Regulation by gut commensal bacteria of carcinoembryonic antigen-related cell adhesion molecule expression in the intestinal epithelium Yasuaki Kitamura1,2, Yoji Murata1*, Jung-ha Park1, Takenori Kotani1, Shinya Imada1, Yasuyuki Saito1, Hideki Okazawa1, Takeshi Azuma2 and Takashi Matozaki1* 1

Division of Molecular and Cellular Signaling, Department of Biochemistry and Molecular Biology, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan 2 Division of Gastroenterology, Department of Internal Medicine, Kobe University Graduate School of Medicine, 7-5-1 Kusunoki-cho, Chuo-ku, Kobe 650-0017, Japan

Carcinoembryonic antigen-related cell adhesion molecule (CEACAM) 1 and CEACAM20, immunoglobulin superfamily members, are predominantly expressed in intestinal epithelial cells (IECs) and co-localized at the apical surface of these cells. We here showed that the expression of mouse CEACAM1 and CEACAM20 at both mRNA and protein levels was markedly reduced in IECs of the small intestine by the treatment of mice with antibiotics against Gram-positive bacteria. The expression of both proteins was also decreased in IECs of the small intestine from germ-free mice, compared with that from control specific-pathogen-free mice. Exposure of intestinal organoids to IFN-c markedly increased the expression of either CEACAM1 or CEACAM20, whereas the exposure to TNF-a increased the expression of the former protein, but not that of the latter. In contrast, the expression of CEACAM20, but not of CEACAM1, in intestinal organoids was markedly increased by exposure to butyrate, a short-chain fatty acid produced by bacterial fermentation in the intestine. Collectively, our results suggest that Gram-positive bacteria promote the mRNA expression of CEACAM1 or CEACAM20 in the small intestine. Inflammatory cytokines or butyrate likely participates in such effects of commensal bacteria.

Introduction The mammalian intestinal tract harbors trillions of commensal bacteria, which are separated from the inner body by a single layer of intestinal epithelial cells (IECs) and form a symbiotic relationship with their host. These bacteria are thought to be important not only for gene expression by the immune cells of many cytokines, but also for that by IECs (Hooper et al. 2001; Kayama & Takeda 2012; Alexander et al. 2014). Indeed, germ-free mice manifested the decreased expression of mRNAs for ZO-1 and occludin (Shimada et al. 2013), both of which are key molecules for the formation and maintenance of epithelial tight junction. Moreover, mice treated with antibiotics exhibited the altered expression pattern of Communicated by: Yoshimi Takai *Correspondence: [email protected] or [email protected]

578

Genes to Cells (2015) 20, 578–589

genes related to cell cycle in IECs with the reduction of epithelial cell proliferation (Reikvam et al. 2011). By contrast, colonization by symbiotic bacteria of germ-free mice promoted the mRNA expression for RegIIIc, a secreted C-type lectin, which has the antimicrobial activity against Gram-positive bacteria, in Paneth cells of the small intestine (Cash et al. 2006). The carcinoembryonic antigen-related cell adhesion molecule (CEACAM) family is a subgroup of the immunoglobulin (Ig) superfamily and consists of 12 or 8 family members in human or mouse, respectively (Kammerer & Zimmermann 2010). Among this family members, Ceacam1 gene is present in both human and mouse (Kammerer & Zimmermann 2010). Ceacam1 mRNA is found in various tissues including the intestine (Zebhauser et al. 2005), and CEACAM1 protein is expressed at the apical surface of IECs (Zalzali et al. 2008). CEACAM1 consists of alternative splicing isoforms in human and mouse (Houde et al. 2003; Beauchemin & Arabzadeh 2013). CEACAM1-L, an

DOI: 10.1111/gtc.12247 © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Regulation of CEACAM expression

isoform of CEACAM1, is a transmembrane protein carrying Ig-like domains in the extracellular region, and the cytoplasmic region of this protein contains two immunoreceptor tyrosine-based inhibitory motifs (ITIMs). Forced expression of CEACAM1-L in colon carcinoma cells inhibited cell proliferation likely through the action of the ITIMs in the cytoplasmic region (Izzi et al. 1999). In contrast, ablation of Ceacam1 gene in mice increased cell proliferation and decreased apoptosis of normal colonic epithelial cells (Leung et al. 2006), suggesting that CEACAM1 participates in the negative regulation of IEC growth. The expression of CEACAM1 is shown to be promoted by infection with Neisseria gonorrhoeae of human ovarian surface epithelial cells (Muenzner et al. 2002). Furthermore, the expression of CEACAM1 is also promoted by the stimulation of colon carcinoma cell lines with TNF-a or IFN-c (Ou et al. 2009). However, it remains unknown whether commensal bacteria or inflammatory cytokine promotes the expression of CEACAM1 in normal IECs. CEACAM20 is another CEACAM family member, of which genomic gene is found in both human and mouse (Kammerer & Zimmermann 2010). From the cDNA sequence, it is predicted that CEACAM20 is a transmembrane protein with four Ig-like domains in its extracellular region. In contrast to CEACAM1, CEACAM20 has an immunoreceptor tyrosine-based activation motif (ITAM) in its cytoplasmic region (Zebhauser et al. 2005). Mouse mRNA for CEACAM20 was shown to be expressed in limited tissues such as the intestine and testis (Zebhauser et al. 2005). CEACAM20, as well as CEACAM1, is also shown to be expressed in human prostate epithelial cells and implicated in prostate morphogenesis in vitro (Zhang et al. 2013). We have recently found that CEACAM20 is predominantly expressed in colonic epithelial cells (Y. Murata and T. Matozaki, a manuscript submitted). However, the expression pattern of, as well as the mechanism for gene expression of, CEACAM20 in IECs remains to be elucidated. Therefore, we have now investigated whether gut commensal bacteria regulate the expression of CEACAM1 or of CEACAM20 in the intestinal epithelium.

