Enhanced sensitivity to DSS colitis caused by a hypomorphic Mbtps1 mutation disrupting the ATF6-driven unfolded protein response

Katharina Brandla,1, Sophie Rutschmanna,1,2, Xiaohong Lia, Xin Dua, Nengming Xiaoa, Bernd Schnablb, David A. Brennerb, and Bruce Beutlera,3
aDepartment

of Genetics, The Scripps Research Institute, 10550 North Torrey Pines Road, La Jolla, CA 92037; and bDepartment of Medicine, University of California at San Diego, 9500 Gilman Drive, La Jolla, CA 92093 Contributed by Bruce Beutler, December 19, 2008 (sent for review December 15, 2008)

Here, we describe an N-ethyl-N-nitrosourea (ENU)-induced missense error in the membrane-bound transcription factor peptidase site 1 (S1P)-encoding gene (Mbtps1) that causes enhanced susceptibility to dextran sodium sulfate (DSS)-induced colitis. S1P cleaves and activates cAMP response element binding protein/ATF transcription factors, the sterol regulatory element-binding proteins (SREBPs), and other proteins of both endogenous and viral origin. Because S1P has a nonredundant function in the ATF6-dependent unfolded protein response (UPR), woodrat mice show diminished levels of major endoplasmic reticulum chaperones GRP78 (BiP) and GRP94 in the colon upon DSS administration. Experiments with bone marrow chimeric mice reveal a requirement for S1P in nonhematopoietic cells, without which a diminished UPR and colitis develop.
ER stress UPR site 1 protease inammatory bowel disease

he unfolded protein response (UPR) is a highly coordinated series of reactions that facilitates the folding, processing, export, and degradation of proteins that are misfolded in the endoplasmic reticulum (ER). An excess of unfolded proteins is perceived by the cell as ER stress. The UPR is initiated through activation of one or more of 3 separate signaling cascades triggered by sensors of ER stress. The sensors are activated by dissociation of UPR-inhibitory glucose-regulated protein 78 (GRP78), also known as Ig heavy chain-binding protein (BiP). The apical proteins of the 3 UPR cascades include inositol-requiring enzymes IRE1 and IRE1 [also known as ER-to-nucleus signaling 1 (ERN1) and ERN2, respectively], pancreatic ER kinase (PERK), and activating transcription factor 6 (ATF6) (1). The UPR alleviates ER stress in 3 ways: by reducing influx of newly synthesized protein into the ER, by causing the degradation of misfolded proteins, and by enhancing chaperone-mediated folding. Enhanced expression of chaperones and degradation of misfolded proteins both require transcriptional activation, which is mediated by all 3 pathways. Activation of PERK induces the translation of mRNA encoding the transcription factor ATF4. The PERK pathway is additionally responsible for inhibition of ER protein influx, which it accomplishes by phosphorylation and inactivation of eukaryotic translation-initiation factor 2 . Activation of IRE1 (which is widely expressed) or IRE1 (which is specifically expressed in intestinal epithelial cells) initiates unconventional splicing of the transcript encoding X-box-binding protein 1 (XBP1) to produce the highly active basic leucine zipper (bZIP) family transcription factor XBP1s (2). Finally, ATF6, a latent bZIP family transcription factor, is converted to an active transcription factor by ER stress-induced intramembrane proteolysis. The closely related ATF6 and ATF6 molecules are encoded by separate genes and constitutively synthesized as type II transmembrane proteins in the ER. ATF6 is believed to be of key importance in the UPR (2). Upon ER stress, ATF6 translocates from the ER to the Golgi, where it is sequentially
3300 3305 PNAS March 3, 2009 vol. 106 no. 9

