AJP - GI AJP: Lung Cellular and Molecular Physiology
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Am J Physiol Gastrointest Liver Physiol 295: G766-G775, 2008. First published August 14, 2008; doi:10.1152/ajpgi.90423.2008
0193-1857/08 $8.00
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/4/G766    most recent
ajpgi.90423.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ohama, T.
Right arrow Articles by Ozaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ohama, T.
Right arrow Articles by Ozaki, H.

INFLAMMATION/IMMUNITY/MEDIATORS

Sphingosine-1-phosphate enhances IL-1β-induced COX-2 expression in mouse intestinal subepithelial myofibroblasts

Takashi Ohama,1,2 Muneyoshi Okada,1 Takahisa Murata,1 David L. Brautigan,2 Masatoshi Hori,1 and Hiroshi Ozaki1

1Department of Veterinary Pharmacology, Graduate School of Agriculture and Life Sciences, The University of Tokyo, Tokyo, Japan; and 2Center for Cell Signaling, University of Virginia, Charlottesville, Virginia

Submitted 10 July 2008 ; accepted in final form 7 August 2008


   ABSTRACT
TOP
 ABSTRACT
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Intestinal subepithelial myofibroblasts (SEMFs) is a specific population of cells involved in intestinal inflammation and carcinogenesis via an elaborate network of cytokines, chemokines and other inflammatory factors, including PGE2. Sphingosine-1-phosphate (S1P) has been implicated as an important mediator of inflammation and cancer and in certain cell types increases cyclooxygenase-2 (COX-2) expression. In the present study, we aimed to assess involvement of S1P in COX-2 expression by SEMFs. Primary SEMFs were obtained from C57BL/6J mouse and their identity was verified by fluorescent staining of specific marker proteins. Expression of S1P receptors 1, 2, 3 and sphingosine kinases 1 and 2 in SEMFs were determined by RT-PCR analysis. COX-2 expression and PGE2 production were assayed by Western blotting and ELISA, respectively. COX-2 mRNA stability was assayed by Northern blotting. S1P produced dose-dependent increase in COX-2 expression, resulting in increased PGE2 release from SEMFs. Using specific inhibitors, we show that actions of p38, ERK, IKK, and PKC were involved in S1P-induced COX-2 expression. On the other hand, p38 and PKC had lesser roles in IL-1β-induced COX-2 expression. Inhibition of sphingosine kinase to block S1P production did not affect IL-1β-induced COX-2 expression, but S1P amplified IL-1β-induced p38 activation and COX-2 expression. PKC inhibition blocked S1P amplified COX-2 expression. S1P addition increased COX-2 mRNA stability. In SEMFs, S1P amplifies IL-1β-induced COX-2 expression through increased mRNA stability. These observations point to involvement of S1P in activation of SEMFs that may contribute to intestinal inflammation and carcinogenesis.

cyclooxygenase 2; sphingosine kinase; curcumin; Protein kinase C; mRNA stability

UNDER THE EPITHELIUM OF THE gastrointestinal tract, intestinal subepithelial myofibroblasts (SEMFs) form a cellular network from the esophagus to the anus. These SEMFs exhibit the ultrastructural features of both fibroblasts and smooth muscle cells and are characterized by positive immunoreactivity for {alpha}-smooth muscle actin ({alpha}-SMA) and vimentin (3, 37). SEMFs are distinguished from smooth muscle cells, which express {alpha}-SMA but are negative for vimentin. SEMFs are classified as members of a family of functionally related cells, including hepatic stellate cells (Ito cells), glomerular mesangial cells, and orbital and synovial fibroblasts (41).

