Loss of NHE8 expression impairs intestinal mucosal integrity

Abstract

The newest member of the Na⁺/H⁺ exchanger (NHE) family, NHE8, is abundantly expressed at the apical membrane of the intestinal epithelia. We previously reported that mucin 2 expression was significantly decreased in the colon in NHE8^−/− mice, suggesting that NHE8 is involved in intestinal mucosal protection. In this study, we further evaluated the role of NHE8 in intestinal epithelial protection after dextran sodium sulfate (DSS) challenge. Compared with wild-type mice, NHE8^−/− mice have increased bacterial adhesion and inflammation, especially in the distal colon. NHE8^−/− mice are also susceptible to DSS treatment. Real-time PCR detected a remarkable increase in the expression of IL-1β, IL-6, TNF-α, and IL-4 in DSS-treated NHE8^−/− mice compared with DSS-treated wild-type littermates. Immunohistochemistry showed a disorganized epithelial layer in the colon of NHE8^−/− mice. Periodic acid-Schiff staining showed a reduction in the number of mature goblet cells and the area of the goblet cell theca in NHE8^−/− mice. Phyloxine/tartrazine staining revealed a decrease in functional Paneth cell population in the NHE8^−/− small intestinal crypt. The expression of enteric defensins was also decreased in NHE8^−/− mice. The reduced mucin production in goblet cells and antimicrobial peptides production in Paneth cells lead to disruption of the intestinal mucosa protection. Therefore, NHE8 may be involved in the establishment of intestinal mucosal integrity by regulating the functions of goblet and Paneth cells.

to date, 10 Na⁺/H⁺ exchanger (NHE) family members have been discovered in mammals. Each member has its unique tissue distribution and cellular localization, inhibitor sensitivities, and physiological regulation (8, 30, 38, 43). These NHEs have multifunctions in pathological and physiological processes, including electroneutral NaCl transport, acid-base regulation, intracellular pH homeostasis, and cell volume regulation (22, 43). Four NHE members were identified in the alimentary tract of mammalian animals and human beings, including NHE1-3 and NHE8. As a basolateral membrane protein, NHE1 regulates intracellular pH and cell volume in the intestinal epithelial cells (4, 31). NHE2, -3, and -8 are expressed at the apical membrane of enterocytes (39, 43). These apically localized NHEs have distinct developmental expression patterns and transporter kinetic characteristics. NHE2 is sensitive to inhibitor HOE-694, and NHE3 is sensitive to inhibitor S-3226. NHE8 is sensitive to both inhibitors (6, 38). NHE2 is vital for parietal cell viability (21), whereas NHE3 is important in intestinal and renal Na⁺ absorption (34, 35). In the absence of NHE2 expression, NHE3 does not exert a compensatory role, and, conversely, NHE2 cannot compensate for the loss of NHE3 (21). However, NHE8 could partially compensate for the loss of NHE2 and/or NHE3 (41). NHE8 is expressed in many tissue types, and it also contributes to Na⁺ absorption in the epithelial cells (9, 39).

We have previously shown that the expression of mucin 2 (Muc2) was reduced in NHE8^−/− mice (42). As a primary secreted mucin, Muc2 is a key structural component of the mucus layer (16). It coats the intestinal epithelia cells and segregates them from the luminal contents, serving as the first line of defense against the commensal and pathogenic microbes (23). Recently, we found that NHE8^−/− mice are prone to developing gastric ulcers (40). Therefore, we suspected that NHE8 might be an important player in the development of intestinal mucosal integrity. In the current study, we evaluated the mucosal integrity in normal NHE8^−/− mice and dextran sodium sulfate (DSS)-treated NHE8^−/− mice. Our observations showed that the intestinal mucosa in NHE8^−/− mice is prone to bacterial adhesion and penetration, which in turn promotes inflammation. We also discovered that both mucus-producing cells (goblet cells) and antibacterial peptide-producing cells (Paneth cells) were decreased in the NHE8^−/− mice. Thus, NHE8 is involved in regulating mucosal homeostasis in the intestine.

MATERIALS AND METHODS

Animals.