Results Localization and expression of CEACAM1 or CEACAM20 in mouse small intestine and colon

We first examined the localization of CEACAM1 and CEACAM20 proteins in the small intestine and colon. Immunohistofluorescence analysis with the

monoclonal antibodies (mAbs) to CEACAM1 showed that CEACAM1 was localized at the apical surface of IECs in both the small intestine and colon (Fig. 1A, upper panels), whereas staining for b-catenin was present at sites of cell–cell adhesion as described previously (Sadakata et al. 2009). Staining for CEACAM1 overlapped with that of ezrin/radixin/moesin-binding phosphoprotein of 50 kDa (EBP50) (Fig. 1B), which is concentrated at the microvilli of IECs (Sadakata et al. 2009), suggesting that CEACAM1 is localized to the microvilli of IECs. We have recently found that CEACAM20 is localized at the apical surface of colonic epithelial cells (Y. Murata and T. Matozaki, a manuscript submitted). By immunohistofluorescence analysis with the polyclonal antibodies (pAbs) to CEACAM20, we indeed confirm such result in colonic sections (Fig. 1A, right lower panels). We also found that CEACAM20 was expressed and localized at the apical surface of IECs in the small intestine (Fig. 1A, left lower panels). Furthermore, the immunoreactivity of CEACAM20 was well co-localized with that for CEACAM1 at the apical surface of IECs (Fig. 1C). Immunoblot analysis also showed that both CEACAM1 and CEACAM20 were expressed in IECs isolated from different parts of mouse small intestine (duodenum, jejunum and ileum) and of mouse colon (cecum, proximal and distal colon) (Fig. 1D). These results thus suggest that both CEACAM1 and CEACAM20 are predominantly expressed in IECs and co-localized to the microvilli of these cells. Role of gut commensal bacteria in the regulation of CEACAM1 or CEACAM20 expression in the intestine

It was previously showed that the expression of CEACAM1 in human ovarian surface epithelial cells was markedly increased by the infection of these cells with Neisseria gonorrhoeae (Muenzner et al. 2002). We thus investigated whether gut commensal bacteria are important for the expression of CEACAM1, as well as of CEACAM20, in IECs. Specific pathogen-free (SPF) mice were provided with a mixture of antibiotics [ampicillin, vancomycin, metronidazole and neomycin] in drinking water for 4 weeks. The reduction of commensal bacteria by the antibiotics was confirmed by the amounts of bacterial 16S rDNA in mice feces (Fig. S1 in Supporting Information). We found that the protein levels of CEACAM1 and CEACAM20 were markedly decreased in IECs of the small intestine from the antibiotic-treated mice,

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Genes to Cells (2015) 20, 578–589

579

Y Kitamura et al. Small intestine

(A)

Colon

CC1/ -catenin

CC20/ -catenin

Small intestine

(B)

Colon

Small intestine

(C)

CC1

CC1

EBP-50

CC20

Merge

Merge

Colon

(D)

kDa 200 116

kDa 116 -CC1

Blot 116 97.5

Blot

-CC1

Blot

-CC20

116 97.5 Blot

-CC20 66

66 45 Blot

- -tubulin

45 Blot

- -tubulin

Figure 1 Localization and expression of CEACAM1 or CEACAM20 in mouse small intestine and colon. (A-C) Cryostat sections of the small intestine and colon from adult mice were subjected to immunofluorescence staining with mAbs to CEACAM1 (CC1, A-C), mAbs to b-catenin (A), pAbs to CEACAM20 (CC20, A and C) and pAbs to EBP-50 (B), and staining with DAPI (B and C, blue). Boxed regions in the left panels are shown at higher magnification in the right panels (A). Merged images are also shown in the lower panels (B and C). Scale bars, 20 lm (A–C). Images are representative from at least three separate experiments. (D) Lysates of IECs from the small intestine (duodenum, jejunum and ileum, left panel) and colon (cecum, proximal and distal colon, right panel) were subjected to immunoblot analysis with mAbs to CEACAM1, pAbs to CEACAM20 or mAbs to b-tubulin (loading control). The positions of molecular size standards are indicated on the left. Data are representative from at least three separate experiments.