It is thought that various factors produced by SEMFs, including growth factors, proinflammatory cytokines, matrix protein, and prostaglandins (PGs), play an important role in mucosal repair, inflammatory response, and carcinogenesis of the intestine (3, 37). PGs are derived from arachidonic acid (AA), which is a polyunsaturated fatty acid (C20:4) released from cell membrane phospholipids by phospholipase A2 (PLA2). AA is converted to PGs via the rate-limiting enzyme cyclooxygenase (COX). Of the two known COX isoforms, COX-1 is constitutively expressed in most tissues, whereas COX-2 is expressed predominantly during acute or chronic inflammation and cancer (40). COX-2 plays a key role in gastrointestinal mucosal defense, inflammation, and carcinogenesis (9, 24, 30, 42). SEMFs are thought to be an important source of PGs in mucosa, because many factors such as cytokines (e.g., IL-1β and IL-17) and bacterial products (e.g., LPS) induce COX-2 expression and PG synthesis in isolated SEMFs (29, 44). Moreover, in the Gi{alpha}2 knockout mouse, which is a model of spontaneous development of colitis, SEMFs exhibit decreased PGE2 production, which may contribute to the pathogenesis of colitis (14). This PGE2-mediated intestinal homeostasis is maintained by preservation of mucosal integrity and downregulation of the immune response via EP4 receptor activation (19). In colon carcinoma, COX-2 expression is located in specific areas of the adenomas, and immunostaining has revealed expression of COX-2 in SEMFs from colon carcinoma patients (1, 17). These findings support the hypothesis that expression of COX-2 by SEMFs is an important factor in the pathological condition of colon mucosa. Therefore it would be useful to have a better understanding of COX-2 expression by SEMFs.

Sphingolipids are a novel class of bioactive lipids that play key roles in the regulation of several cellular processes including growth, differentiation, stress response, and apoptosis (38). Studies indicate that sphingosine can induce production of the inflammatory mediator PGE2 (4, 5). More recent studies have implicated a regulatory function of sphingosine-1-phosphate (S1P), the phosphorylated form of sphingosine, in PGE2 production via induction of COX-2 expression (35, 36). S1P is produced by phosphorylation of sphingosine catalyzed by sphingosine kinase (SphK), a highly conserved enzyme that is activated by many agonists and stimuli. Diverse S1P-induced responses are mediated by S1P binding to a family of G protein-coupled receptors (S1P1–5), and it is also known that some biological activities of S1P are mediated by its intracellular actions (38, 39). For instance, contraction of gastric smooth muscle cells is mediated by receptor activation (16, 45); on the other hand, contraction of airway smooth muscle cells is via intracellular actions of internalized S1P (2). Many signaling enzymes including ERK, p38, JNK, PI3K/Akt, PKC, and src family tyrosine kinase are involved in S1P-induced COX-2 expression in various cell types (15, 21, 31). It is reported that S1P is upregulated in colon carcinogenesis (20) and that knockout of SphK1 results in reduction of adenoma size (23). Recently Maines et al. (27) observed the suppression of mouse colonic inflammation by orally available SphK inhibitors. These studies suggest the possible involvement of S1P in COX-2 expression and pathological conditions of colon mucosa, such as inflammation and cancer, but the mechanisms producing these effects are unclear. Therefore, to clarify the role of S1P in COX-2 expression in SEMFs would be quite useful for a better understanding of the pathology of intestinal inflammation and carcinogenesis.

In the present study, we sought to determine the role of S1P in controlling expression of COX-2 in mouse SEMFs. Moreover we tested the effect of S1P on IL-1β signaling that is known to induce COX-2 expression and PGE2 production in SEMFs (10, 29).


   MATERIALS AND METHODS
TOP
 ABSTRACT
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Culture of mouse colonic subepithelial myofibroblasts. All experiments and animal care in this study were performed in accordance with the Guide to Animal Use and Care of the University of Tokyo and permitted by the committee. The notification number is P07-085. Female C57BL/6J mice were purchased from Charles River Japan and Jackson Laboratory USA.

Isolated primary SEMF cultures were prepared using a method described by Mahida et al. (26) with some modification. Briefly, a segment of the proximal colon was detached from the mesenterium and was placed in sterile Hanks’ balanced salt solution. Smooth muscle layers were detached from the mucosal layer. The mucosal samples were completely denuded of epithelial cells by three-times-repeated 30-min incubation in 1 µM EDTA-Hanks solution at 37°C. The deepithelialized mucosal samples were cultured in DMEM (GIBCO) containing 10% fetal bovine serum (FBS) at 37°C in a 5% CO2 atmosphere. The denuded tissues were maintained in culture for up to 4 wk, and the myofibroblasts that migrated from the explanted cultured mucosal tissues formed colonies. All culture media were supplemented with 100 U/ml penicillin G, 100 µg/ml streptomycin sulfate, and 0.25 µg/ml amphotericin B. The experiments were performed using cells in passages 2–6. For all experiments, cells were grown to 70–80% confluency and were cultured in DMEM without FBS for 48 h.

