Inhibition of autotaxin alleviates inflammation and increases the expression of sodium-dependent glucose cotransporter 1 and Na⁺/H⁺ exchanger 3 in SAMP1/Fc mice

Abstract

Crohn’s disease (CD) is a chronic, relapsing, inflammatory disease that is often associated with malnutrition because of inflammation in the small intestine. Autotaxin (ATX) is a secreted enzyme that produces extracellular lysophosphatidic acid. Increasing evidence suggests that ATX is upregulated during inflammation, and inhibition of ATX has been effective in attenuating chronic inflammatory conditions, such as arthritis and pulmonary fibrosis. This study aims to determine whether inhibition of ATX alleviates CD-associated inflammation and malnutrition by using SAMP1/Fc mice, a model of CD-like ileitis. SAMP1/Fc mice were treated the ATX inhibitor PF-8380 for 4 wk. Inhibition of ATX led to increased weight gain in SAMP1/Fc mice, decreased T helper 2 cytokine expression, including IL-4, IL-5, and IL-13, and attenuated immune cell migration. SAMP1/Fc mice have low expression of Na⁺-dependent glucose transporter 1 (SGLT1), suggesting impaired nutrient absorption associated with ileitis. PF-8380 treatment significantly enhanced SGLT1 expression in SAMP1/Fc mice, which could reflect the increased weight changes. However, IL-4 or IL-13 did not alter SGLT1 expression in Caco-2 cells, ruling out their direct effects on SGLT1 expression. Immunofluorescence analysis showed that the expression of sucrase-isomaltase, a marker for intestinal epithelial cell (IEC) differentiation, was decreased in inflamed regions of SAMP1/Fc mice, which was partially restored by PF-8380. Moreover, expression of Na⁺/H⁺ exchanger 3 was also improved by PF-8380, suggesting that suppression of inflammation by PF-8380 enhanced IEC differentiation. Our study therefore suggests that ATX is a potential target for treating intestinal inflammation and restoration of the absorptive function of the intestine.

NEW & NOTEWORTHY This study is the first, to our knowledge, to determine whether autotoxin (ATX) inhibition improves inflammation and body weights in SAMP1/Fc mice, a mouse model of ileitis. ATX inhibition increased body weights of SAMP1/Fc mice and increased Na⁺-dependent glucose transporter 1 (SGLT1) expression. Increased SGLT1 expression in the inflamed regions was not a direct effect of cytokines but an indirect effect of increased epithelial cell differentiation upon ATX inhibition.

INTRODUCTION

Inflammatory bowel diseases (IBDs), including Crohn’s disease (CD) and ulcerative colitis, are a debilitating chronic disorder affecting more than 2.5 million individuals in Western countries (50). The etiology of IBD is not yet fully understood despite significant advances in recent years with identification of more than 100 susceptible genes (65). It is widely accepted that IBD is a condition that primarily involves a hyperimmune response in genetically susceptible individuals when mucosal barrier integrity and/or gut microbiota are perturbed because of changes in environmental factors (57). Maintaining a homeostatic state of the intestinal epithelial cells (IECs), microflora populations, and immune responses is the key for a healthy gut and for the healing of IBD. CD4⁺ T-cell species, including T helper 1 (Th1), Th17, and T regulatory cells, are the most influential in the pathogenesis of IBD (8, 42). CD displays as mucosal ulceration and inflammation most commonly in the distal small intestine and sometimes in the colon. Given the pivotal roles of aberrant immune responses in the pathogenesis and disease persistence of IBD, immunosuppressors, including biological drugs that inhibit the activity of TNF-α (11), IL-12 (46), or IL-23 (48) are commonly used in IBD treatment in clinics. In addition to direct suppression of proinflammatory factor production, drugs that suppress immune cell homing to the gut have shown promising outcomes in the treatment of IBD (18, 21, 28, 58).