580

Genes to Cells (2015) 20, 578–589

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Regulation of CEACAM expression

compared with those for the untreated control mice, whereas the protein levels of EBP-50 were not different between these two groups (Fig. 2A). By contrast, the protein levels of these CEACAMs in colonic epithelial cells were not changed by the treatment with antibiotics (Fig. 2A). Immunostaining showed that the immunoreactivity of CEACAM1 or CEACAM20 was markedly lower in the ileum of the antibiotictreated mice than those apparent with the untreated mice, whereas EBP-50 immunoreactivity was not

different between two groups (Fig. 2B). Consistently, the levels of Ceacam1 and Ceacam20 mRNAs were significantly decreased in IECs of the small intestine from the antibiotic-treated mice, compared with those for the untreated control mice (Fig. 2C). Similar to the antibiotic-treated mice, germ-free mice manifested a marked decrease in the protein levels for CEACAM1 and CEACAM20 in IECs of the small intestine (Fig. 3A). In contrast, the protein levels of these CEACAMs in colonic epithelial cells

Relative intensity

(A)

Blot

-CC1 -CC20

Water Abx

1.5 1.0 0.5 0

-EBP-50

Water

(B)

Abx

CC1

CC20

Colon

(C)

Relative mRNA (Normalized to GAPDH)

Small intestine

- -tubulin

Water Abx 1.5 1.0 0.5 0

EBP-50

Figure 2 Role of gut commensal bacteria in the regulation of CEACAM1 or CEACAM20 expression in the intestine. (A) Mice were untreated (water) or treated with antibiotics (Abx) including ampicillin (1 g/L), vancomycin (0.5 g/L), metronidazole (1 g/ L) and neomycin (1 g/L) for 4 weeks, after which lysates of IECs from the small intestine and colon were subjected to immunoblot analysis with mAbs to CEACAM1 (CC1), pAbs to CEACAM20 (CC20), pAbs to EBP-50 or mAbs to b-tubulin (left panel). Immunoblots were subjected to densitometric analysis, and the ratio of the intensity of the band for CEACAM1, CEACAM20 or EBP-50 to that for b-tubulin was calculated. Data are expressed relative to the corresponding value for the untreated mice (right panel). (B) Cryostat sections of the ileum from mice untreated or treated with antibiotics as in (A) were subjected to immunofluorescence staining with mAbs to CEACAM1 (upper panels), pAbs to CEACAM20 (middle panels), or pAbs to EBP-50 (lower panels) and staining with DAPI (blue). Boxed regions in the left panels are shown at higher magnification in the right panels. Images are representative from at least three separate experiments. Scale bars, 20 lm. (C) Mice were untreated or treated with antibiotics as in (A), and the expression of Ceacam1 or Ceacam20 mRNA in IECs from the small intestine was then evaluated by quantitative PCR. The expression level of each mRNA was normalized to that of glyceraldehydes-3-phosphate dehydrogenase (GAPDH). Data are expressed relative to the corresponding value for the untreated mice. Data are means  SE from four (A) or three (C) separate experiments (A, n = 4 for each condition; C, n = 3 for each condition). *P < 0.05; **P < 0.01; ***P < 0.001 (Student’s t-test). © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Genes to Cells (2015) 20, 578–589

581

Y Kitamura et al.

(A)

Relative intensity

SPF GF

-CC20

1.0 0.5 0

- -tubulin

(B)

Small intestine SPF

GF

CC1

Colon (C)

Relative mRNA (Normalized to GAPDH)

Blot

-CC1

1.5

1.5

SPF GF

1.0 0.5 0

CC20

Figure 3 Reduced expression of CEACAM1 or CEACAM20 in the small intestine of germ-free mice. (A) Lysates of IECs from the small intestine and colon of germ-free (GF) or specific pathogen-free (SPF) mice were subjected to immunoblot analysis with mAbs to CEACAM1 (CC1), pAbs to CEACAM20 (CC20) or mAbs to b-tubulin (left panel). Immunoblots were subjected to densitometric analysis, and the ratio of the intensity of the band for CEACAM1 or CEACAM20 to that for b-tubulin was calculated. Data are expressed relative to the corresponding value for SPF mice (right panel). (B) Cryostat sections of the ileum from GF or SPF mice were subjected to immunofluorescence staining with mAbs to CEACAM1 or pAbs to CEACAM20, and staining with DAPI (Blue). Images are representative from at least three separate experiments. Scale bar, 100 lm. (C) The expression of Ceacam1 or Ceacam20 mRNA in IECs from the small intestine of GF and SPF mice was evaluated by quantitative PCR. The expression level of each mRNA was normalized to that of GAPDH. Data are expressed relative to the corresponding value for SPF mice. Data are means  SE from five (A) or three (C) separate experiments (A, n = 5 for each condition; C, n = 3 for each condition). **P < 0.01; ***P < 0.001 (Student’s t-test).