NHE8^−/− mice were generated using NHE8^+/− breeding pairs in a Swiss Webster background as previously described (40). Male mice (8–12 wk old) were used in this study. For the DSS study, wild-type and NHE8^−/− mice were fed with 3% DSS (Affymetrix, Cleveland, OH) for 7 days. Body weight, morbidity, stool consistency, and occult blood in the stool were recorded daily. On the last day, mice were killed, and intestinal mesenteric lymph nodes (MLN), blood, spleen, and liver were collected by asepsis surgery. The ileum, proximal colon, and distal colon were also collected. Disease activity index (DAI) was assessed by an investigator blinded to the experiment according to a combined score method (5, 33). The animal experiment protocol was approved by the University of Arizona Institutional Animal Care and Use Committee.

Gene expression analysis by real-time PCR.

Total RNA was isolated from tissues using TRIzol reagent (Invitrogen, Carlsbad, CA) according to the manufacturer's instructions. RNA (500 ng) was reverse transcripted using Moloney murine leukemia virus reverse transcriptase (Promega, Madison, WI). Cytokine gene expression was analyzed using TaqMan probes (Applied Biosystems, Foster City, CA). Antimicrobial peptides were analyzed by the SYBR Green method (11) using specific primers for defensins (Table 1). The 2^−ΔΔC_T method was applied to calculate gene expression level (25, 37).

Table 1.

Specific primers for quantitative PCR analysis

	Primer	Sequence (5′-3′)	Reference No.
Bacterial group
Eubacteria (total bacteria)	EubacF	TCCTACGGGAGGCAGCAGT	29
	EubacR	GGACTACCAGGGTATCTAATCCTGTT
Bacteroidetes	Bac934F	GGARCATGTGGTTTAATTCGATGAT	12
	Bac1060R	AGCTGACGACAACCATGCAG
Firmicutes	Firm934F	GGAGYATGTGGTTTAATTCGAAGCA	12
	Firm1060R	AGCTGACGACAACCATGCAC
Segmented filament bacteria	SFB736F	GACGCTGAGGCATGAGAGCAT	2
	SFB844R	GACGGCACGGATTGTTATTCA
Lactobacillus	LactoF	GAGGCAGCAGTAGGGAATCTTC	32
	LactoR	GGCCAGTTACTACCTCTATCCTTCTTC
Antimicrobial peptides
Global defensins	Global-defF	GGTGATCATCAGACCCCAGCATCAGT	11
	Global-defR	AAGAGACTAAAACTGAGGAGCAGC
Cryptdin-1	CryptF	TCAAGAGGCTGCAAAGGAAGAGAAC	17
	CryptR	TGGTCTCCATGTTCAGCGACAGC
Cryptdin-related sequence	CRS4CF	GCATGGAATCTGGGTCAAGATAAC	17
	CRS4CR	AGAAGGAAGAGCAATCAAGGCTAAG

Bacterial adhesion quantification.

About 5 mm of tissues from the middle ileum, proximal colon, and distal colon were sampled. Luminal contents were removed gently and rinsed with sterilized PBS. Total DNA and RNA were extracted using an AllPrep DNA/RNA Mini Kit (Qiagen, Valencia, CA). Quantitative analysis of adhesive bacteria was performed using specific primers (Table 1) following the method described previously (24). Quantitative PCR assay was carried out using iQ SYBR Green Supermix (Bio-Rad, Hercules, CA) on the real-time PCR detection system (model CFX96; Bio-Rad).

Histological score of the intestinal tissues.

Standard hematoxylin and eosin (H&E) staining was performed for histological assessment. All sections of the ileum, proximal colon, and distal colon were evaluated in a blinded fashion by two independent observers. Images were taken with a Nikon Digital Sight DS-Fil camera and NIS-Element software (Nikon Instruments, Melville, NY). The sum of scores for inflammatory cell infiltration (score, 0–4), destruction of architecture (score, 0 or 3–4), crypt loss (score, 0 or 3–4), and mucosa thickening (score, 0–4) was used for statistical analysis by the modified method (15). The scores were as follows: immune cell infiltration (0–4): 0 = none or occasional inflammatory cells in the lamina propria, 1 = increased number of inflammatory cells in the lamina propria, 2 = high numbers of inflammatory cells in the lamina propria, 3 = confluent inflammatory cells in the lamina propria, and 4 = confluent inflammatory cells in the lamina propria, extending into the submucosa; destruction of architecture (0 or 3–4): 0 = normal epithelial structure, 3 = epithelial architecture with altered crypt structure and irregular luminal surface, 4 = epithelial architecture with gross structural changes including ulcers; crypt loss (0 or 3–4): 0 = normal crypt frequency, 3 = 5–20% decrease of crypts, 4 = >20% loss of crypts; mucosa thickening (0–4): 0 = normal thickness, 1 = 20–39.9% increase in thickness, 2 = 40–79.9% increase in thickness, 3 = 80–100% increase in thickness, and 4 = >100% increase in thickness.