Autotaxin (ATX) is a unique ecto-nucleotide pyrophosphatase/phosphodiesterase with intrinsic lysophospholipase D activity, which coverts lysophosphatidylcholine to lysophosphatidic acid (LPA). It is the major producer of extracellular LPA and is essential for vascular and neural development (64). ATX is expressed in many organs and tissues, especially high in the brain, spinal cord, lymph nodes, and adipose tissue (16, 26, 29, 30, 54). Increasing evidence points to an important role of ATX in transendothelial migration of T cells. ATX secreted from high endothelial venules (HEVs) interacts with T cells to promote T cell migration across HEVs, thereby facilitating extravasation of T cells to secondary lymphoid organs (29). Consistently, ATX expression by HEVs in the intestine has been reported (26). Interestingly, enteroendocrine cells in human but not rodent intestines express ATX (7). ATX is expressed as a pre-proenzyme, which is secreted upon proteolytic cleavage of the N-terminal signal peptide and further trimmed by a protease (27). Secreted ATX binds to integrin or heparan sulfate on the cell surface that enables the localization of LPA-producing ATX in the vicinity of LPA receptors of the target cells (17, 25). LPA mediates a diverse range of growth factor-like effects, including cell migration, proliferation, and cytokine induction, via a family of G protein-coupled receptors, LPA1–LPA6 (10). Our previous studies have demonstrated the causal link between LPA2 and colorectal cancer in rodents (39, 40). On the other hand, LPA1 is important in maintenance of the intestinal epithelial barrier functions and integrity (33, 38). Moreover, the effect of LPA on membrane transporters, including the cystic fibrosis transmembrane conductance regulator and Na⁺/H⁺ exchanger (NHE3) have been reported (35, 41). Therefore, the effects of LPA in the intestinal tract are complex, and how ATX inhibition impacts intestinal physiology and pathology is not readily apparent. There are ongoing efforts to develop ATX inhibitors to treat chronic inflammatory diseases, including rheumatoid arthritis, fibrosis, and cancer (53). A recent study showed that inhibition of ATX with a nonspecific ATX inhibitor bithionol ameliorated dextran sodium sulfate-induced colitis (26). Although this study suggested anti-inflammatory effects of bithionol, bithionol is an antiparasitic drug approved for human use as a second-line oral medication for the treatment of helminthic infections (1). Therefore, further studies are needed to understand fully the therapeutic potential of ATX inhibition for the treatment of gastrointestinal diseases.

In this study, we determined whether inhibition of ATX activity by using a small molecular inhibitor ameliorates inflammation in SAMP1/Fc mice that elicit CD-like ileitis with mucosal ulceration and inflammation in the distal small intestine (55). We found that inhibition of ATX attenuated the homing of immune cells into the gut and decreased the production of proinflammatory cytokines. ATX inhibition resulted in increased body-weight gain in SAMP1/Fc mice, suggesting improved nutrient absorptive process. Our study further showed that inhibition of ATX improved IEC differentiation and partially restored membrane expression of Na⁺-dependent glucose transporter 1 (SGLT1) and NHE3 in the ileum.

MATERIALS AND METHODS

Animal treatment.

Five- to seven-week-old male SAMP1/Fc (also known as SAMP1/YitFcsJ or SAMP1/YitFc) and male AKR mice were purchased from the Jackson Laboratory (Bar Harbor, ME). Mice were maintained at a local animal facility until the age of 8 wk. The ATX inhibitor PF-8380, solubilized in 65% DMSO-H₂O, was given orally at the dose of 30 mg/kg body wt daily for 4 wk. Control animals received the same volume of DMSO-H₂O as the control. Body weight changes were monitored daily. At the end of the treatment mice were euthanized, and a 10-cm segment of the distal ileum proximal to the cecum was isolated. The intestine was opened longitudinally, and approximately one-fourth of the intestine was excised along the longitudinal direction for RNA extraction. The rest of the tissue was made into a Swiss roll, fixed in 10% formalin, and subjected to the preparation of paraffin sections. All animal experiments were carried out under approval by the Institutional Animal Care and Use Committee of the Atlanta Veterans Administration Medical Center and Emory University and in accordance with the NIH’s Guide for the Care and Use of Laboratory Animals.

ATX activity assay.

Blood was collected from SAMP1/Fc and AKR mice at 16–18 h after the last PF-8380 or control treatment. Serum was then prepared, and ATX activity was measured by using Autotaxin Activity Assay kit (Echelon Biosciences, Inc., Salt Lake, UT) according to the manufacturer’s instructions.

Immunofluorescence and immunohistochemistry.

Paraffin-embedded intestinal tissue sections were deparaffinized and dehydrated in a graded series of xylene and ethanol. Antigen unmasking was performed through pressure cooker treatment in sodium citrate buffer (10 mM sodium citrate, 0.05% Tween 20, pH 6.0). Immunofluorescence staining procedures were as previously described (22). Briefly, tissue sections were permeated with PBS containing 0.2% Triton-X 100 for 20 min, followed by washes. Then tissue sections were blocked with 5% goat serum before they were incubated with rat anti-CD3 (eBioscience), mouse anti-sucrase-isomaltase (SI) (Santa Cruz Biotechnology), rabbit anti-SGLT1 (Abcam), or rabbit anti-NHE3 (Alpha Diagnostics) antibody for 1 h at room temperature. After three washes with PBS, sections were incubated with Alexa Fluor 488- or Alexa Fluor 555-conjugated goat anti-rabbit IgG and/or Hoechst 33342 for 30 min at room temperature. ATX was stained with Alexa Fluor 647-conjugated mouse anti-ATX antibody (5H3) (Abcam). After three washes with PBS, the specimens were mounted with ProLong Gold Antifade Reagent (Invitrogen) and observed under a Nikon fluorescence microscope. For immunohistochemistry, tissue sections were incubated with 3% H₂O₂-methanol solution before incubation with rat anti-CD45 antibody (BioLegend) overnight at 4°C.