were not different between germ-free mice and SPF mice (Fig. 3A). Immunostaining also showed the marked reduction of CEACAM1 and CEACAM20 in the apical surface of the ileum from germ-free mice (Fig. 3B). Furthermore, the levels of Ceacam1 and Ceacam20 mRNAs were significantly decreased in IECs of the small intestine from germ-free mice, compared with those for the control SPF mice (Fig. 3C). These results thus suggest that gut commensal bacteria promote the expression of 582

Genes to Cells (2015) 20, 578–589

CEACAM1 and CEACAM20 in IECs of the small intestine. Involvement of Gram-positive bacteria in the expression of CEACAM1 or CEACAM20 in the small intestine

We next examined which antibiotics in the mixture are indeed effective for the suppression of CEACAM expression. SPF mice were treated for 4 weeks with

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Regulation of CEACAM expression

Gram-positive, bacteria (Ivanov et al. 2008; Puhl et al. 2012). Thus, Gram-positive commensal bacteria are likely important for promoting the expression of CEACAM1 or CEACAM20 in the small intestine.

ampicillin, vancomycin, metronidazole or neomycin individually in drinking water. The protein expression levels of CEACAM1 and CEACAM20 in IECs of the small intestine were markedly decreased by treatment with either ampicillin or vancomycin, compared with those for the control (Fig. 4A), whereas metronidazole failed to affect, neomycin slightly increased (although not significantly), the protein levels of these CEACAMs (Fig. 4A). Immunostaining of the ileum for CEACAM1 and CEACAM20 also confirmed such an effect of vancomycin or neomycin (Fig. 4B). Consistently, treatment with vancomycin, but not with neomycin, significantly reduced the mRNA levels of CEACAM1 and CEACAM20 in IECs of the small intestine (Fig. 4C). Ampicillin and vancomycin are thought to be effective for Gram-positive bacteria, whereas neomycin is preferably effective for Gram-negative, but not for

Gut commensal bacteria are thought to promote the expression by immune cells of cytokines, such as TNF-a or IFN-c, in the intestine (Kayama & Takeda 2012; Alexander et al. 2014). Quantitative polymerase chain reaction (PCR) analysis showed that the treatment of mice with vancomycin indeed resulted in a significant decrease in the mRNA expression of TNF-a and IFN-c in the small intestine (Fig. 5A). Given that the expression of CEACAM1 or CEACAM20 was markedly reduced in the small intestine

Relative intensity

Small intestine

(A)

-CC1

Blot

Role of TNF-a and IFN-c in the expression of CEACAM1 or CEACAM20 in intestinal organoids

-CC20

3 2 1 0

- -tubulin CC1

Water

VCM

NM

(C)

CC1

CC20

Relative mRNA (Normalized to GAPDH)

(B)

CC20

1.5 1.0 0.5 0

Ceacam1

Ceacam20

Figure 4 Involvement of Gram-positive bacteria in the expression of CEACAM1 or CEACAM20 in the small intestine. (A) Mice were untreated (Water) or treated with either ampicillin (ABPC, 1 g/L), vancomycin (VCM, 0.5 g/L), metronidazole (MNZ, 1 g/L) or neomycin (NM, 1 g/L) for 4 weeks, after which lysates of IECs from the small intestine were subjected to immunoblot analysis with mAbs to CEACAM1 (CC1), pAbs to CEACAM20 (CC20) or mAbs to b-tubulin (left panel). Immunoblots were subjected to densitometric analysis, and the ratio of the intensity of the band for CEACAM1 or CEACAM20 to that for b-tubulin was calculated. Data are expressed relative to the value corresponding to the untreated mice (right panel). (B) Cryostat sections of the ileum from mice untreated or treated with either vancomycin or neomycin as in (A) were subjected to immunofluorescence staining with mAbs to CEACAM1 or pAbs to CEACAM 20, and staining with DAPI (Blue). Images are representative from three separate experiments. Scale bar, 100 lm. (C) Mice were untreated or treated with either vancomycin or neomycin as in (A), and the expression of Ceacam1 or Ceacam20 mRNA in IECs from the small intestine was then evaluated by quantitative PCR. The expression level of each mRNA was normalized to that of GAPDH. Data are expressed relative to the corresponding value for the untreated mice. Data are means  SE from three independent experiments (A and C, n = 3 for each condition). *P < 0.05; **P < 0.01 (Student’s t-test). © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Genes to Cells (2015) 20, 578–589

583

Y Kitamura et al.