Because SEMFs are characterized by positive immunoreactivity for {alpha}-SMA and vimentin (3, 37), to verify the origin of the SEMFs and the homogeneity of the cultures we assayed expression of {alpha}-SMA and vimentin.

Fluorescent staining. To visualize smooth muscle-specific {alpha}-actin ({alpha}-SMA) and vimentin, mouse SEMFs were grown on glass coverslips. The cells were fixed with 4% formaldehyde in 0.05% Tween containing Tris-buffered saline (TBS) for 10 min at 37°C and then were permeabilized with 0.2% Triton X-100 for 30 s. Afterward, the cells were incubated with anti {alpha}-SMA antibody (1:200 dilution; Sigma) or anti-vimentin (1:200 dilution; Sigma) as a first antibody for 1 h at 37°C. Then the cells were incubated with Alexa 568 conjugated anti-mouse (for {alpha}-SMA) or goat (for vimentin) IgG as a second antibody for 1 h at 37°C. The images were captured by using a confocal laser scanning microscope (LSM510; Zeiss).

RT-PCR analysis. Total RNA was extracted from the SEMFs by the acid guanidinium isothiocyanate-phenol-chloroform method, and the concentration of RNA was adjusted to 0.5 µg/µl by use of RNase-free distilled water. Semiquantitative RT-PCR was performed as described elsewhere (33). The oligonucleotide sequences shown in Table 1 were used as primers to amplify the target S1P receptor isoforms and SphK genes. Specific primers were designed for each S1P receptor isoform, based on the conserved sequences in mouse cDNAs.


View this table:
[in this window]
[in a new window]

  		Table 1. Oligonucleotide primers for RT-PCR and suitable productive size

 
The PCR products were separated by electrophoresis on 2% agarose gel containing <0.00002% ethidium bromide. The possibility of DNA contamination was excluded by performing PCR using total RNA without the reverse transcription step. The resultant fluorescent bands were visualized by using an ultraviolet transilluminator and FAS-III (Toyobo).

Western blotting. Western blot analysis was performed as previously described (34). Briefly, to the homogenizing solution we added the homogenizing buffer containing 60 mM β-glycerophosphate, 2 mM EGTA, 0.5% Nonidet P-40, 0.2% SDS, 100 mM NaF, 1 mM Na3VO4, 50 mM Tris·HCl (pH 8.3), and Roche's complete protease inhibitor cocktail. Membranes were incubated in a blocking buffer consisting of TBS containing 5% nonfat skim milk for 30 min at room temperature. Membranes were incubated with one of the following primary antibodies in the blocking buffer overnight at 4°C: anti-COX-2 antibody (1:2,000 dilution, Cayman), anti-p38 antibody (1:1,000 dilution), anti-phospho-p38 antibody (1:500 dilution), anti-ERK antibody (1:1,000 dilution), anti-phospho-ERK antibody (1:1,000 dilution), or anti-β actin (1:1,000 dilution). These antibodies are purchased from Cell Signaling. Reactions were detected by using an ECL plus Western Blotting Detection System (Amersham Biosciences). Bands were visualized via an LAS-1000 mini luminescence imager (Fujifilm).

PGE2 production. Levels of PGE2 in culture medium were measured by using a PGE2 Quantikine ELISA kit (R&D Systems) as described elsewhere (32). SEMFs were grown in a dish (diameter 60 mm) to 70–80% confluency and were then cultured in 2 ml DMEM for 48 h. The medium was then changed, followed immediately by addition of S1P and further incubation of the SEMFs for 6 h. The PGE2 level in the culture medium was then assayed by the method recommended by the manufacturer.

COX-2 mRNA stability assay. SEMFs were incubated with IL-1β (10 µg/ml) or IL-1β plus S1P (10 µM) for 2 h. Then cells were treated with the medium containing RNA polymerase II transcriptional inhibitor DRB (5,6 dichloro-β-D-ribofulanosylbenzimidazole) (50 µM) plus IL-β or DRB plus IL-1β and S1P. Cultures were frozen at 0, 1, and 2 h after DRB treatments. Total RNA was extracted as mentioned under RT-PCR analysis above, and COX-2 mRNA levels were determined by Northern blotting.