cleaved into an active transcription factor by the site 1 protease (S1P) and site 2 protease (S2P) (3). Upon translocation to the nucleus, active ATF6 binds ER stress-responsive elements to activate transcription of genes encoding the chaperone proteins GRP78 and GRP94 (1), the CCAAT/enhancer-binding protein homologous protein (CHOP), and XBP1 (2). S1P is therefore required for the ATF6-mediated UPR to occur. Recent work suggests a role for ER stress in inflammatory bowel disease (IBD). For example, a mutation of the Muc2 gene, which encodes the abundant and highly glycosylated mucin-2 protein, leads to ER stress and spontaneous colitis in mice (4). Mice lacking IRE1 are more susceptible to dextran sodium sulfate (DSS)-induced colitis (13). Mice lacking XBP1 in epithelial cells display a loss of secretory paneth cells and goblet cells in the intestinal epithelia, as well as spontaneous inflammation in the ileum. Moreover, single-nucleotide polymorphisms in the XBP1 locus have been correlated with IBD in humans (6). There has, however, been no indication that the ATF6-mediated UPR is essential for intestinal homeostasis and prevention of IBD. Here, we describe an N-ethyl-N-nitrosourea (ENU)-induced hypomorphic allele of the membrane-bound transcription factor S1P-encoding gene, Mbtps1. The mutant strain was named woodrat (wrt) because of an effect of the mutation on coat color, to be described in a separate publication. S1P is an ER/early Golgi membrane-anchored, calcium-dependent serine protease that belongs to the family of precursor convertases (7). These enzymes are responsible for intracellular proteolysis of inactive precursors to produce active peptides (8). S1P cleaves and activates cAMP response element binding protein (CREB)/ATF transcription factors, the sterol regulatory element-binding proteins (SREBPs), and other proteins of both endogenous and viral origin (3, 9). The biological function of S1P has not been discovered in full, because targeted disruption of Mbtps1 in the mouse prevents normal epiblast formation and subsequent implantation of the embryo (10), resulting in lethality at an early developmental stage (11). Our studies implicate S1P and the ATF6 transcription factors in the management of ER stress within intestinal epithelial cells, and show that the S1P3ATF6 axis is required to prevent DSS-induced colitis.
Author contributions: K.B., S.R., D.A.B., and B.B. designed research; K.B., S.R., X.L., X.D., N.X., and B.S. performed research; K.B., S.R., and B.B. analyzed data; and K.B., S.R., and B.B. wrote the paper. The authors declare no conict of interest.
1K.B.

and S.R. contributed equally to this work.

2Present

address: Department of Immunology, Imperial College, Du Cane Road, London W12 0NN, United Kingdom. whom correspondence should be addressed. E-mail: bruce@scripps.edu.

3To

This article contains supporting information online at www.pnas.org/cgi/content/full/ 0813036106/DCSupplemental. 2009 by The National Academy of Sciences of the USA

www.pnas.org cgi doi 10.1073 pnas.0813036106

Fig. 1. The hypopigmented woodrat mutant mice have a point mutation in Mbtps1. (A) The woodrat coat color phenotype at 4 months postpartum. (B) DNA sequence taken from exon 13 of Mbtps1, the gene coding for S1P, in a C57BL/6 WT control (Upper) and a woodrat homozygous mouse (Lower). The mutation is an A-to-G transition that results in a Y-to-C substitution at amino acid 496.

Results The recessive woodrat mutation was induced with ENU on a pure C57BL/6 background and detected as a visible phenovariant in one of more than 88,500 G3 animals examined to date. Modest hypopigmentation was first observed in woodrat homozygotes 8 days postpartum, and it progressed to yield a coat made of a combination of white and black hairs by weaning age, varying in pigment content along the length of each shaft. In adults, the coat was more homogenous, incorporating hairs with a white base and a gray/brown tip (Fig. 1A). The woodrat stock was amplified by breeding homozygous males to heterozygous females, and positionally cloned by outcrossing homozygous males to C3H/HeN strain females and intercrossing the F1 hybrid population. On 14 meioses, the mutation was mapped to chromosome 8 with a peak LOD score of 3, and on 336 meioses it was mapped to a 1.2-Mb interval between 122 Mb and 123.2 Mb from the centromere. The critical region contained 20 annotated genes, all of which were fully sequenced at the cDNA or genomic level. A single missense mutation was found in exon 13 of the 24-exon Mbtps1 gene (GenBank accession no. NM_01970). The mutation was an A-to-G transition at nucleotide 2041, predicting the amino acid substitution Y496C, located on the C-terminal side of the protease domain (Fig. 1B). We crossed heterozygotes for a gene trap-inactivated Mbtps1 allele (Mbtps1 /trap) with heterozygotes for the wrt Mbtps1 allele (Mbtps1 /wrt). We also evaluated the progeny of Mbtps1 /wrt Mbtps1 /wrt crosses and Mbtps1wrt/wrt Mbtps1 /wrt crosses [supporting information (SI) Tables S1 and S2]. The Mbtps /trap genotype and the Mbtps1wrt/wrt genotype conferred comparably
Brandl et al.