Fluorescence in situ hybridization staining and immunostaining.

Ileum, proximal colon, and distal colon were collected and fixed in Carnoy's fixative. For fluorescence in situ hybridization (FISH) staining, paraffin-embedded sections were hybridized with bacterial 16S rRNA probe (EUB 338), and DNA was stained by DAPI according to a previously described method (16). For Muc2 staining, paraffin-embedded sections were stained with anti-MUC2C3 antiserum following a previously described method (16). Images were captured with an Nikon Eclipse E1000 microscope using a Plan Fluor ×40/0.75 objective and the NIS Element software (Nikon).

Goblet cell counting and theca area measurement.

Ileum, proximal colon, and distal colon were collected. To count the number of goblet cells, 4- to 5-μm-thick sections were stained with Periodic acid-Schiff's reagent (PAS; Electron Microscopy Sciences, Hatfield, PA). Images were captured with a Nikon Digital Sight DS-Fil camera. Goblet cells were counted for villus-crypt axis of ileum, single crypt of proximal colon, and a defined distance (100 μm) from the surface epithelium of longitudinally cut crypts of distal colon. At least eight crypts were analyzed for each intestinal segment of individual mouse. Mucus-filled theca areas of goblet cells were measured on pictures of PAS-stained sections using the Volocity software (version 6.1; Perkin-Elmer). Ten randomly selected goblet cells (5 goblet cells/section in 2 sections) were measured per mouse sample.

Paneth cell and crypt cell counting in ileum section.

Ileum sections were stained with H&E and phyloxine/tartrazine (PT). Images were captured with a Nikon Digital Sight DS-Fil camera. PT-positive Paneth cells and crypt cells in the same crypt were counted. The proportion of Paneth cells in each crypt was calculated as (PT-positive cells/crypt cells in the same crypt) × 100%. At least eight crypts were analyzed for each mouse.

Cell proliferation assay.

Mice were injected intraperitoneally with 5-bromo-2′-deoxyuridine (BrdU; 50 mg/kg body wt) 2 h before tissue collection. Sections were stained with a BrdU staining kit (Invitrogen) following the manufacturer's manual. Pictures were captured by the method described above. BrdU-positive cells were counted for villus-crypt axis of ileum and for per one vision field of proximal colon and distal colon, respectively.

Statistical analysis.

Unpaired Student's t-test was used to compare values of the experimental data between two groups. One-way ANOVA was used to compare values of the experimental data among multiple groups. A P value <0.05 was considered to be significant.

RESULTS

Clinical symptom evaluation in DSS-treated mice.

Both DSS-treated NHE8^−/− and wild-type mice exhibited clinical symptoms, including body weight loss, bloody stools, and lack of activities. As indicated in Fig. 1, DSS-treated NHE8^−/− mice lost more body weight than wild-type mice after 6 days of DSS treatment (Fig. 1A). DAI was significantly higher in NHE8^−/− mice after 3 days of DSS treatment, and it remained higher through the rest of the treatment (Fig. 1B).

Fig. 1.

Body weight change and disease activity index (DAI) in dextran sodium sulfate (DSS)-treated mice. Mice were fed with 3% DSS for 7 days. Body weight, morbidity, stool consistency, and occult blood in the stool were recorded daily. WT, wild-type mice; KO, Na⁺/H⁺ exchanger 8-deficient (NHE8^−/−) mice. Values are means ± SE from at least 8 mice. #P < 0.05 between WT and NHE8^−/− mice. A: summary of body weight change. B: summary of DAI.

Bacterial translocation in NHE8^−/− mice.