Northern blotting. Mouse COX-2 cDNA probe was purchased from R&D Systems, and mouse GAPDH cDNA probe was kindly provided by Dr. Wotton (University of Virginia). Labeling of probes with digoxigenin (DIG) and detection of mRNA were performed by using DIG labeling kit (Roche Applied Science) following the manufacturer's instructions. Briefly 0.2 and 1 µg of total RNA were applied to Zeta-Probe GT membrane (Bio-Rad Laboratories) by using a dot-blot system (Microsample Filtration Manifold; Schleicher & Schuell) and cross-linked to the membrane by UV cross-linker (UVC500; Hoefer). Then the membrane was treated with 25 ng/ml of DIG-labeled COX-2 probe in DIG easy hybrid solution (Roche Applied Science) at 42°C for 16 h. The bands were detected by anti-DIG antibodies (Roche Applied Science). The band density was normalized with GAPDH.

Statistical analysis. Results are expressed as means ± SE. Comparisons between the control and test groups were performed by one-way ANOVA, followed by Dunnett's multiple-comparison test. For all analyses, a probability value of P < 0.05 was considered to indicate statistical significance.


   RESULTS
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Characterization of isolated mouse SEMFs. Primary SEMFs were prepared by recovery of cells that migrate out of excised intestinal mucosa segments according to a method reported by Mahida et al. (26). SEMFs are characterized by positive immunoreactivity for both {alpha}-SMA and vimentin (3, 37) that distinguish them from other cells. Cells in our cultures clearly exhibited immunoreactivity for {alpha}-SMA and vimentin, verifying that they were SEMFs (Fig. 1).


Figure 1
View larger version (27K):
[in this window]
[in a new window]

  		Fig. 1. Expression of {alpha}-smooth muscle actin ({alpha}-SMA) and vimentin in primary cultured subepithelial myofibroblasts (SEMFs) isolated from mouse colon. Mouse SEMFs were immunostained for {alpha}-SMA (A) and vimentin (B). 4,6-Diamidino-2-phenylindole (DAPI) was used for nuclear labeling. Bars indicate 50 µm.

 
We assayed SEMFs for 5 types of S1P receptors (S1P1–5) and two types of SphK (SphK1 and 2) that phosphorylate sphingosine to produce S1P. To test expression, specific DNA primers for each S1P receptor and sphingosine kinase isoform were designed based on the conserved sequences in mouse, as described in MATERIAL AND METHODS and Table 1, and RT-PCR was performed. We observed abundant expression of mRNA for S1P1, S1P2, and S1P3 and weak but detectable expression of S1P5 (Fig. 2A). Although S1P4 was detected in liver tissue, we did not detect S1P4 mRNA in the SEMFs. These results indicate that S1P probably binds multiple receptors on SEMFs. Moreover, expression of SphK1 and SphK2 were observed in mouse SEMFs (Fig. 2B), suggesting that S1P production by sphingosine metabolism can be triggered by SEMFs.


Figure 2
View larger version (14K):
[in this window]
[in a new window]

  		Fig. 2. Isoforms of sphingosine-1-phosphate (S1P) receptors and sphingosine kinases expressed in mouse SEMFs. Total RNA isolated from cultured SEMFs was reversibly transcribed, and cDNA was amplified by PCR using specific primers for S1P receptors (A) and sphingosine kinases (SphK; B). Experiments were performed with (+) or without (–) the reverse transcriptase (RT) reaction step. S1P4 and S1P5 mRNA were also detected in total RNA extracted from liver.

 
S1P stimulates COX-2 expression and PGE2 production in isolated mouse SEMFs. We investigated the effects of S1P on COX-2 expression and PGE2 formation in isolated mouse SEMFs, because PGE2 has emerged as the prototypical eicosanoid with key functions in inflammatory responses and carcinogenesis (9, 24, 30). As shown in Fig. 3, A and B, S1P elicited dose (0.1–10 µM)- and time (1–12 h)-dependent COX-2 protein expression. COX-2 expression reached its maximum after 6 h of incubation with S1P, and 1 µM of S1P induced maximum COX-2 expression. Moreover, SEMFs incubated with S1P for 6 h secreted PGE2 into the culture medium, presumably as a product of COX-2 (Fig. 3C).