Relative mRNA (Normalized to GAPDH)

(A)

(B) kDa 200

4 3

Water VCM

2

116 Blot: -CC1 116 97.5

1 Blot: -CC20 66

0

Blot: - -tubulin

(D) 4

-CC1 Blot

3 2 1

-CC20 - -tubulin

0

Ceacam1 Ceacam20

Relative intensity

Relative mRNA (Normalized to GAPDH)

(C)

6 4 2 0

CC1

CC20

Figure 5 Role of TNF-a and IFN-c in the expression of CEACAM1 or CEACAM20 in intestinal organoids. (A) Mice were untreated (water) or treated with vancomycin (VCM, 0.5 g/L) for 4 weeks. The expression of Tnfa or Ifng mRNAs in the small intestine was then evaluated by quantitative PCR. The expression level of each mRNA was normalized to that of GAPDH. Data are expressed relative to the corresponding value for the untreated mice. (B) Lysates of mouse intestinal organoid were subjected to immunoblot analysis with mAbs to CEACAM1 (CC1), pAbs to CEACAM20 (CC20) or mAbs to b-tubulin. The positions of molecular size standards are indicated on the left. (C) Mouse intestinal organoids were stimulated with vehicle (-), TNF-a (100 ng/mL) or IFN-c (100 ng/mL) for 12 h. The expression of Ceacam1 or Ceacam20 mRNA was then evaluated by quantitative PCR. The expression level of each mRNA was normalized as in (A). Data are expressed relative to the corresponding value for the control organoids. (D) Mouse intestinal organoids were stimulated for 24 h as in (C) and were then subjected to immunoblot analysis as in (B) (upper panel). Immunoblots were subjected to densitometric analysis, and the ratio of the intensity of the band for CEACAM1 or CEACAM20 to that for b-tubulin was calculated. Data are expressed relative to the corresponding value for the control organoids (lower panel). Data are means  SE from three (A), four (C) or five (D) independent experiments (A, n = 3 for each condition). *P < 0.05; ***P < 0.001 (Student’s t-test).

584

Genes to Cells (2015) 20, 578–589

of germ-free mice or mice treated with antibiotics, TNF-a or IFN-c likely promotes the expression of these CEACAM family members in IECs. We therefore investigated whether TNF-a or IFN-c promotes the expression of mRNA for CEACAM1, as well as for CEACAM20, in intestinal organoids prepared from mouse small intestine. We confirmed by immunoblot analysis that both CEACAM1 and CEACAM20 proteins are endogenously expressed in intestinal organoids (Fig. 5B). Intestinal organoids were stimulated with either TNF-a or IFN-c for 12 h. Quantitative PCR analysis showed that the mRNA expression levels for both CEACAM1 and CEACAM20 were significantly increased by the exposure of intestinal organoids to IFN-c (Fig. 5C). By contrast, the exposure of intestinal organoids to TNF-a had no effect on the mRNA expression of CEACAM20, whereas it markedly promoted that of CEACAM1 (Fig. 5C). Consistent with these results, the protein levels of CEACAM1 and CEACAM20 in intestinal organoids were increased by their exposure to IFN-c (Fig. 5D). Stimulation of intestinal organoids with TNF-a also tended to promote the expression level of CEACAM1 protein, whereas it had no effect on the expression level of CEACAM20 protein (Fig. 5D). These results suggest that IFN-c promotes the mRNA expression for both CEACAM1 and CEACAM20 in IECs of the small intestine. However, TNF-a likely promotes the mRNA expression for CEACAM1, but not for CEACAM20. Role of short-chain fatty acids in the expression of CEACAM1 or CEACAM20 in intestinal organoids

Gut commensal bacteria participate in the production of short-chain fatty acids (SCFAs), such as acetate, propionate and butyrate, by fermentation of dietary fiber (Guilloteau et al. 2010). Indeed, the levels of SCFAs were markedly decreased in the intestine of antibiotic-treated or germ-free mice compared with those for control mice (Wichmann et al. 2013). Given that the expression of CEACAM1 or CEACAM20 was markedly reduced in the small intestine of germ-free mice and mice treated with antibiotics, SCFAs likely regulate the expression of CEACAM1 or CEACAM20 in IECs. We therefore investigated the effect of acetate, propionate or butyrate on the expression of CEACAM1 or CEACAM20 in intestinal organoids. The expression level of Ceacam20 mRNA was significantly increased by the exposure of intestinal organoids to butyrate, but not acetate or propionate (Fig. 6A). Treatment with a mixture of

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Regulation of CEACAM expression

We further investigated the molecular mechanism by which butyrate promotes CEACAM20 expression in intestinal organoids. Butyrate is thought to inhibit the activity of histone deacetylases (HDACs) that participate in the regulation of gene expression through deacetylation of histones or transcription factors (Waldecker et al. 2008). Treatment of intestinal organoids with trichostatin A (TSA), an inhibitor for HDACs, promoted the mRNA expression for CEACAM20 and such an effect was also comparable to that apparent with butyrate (Fig. 6C), suggesting that the promotion by butyrate of Ceacam20 mRNA expression is

acetate, propionate and butyrate had the same effect on the level of Ceacam20 mRNA in intestinal organoids as that apparent with butyrate alone (Fig. 6A). Moreover, the protein level of CEACAM20 was also increased by the exposure of intestinal organoids to butyrate (Fig. 6B). In contrast, either the mRNA or protein level of CEACAM1 was not changed by the exposure of intestinal organoids to any SCFAs or to their mixture (Fig. 6A, B). Collectively, these results suggest that butyrate participates in the promotion of CEACAM20 expression in IECs of the small intestine. (B) 5

-CC1 Blot

3 2 1 0

-CC20 - -tubulin

Ceacam1

Relative mRNA (Normalized to GAPDH)

(C)

Ceacam20

Relative intensity

N.S.