Bacterial translocation is defined by the presence of bacterial DNA in MLNs and blood (10). Bacterial translocation was evaluated by testing the presence of bacterial DNA using the PCR method. As shown in Table 2, a substantial proportion of NHE8^−/− mice was subjected to bacterial penetration in their blood, spleen, and/or liver. Under normal conditions, bacterial translocation was detected in MLNs, blood, spleen, and liver in some NHE8^−/− mice but was only detected in the MLNs in very few wild-type mice. DSS treatment increased bacterial infiltration and translocation both in wild-type mice and NHE8^−/− mice, with a higher incidence of bacterial translocation observed in NHE8^−/− mice.

Table 2.

Bacterial DNA translocation at laparotomy

	MLNs	Blood	Spleen	Liver
WT	1/8 (12.5)	0/8 (0)	0/8 (0)	0/8 (0)
KO	3/8 (37.5)	1/8 (12.5)	2/8 (25)	1/8 (12.5)
WT + DSS	4/8 (50)	2/8 (25)	1/8 (12.5)	1/8 (12.5)
KO + DSS	7/8 (87.5)	5/8 (62.5)	5/8 (62.5)	2/8 (25)

Units are no. of mice with percent in parentheses.

WT, wild-type mice; KO, Na⁺/H⁺ exchanger 8-deficient mice; MLNs, mesenteric lymph nodes; DSS, dextran sodium sulfate.

Cytokine expression and histological score of inflammation.

To test if loss of NHE8 gene impairs intestinal mucosal protection, cytokine gene expression was evaluated in wild-type mice and NHE8^−/− mice. As shown in Fig. 2A, only IL-1β expression was elevated in the ileum in NHE8^−/− mice compared with wild-type mice. The expression of IL-1β, IL-4, IL-6, and TNF-α was significantly increased in the colon in NHE8^−/− mice compared with wild-type mice. DSS treatment did not affect cytokine expression in the ileum in neither NHE8^−/− mice nor wild-type mice, but significantly increased the colonic expression of IL-1β, IL-4, IL-6, TNF-α, and IL-2 in NHE8^−/− mice and wild-type mice. The colonic cytokine expression increase induced by DSS treatment was much higher in NHE8^−/− mice than that in wild-type mice (Fig. 2, B and C). Histological examination agreed with the cytokine expression pattern. As indicated in Fig. 3, only small changes were seen in the ileum in DSS-treated NHE8^−/− mice, but no changes were seen in DSS-treated wild-type mice (Fig. 3A). Instead, extensive ulceration of the epithelial layer, edema, crypt damage, and infiltration of granulocytes and mononuclear cells in the mucosa were seen in the colon in DSS-treated mice with more severe changes detected in DSS-treated NHE8^−/− mice (Fig. 3, B and C). Histological inflammation score also reflected the severity of DSS damage in NHE8^−/− mice. The inflammation scores in DSS-treated NHE8^−/− mice were statistically higher than in DSS-treated wild-type mice (Fig. 3D).

Fig. 2.

Inflammatory cytokine gene expression analysis. Intestinal segments (ileum, proximal colon, and distal colon) were collected and used for total RNA isolation. Real-time PCR was used to compare cytokine gene expression in the intestine. Values are means ± SE from 8 mice. #P < 0.05 and *P < 0.01 between WT mice and NHE8^−/− mice, DSS-treated WT mice and DSS-treated NHE8^−/− mice, WT mice and DSS-treated WT mice, and NHE8^−/− mice and DSS-treated NHE8^−/− mice. A: cytokine expression in the ileum. B: cytokine expression in the proximal colon. C: cytokine expression in the distal colon.

Fig. 3.

Tissue hematoxylin and eosin (H&E) stain and histology score. Mice were harvested, and intestinal tissues were collected for standard H&E staining. Sections were used for histological assessment. Data are presented as means ± SE from 8 mice. #P < 0.05 and *P < 0.01 between WT mice and NHE8^−/− mice, DSS-treated WT mice and DSS-treated NHE8^−/− mice, WT mice and DSS-treated WT mice, and NHE8^−/− mice and DSS-treated NHE8^−/− mice. A: ileum H&E stain. B: proximal colon H&E stain. C: distal colon H&E stain. D: histology score for ileum, proximal colon, and distal colon.