Western blotting and cell surface biotinylation.

Caco-2 cells were grown on Transwell filters (Corning) in DMEM supplemented with 1 mM sodium pyruvate, 50 U/ml penicillin, 50 μg/ml streptomycin, and 10% FBS in a 5% CO₂ humidified incubator at 37°C. Cells were cultured for 2–3 wk postconfluence to ensure full differentiation. To determine SGLT1 protein expression, Caco-2 cells were treated from the basolateral side with IL-4, IL-13, or IL-4 + IL-13 for the indicated time. Cells were then lysed in in lysis buffer (Cell Signaling Technology) containing 20 mM Tris·HCl (pH 7.5), 150 mM NaCl, 1 mM β-glycerophosphate, 2.5 mM sodium pyrophosphate, 1 mM Na₂EDTA, 1 mM EGTA, 1 mM Na₃VO₄, 1 μg/ml leupeptin, 1% Triton X-100, and protease inhibitor mixture tablets (Roche). The crude lysate was sonicated for 2 × 15 s and spun at 14,000 g for 15 min. Protein concentration was determined by the bicinchoninic acid assay (Sigma). A total of 40 μg of protein lysates was loaded on SDS-PAGE gel, separated, and transferred to nitrocellulose membrane for immunoblotting with a rabbit anti-SGLT1 antibody (Novus Biologicals). Densitometric analysis was performed using ImageJ software (National Institutes of Health, Bethesda, MD).

Surface biotinylation of SGLT1 was performed as previously described (24). Briefly, Caco-2 treated with IL-4, IL-13, or IL-4 + IL-13 were rinsed twice in cold PBS and incubated for 10 min in borate buffer (154 mM NaCl, 7.2 mM KCl, 1.8 mM CaCl₂, and 10 mM H₃BO₃, pH 9.0). Cells were then incubated for 40 min with 0.5 mg/ml NHS-SS-biotin (Pierce) in borate buffer. Unbound NHS-SS-biotin was quenched with Tris buffer (20 mM Tris, 120 mM NaCl, pH 7.4). Cells were then rinsed with PBS, scraped, and lysed in the lysis buffer described above. An aliquot of supernatant was retained as the total fraction representing the total cellular NHE3, and 200 µg of lysate were then incubated with streptavidin-agarose beads (Pierce) for 2 h. The strepavidin-agarose beads were washed three times in lysis buffer and twice in PBS. All the above procedures were performed at 4°C or on ice. Biotinylated surface proteins were then eluted by boiling the beads at 95°C for 10 min. Dilutions of the total and surface SGLT1 were resolved by SDS-PAGE and immunoblotted with anti-SGLT1 antibody.

Quantitative RT-PCR.

Total RNA was extracted from the intestinal tissues using RNeasy Mini kit (Qiagen). Three micrograms of total RNA were used for cDNA synthesis using the First Strand cDNA Synthesis kit (Invitrogen) according to the manufacturer’s instructions. Quantitative PCR was performed using iQ SYBR Green Supermix (Bio-Rad) on the Eppendorf Mastercycler Realplex. PCR primer sequences are listed in Table 1.

Table 1.

Primers used for real-time PCR

	Primer
ATX
Forward	TCT AGC ATC CCA GAG CAC CT
Reverse	CGT TTG AAG GCA GGG TAC AT
IL-4
Forward	GCT ATG TGT CAT CCT GCT CTT C
Reverse	GGC GTC CCT TCT CCT GTG
IL-5
Forward	GCT TCT GCA CTT GAG TGT TCT G
Reverse	CCT CAT CGT CTC ATT GCT TGT C
IL-13
Forward	TTG CTT GCC TTG GTG GTC TC
Reverse	GGG AGT CTG GTC TTG TGT GAT G
TNF-α
Forward	TGG TAG CAA ACC ACC AAG TG
Reverse	AGA TAG CAA ATC GGC TGA CG
IFN-γ
Forward	GCC AAG TTT GAG GTC AAC AAC
Reverse	CCG AAT CAG CAG CGA CTC
NHERF1
Forward	CAG GAG AAG TCC GAA CAA GC
Reverse	GAT GAA CTG GCC TGG CTT AG
NHERF2
Forward	CTC GAG CTA CTC GGG TCA AC
Reverse	ACC CAG TGG ACA AGA CAA GG
NHERF3
Forward	CGA TAG CTA CGG CTT TCA CC
Reverse	GGT TCC TCT TGC ACG TTC TC
β-Actin
Forward	CTG TCC CTG TAT GCC TCT G
Reverse	ATG TCA CGC ACG ATT TCC

ATX, autotaxin.