diminished intrauterine viability and the Mbtps1wrt/trap genotype was embryonic lethal. Because a single copy of the gene (Mbtps1 /trap) permitted diminished survival to term and caused no visible phenotype, whereas Mbtps1wrt/wrt also permitted diminished survival to term but did cause a visible phenotype, we infer that the Mbtps1wrt allele encodes a protein with 50% of WT activity. Because Mbtps1wrt/trap is insufficient to support survival to term, the minimum viable level of enzyme activity for intrauterine survival through term probably lies between 25% and 50%. Mbtps1wrt/wrt mice were subjected to numerous studies of immunological function, which included a test of their response to orally administered DSS. After administration of 2% DSS in the drinking water for 7 days, Mbtps1wrt/wrt mice showed severe weight loss compared with wild-type (WT) controls, consistent with an exaggerated colitic response (Fig. 2A). Antibiotic treatment abrogated the difference between DSS-induced weight loss in WT vs. Mbtps1wrt/wrt mice, indicating that the enhanced
PNAS March 3, 2009 vol. 106 no. 9 3301

IMMUNOLOGY

Fig. 2. Increased susceptibility of woodrat mice to DSS-induced colitis. (A) Percent initial weight of WT and Mbtps1wrt/wrt mice after administration of 2% DSS for 7 days. C57BL/6 mice receiving 0% DSS served as controls. Each group, n 9 20. *, P 0.05; **, P 0.005, 2-tailed MannWhitney test for WT and Mbtps1wrt/wrt. (B) C57BL/6 and Mbtps1wrt/wrt were treated with antibiotics in drinking water 5 days before and during the administration of DSS (2%). Weight loss was determined daily, n 5 each group. (C) Photograph of representative colons and cecums from untreated C57BL/6 mice (top), CD57BL/6 mice treated for 9 days with 2% DSS (middle), and Mbtps1wrt/wrt mice treated for 9 days with 2% DSS (bottom). (D) Representative photomicrographs of colons from WT (C57BL/6) and Mbtps1wrt/wrt mice at days 0 (Upper) and 7 (Lower) of DSS administration. (H&E; magnication: 100 .) (E) Concentrations of IL-1 and IL-6 in supernatants from distal colonic explants cultured for 24 h from WT (C57BL/6) and Mbtps1wrt/wrt mice either before or after 7 days administration of 2% DSS, n 6 7 each group. *, P 0.05, Students t test; n.s., not signicant. (F) Plasma FITC-dextran concentrations in WT (C57BL/6) and Mbtps1wrt/wrt mice 4 h following oral gavage (40 mg/100 g body weight) are shown. Mice with damaged epithelial barrier (C57BL/6 mice, receiving 9 days of 2% DSS) served as a positive control, n 7 8 for WT and Mbtps1wrt/wrt; n 3 for C57BL/6 with 2% DSS; n.s., not signicant.

Fig. 3. Reduced ER stress response in the colon of woodrat mice after DSS administration. Protein extracts from the distal colon of WT (C57BL/6), Mbtps1 /wrt, and Mbtps1wrt/wrt mice before (A and B) and after (C and D) administration of 2% DSS in the drinking water for 3 days were analyzed by Western blotting (A and C) and quantitative image analysis (B and D) with GRP78- and GRP94-specic antibodies. -actin was used as a loading control. n 6 14 (B), n 1114 (D). *, P 0.05; **, P 0.005, Students t test for WT and Mbtps1wrt/wrt; n.s., not signicant.