BrdU incorporation in intestinal epithelial cells.

An active and persistent inflammatory response is characterized by enhanced cell proliferation (13). To confirm if loss of NHE8 results in abnormal intestinal proliferation, NHE8^−/− mice and their wild-type littermates were injected intraperitoneally with BrdU. As shown in Fig. 4A, the proliferation zone of the intestinal epithelium in wild-type mice was limited to the bottom of the crypts, whereas the proliferation zone was expanded in the entire crypt region in NHE8^−/− mice. The number of BrdU-positive cells in NHE8^−/− mice was also significantly increased per villus-crypt axis in the ileum or per field of vision in the colon compared with wild-type mice (Fig. 4B). These results suggest that hyperproliferation is present in the intestine in the absence of NHE8 function.

Fig. 4.

Proliferation in the intestine in WT mice and NHE8^−/− mice. Mice were injected with BrdU at a dose of 50 mg/kg body wt. Intestinal tissues were collected 2 h after injection. Tissue sections were stained with a BrdU staining kit. BrdU-positive cells were counted for villus-crypt axis of ileum and for per one vision field of proximal colon and distal colon, respectively. Data are presented as means ± SE from 3 mice. *P < 0.01 between WT mice and NHE8^−/− mice. A: representative 5-bromo-2′-deoxyuridine (BrdU) immunostaining in the ileum, proximal colon, and distal colon in WT and NHE8^−/− mice. B: summary graph of BrdU-positive cells per villus-crypt axis in ileum, per field of vision in proximal colon and distal colon.

Abundance of bacterial adhesion at the mucosal surface in NHE8^−/− mice.

In general, severe inflammation correlates with a large number of bacteria in contact with the epithelial layer (20). To determine if the extensive inflammation observed in DSS-treated NHE8^−/− mice involves an altered bacterial adhesion, quantitative PCR was used to analyze the major bacteria groups attached to the intestinal epithelial surface. As shown in Fig. 5, bacteria burden adhered to ileal and colonic mucosal surface were increased in NHE8^−/− mice and in DSS-treated wild-type and NHE8^−/− mice. In the ileum, the total bacteria and primary bacterial groups in DSS-treated NHE8^−/− mice were all remarkably increased with the exception of Lactobacillus species and segmented filament bacteria. In the colon, the increased bacteria was mainly associated with Bacteroidetes, Firmicutes, Lactobacillus bacteria, and segment filament bacteria. Interestingly, the dysbiosis of the bacteria flora in DSS-treated NHE8^−/− mice was different between the ileum and colon, where DSS treatment increased the adhesion of Bacteroidetes and Firmicutes in the colon but decreased the adhesion of Lactobacillus bacteria in the ileum.

Fig. 5.

Quantitative analysis of bacterial adhesion. Real-time PCR was used to compare bacterial DNA abundance in tissue samples. Eubac, eubacteria; Firm, Firmicutes; Bac, Bacteroidetes; Lac, Lactobacillus; SFB, segmented filament bacteria. Data are presented as means ± SE from 8 mice. #P < 0.05 and *P < 0.01 between WT mice and NHE8^−/− mice, DSS-treated WT mice and DSS-treated NHE8^−/− mice, WT mice and DSS-treated WT mice, and NHE8^−/− mice and DSS-treated NHE8^−/− mice. A: bacterial adhesion in the ileum. B: bacterial adhesion in the proximal colon. C: bacterial adhesion in the distal colon. Intestinal segments were collected and used for DNA isolation.

FISH stain of bacteria in the distal colon in mice.

Because dramatic changes were observed in the colon, we examined bacterial loads by utilizing the FISH staining method. As shown in Fig. 6, bacteria were only detected in the outer mucus layer and were separated from the inner layer of the distal colon in wild-type mice. However, increasing amounts of bacteria were seen in contact with the epithelia in the colon of NHE8^−/− mice. Upon DSS treatment, bacteria were seen in contact with the epithelia, and sometimes they were found in the upper part of the crypt in wild-type mice, while large amounts of bacteria were found inside colonic tissue and/or cells in the area with ulcers in DSS-treated NHE8^−/− mice.

Fig. 6.