Statistical analysis.

Statistical significance was assessed by Student’s t-test or analysis of variance. Results are presented as means ± SE. P value < 0.05 was considered significant.

RESULTS

Inhibition of ATX attenuates ileitis in SAMP1/Fc mice.

Before evaluating the potential role of ATX in the pathogenesis of ileitis, we first determined whether ATX expression is altered in the ilea of SAMP1/Fc mice. Previous studies have shown that ileal inflammation is not observed in SAMP1/Fc mice until they are several weeks old (49, 55). Thus, we compared ATX mRNA expression in 4-, 8-, and 24-wk-old SAMP1/Fc and AKR mice. ATX mRNA expression in 4-wk-old SAMP1/Fc mice lacking signs of ileitis was similar to that in age-matched AKR mice. On the contrary, a ~threefold increase in ATX mRNA level was observed at 8 and 24 wk of age (Fig. 1A), suggesting that the onset of inflammation precedes the induction of ATX expression. To determine whether inhibition of ATX attenuates the progression of ileitis, SAMP1/Fc mice and age-matched control AKR mice were administered with PF-8380 (30 mg/kg body wt) or DMSO-H₂O as a control for 4 wk. Because the ileitis set in around 4–6 wk (55), mice were treated starting at 8 wk of age, which is still at the early stage of disease. Because patients with CD, in particular children, are at a high risk for malnutrition and growth deficiency (20, 63), we determined whether PF-8380 alters body weights in SAMP1/Fc mice. Oral administration of the vehicle, DMSO-H₂O, initially decreased body weights for several days in both AKR and SAMP1/Fc mice, but by the end of 4 wk both AKR and SAMP1/Fc mice showed increased weight (Fig. 1, B–C). PF-8380, on the other hand, resulted in a significant weight increase in SAMP1/Fc mice (Fig. 1B) but not in AKR mice (Fig. 1C). The differences in the weight changes between the PF-8380- and control-treated SAMP1/Fc mice were statistically significant during the first week of treatment. In the following 3 wk, the average weight remained greater in the PF-8380 group relative to the control group, but the statistical significance was not consistently observed. We speculate that the lack of the consistent statistical significance could have been because of the small sample sizes. ATX activity assay showed that at the end of the 4-wk-long PF-8380 treatment, ATX activity was decreased by 35% in SAMP1/Fc mice, although no significant change was observed in AKR mice (Fig. 1D). To determine the effect of PF-8380 on inflammation in SAMP1/Fc mice, we assessed cytokine expression before and after PF-8380 treatment. Consistent with a previous report (55), 12-wk-old control-treated SAMP1/Fc mice at the end of experiment showed a significant increase in IL-4 (~4-fold), IL-5 (~24-fold), and IL-13 (~50-fold) mRNA expression compared with AKR mice (Fig. 2A). Importantly, PF-8380 treatment markedly decreased IL-4, IL-5, and IL-13 expression in SAMP1/Fc mice (Fig. 2A). There was no significant difference in TNF-α and IFN-γ mRNA expression between SAMP1/Fc and AKR mice, consistent with previous findings that TNF-α and IFN-γ expression in SAMP1/Fc mice is not altered following long-standing inflammation (2). However, we observed a significant reduction in TNF-α mRNA expression by PF-8380. Because inflammation in SAMP1/Fc mice is associated with T-cell activation (47), we determined whether PF-8380 altered immune cell infiltration in SAMP1/Fc mice. Immunohistochemical staining of CD45 showed the number of immune cells was significantly lower in PF-8380-treated SAMP1/Fc mice than control-treated SAMP1/Fc (Fig. 2B). Immunofluorescence staining of CD3 further demonstrated decreased T-cell number by PF-8380 treatment (Fig. 2C). Our findings demonstrate that inhibition of ATX activity in SAMP1/Fc mice reduces intestinal inflammation by impeding immune cell infiltration and thus facilitating weight gain.

Fig. 1.