vulnerability of Mbtps1wrt/wrt mice to DSS is dependent upon commensal flora (Fig. 2B). Macroscopic analysis revealed bleeding in the colons of Mbtps1wrt/wrt mice in contrast to WT mice receiving DSS for 9 days. In addition, the colons of Mbtps1wrt/wrt mice were thickened and shortened, indicating severe edema and inflammation (Fig. 2C). H&E staining of histologic sections taken from the colons of Mbtps1wrt/wrt and WT mice after 7 days of 2% DSS showed dramatic leukocyte recruitment and crypt loss in the mutant strain compared with WT control mice (Fig. 2D). No significant abnormalities were observed in 6- to 8-weekold Mbtps1wrt/wrt controls receiving no DSS. In addition, Mbtps1wrt/wrt mice showed higher levels of IL-1 and IL-6 within epithelium and lamina propria compared with WT mice before and after administration of DSS for 7 days, consistent with mild constitutive colonic inflammation (Fig. 2E). Gut epithelial barrier dysfunction can permit luminal antigens to enter the subepithelial tissues, resulting in the recruitment and activation of leukocytes (12), and potentially leading to IBD. To determine whether the epithelial barrier is intact in Mbtps1wrt/wrt mice, we assessed intestinal permeability by using an FITCdextran absorption assay (4, 12). As a positive control, WT mice receiving 2% DSS for 9 days were used. No abnormality of epithelial permeability could be observed in Mbtps1wrt/wrt mice, suggesting an intact epithelial barrier in Mbtps1wrt/wrt mice (Fig. 2F). To investigate whether the UPR might occur in response to DSS-induced colitis, and whether it might be altered in Mbtps1wrt/wrt mice, GRP78 and GRP94 protein levels were measured in the colons of WT, Mbtps1 /wrt, and Mbtps1wrt/wrt mice before and after administration of DSS. Colons of WT, Mbtps1 /wrt, and Mbtps1wrt/wrt mice showed similar GRP78 and GRP94 levels before induction of colitis with DSS (Fig. 3 A and B). Upon administration of DSS for 3 days, Mbtps1wrt/wrt mice showed significantly lower levels of GRP78 and GRP94 than WT mice. Intermediate expression was observed in Mbtps1 /wrt mice (Fig. 3 C and D). Because ER stress could occur in both hematopoietic and nonhematopoietic cells in the setting of DSS-induced colitis, we generated bone marrow (BM) chimeric mice in which donors and/or recipients were either WT (CD45.1) or Mbtps1wrt/wrt
3302 www.pnas.org cgi doi 10.1073 pnas.0813036106

(CD45.2). Mice were studied 7 weeks after transplantation and were found to be fully chimerized, as assessed by surface staining of BM cells with CD45.1- and CD45.2-specific antibodies (Table S3). Colitis was then induced with DSS for a period of 7 days, and colons were harvested. Western blot analysis was performed to detect GRP78 and GRP94 proteins within inflamed colonic tissues. WT mice reconstituted with Mbtps1wrt/wrt hematopoietic cells expressed GRP78 and GRP94 at levels similar to WT mice reconstituted with WT BM. In contrast, Mbtps1wrt/wrt mice exhibited markedly diminished GRP78 and GRP94 levels, regardless of whether the BM donor was WT or Mbtps1wrt/wrt (Fig. 4A). Immunohistological analysis of different BM chimeric mice also indicated that the UPR was chiefly induced within cells of the colonic epithelium rather than within cells of hematopoietic origin (Fig. 4B). To determine whether epithelial cells or hematopoietic cells of the Mbtps1wrt/wrt genotype were permissive for DSS-induced colitis, chimeric mice were given 2% DSS in the drinking water for 7 days, and weight loss was determined on a daily basis (Fig. 4C). Chimeric mice with Mbtps1wrt/wrt hematopoietic cells showed weight loss similar to control mice receiving transplants of WT BM, whereas Mbtps1wrt/wrt recipients showed significant weight loss, regardless of donor Mbtps1 genotype. These findings indicate that WT Mbtps1 expression in nonhematopoietic cells is essential for S1P-mediated protection in the DSS colitis model. Because the data from the BM chimeric mice suggest reduced ER stress response in the nonhematopoietic compartment, we next isolated colonic epithelial cells from WT and Mbtps1wrt/wrt mice and determined GRP78 and GRP94 expression levels. GRP78 and GRP94 mRNA levels were significantly reduced in epithelial cells from Mbtps1wrt/wrt mice (Fig. 4D). These data suggest that an inadequate UPR in the colonic epithelial compartment may permit DSS-induced inflammation of the colon, thus linking cell-specific ER stress to the induction of organspecific disease. To test the effects of a well-validated form of ER stress on colonic epithelial integrity in woodrat mice, we challenged mice by i.p. injection of tunicamycin. Tunicamycin was found to be markedly more toxic to Mbtps1wrt/wrt animals than WT controls. Whereas 10 of 11 WT mice survived tunicamycin challenge, 10 of 11 woodrat homozygotes succumbed within 30 h (Fig. 4E). Analysis of liver and pancreas did not show any histological abnormalities. However, histological studies revealed severe colitis in Mbtps1wrt/wrt mice but not in WT mice (Fig. 4F), suggesting that integrity of the S1P3ATF63UPR axis is required to protect the colonic epithelium from tunicamycin-induced ER stress. Discussion Here, we have shown that a missense mutation causing impairment of the S1P-mediated, ATF6-dependent UPR in colonic epithelial cells yields increased susceptibility to DSS-induced colitis. Upon induction of colitis, the expression of the ER resident chaperones GRP78 and GRP94 is significantly diminished in the colons of Mbtps1wrt/wrt mice compared with WT controls. This reflects a defective UPR in nonhematopoietic cells. A link between colitis and ER stress has been suggested in several recent studies (4, 6, 13). On the one hand, excessive stress, caused by misfolding of a protein synthesized in abundance by epithelial cells, can lead to colitis (4). On the other hand, inadequacy of the UPR can cause enhanced susceptibility to colitis. Hence, mice lacking XBP1 in intestinal epithelial cells show spontaneous enteritis and increased susceptibility to DSSinduced colitis (6). Deficiency of certain proteins required for the UPR can eventuate ER stress, with consequent up-regulation of other UPR components. For example, XBP1 deficiency in intestinal epithelial cells results in IRE1 hyperactivation and resulting
Brandl et al.