Figure 3
View larger version (19K):
[in this window]
[in a new window]

  		Fig. 3. S1P-induced COX-2 protein expression and PGE2 production in SEMFs. A: mouse SEMFs were incubated with 10 µM of S1P for indicated periods, and COX-2 protein expression was assayed by immunoblot analysis. B: effects of various concentration of S1P (4 h) on COX-2 protein expression of SEMFs. C: PGE2 level in culture medium was analyzed by ELISA after culture with or without S1P (10 µM) for 6 h. Results are expressed as means ± SE of 4–8 experiments. *Significantly different from control, P < 0.05.

 
S1P-induced COX-2 expression is mediated by MAPKs, PKC, and IKK in isolated mouse SEMFs. Multiple signaling pathways have been implicated in the regulation of COX-2 expression in different cell types (15, 21, 22). Therefore we used kinase inhibitors to examine the involvement of MAPKs (p38 and ERK), PKC, and IKK pathways in S1P-induced COX-2 expression in mouse SEMFs. Upregulation of COX-2 expression by S1P was effectively blocked by SB203580 (p38 inhibitor) and PD98059 (MEK/ERK pathway inhibitor) (Fig. 4A), suggesting that both p38 and ERK are required for S1P-induced COX-2 expression in mouse SEMFs. Bisindolylmaleimide (BIS; PKC inhibitor) (Fig. 4B) and curcumin (IKK inhibitor) (Fig. 4C) partially but significantly inhibited COX-2 upregulation by S1P, suggesting that PKC and IKK activation is necessary for a maximum response. Because it is known that S1P receptors are coupled with pertussis toxin (PTX)-sensitive G protein, we examined the effect of PTX on S1P-induced COX-2 expression in mouse SEMFs. As shown in Fig. 4D, PTX completely blocked upregulation of COX-2, supporting the idea that S1P induced COX-2 protein expression is mediated through S1P receptors.


Figure 4
View larger version (22K):
[in this window]
[in a new window]

  		Fig. 4. Mechanism of S1P-induced COX-2 expression. Mouse SEMFs were incubated with 30 µM of PD98059 (MEK/ERK inhibitor) or 5 µM of SB203580 (p38 inhibitor) (A), 5 µM bisindolylmaleimide (BIS; PKC inhibitor) (B), or 50 µM curcumin (IKK inhibitor) (C) for 30 min or pertussis toxin (PTX; 100 ng/ml) (D) for 16 h before and during incubation with S1P (10 µM) for 4 h. COX-2 protein expression was assayed by immunoblot analysis. Results are expressed as means ± SE of 4 experiments. *Significantly different from S1P treatment, P < 0.05.

 
S1P synthesis is not required for IL-1β-induced COX-2 expression in mouse SEMFs. We observed time-dependent enhancement of COX-2 expression by IL-1β treatment in isolated mouse SEMFs. IL-1β increased the level of PGE2, the product of COX-2 reaction, in the culture medium of mouse SEMFs (Fig. 5A). Previous reports demonstrated the involvement of MAPKs, PKC, and IKK in IL-1β-induced COX-2 expression pathway in a human SEMF cell line (10, 29). We examined the effects of kinase inhibitors on IL-1β-induced COX-2 expression in mouse primary SEMFs. SB203580 and PD98059 partially but significantly blocked COX-2 upregulation (Fig. 5B), suggesting that the p38 and ERK pathways are necessary for maximum COX-2 expression by IL-1β. On the other hand, because pretreatment with BIS failed to inhibit increase of COX-2 expression (Fig. 5C), PKC activation is dispensable for IL-1β-induced COX-2 expression. Curcumin blocked COX-2 upregulation (Fig. 5D), and, even though this is not a specific inhibitor, the result suggests IKK/NF-{kappa}B signaling is critical for IL-1β-induced COX-2 expression in mouse SEMFs.


Figure 5
View larger version (29K):
[in this window]
[in a new window]

  		Fig. 5. Mechanism of IL-1β-induced COX-2 expression. A: mouse SEMFs were incubated with or without IL-1β (10 ng/ml) for 6 h, and PGE2 level in culture medium was analyzed by ELISA. Mouse SEMFs were incubated with 30 µM of PD98059 (MEK/ERK inhibitor) or 5 µM of SB203580 (p38 inhibitor) (B), 5 µM bisindolylmaleimide (PKC inhibitor) (C), 50 µM curcumin (IKK inhibitor) (D) for 30 min, 10 µM of N,N-dimethylsphingosine (DMS) or SK inhibitor (E) for 40 min, or 100 ng/ml PTX for 16 h (F) before and during incubation with IL-1β (10 ng/ml) for 4 h. COX-2 protein expression was assayed by immunoblot analysis. Results are expressed as means ± SE of 4–8 experiments. *Significantly different from control for A and from IL-1β treatment for B–D, P < 0.05.