Bacteria localization in the colonic tissue in mice. Distal colon was sectioned and detected by the general bacterial probe EUB338-Alexa Fluor 555 (red). Nuclei were counterstained with Hoechst DNA stain (blue). Arrows indicate infiltrated bacteria in the epithelial cells. Images were retrieved with ×40 objective.

PAS stain and PAS-positive cell counting.

Bacterial penetration in the inner mucus layer appears to be associated with a compromised mucosal integrity (15). Previously, we reported that Muc2 gene expression was reduced in the colon of NHE8^−/− mice (42). Here, we further compared mucin expression in the entire intestinal tract in wild-type mice and NHE8^−/− mice using the PAS technique. As shown in Fig. 7, A–C, a significant reduction of PAS-positive goblet cells was observed in the ileum and the colon in NHE8^−/− mice compared with that of wild-type mice, as well as decreased staining in the goblet cell theca of NHE8^−/− mice, indicating less stored mucus. The reduced Muc2 production was also confirmed by Muc2 immunohistochemical stain (Fig. 7D).

Fig. 7.

Periodic acid-Schiff (PAS) stain of goblet cells, PAS-positive cell counting, and mucin 2 (Muc2) staining. Intestinal tissues were collected from mice. Tissue sections were stained with PAS and then observed under a microscope. Goblet cells were counted for villus-crypt axis of ileum, single crypt of proximal colon, and a defined distance (100 μm) from the surface epithelium of longitudinally cut crypts of distal colon. Mucus-filled theca areas were measured in goblet cells on PAS-stained sections using the Volocity software. Data are presented as means ± SE from 8 mice. *P < 0.01 between WT mice and NHE8^−/− mice. A: representative PAS stain in the ileum, proximal colon, and distal colon. B: summary results of PAS-positive goblet cell counting. C: summary data of the area of the goblet cell theca. D: Muc2 immunohistochemical staining.

Reduced intestinal defensins expression in NHE8^−/− mice.

To explore other possible mechanism(s) by which NHE8^−/− mice were vulnerable to bacteria penetration and inflammation, we studied Paneth cell function by staining the secreted granules of Paneth cells and detecting the mRNA expression level of defensins in NHE8^−/− mice and wild-type mice. As shown in Fig. 8A, Paneth cell granules stained by PT were decreased in the crypt in NHE8^−/− mice compared with that in wild-type mice. The number of PT-positive cells in the crypts was also significantly reduced in NHE8^−/− mice (Fig. 8B). At the gene expression level, the expression of global defensins was decreased in NHE8^−/− mice. This change was primarily associated with a remarkable decrease in expression of cryptdin 1 and cryptdin-related sequence-4c (Fig. 8C).

Fig. 8.

Paneth cell counting and defensins expression. Ileal tissue was collected from mice, and tissue sections were stained with H&E and phyloxine/tartrazine (PT). PT-positive Paneth cells and crypt cells in the same crypt were counted. The proportion of Paneth cells in each crypt was calculated as (PT-positive cells/crypt cells in the same crypt) × 100%. Ileal RNA was also isolated and used for defensins gene expression. Data are presented means ± SE from 8 mice. *P < 0.01 between WT mice and NHE8^−/− mice. A: representative section stained with PT. B: PT-positive Paneth cell proportion in the crypts. C: expression of the defensins genes.

DISCUSSION

The digestive tract is colonized by a huge number of microbes and is constantly exposed to ingested antigens and potential pathogens. The functional integrity of the intestinal mucosa depends on the coordinated regulation of the mucus layer, the intercellular tight junction, the epithelium cells, and the host innate and adoptive immune response (7). The mucus layer overlying the epithelium is secreted by the goblet cells. This mucus layer in colon is made up of an outer, loosely adherent layer and an inner, firmly adherent layer. Gut microbes are mainly present in the outer layer and are absent from the inner layer (19). In the small intestine the mucus makes up a diffusion barrier that largely contains various host defense molecules produced by goblet cells, Paneth cells, and absorptive enterocytes restricting bacteria to be in close contact with the epithelium (3). Disruption in the intestinal mucus barrier could result in inflammation and epithelial cell damage (27). Intestinal inflammation is associated with bacteria penetration (14). In this current study, we found that the increased bacterial adhesion in NHE8^−/− mice was correlated with elevated inflammatory cytokine expression. As demonstrated by others (16), the inner layer of normal healthy colon forms a barrier that separates bacteria from the epithelium. Muc2, as a primary structural component of the mucosal layer, plays an important role in maintaining the integrity of protective mucus barriers (15, 16). Loss of Muc2 expression in mice displayed a diminished intestinal mucus layer and increased permeability and bacterial adherence to the epithelial cell surface (36). We previously reported that NHE8^−/− mice have reduced Muc2 gene expression and were susceptible to developing gastric ulcers (42). Here, we further confirmed that loss of NHE8 function in mice resulted in increased inflammatory cytokine expression in the ileum and in the colon. In correlation with the inflammation, a high proliferation rate was also detected in the ileum and the colon in NHE8^−/− mice. These observations suggested that NHE8 function is involved in mucosal protection in the intestine.