Fig. 4. ER stress in nonhematopoietic cells correlates with the colitis phenotype. (A) Protein extracts from the distal colon of chimeric mice treated with 2% DSS for 7 days were analyzed by Western blotting with GRP78- and GRP94-specic antibodies. -Actin was used as a loading control. Shown are representative extracts from each donorrecipient combination (n 5). (B) Immunohistochemical detection of GRP78 (brown) in parafn-embedded distal colonic sections in different BM chimeric mice. (C) Weight loss in BM chimeric mice was measured daily upon administration of 2% DSS in the drinking water. C57BL/6 mice receiving 0% DSS served as controls. n 5 for WT3woodrat (wrt), woodrat3 WT; WT3 WT and for woodrat3woodrat. Statistical analyses of differences between groups at given days are shown in Table S6. (D) mRNA was extracted from colonic epithelial cells from WT (C57BL/6) and Mbtps1wrt/wrt mice. GRP78 and GRP94 expression was examined by quantitative real-time PCR. Expression levels were normalized to -actin, n 9 10 each group. **, P 0.005; ***, P 0.0005, Students t test. (E) Tunicamycin was injected i.p. into WT (C57BL/6) and Mbtps1wrt/wrt mice, and survival was monitored for 96 h, n 11 each group. ***, P 0.0005, log-rank test. (F) Two representative colonic sections from the distal part of WT and Mbtps1wrt/wrt mice (H&E-stained) are shown. (Magnication: B and F, 200 .)

elevation in cellular content of major ER chaperones, such as GRP78 and GRP94. Mice lacking IRE1 , specifically expressed in intestinal epithelial cells, showed higher GRP78 levels in their colons compared with WT mice (13). However, this observation and others (1416) emphasize the redundancy of the IRE13XBP1 pathway for the induction of major ER chaperones, such as GRP78 and GRP94 in mammalian cells. Moreover, ATF6 cleavage is required for IRE1 dependent induction of UPR-related transcriptional events (15). Lack of IRE1 or XBP1 might therefore lead to a compensatory induction of the ATF6-dependent UPR, explaining elevated levels of GRP78 and GRP94 in those mouse models. Studies in ATF6 null mouse embryonic fibroblasts showed that induction of major ER resident molecular chaperones was impaired (17, 18). In the present study, we have shown that relatively severe deficiency of the ATF6-processing enzyme S1P results in lower protein levels of major ER chaperones in the colon under conditions of ER stress, although not under basal conditions. We infer that in Mbtps1wrt/wrt mice, ATF6 processing is sufficient to support chaperone biosynthesis in the absence of stress but insufficient to allow intestinal epithelial cells to generate a normal UPR, which entails dramatic up-regulation of chaperone production.
Brandl et al.