PF-8380 (PF) treatment inhibits autotaxin (ATX) activity and increases weight gain of SAMP1/Fc mice. Ileal scrapes were isolated from 4-, 8-, and 24-wk-old AKR and SAMP1/Fc mice, and the ATX mRNA expression was determined by quantitative RT-PCR (A). Data are presented as mean ± SE n = 3. *P < 0.01. Weight changes of SAMP1/Fc (B) and AKR (C) mice were monitored on daily basis beginning the first day of the 4-wk treatment by PF or control (Con). Weight gain is presented as the percentage of changes by comparing to the weight obtained on day 1 (set as 100%). n = 5. *P < 0.01. ATX activity was measured using serum prepared from AKR and SAMP1/Fc mice that were orally administered with PF or Con for 4 wk (D). FS-3, fluorogenic ATX substrate. Data are presented as mean ± SE n = 5. *P < 0.01.

Fig. 2.

Inhibition of autotaxin (ATX) attenuates inflammatory cytokine expression and immune cell migration in SAMP1/Fc mice. Ileal tissues were collected from AKR and SAMP1/Fc mice that were treated with PF-8380 (PF) or control (Con), and the mRNA expression of inflammatory cytokines was determined by quantitative RT-PCR (A). Data represent mean ± SE n = 5. *P < 0.01; **P < 0.05. Immunohistochemical staining of CD45 (B) and immunofluorescence staining of CD3 (C) in PF- and Con-treated SAMP1/Fc mice. Scale bar, 50 µm.

Inhibition of ATX increases SGLT1 expression in enterocytes.

SGLT1 expressed at the luminal membrane of enterocytes transports glucose from digested food across the intestinal brush border (66, 69). It has previously been shown that intestinal glucose absorption is inhibited in an inflamed state (61). Because PF-8380 treatment increased body weights of SAMP1/Fc mice, we determined whether the SGLT1 expression was altered in the intestine of SAMP1/Fc mice. SGLT1 mRNA and protein expression was markedly lower in SAMP1/Fc mice compared with AKR mice (Fig. 3, A–B). The inhibition of ATX by PF-8380 did not significantly modulate SGLT1 expression in AKR mice, but the expression levels of SGLT1 mRNA and protein in SAMP1/Fc mice were elevated by PF-8380 (Fig. 3, A–B). Given the heterogeneous conditions of ileitis in SAMP1/Fc mice, we performed immunofluorescence staining to confirm the changes in SGLT1 expression between PF-8380 and control-treated mice. In AKR mice, SGLT1 was expressed at the microvillar membrane of intestinal villi (Fig. 3C). In striking contrast, the expression of SGLT1 was completely absent in highly inflamed regions, identified by epithelial blunting, in SAMP1/Fc mice. Immunofluorescence signal of SGLT1 was detected in normal-looking intestinal villi of SAMP1/Fc mice, although the fluorescence intensity was markedly lower compared with AKR mouse villi. PF-8380 did not change SGLT1 expression in AKR mice (Fig. 3C). On the contrary, there was a significant increase in SGLT1 immunofluorescence signals in normal-looking villi of SAMP1/Fc mice treated with PF-8380, suggesting that the inhibition of ATX increased SGLT1 expression. Similarly, PF-8380 treatment increased SGLT1 expression in the inflamed mucosa (Fig. 3C, arrow), although the SGLT1 expression remained relatively low.

Fig. 3.

Inhibition of autotaxin (ATX) upregulates Na⁺-dependent glucose transporter 1 (SGLT1) expression in the intestine of SAMP1/Fc mice. AKR and SAMP1/Fc mice were treated with PF-8380 (PF) or control (Con). A: SGLT1 mRNA expression was determined by quantitative RT-PCR. The mRNA expression data are presented as mean ± SE n = 5. *P < 0.01. B: SGLT1 protein expression was determined by immunoblotting. RI is expressed relative to AKR Con. #P < 0.01 compared with AKR Con. *P < 0.01 compared with SAMP1/Fc Con. C: representative images are shown for normal-looking villi and severely inflamed regions in SAMP1/Fc mice. Arrow indicates membrane expression of SGLT1. Scale bar, 50 µm. RI, relative intensity determined by densitometry analysis.

Cytokines do not alter SGLT1 expression in Caco-2 cells.