4

4 3 2 1 0

CC1 CC20

(D) 4 3 2 1 0

Relative mRNA (Normalized to GAPDH)

Relative mRNA (Normalized to GAPDH)

(A)

1.5 1.0 0.5 0

+ Butyrate

Figure 6 Role of short-chain fatty acids in the expression of CEACAM1 or CEACAM20 in intestinal organoids. (A) Mouse intestinal organoids were stimulated with vehicle (-), acetate (1 mM), propionate (1 mM), butyrate (1 mM) or a mixture of these three SCFAs (mix, 1 mM each) for 12 h. The expression of Ceacam1 or Ceacam20 mRNA was then evaluated by quantitative PCR. The expression level of each mRNA was normalized to that of GAPDH. Data are expressed relative to the corresponding value for the control organoids. (B) Mouse intestinal organoids were stimulated with vehicle (-) or butyrate (1 mM) for 24 h and were then subjected to immunoblot analysis with mAbs to CEACAM1 (CC1), pAbs to CEACAM20 (CC20) or mAbs to b-tubulin (left panel). Immunoblots were subjected to densitometric analysis, and the ratio of the intensity of the band for CEACAM1 or CEACAM20 to that for b-tubulin was calculated. Data are expressed relative to the corresponding value for the control organoids (right panel). (C) Mouse intestinal organoids were stimulated with vehicle (-), butyrate (0.2 mM) or trichostatin A (TSA, 50 nM) for 12 h. The expression of Ceacam20 mRNA was then evaluated by quantitative PCR. The expression level of Ceacam20 mRNA was normalized and data are expressed as in (A). (D) Mouse intestinal organoids were treated with butyrate (0.2 mM) in the presence of either vehicle (-), a p38 MAPK inhibitor (SB203580, 20 lM), a JNK inhibitor (SP600125, 10 lM) or a MEK inhibitor (PD98059, 100 lM) for 12 h. The expression of Ceacam20 mRNA was then evaluated by quantitative PCR. The expression level of Ceacam20 mRNA was normalized and data are expressed as in (A). Data are means  SE from three (A, C and D) or five (B) independent experiments. *P < 0.05; **P < 0.01 (A and C, one-way ANOVA followed by Tukey’s post hoc test; B and D, Student’s t-test). N.S., not significant. © 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Genes to Cells (2015) 20, 578–589

585

Y Kitamura et al.

attributed to the inhibition of HDACs. In addition, HDAC inhibitors such as butyrate or TSA are shown to activate mitogen-activated protein kinases (MAPKs), such as extracellular signal-regulated kinase-1/2 (Erk-1/2), p38 MAPK and c-Jun amino-terminal kinase (JNK) (Ding et al. 2001; Mandal et al. 2001; Zhong et al. 2003). The expression by butyrate of Ceacam20 mRNA in intestinal organoids was prevented by treatment with SP600125 (an inhibitor for JNK), but not that with PD98059 (an inhibitor for mitogen-activated protein kinase/extracellular signal-regulated kinase kinase) nor SB203580 (an inhibitor for p38) (Fig. 6D). These results suggest that the activation of JNK is involved in the promotion by butyrate of Ceacam20 mRNA expression in intestinal organoids.

Discussion We here showed that CEACAM1 and CEACAM20 proteins are expressed and localized at the apical surface of IECs in the small intestine and colon. The mRNA and protein levels of CEACAM1 and CEACAM20 in IECs of the small intestine were markedly reduced in germ-free mice and in mice treated with antibiotics effective against Gram-positive but not Gram-negative bacteria, suggesting that Gram-positive bacteria promote the expression of these CEACAMs in IECs of the small intestine. We also showed that the expression levels of CEACAM1 and CEACAM20 in intestinal organoids were increased by exposure to IFN-c. In contrast, stimulation with TNF-a promoted the expression of CEACAM1. Indeed, treatment of mice with vancomycin (effective against Gram-positive bacteria) reduced the expression of Ifng or Tnfa mRNA in the small intestine. Thus, Gram-positive bacteria are likely to promote the expression of CEACAM1 or CEACAM20 in IECs of the small intestine, at least in part, through the action of these inflammatory cytokines. Interestingly, stimulation with butyrate, not other short-chain fatty acids, promoted the expression of CEACAM20 in intestinal organoids. Indeed, Gram-positive bacteria, such as Clostridium species, are thought to contribute to the production of butyrate in the intestine (Guilloteau et al. 2010). Therefore, butyrate is important for promotion by Gram-positive bacteria of CEACAM20 expression in IECs of the small intestine. The physiological significance for expression by commensal bacteria of CEACAM1 or CEACAM20 in the small intestine remains unclear. However, it is known that commensal bacteria participate in the regulation of IEC proliferation in the small intestine 586