Other NHE isoforms, such as NHE3, have been shown to be involved in mucosal protection. Mice lacking NHE3 expression were susceptible to DSS-induced epithelial injury (18). In our current study, we also showed that NHE8^−/− mice were vulnerable to DSS treatment. DSS-treated NHE8^−/− mice developed rapid body weight loss, higher DAI, frequent bacterial translocation, and inflammatory response. Thus, NHE8 is important for protecting intestinal epithelia from bacterial invasion. Because no obvious alteration in the expression of claudin 3, a tight junction protein, was observed in NHE8^−/− mice, and because DSS treatment reduced claudin 3 expression by one-half in both wild-type mice and NHE8^−/− mice (data not shown), the susceptibility of NHE8^−/− mice to DSS treatment is unlikely the result of abnormal tight junction expression. In fact, our findings in this study and previous reports confirmed that the impaired mucin synthesis in NHE8^−/− mice is due to the decrease of goblet cell function (42). These results may at least partly explain the higher recurrence of bacteria translocation in NHE8^−/− mice.

Antimicrobial peptides (AMPs) are produced by Paneth cells and are a vital host defense molecule of the first line of gut defense, which could effectively kill the bacteria that get into the mucus and close to the epithelium. AMPs are released on mucosal surfaces as effectors of innate immunity (28). A recent study indicated that α-defensins of Paneth cells secreted in the small intestinal lumen persist as intact and functional forms throughout the intestinal tract (26). Paneth cell dysfunction induced by loss of autophagy-related 16-like (ATG16L1) gene expression contributes to spontaneous intestinal inflammation (1). In NHE8^−/− mice, the expression of AMPs in Paneth cells was significantly decreased, suggesting an important role of NHE8 in Paneth cell function. Taken together, the increased susceptibility to DSS in NHE8^−/− mice may be due to the deficiency of AMPs and mucin production in the intestine. Therefore, NHE8 may affect the functions of goblet cells and Paneth cells by modulating their maturation and/or differentiation, but more studies will be necessary to address this question.

In conclusion, this study reports for the first time that loss of NHE8 function in mice resulted in impaired intestinal mucosal integrity. NHE8^−/− mice not only displayed a significant decrease in goblet cells and Muc2 production but also a remarkable decrease in antimicrobial peptide production. The reduced mucin layer and the compromised antibacterial capacity in NHE8^−/− mice resulted in a significant increase in bacteria load that may contribute to bacteria penetration, bacteria translocation, and subsequent intestinal inflammation. Therefore, NHE8 is involved in maintaining mucosal integrity via its roles in modulating the functions of both goblet cells and Paneth cells.

GRANTS

This work was funded by National Institute of Diabetes and Digestive and Kidney Diseases Grant R01-DK-073638, the National Natural Science Foundation of China (No. 81172600, 81272101), and the Swedish Research Council.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: A.W., H.X., and F.K.G. conception and design of research; A.W., J.L., Y.Z., M.E.J., and H.X. performed experiments; A.W., J.L., Y.Z., M.E.J., and H.X. analyzed data; A.W., J.L., Y.Z., M.E.J., and H.X. interpreted results of experiments; A.W., M.E.J., and H.X. prepared figures; A.W. and H.X. drafted manuscript; H.X. and F.K.G. edited and revised manuscript; H.X. and F.K.G. approved final version of manuscript.