The consequences of impaired induction of ER chaperones in Mbtps1wrt/wrt epithelial cells would likely include apoptosis, mediated by the IRE13JNK3caspase 12 pathway (1). Indeed, induction of ER stress with tunicamycin in ATF6 -deficient mouse embryonic fibroblasts leads to cell death (18), suggesting that resolution of ER stress depends upon the ATF6 -mediated UPR. The death of epithelial cells would in turn disrupt barrier function together with its innate defense mechanisms and trigger an inflammatory response (6, 1921). The hypomorphic S1P variant present in woodrat mice may also fail to cleave other cellular targets, including ER resident transmembrane bZIP transcription factors. They include the SREBPs, ATF6 , CREbinding protein H (CREBH; also known as CREB3l3), old astrocyte specifically induced substance (OASIS; also known as CREB3l1), leucine zipper protein (lZIP; also known as CREB3), BBF2H7 (also known as CREB3l2), and CREB4 (also known as CREB3l4 or Tisp40). All of these [e.g., CREBH (22), OASIS (23), CREB3 (24), BBF2H7 (25), and CREB3l4 (26)] may at least potentially participate in the UPR. Further work will be required to establish whether any of these bZIP transcription factors contribute to intestinal homeostasis. Materials and Methods
ENU Mutagenesis. C57BL/6J and C57BL/6J Ly5.1 congenics were purchased from the Jackson Laboratories. All mice were kept and bred in The Scripps
PNAS March 3, 2009 vol. 106 no. 9 3303

IMMUNOLOGY

Research Institute Vivarium under the supervision of the Department of Animal Resources. All animal procedures were approved and performed according to institutional guidelines for animal care. C57BL/6J mice were used for ENU mutagenesis to generate the woodrat strain, as described at http:// mutagenetix.scripps.edu. Mapping and Sequencing. wrt/wrt males were mated to C3H/HeN females, and the offspring were intercrossed. The progeny of the F2 generation were phenotyped and genotyped for mapping by analyzing 59 microsatellite markers for genome-wide linkage analysis. For high-resolution mapping, 13 publicly available and 7 previously undescribed markers were used (Tables S4 and S5). Because the mutation abolishes an AIII endonuclease restriction site, genotyping is performed by amplifying genomic DNA with the primers ACACCCTCAGTGCCAGCCAGGAC and ACAAAGTCTGGGCCACGTCACACAG, and digesting the PCR product with the endonuclease AIII. Gene-Trap Vector and ES Cell Culture. The ES cell line PST115 with gene-trap vector specically inactivating the Mbtps1 allele was created by BayGenomics and purchased from Mutant Mouse Regional Resource Centers. Chimera Mouse Production, Breeding, and Genotyping. Blastocysts were harvested from C57BL/6J embryo donor mice mated with B6D2F1 males (C57BL/ 6J DBA2/J hybrids) and injected with PST115 cells. Injected blastocysts were transplanted into pseudopregnant CD-1 recipients to produce chimeric mice. Chimeric male mice were bred to wrt/ females, and agouti offspring were genotyped for gene-trap and Mbtps1 alleles. Induction of DSS Colitis. Sex- and age-matched littermates received 2% DSS (MP Biomedicals) in drinking water for 7 days. Weight was recorded daily. For gut commensal depletion, animals were provided with ampicillin (1 g/L; Sigma), vancomycin (0.5 g/L; Sigma), neomycin sulfate (1 g/L; Sigma), and metronidazole (1 g/L; Baxter) in drinking water for at least 5 days. After that, DSS (2%) was added to the antibiotic-containing water. In Vivo Induction of UPR. For in vivo susceptibility experiments, 11 mice per genotype were injected (i.p.) with 2 mg/kg body weight of tunicamycin (Sigma) solubilized in DMSO and diluted in DMEM, 3% FCS, and 2% penicillin/ streptomycin. Survival was monitored over 96 h. Histology and Immunohistochemistry. Freshly isolated colon was xed in formalin and embedded in parafn. Immunohistochemical staining for GRP78 (Santa Cruz Biotechnology) was performed by using the DAKO Envision System for goat primary antibodies according to the manufacturers protocol. GRP78 antibody was diluted 1:200 in PBS containing 1% BSA. Control slides were stained with goat IgG instead of the primary antibody and did not show any positive staining. H&E staining was performed by using a standard protocol. Real-Time PCR analysis. Colonic RNA was isolated by using TRIzol reagent (Invitrogen). DNase-treated RNA underwent randomly primed cDNA synthesis and real-time PCR analysis. SYBR Green-based real-time PCR was performed by using the DyNAmo SYBR Green qPCR Kit (Finnzymes). GRP78- and GRP94specic primers were obtained from Qiagen, and signals were normalized to
1. Ron D, Walter P (2007) Signal integration in the endoplasmic reticulum unfolded protein response. Nat Rev Mol Cell Biol 8:519 529. 2. Todd DJ, Lee AH, Glimcher LH (2008) The endoplasmic reticulum stress response in immunity and autoimmunity. Nat Rev Immunol 8:663 674. 3. Ye J, et al. (2000) ER stress induces cleavage of membrane-bound ATF6 by the same proteases that process SREBPs. Mol Cell 6:13551364. 4. Heazlewood CK, et al. (2008) Aberrant mucin assembly in mice causes endoplasmic reticulum stress and spontaneous inammation resembling ulcerative colitis. PLoS Med 5:e54. 5. Bertolotti A, Zhang Y, Hendershot LM, Harding HP, Ron D (2000) Dynamic interaction of BiP and ER stress transducers in the unfolded-protein response. Nat Cell Biol 2:326 332. 6. Kaser A, et al. (2008) XBP1 links ER stress to intestinal inammation and confers genetic risk for human inammatory bowel disease. Cell 134:743756. 7. Seidah NG, Chretien M (1999) Proprotein and prohormone convertases: A family of subtilases generating diverse bioactive polypeptides. Brain Res 848:45 62. 8. Seidah NG, et al. (1999) Mammalian subtilisin/kexin isozyme SKI-1: A widely expressed proprotein convertase with a unique cleavage specicity and cellular localization. Proc Natl Acad Sci USA 96:13211326. 9. Sakai J, et al. (1998) Molecular identication of the sterol-regulated luminal protease that cleaves SREBPs and controls lipid composition of animal cells. Mol Cell 2:505514.