Genes to Cells (2015) 20, 578–589

(Khoury et al. 1969; Jones et al. 2013). In particular, we have recently found that Gram-positive bacteria are important for such function of commensal bacteria (J.-H. Park and T. Matozaki, unpublished data). By contrast, CEACAM1 is thought to prevent proliferation or survival of IECs through the protein tyrosine phosphatase, Shp1, bound to the ITIMs in the cytoplasmic region of this protein (Izzi et al. 1999; Leung et al. 2006). Although the physiological function of CEACAM20 in the intestine remains poorly understand, CEACAM20 is implicated to promote proliferation or differentiation of prostate epithelial cells (Zhang et al. 2013). In addition, the cytoplasmic region of CEACAM20 contains the ITAM that binds the protein tyrosine kinase (Ivashkiv 2009). Thus, CEACAM20 and CEACAM1 likely play a positive or a negative role in IEC proliferation, respectively. Furthermore, commensal bacteria regulate the proliferation of IECs likely through promoting the expression of these two CEACAMs in IECs. The molecular mechanism by which IFN-c or TNF-a regulates the expression of Ceacam1 or Ceacam20 mRNA in IECs also remains unknown. However, it was showed that the transcription factor IRF-1 interacts with the interferon-sensitive response element, which is indeed present in the Ceacam1 promoter, and that IRF-1 indeed participates in IFNc-induced expression of Ceacam1 mRNA in human colon carcinoma cells (Chen et al. 1996). Moreover, the activation of transcription factor NF-jB was shown to be important for the TNF-a-induced expression of Ceacam1 mRNA in ovarian surface epithelial cells (Muenzner et al. 2002). Thus, it is likely that IRF-1 and NF-jB participate in the promotion of CEACAM1 or CEACAM20 expression in normal IECs by IFN-c or TNF-a, respectively. In this study, we also investigated the molecular mechanism by which butyrate promotes the mRNA expression of CEACAM20 in IECs. We showed that the treatment of intestinal organoids with TSA, a HDAC inhibitor, promoted the expression of Ceacam20 mRNA in intestinal organoids. In contrast, an inhibitor of JNK prevented the butyrate-induced mRNA expression of CEACAM20. Given that butyrate is an inhibitor for HDACs (Waldecker et al. 2008) and the inhibition of HDAC promotes activation of JNK (Mandal et al. 2001; Zhong et al. 2003), butyrate promotes the expression of CEACAM20 in IECs likely through inhibiting HDACs and thereby activating JNK. JNK is known to phosphorylate and activate the transcription factor c-Jun. c-Jun induces transcription of genes that contain a binding site for the activation

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Regulation of CEACAM expression

protein-1 (AP-1) complex composed of members of Jun and Fos families (Karin 1995). The JNK/AP-1 signaling is also shown to participate in butyrate-dependent gene expression in colon carcinoma cells (Mandal et al. 2001). Therefore, the JNK/AP-1 signaling is likely to be important for the promotion by butyrate of CEACAM20 expression in normal IECs.

Experimental procedures Antibodies and reagents A rat mAb (for immunofluorescence staining) to and a mouse mAb (for immunoblot analysis) to CEACAM1 were from R&D Systems (Minneapolis, MN) and BD Biosciences (San Diego, CA), respectively. A mouse mAb to b-catenin was from BD Biosciences. A mouse mAb to b-tubulin was from Sigma-Aldrich (St. Louis, MO). Rabbit pAbs specific for mouse CEACAM20 were generated against glutathione S-transferase (GST) fusion proteins containing the cytoplasmic region (amino acids 485–577) of CEACAM20 (GST-CEACAM20-CP). Rabbits were injected with GST-CEACAM20CP proteins, and the resulting pAbs to CEACAM20 were purified from serum by the use of columns containing GST or GST-CEACAM20-CP proteins immobilized on CNBrSepharose (GE Healthcare; Waukesha, WI) as described previously (Mori et al. 2010). Rabbit pAbs to EBP-50 were from Abcam (Cambridge, UK). Secondary antibodies labeled with Cy3 or Alexa488 for immunofluorescence analysis were obtained from Jackson ImmunoResearch (West Grove, PA) or ThermoFisher (Waltham, MA), respectively, and 4’,6-diamino-2-phenylindole (DAPI) was from Nacalai tesque (Kyoto, Japan). Horseradish peroxidase-conjugated goat secondary polyclonal antibodies to rabbit or mouse IgG for immunoblot analysis were from Jackson ImmunoResearch. TNF-a and IFN-c were from Peprotech (Rocky Hill, NJ) and ThermoFisher, respectively. TSA was from Sigma-Aldrich. SB203580 and PD98059 were from EMD Millipore (Billerica, MA). SP600125 was from Wako (Osaka, Japan).

brief, the freshly isolated duodenum, jejunum, ileum, cecum, and proximal and distal colon were washed with phosphatebuffered saline (PBS), cut into small pieces, washed three times with Hanks’ balanced salt solution (HBSS) containing 1% fetal bovine serum and 25 mM 4-(2-hydroxyethyl) piperazine-1ethanesulfonic acid (HEPES)-NaOH (pH 7.5) and then incubated four times on a rolling platform for 15 min at room temperature in HBSS containing 50 mM ethylenediaminetetraacetic acid (EDTA) and 25 mM HEPES-NaOH (pH 7.5). The tissue debris was removed, and IECs in the resulting supernatant were isolated by centrifugation at 2509 g for 10 min at 4°C and washed three times with PBS.