Our results in Fig. 3 demonstrated that PF-8380 decreased IL-4, IL-5, and IL-13 expression in SAMP1/Fc mice. It was shown previously that IL-4 and IL-13 decrease Na⁺-linked glucose absorption in mouse intestine (44). This led to the hypothesis that elevated Th2 immune response suppresses SGLT-1 expression in SAMP1/Fc mice. To test this hypothesis, we employed intestinal epithelial Caco-2 cells as an in vitro model to determine the effect of IL-4 and IL-13 on SGLT1 expression. Caco-2 monolayers were treated with IL-4, IL-13 individually, or together from the basolateral side. However, we did not observe significant changes in SGLT1 protein expression in Caco-2 cells (Fig. 4A). Because SGLT1 is regulated by trafficking to the plasma membrane (66), we determined SGLT1 expression on the plasma membrane by surface biotinylation but again observed no difference in SGLT1 expression in the plasma membrane by IL-4 or IL-13 (Fig. 4B).

Fig. 4.

Inhibition of autotaxin (ATX) increases membrane expression of sucrase-isomaltase (SI) in intestinal epithelial cells in SAMP1/Fc mice. A: filter-grown Caco-2 cells were treated from the basolateral side with 50 ng/ml IL-4, IL-13, or IL-4 + IL-13 for 8 or 24 h. The total protein expression of Na⁺-dependent glucose transporter 1 (SGLT1) was then determined by Western blotting. B: caco-2 cells were treated with 50 ng/ml IL-4, IL-13 or IL-4 + IL-13 for 24 h, and the membrane expression of SGLT1 was determined by cell surface biotinylation assay. Ileal sections of PF-8380 (PF)- or control (Con)-treated AKR (C) and SAMP1/Fc (D) mice were stained for sucrase-isomaltase (SI; green). Nuclei were stained with DAPI (blue), and magnifications at ×100 and ×200 are shown. Scale bar, 50 µm.

Inhibition of ATX improves epithelial differentiation in SAMP1/Fc mice.

The expression of membrane transporters in IECs is often associated with cell polarization states. Having been unable to demonstrate the direct effect of IL-4 and IL-13 on SGLT1 expression, we sought to determine whether inhibition of ATX modulates epithelial cell differentiation in the ilea of SAMP1/Fc mice using sucrase-isomaltase (SI) as a marker of IEC differentiation (43, 62). AKR mouse ilea displaced the classical pattern of SI expression along the villus-crypt axis (Fig. 4C), but SI expression was hardly visible in inflamed ileal sections of SAMP1/Fc mice (Fig. 4D, ×100, top). At a high magnification (×200), we were able to detect weak staining at the apical membrane. Unexpectedly, we observed rather intense fluorescence signals that overlapped with DAPI signals in inflamed villi of SAMP1/Fc mice. A search of literatures did not reveal any previous findings of SI in the nucleus, but the nuclear SGLT1 signals were observed only in SAMP1/Fc mouse intestine and not in AKR mouse intestine. Whether these are nonspecific labeling with anti-SI antibody or a result of an unknown processing of SGLT1 in SMAP1/Fc mice is not clear and needs additional investigation. Regardless, PF-8380 treatment resulted in a significant increase in SI expression at the luminal surface of IECs in SAMP1/Fc mice (Fig. 4D, bottom). PF-8380 did not significantly change SI expression in AKR mice. These findings thus implicate that suppressing ATX activity improves IEC differentiation and enhances SGLT1 expression.

NHE3 expression is upregulated following ATX inhibition in SAMP1/Fc mice.

NHE3 residing at the brush border membrane of IECs mediates Na⁺ and fluid absorption (23, 24). Considering that patients with CD frequently encounter diarrhea and previous findings of decreased NHE3 expression associated with IBD (60), we next determined whether NHE3 expression is downregulated in SAMP1/Fc mice and if inhibition of ATX restores NHE3 expression. We stripped the blot used in Fig. 3 and probed with anti-NHE3 antibody. Immunoblotting of ileal mucosal lysates of AKR and SAMP1/Fc mice did not show a significant difference in NHE3 protein expression (Fig. 5A). Immunofluorescence staining of the ileum showed that NHE3 expression was comparable between AKR mice and normal-looking villi of SAMP1/Fc mice (Fig. 5B). However, the expression level of NHE3 in inflamed regions of SAMP1/Fc mice was markedly decreased. It is worth noting that in contrast to the brush border membrane expression in normal-looking villi, NHE3 showed intracellular localization in epithelial cells of moderately inflamed areas, indicating that NHE3 underwent internalization under inflammatory conditions (Fig. 5C). Similar to the effect on SGTL1 expression, PF-8380 did not have an observable effect on NHE3 expression in AKR mice. However, treatment of SAMP1/Fc mice with PF-8380 resulted in a robust increase in NHE3 expression in the plasma membrane (Fig. 5B). Interestingly, increased NHE3 expression in PF-8380-treated SAMP1/Fc mice was observed in the mid segment of villi, suggesting that increased NHE3 expression is likely secondary to IEC differentiation upon ATX inhibition. Previous studies have shown the importance of the NHERF family of proteins in NHE3 regulation (9, 24). Comparing NHERF1–3 mRNA expression showed that although NHERF1 and NHERF2 mRNA expression was not different between AKR and SAMP1/Fc mice, NHERF3 mRNA level was significantly lower in SAMP1/Fc intestine (Fig. 5D). Moreover, PF-8380 markedly increased NHERF3 expression in SAMP1/Fc intestine.

