D. C. Dias, K. Yamoah, and K. Wu. 2004. Nedd8 on signaling pathways. These compounds have been used therapeu- Cullin: building an expressway to protein destruction. Oncogene 23: 1985–1997. tically (by intraluminal instillation) to dampen intestinal inflam- 13. Liakopoulos, D., G. Doenges, K. Matuschewski, and S. Jentsch. 1998. A novel protein modification pathway related to the ubiquitin system. EMBO J. 17: mation in inflammatory bowl diseases (8, 9, 75) and have been 2208–2214. shown to block NF-␬B activation in vitro (28, 29). 14. Kamitani, T., K. Kito, H. P. Nguyen, and E. T. Yeh. 1997. Characterization of Our observations are likely limited to effects of large bacterial com- NEDD8, a developmentally down-regulated ubiquitin-like protein. J. Biol. Chem. 272: 28557–28562. munities with substantial metabolic outputs necessary to ferment suf- 15. Parry, G., and M. Estelle. 2004. Regulation of cullin-based ubiquitin ligases by ficient quantities of organic acids and perhaps other products. As the Nedd8/RUB ubiquitin-like proteins. Semin. Cell Dev. Biol. 15: 221–229. 16. Ohh, M., W. Y. Kim, J. J. Moslehi, Y. Chen, V. Chau, M. A. Read, and such, our observations probably do not reflect events in most W. G. Kaelin, Jr. 2002. An intact NEDD8 pathway is required for Cullin-depen- acute enteric infections. Invasive enteric pathogens such as Sal- dent ubiquitylation in mammalian cells. EMBO Rep. 3: 177–182. 546 BUTYRATE EFFECTS ON EPITHELIAL SIGNALING

17. Read, M. A., J. E. Brownell, T. B. Gladysheva, M. Hottelet, L. A. Parent, 45. Kamata, H., S. Honda, S. Maeda, L. Chang, H. Hirata, and M. Karin. 2005. M. B. Coggins, J. W. Pierce, V. N. Podust, R. S. Luo, V. Chau, and Reactive oxygen species promote TNF␣-induced death and sustained JNK acti- V. J. Palombella. 2000. Nedd8 modification of cul-1 activates SCF␤(TrCP)-depen- vation by inhibiting MAP kinase phosphatases. Cell 120: 649–661. dent ubiquitination of I␬B␣. Mol. Cell. Biol. 20: 2326–2333. 46. Mahadev, K., X. Wu, A. Zilbering, L. Zhu, J. T. Lawrence, and B. J. Goldstein. 18. Furukawa, M., Y. Zhang, J. McCarville, T. Ohta, and Y. Xiong. 2000. The CUL1 2001. Hydrogen peroxide generated during cellular insulin stimulation is integral C-terminal sequence and ROC1 are required for efficient nuclear accumulation, to activation of the distal insulin signaling cascade in 3T3–L1 adipocytes. J. Biol. NEDD8 modification, and ubiquitin ligase activity of CUL1. Mol. Cell. Biol. 20: Chem. 276: 48662–48669. 8185–8197. 47. Mahadev, K., A. Zilbering, L. Zhu, and B. J. Goldstein. 2001. Insulin-stimulated 19. Amir, R. E., K. Iwai, and A. Ciechanover. 2002. The NEDD8 pathway is essen- hydrogen peroxide reversibly inhibits protein-tyrosine phosphatase 1b in vivo and ␤ tial for SCF -TrCP-mediated ubiquitination and processing of the NF-␬B precur- enhances the early insulin action cascade. J. Biol. Chem. 276: 21938–21942. sor p105. J. Biol. Chem. 277: 23253–23259. 48. Chen, K. C., Y. Zhou, K. Xing, K. Krysan, and M. F. Lou. 2004. Platelet derived 20. Amir, R. E., H. Haecker, M. Karin, and A. Ciechanover. 2004. Mechanism of growth factor (PDGF)-induced reactive oxygen species in the lens epithelial cells: processing of the NF-␬B2 p100 precursor: identification of the specific polyu- the redox signaling. Exp. Eye Res. 78: 1057–1067. biquitin chain-anchoring lysine residue and analysis of the role of NEDD8-mod- 49. Goldstein, B. J., K. Mahadev, X. Wu, L. Zhu, and H. Motoshima. 2005. Role of ␤-TrCP ification on the SCF ubiquitin ligase. Oncogene 23: 2540–2547. insulin-induced reactive oxygen species in the insulin signaling pathway. Anti- 21. Collier-Hyams, L. S., V. Sloane, B. C. Batten, and A. S. Neish. 