-actin. Normalized data were used to quantitate relative levels of GRP78 and GRP94 by using Ct analysis. Western Blot Analysis. Colons were ushed with PBS, and a distal piece (0.5 cm) was homogenized in RIPA buffer. Protein content was determined by BCA assay (Pierce), and protein samples were analyzed by reducing 4 12% SDS/ PAGE (Nupage Bis-Tris-Gel; Invitrogen) and were detected with anti-GRP78 (Santa Cruz Biotechnology) and anti-GRP94 (Cell Signaling Technology) and anti- -actin (Cell Signaling Technology) antibodies. Western blot signals were analyzed by densitometric measurements and subsequently quantied using the National Institutes of Health Image software program. Generation of BM Chimeric Mice. Generation of BM chimeric mice was performed as described recently (19, 20). Recipient WT (CD45.1) or Mbtps1wrt/wrt (CD45.2) mice were lethally irradiated with 950 rad using a 137Cs source and injected intravenously 23 h later with 5 106 BM cells derived from the tibia and femurs of the respective donors. Seven weeks after engrafting, reconstitution was assessed by FACS analysis of bone marrow. After red blood cell lysis, BM cells were stained with FITC-labeled CD45.1 antibody and PerCPCy5.5-labeled CD45.2 antibody. Cell suspensions were analyzed on a BD FACSCalibur. Epithelial Cell Isolation. Colons of WT and Mbtps1wrt/wrt mice were opened longitudinally, rinsed with cold PBS, and cut into pieces of 1-cm lengths. Colonic pieces were placed in ice-cold CMF-HBSS (Gibco) with 2% FCS. After vigorous shaking the supernatant was discarded, and tissue was incubated for 20 min in CMF-HBSS with 10% FCS, 1 mM EDTA, and 1 mM DTT at 37 C with shaking (170 rpm). This step was repeated 3 times, and supernatant was collected. Supernatant was washed 3 times, and cell suspension was sieved through a cell strainer (100 m; BD). Cell suspension was lysed in RIPA buffer for protein extraction or in Trizol for RNA extraction. FITC-Dextran Permeability Assay. Intestinal permeability was assessed by luminal enteral administration of FITC-dextran 4000 (Sigma), a nonmetabolizable macromolecule that is used as a permeability probe. All mice were gavaged with FITC-dextran (40 mg/100 g body weight) 4 h before killing. Whole blood was obtained by cardiac puncture at the time of killing, and FITC-dextran measurements were performed in duplicate by uorometry. Dilutions of FITC-dextran in PBS were used as a standard curve, and absorption of 100 L of serum or standard was measured in a uorometer at 488 nm. Colon Culture and ELISA. One-centimeter segments of the distal part of colon were washed in cold PBS and cultured in a 24-well, at-bottom culture plates (Falcon) in serum-free RPMI medium 1640 (Gibco) supplemented with penicillin and streptomycin. After 24 h, supernatant uid was collected and analyzed for IL-6 and IL-1 production by using ELISA kits from BD (IL-1 ) and eBioscience (IL-6). Statistical Analysis. Statistical analysis was performed on Prism software. All P values 0.05 were considered signicant. Error bars denote the SEMs. ACKNOWLEDGMENTS. We thank Kosuke Fujimoto, Sara Kalina, and Navdeep Singh for excellent technical help, and W. Falk and C. Hofmann (University of Regensburg, Germany) for helpful discussions. This research was supported by the Alexander von Humboldt Foundation through a Feodor Lynen postdoctoral fellowship (to K.B.). S.R. was supported by a Human Frontier postdoctoral fellowship.
10. Mitchell KJ, et al. (2001) Functional analysis of secreted and transmembrane proteins critical to mouse development. Nat Genet 28:241249. 11. Yang J, et al. (2001) Decreased lipid synthesis in livers of mice with disrupted Site-1 protease gene. Proc Natl Acad Sci USA 98:1360713612. 12. Laukoetter MG, et al. (2007) JAM-A regulates permeability and inammation in the intestine in vivo. J Exp Med 204:30673076. 13. Bertolotti A, et al. (2001) Increased sensitivity to dextran sodium sulfate colitis in IRE1beta-decient mice. J Clin Invest 107:585593. 14. Lee AH, Iwakoshi NN, Glimcher LH (2003) XBP-1 regulates a subset of endoplasmic reticulum resident chaperone genes in the unfolded protein response. Mol Cell Biol 23:7448 7459. 15. Lee K, et al. (2002) IRE1-mediated unconventional mRNA splicing and S2P-mediated ATF6 cleavage merge to regulate XBP1 in signaling the unfolded protein response. Genes Dev 16:452 466. 16. Yoshida H, et al. (2003) A time-dependent phase shift in the mammalian unfolded protein response. Dev Cell 4:265271. 17. Wu J, et al. (2007) ATF6alpha optimizes long-term endoplasmic reticulum function to protect cells from chronic stress. Dev Cell 13:351364. 18. Yamamoto K, Yoshida H, Kokame K, Kaufman RJ, Mori K (2004) Differential contributions of ATF6 and XBP1 to the activation of endoplasmic reticulum stress-responsive cis-acting elements ERSE, UPRE and ERSE-II. J Biochem 136:343350.