Fig. 5.

Inhibition of autotaxin (ATX) enhances Na⁺/H⁺ exchanger 3 (NHE3) membrane expression in SAMP1/Fc mice. A: NHE3 protein expression was determined by immunoblotting of ileal lysates. B: ileal sections of PF-8380 (PF)- or control (Con)-treated AKR and SAMP1/Fc mice were stained for NHE3 (green). Representative expression of NHE3 in normal-looking villi and severely inflamed regions in SAMP1/Fc mice are shown. Arrow indicates membrane expression of NHE3. C: subcellular localization of NHE3 (green) in normal-looking villi and moderately inflamed regions in SAMP1/Fc mice is shown at high magnification. Nuclei were stained with DAPI (blue). Arrow head denotes intracellular expression of NHE3. Scale bar, 50 µm. D: expression of NHERF1–3 mRNA was determined by quantitative RT-PCR in ileal mucosa of AKR and SAMP1/Fc mice treated with PF-8380 or control. n = 4. *P < 0.01. RI, relative intensity determined by densitometry analysis.

DISCUSSION

The hallmark of active IBD is an aberrant infiltration by immune cells that results in chronic and uncontrolled inflammation in intestinal mucosa. Suppression of immune responses and immune cell infiltration into the gut has been proven effective in ameliorating IBD-associated symptoms (48). By using SAMP1/Fc mice as a model of CD, we demonstrated that inhibition of ATX by PF-8380 decreased the expression of proinflammatory cytokines and attenuated immune cell homing in gut mucosa. Inhibition of ATX activity also partially restored epithelial cell differentiation as well as the expression of intestinal membrane transporters, SGLT1 and NHE3. Our study thus implicates that ATX is a potential target for treating IBD.

Recent studies have shown upregulation of ATX expression in inflammatory conditions, including cancer, arthritis, and multiple sclerosis (6, 12, 52, 70). Accumulating evidence demonstrates that inhibition of ATX ameliorates chronic inflammation (3). A recent study has shown increased ATX mRNA expression in intestinal biopsies from patients with IBD (26). We found that ATX expression was not different between AKR and SAMP1/Fc mice before the onset of ileitis but increased in SAMP1/Fc mice compared with control mice at 8 wk and remained elevated at 16 wk, suggesting that increased inflammation in the intestine elicits ATX induction. Consistently, it was shown previously that TNF-α and liposaccharide induce ATX expression (36, 67).

PF-8380 is one of the ATX inhibitors that works both in vitro and in vivo. We found that the effects of PF-8380 on serum ATX activity was less than 50%. PF-8380 is an orally bioavailable small molecule that displays >95% reduction in plasma LPA at 3 h after oral administration. However, the clearance of plasma PF-8380 is relatively quick with <20% remaining after 8 h (19). In our study, ATX activity was determined in blood collected 16–18 h after last administration of PF-8380, which probably explains the 35% decrease in ATX activity in SAMP1/Fc mice. Nevertheless, the inhibition of ATX by PF-8380 decreased proinflammatory cytokine expression and T-cell infiltration, suggesting that increased ATX function exacerbates inflammation.

Compared with AKR mice, SAMP1/Fc mice had increased Th2 cytokines, IL-4, IL-5, and IL-13. This feature of predominant Th2 cell polarization in SAMP1/Fc mice is consistent with previous reports (49, 55). We found that PF-8380 decreased expression of IL-4, IL-5, and IL-13 in SAMP1/Fc mice and mitigated immune cell infiltration. In contrast, TNF-α and INF-γ expression in SAMP1/Fc did not differ from that of AKR, which is consistent with previous findings (4). Development of ileitis in SAMP1/Fc mice involves active infiltration of lymphocytes (31). Accompanying the decreased cytokine expression by PF-8380 treatment was a marked reduction in CD3⁺ T cell infiltration in the ilea of SAMP1/Fc mice. Recent studies showed that ATX secreted by HEVs binds T lymphocytes through integrin interaction and promotes T-cell migration across HEVs (26, 29). Therefore, decreased T-cell infiltration in PF-8380-treated SAMP1/Fc mice is consistent with these earlier studies. However, the impact of ATX on lymphocyte extravasation into lymphatic tissues is not yet clear. Although intravenous injection of inactive ATX resulted in decreased lymphocyte numbers into lymph nodes and Peyer’s patch (29), depletion of circulating ATX using anti-ATX antibodies has no effect on lymphocyte trafficking to the spleen and lymphatic tissues (51). Nevertheless, our current findings are in line with the previous study that found inhibition of ATX by the nonspecific ATX inhibitor bithionol decreased lymphocyte migration, ameliorating dextran sodium sulfate-induced colitis and CD4⁺ T cell transfer-mediated ileocolitis (26).