2005. Cutting oxid. Redox Signal 7: 1021–1031. edge: bacterial modulation of epithelial signaling via changes in neddylation of 50. Lambeth, J. D. 2004. NOX enzymes and the biology of reactive oxygen. Nat. cullin-1. J. Immunol. 175: 4194–4198. Rev. Immunol. 4: 181–189. 22. Kumar, A., H. Wu, L. S. Collier-Hyams, J. M. Hansen, T. Li, K. Yamoah, 51. Terada, L. S. 2006. Specificity in reactive oxidant signaling: think globally, act Z. Q. Pan, D. P. Jones, and A. S. Neish. 2007. Commensal bacteria modulate locally. J. Cell Biol. 174: 615–623. Cullin-dependent signaling via generation of reactive oxygen species. EMBO J. 52. Tonks, N. K. 2005. Redox redux: revisiting PTPs and the control of cell signal- 26: 4457–4466. ing. Cell 121: 667–670. 23. McSorley, S. J., B. D. Ehst, Y. Yu, and A. T. Gewirtz. 2002. Bacterial flagellin ϩ 53. Chiarugi, P., and F. Buricchi. 2007. Protein tyrosine phosphorylation and revers- is an effective adjuvant for CD4 T cells in vivo. J. Immunol. 169: 3914–3919. ible oxidation: two cross-talking posttranslation modifications. Antioxid. Redox 24. Zeng, H., A. Q. Carlson, Y. Guo, Y. Yu, L. S. Collier-Hyams, J. L. Madara,

Signal 9: 1–24. Downloaded from A. T. Gewirtz, and A. S. Neish. 2003. Flagellin is the major proinflammatory determinant of enteropathogenic Salmonella. J. Immunol. 171: 3668–3674. 54. Seo, J. H., Y. Ahn, S. R. Lee, C. Yeol Yeo, and K. Chung Hur. 2005. The major 25. Wang, H., and J. A. Joseph. 1999. Quantifying cellular oxidative stress by di- target of the endogenously generated reactive oxygen species in response to in- chlorofluorescein assay using microplate reader. Free Radical Biol. Med. 27: sulin stimulation is phosphatase and tensin homolog and not phosphoinositide-3 612–616. kinase (PI-3 kinase) in the PI-3 kinase/Akt pathway. Mol. Biol. Cell 16: 348–357. 26. Halvey, P. J., W. H. Watson, J. M. Hansen, Y. M. Go, A. Samali, and D. P. Jones. 55. Veal, E. A., A. M. Day, and B. A. Morgan. 2007. Hydrogen peroxide sensing and 2005. Compartmental oxidation of thiol-disulphide redox couples during epider- signaling. Mol. Cell 26: 1–14. mal growth factor signalling. Biochem. J. 386: 215–219. 56. Bossis, G., and F. Melchior. 2006. Regulation of SUMOylation by reversible oxidation of SUMO conjugating enzymes. Mol. Cell 21: 349–357. 27. Watson, W. H., and D. P. Jones. 2003. Oxidation of nuclear thioredoxin during http://www.jimmunol.org/ oxidative stress. FEBS Lett. 543: 144–147. 57. Apel, K., and H. Hirt. 2004. Reactive oxygen species: metabolism, oxidative 28. Segain, J. P., D. Raingeard de la Bletiere, A. Bourreille, V. Leray, N. Gervois, stress, and signal transduction. Annu. Rev. Plant Biol. 55: 373–399. C. Rosales, L. Ferrier, C. Bonnet, H. M. Blottiere, and J. P. Galmiche. 2000. 58. Jeng, J. H., M. Y. Kuo, P. H. Lee, Y. J. Wang, M. Y. Lee, J. J. Lee, B. R. Lin, Butyrate inhibits inflammatory responses through NF␬B inhibition: implications T. F. Tai, and M. C. Chang. 2006. Toxic and metabolic effect of sodium butyrate for Crohn’s disease. Gut 47: 397–403. on SAS tongue cancer cells: role of deregulation and redox changes. 29. Yin, L., G. Laevsky, and C. Giardina. 2001. Butyrate suppression of colonocyte Toxicology 223: 235–247. NF-␬B activation and cellular proteasome activity. J. Biol. Chem. 276: 59. Hsiao, C. H., W. Li, T. F. Lou, B. S. Baliga, and B. S. Pace. 2006. Fetal hemo- 44641–44646. globin induction by histone deacetylase inhibitors involves generation of reactive 30. Nugent, S. G., D. Kumar, D. S. Rampton, and D. F. Evans. 2001. Intestinal oxygen species. Exp. Hematol. 34: 264–273. luminal pH in inflammatory bowel disease: possible determinants and implica- 60. Velazquez, O. C., H. M. Lederer, and J. L. Rombeau. 1997. Butyrate and the tions for therapy with aminosalicylates and other drugs. Gut 48: 571–577. colonocyte: production, absorption, metabolism, and therapeutic implications.