3304

www.pnas.org cgi doi 10.1073 pnas.0813036106

Brandl et al.

19. Brandl K, et al. (2008) Vancomycin-resistant enterococci exploit antibiotic-induced innate immune decits. Nature 455:804 807. 20. Brandl K, Plitas G, Schnabl B, Dematteo RP, Pamer EG (2007) MyD88-mediated signals induce the bactericidal lectin RegIII{gamma} and protect mice against intestinal Listeria monocytogenes infection. J Exp Med 204:18911900. 21. Sansonetti PJ (2004) War and peace at mucosal surfaces. Nat Rev Immunol 4:953964. 22. Luebke-Wheeler J, et al. (2008) Hepatocyte nuclear factor 4alpha is implicated in endoplasmic reticulum stress-induced acute phase response by regulating expression of cyclic adenosine monophosphate responsive element binding protein H. Hepatology 48:12421250.

23. Murakami T, et al. (2006) Cleavage of the membrane-bound transcription factor OASIS in response to endoplasmic reticulum stress. J Neurochem 96:1090 1100. 24. Liang G, et al. (2006) Luman/CREB3 induces transcription of the endoplasmic reticulum (ER) stress response protein Herp through an ER stress response element. Mol Cell Biol 26:7999 8010. 25. Kondo S, et al. (2007) BBF2H7, a novel transmembrane bZIP transcription factor, is a new type of endoplasmic reticulum stress transducer. Mol Cell Biol 27:1716 1729. 26. Nagamori I, et al. (2006) The testes-specic bZip type transcription factor Tisp40 plays a role in ER stress responses and chromatin packaging during spermiogenesis. Genes Cells 11:11611171.

Brandl et al.

PNAS

March 3, 2009

vol. 106

no. 9

3305

IMMUNOLOGY