Malnutrition is a major complication of IBD, especially in CD (59). Malabsorption is one of the primary factors that cause malnutrition, but the mechanisms underlying the inefficient absorption of macro- and micronutrients is not fully understood. Existing data link malabsorption to inflammatory cytokines and epithelial alterations, such as impaired epithelial transport and loss of epithelial integrity (59). We demonstrated in this study that epithelial cell differentiation was significantly compromised in the moderate to severe inflamed regions of SAMP1/Fc mouse ileum, as indicated by the reduced membrane expression of SI. SGLT1 and NHE3 are primarily expressed in the brush border membrane of the villus epithelia under physiological settings. Consistent with the loss of epithelial cell polarization, SGLT1 and NHE3 membrane expression was reduced in inflamed regions. Although previous studies have shown regulation of glucose transporters by cytokines (5, 45), our in vitro experiment did not demonstrate the direct effect of IL-4 or IL-13 on SGLT1 expression in Caco-2 cells. However, we cannot rule out the possibility that IL-4 or IL-13 contribute to the downregulation of SGLT1 by impairing epithelial cell polarization in vivo. In partial support of this possibility, IL-4 or IL-13 can perturb epithelial barrier function in human colonic and airway epithelial cells (13, 56), and IL-13 alters human airway epithelial cell differentiation (32).

Our study showed that NHERF3 expression was decreased in SAMP1/Fc mice without significant changes in NHERF1 or NHERF2 expression. A recent study has shown that NHERF3 expression is decreased in inflamed human and mouse intestines (68). Moreover, NHERF3 silencing resulted in decreased NHE3 expression in Caco-2BBE cells. Therefore, our finding of decreased NHERF3 appears in an agreement with this earlier study. A recent study showed that NHERF3 expression is downregulated in IL-10^−/− mice, suggesting that NHERF3 downregulation may cause NHE3 dysfunction in these mice (34). How NHERF3 is downregulated under inflammatory conditions is not yet known, but because PF-8380 decreased proinflammatory cytokine expression while increasing NHERF3 expression in SAMP1/Fc mice, we suggest that NHE3 is regulated in part via changes in NHERF3 expression in SAMP1/Fc mice.

Another intriguing finding of our study is that inhibition of ATX enhanced IEC differentiation as shown by increased luminal membrane expression of SI. LPA has varying effects on neuronal and endothelial cell differentiation (14, 15, 37), but its effect on IEC differentiation is not known. Our previous studies showed that defective LPA1 signaling decreases IEC migration, although perturbing epithelial barrier function (33, 38). Because cell migration is intrinsically coupled with differentiation, it is possible that LPA exerts its effect on IEC differentiation. Future studies are needed to assess whether the improvement in IEC differentiation in the current study is a direct result of decreased LPA-mediated signaling or a consequence of decreased inflammation. We did not examine the expression of other nutrient and electrolyte transporters, but it is reasonable to expect changes in expression of other transporters based on the weight gain upon ATX inhibition.

In summary, we have demonstrated that inhibition of ATX activity alleviates inflammation and improves IEC integrity in SAMP1/Fc mice. Our study demonstrated that ATX inhibition improves weight gain in an experimental model of CD that is associated with increased SGLT1 and NHE3 expression. Our study implicates that inhibiting ATX activity may help to improve CD-associated inflammation and malnutrition.

GRANTS

This work was supported by the Veterans Affairs Merit Award I01BX002540 and National Institutes of Health Grant DK-107719 (to C. C. Yun). Confocal microscopic analyses were supported in part by the Integrated Cellular Imaging Shared Resources of Winship Cancer Institute of Emory University and NIH/National Cancer Institute under Award No. P30-CA-138292.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

P.H., A.H., F.C., and C.C.Y. conceived and designed research; P.H., A.H., and S.L. performed experiments; P.H., A.H., and S.L. analyzed data; P.H., A.H., and C.C.Y. interpreted results of experiments; P.H. and S.L. prepared figures; C.C.Y. edited and revised manuscript; P.H. drafted manuscript; P.H. and C.C.Y. approved final version of manuscript.