31. Weng, M., W. A. Walker, and I. R. Sanderson. 2007. Butyrate regulates the Adv. Exp. Med. Biol. 427: 123–134. by guest on September 24, 2021 expression of pathogen-triggered IL-8 in intestinal epithelia. Pediatr. Res. 62: 61. Stein, J., M. Zores, and O. Schroder. 2000. Short-chain fatty acid (SCFA) uptake 542–546. into Caco-2 cells by a pH-dependent and carrier mediated transport mechanism. 32. Collier-Hyams, L. S., and A. S. Neish. 2005. Innate immune relationship between Eur. J. Nutr. 39: 121–125. commensal flora and the mammalian intestinal epithelium. Cell Mol. Life Sci. 62: 62. Charney, A. N., L. Micic, and R. W. Egnor. 1998. Nonionic diffusion of short- 1339–1348. chain fatty acids across rat colon. Am. J. Physiol. 274: G518–G524. 33. Neish, A. S., A. T. Gewirtz, H. Zeng, A. N. Young, M. E. Hobert, V. Karmali, 63. Cotran, R., V. Kumar, and T. Collins. 1999. The Pathologic Basis of Disease. A. S. Rao, and J. L. Madara. 2000. Prokaryotic regulation of epithelial responses Saunders, Philadelphia. by inhibition of I␬B-␣ ubiquitination. Science 289: 1560–1563. 64. Clark, D. A., J. E. Thompson, L. B. Weiner, J. A. McMillan, A. J. Schneider, and 34. Wada, H., E. T. Yeh, and T. Kamitani. 2000. A dominant-negative UBC12 mu- J. E. Rokahr. 1985. Necrotizing enterocolitis: intraluminal biochemistry in human tant sequesters NEDD8 and inhibits NEDD8 conjugation in vivo. J. Biol. Chem. neonates and a rabbit model. Pediatr. Res. 19: 919–921. 275: 17008–17015. 65. Sanderson, I. R. 1999. The physicochemical environment of the neonatal intes- 35. Cardozo, T., and M. Pagano. 2004. The SCF ubiquitin ligase: insights into a tine. Am. J. Clin. Nutr. 69: 1028S–1034S. molecular machine. Nat. Rev. Mol. Cell. Biol. 5: 739–751. 66. Nafday, S. M., W. Chen, L. Peng, M. W. Babyatsky, I. R. Holzman, and J. Lin. 36. Lyapina, S., G. Cope, A. Shevchenko, G. Serino, T. Tsuge, C. Zhou, D. A. Wolf, 2005. Short-chain fatty acids induce colonic mucosal injury in rats with various N. Wei, and R. J. Deshaies. 2001. Promotion of NEDD-CUL1 conjugate cleavage postnatal ages. Pediatr. Res. 57: 201–204. by COP9 signalosome. Science 292: 1382–1385. 67. McCafferty, D. M., and I. J. Zeitlin. 1989. Short chain fatty acid-induced colitis 37. Bostick, M., S. R. Lochhead, A. Honda, S. Palmer, and J. Callis. 2004. Related in mice. Int. J. Tissue React. 11: 165–168. to ubiquitin 1 and 2 are redundant and essential and regulate vegetative growth, 68. Teitelbaum, J. E., and W. A. Walker. 2002. Nutritional impact of pre- and pro- auxin signaling, and ethylene production in Arabidopsis. Plant Cell 16: biotics as protective gastrointestinal organisms. Annu. Rev. Nutr. 22: 107–138. 2418–2432. 