Immunoblot analysis Cells were washed with ice-cold PBS and then lysed with RIPA buffer (50 mM Tris–HCl, pH 8.0, 150 mM NaCl, 1 mM EDTA, 1% Triton X-100, 0.5% sodium deoxycholate and 0.1% sodium dodecyl sulfate) containing protease inhibitor cocktail for mammalian (Nacalai tesque). The lysates were centrifuged at 17,5009 g for 15 min at 4°C, and the resulting supernatants were subjected to immunoblot analysis as previously described (Murata et al. 2010).

Immunofluorescence analysis The small intestine and colon were fixed for 2 h on ice with 4% paraformaldehyde (PFA) in PBS, transferred to a series of sucrose solutions (7, 20 and 30% [w/v], sequentially) in PBS for cryoprotection, embedded in OCT compound (Sakura, Tokyo, Japan) and rapidly frozen in liquid nitrogen. Frozen sections with a thickness of 5 lm were prepared with a cryostat, mounted on glass slides and air-dried. The sections were then subjected to immunofluorescence analysis with primary antibodies and fluorescent dye-labeled secondary antibodies as described previously (Yamashita et al. 2014). Images were obtained with a confocal laser scanning microscope (LSM710, Zeiss).

Depletion of gut commensal bacteria Mice Wild-type C57BL/6J mice were obtained from CLEA Japan, Inc (Tokyo, Japan) and maintained in the Institute for Experimental Animals at Kobe University Graduate School of Medicine under SPF conditions. Germ-free mice (C57BL/6J background) were provided form CLEA Japan, Inc. All animal experiments were carried out according to the guidelines of the Animal Care and Experimentation Committee of Kobe University.

Isolation of mouse IECs The isolation of mouse IECs was carried out with a slight modification as previously described (Murata et al. 2010). In

Gut commensal bacteria were depleted as previously described (Rakoff-Nahoum et al. 2004). Four-week-old mice were treated with a mixture of ampicillin (Wako; 1 g/L), neomycin sulfate (Nacalai tesque; 1 g/L), metronidazole (Wako; 1 g/L) and vancomycin (Wako; 500 mg/L), or with each antibiotic in drinking water during the course of experiment.

Isolation of RNA and quantitative PCR analysis Total RNA was isolated from the freshly isolated whole small intestine, IECs and intestinal organoids by the use of Sepasol RNA I (Nacalai tesque) or RNeasy mini kit (QIAGEN, Hilden, Germany), and the first-strand cDNA was synthesized from 0.5 lg of isolated total RNA by the use of QuantiTect

© 2015 The Molecular Biology Society of Japan and Wiley Publishing Asia Pty Ltd

Genes to Cells (2015) 20, 578–589

587

Y Kitamura et al. Reverse Transcription kit (QIAGEN) according to the manufacturer’s instructions. cDNA fragments of interest were amplified by FastStart SYBR Green Master on LightCycler 480 (Roche Applied Science, Penzberg, Germany) according to the manufacturer’s instructions. The amplification results were analyzed by the use of LightCycler 480 software (Roche Applied Science) and then normalized to the expression level of glyceraldehydes-3-phosphate dehydrogenase (GAPDH). The primers used for quantitative real-time PCR were as follows: Ceacam1, 5’-AATCTGCCCCTGGCGCTTGGAGCC-3’ and 5’-AAATCGCACAGTCGCCTGAGTACG-3’; Ceacam20, 5’-GTCCAACCCTGTCACCAACT-3’ and 5’-CCAAGGT TGGGAACTCGATA-3’; Ifng, 5’-ATGAACGCTACACACT GCATC-3’ and 5’-CCATCCTTTTGCCAGTTCCTC-3’; Tnfa, 5’-CCCTCACACTCAGATCATCTTCT-3’ and 5’GCTACGACGTGGGCTACAG-3’; Gapdh, 5’-AGGTCGGT GTGAACGGATTTG-3’ and 5’-TGTAGACCATGTAGTT GAGGTCA-3’.

Statistical analysis Data are presented as means  SE and were analyzed by Student’s t-test for comparing the difference between two groups or by one-way ANOVA followed by Tukey’s post hoc test for comparisons among three or more groups. A P value of

Regulation by gut commensal bacteria of carcinoembryonic antigen-related cell adhesion molecule expression in the intestinal epithelium.

Carcinoembryonic antigen-related cell adhesion molecule (CEACAM) 1 and CEACAM20, immunoglobulin superfamily members, are predominantly expressed in in...
2MB Sizes 0 Downloads 11 Views