69. Sartor, R. B. 2004. Therapeutic manipulation of the enteric microflora in inflam- 38. Kurz, T., L. Pintard, J. H. Willis, D. R. Hamill, P. Gonczy, M. Peter, and matory bowel diseases: antibiotics, probiotics, and prebiotics. Gastroenterology B. Bowerman. 2002. Cytoskeletal regulation by the Nedd8 ubiquitin-like protein 126: 1620–1633. modification pathway. Science 295: 1294–1298. 39. Ou, C. Y., Y. F. Lin, Y. J. Chen, and C. T. Chien. 2002. Distinct protein deg- 70. Klein, G., A. Pack, C. Bonaparte, and G. Reuter. 1998. Taxonomy and physiol- radation mechanisms mediated by Cul1 and Cul3 controlling Ci stability in Dro- ogy of probiotic lactic acid bacteria. Int. J. Food Microbiol. 41: 103–125. sophila eye development. Genes Dev. 16: 2403–2414. 71. Flint, H. J., S. H. Duncan, K. P. Scott, and P. Louis. 2007. Interactions and 40. Tateishi, K., M. Omata, K. Tanaka, and T. Chiba. 2001. The NEDD8 system is competition within the microbial community of the human colon: links between essential for cell cycle progression and morphogenetic pathway in mice. J. Cell diet and health. Environ. Microbiol. 9: 1101–1111. Biol. 155: 571–579. 72. Madsen, K. L., J. S. Doyle, L. D. Jewell, M. M. Tavernini, and R. N. Fedorak. 41. Cope, G. A., and R. J. Deshaies. 2003. COP9 signalosome: a multifunctional 1999. Lactobacillus species prevents colitis in interleukin 10 gene-deficient mice. regulator of SCF and other cullin-based ubiquitin ligases. Cell 114: 663–671. Gastroenterology 116: 1107–1114. 42. Groisman, R., J. Polanowska, I. Kuraoka, J. Sawada, M. Saijo, R. Drapkin, 73. Pena, J. A., and J. Versalovic. 2003. Lactobacillus rhamnosus GG decreases A. F. Kisselev, K. Tanaka, and Y. Nakatani. 2003. The ubiquitin ligase activity TNF-␣ production in lipopolysaccharide-activated murine macrophages by a con- in the DDB2 and CSA complexes is differentially regulated by the COP9 signa- tact-independent mechanism. Cell Microbiol. 5: 277–285. losome in response to DNA damage. Cell 113: 357–367. 74. Menard, S., C. Candalh, J. C. Bambou, K. Terpend, N. Cerf-Bensussan, and 43. Wang, X., S. Feng, N. Nakayama, W. L. Crosby, V. Irish, X. W. Deng, and M. Heyman. 2004. Lactic acid bacteria secrete metabolites retaining anti- N. Wei. 2003. The COP9 signalosome interacts with SCF UFO and participates inflammatory properties after intestinal transport. Gut 53: 821–828. in Arabidopsis flower development. Plant Cell 15: 1071–1082. 75. Scheppach, W., and F. Weiler. 2004. The butyrate story: old wine in new bottles? 44. Khoury, J., J. C. Ibla, A. S. Neish, and S. P. Colgan. 2007. Antiinflammatory Curr. Opin. Clin. Nutr. Metab. Care 7: 563–567. adaptation to hypoxia through adenosine-mediated Cullin-1 deneddylation. 76. McFall-Ngai, M. J. 2002. Unseen forces: the influence of bacteria on animal J. Clin. Invest. 117: 703–711. development. Dev. Biol. 242: 1–14.