 
It is known that S1P is involved in COX-2 expression induced by cytokines, such as IL-1β and TNF-{alpha}, in various cell types (6, 35, 36). To test this point, we treated cells with sphingosine kinase inhibitors to block S1P synthesis. Two types of sphingosine kinase (SK) inhibitors, 2-(p-hydroxyanilino)-4-(p-chlorophenyl) thiazole and N,N-dimethylsphingosine failed to inhibit the IL-1β-induced COX-2 expression (Fig. 5E). Moreover, pretreatment with PTX that blocks S1P-induced COX-2 expression did not inhibit COX-2 upregulation by IL-1β (Fig. 5F). These results indicate that S1P production is not necessary for IL-1β-induced COX-2 expression in mouse SEMFs.

S1P amplifies IL-1β-induced COX-2 expression. We examined whether simultaneous treatments with S1P and IL-1β affected COX-2 expression. Costimulation with S1P enhanced IL-1β-induced COX-2 expression (Fig. 6A). We examined the phosphorylation of p38 and ERK and found both S1P and IL-1β induced phosphorylation of p38 and simultaneous treatment with S1P and IL-1β clearly enhanced phosphorylation of p38 (Fig. 6B). On the other hand, the level of ERK phosphorylation by simultaneous treatment with S1P and IL-1β was almost the same as with S1P alone. Because PKC activation is involved in S1P-induced, but not in IL-1β-induced, COX-2 expression, we examined the effects of PKC inhibitors on simultaneous treatment with S1P and IL-1β. Pretreatment with PKC inhibitors BIS or Go6976 blocked S1P-enhanced COX-2 expression to the level of IL-1β alone (Fig. 6C). These results indicate the involvement of p38 and PKC activation in S1P-amplified expression of COX-2.


Figure 6
View larger version (12K):
[in this window]
[in a new window]

  		Fig. 6. S1P amplifies IL-1β-induced COX-2 expression. A: mouse SEMFs were incubated with S1P (10 µM), IL-1β (10 ng/ml), or S1P plus IL-1 for 4 h, and COX-2 protein expression was assayed by immunoblot analysis. B: SEMFs were stimulated by S1P, IL-1β, or S1P plus IL-1β for 15 min, and phosphorylation of p38 and ERK were detected by immunoblot analysis. The data are representative of 3 independent experiments. C: 5 µM BIS or 1 µM Go6976 (PKC inhibitors) were treated for 30 min before and during incubation with S1P, IL-1β, or S1P plus IL-1β for 4 h. COX-2 protein expression was assayed by immunoblot analysis. Results are expressed as means ± SE of 4 experiments for A and C. *Significantly different, P < 0.05.

 
S1P stabilizes COX-2 mRNA. It is known that p38 and PKC activation lead to COX-2 mRNA stabilization through 3'-untranslated region (UTR) recognition by RNA binding protein complex including human antigen R (HuR) (8, 12, 13, 25). COX-2 mRNA stability assay was performed using mouse primary SEMFs treated with IL-1β or IL-1β plus S1P. After COX-2 mRNA expression was activated, RNA polymerase II transcriptional inhibitor DRB was added to stop transcription. COX-2 mRNA level was detected at 0, 1, and 2 h after DRB treatment by Northern blot. As shown in Fig. 7, COX-2 mRNA half-life was clearly prolonged from 1.4 h for IL-1β alone to 4.5 h for IL-1β plus S1P. We concluded that S1P increased COX-2 mRNA stabilization to amplify COX-2 protein expression in mouse SEMFs.


Figure 7
View larger version (23K):
[in this window]
[in a new window]

  		Fig. 7. S1P increases COX-2 mRNA stability. Effect of S1P on COX-2 mRNA stability was assayed as indicated in MATERIAL AND METHODS. Isolated mouse SEMFs were treated with IL-1β or IL-1β plus S1P for 2 h to obtain COX-2 mRNA expression and then RNA polymerase II transcriptional inhibitor DRB (5,6 dichloro-β-D-ribofulanosylbenzimidazole) were added to stop transcription. COX-2 and GAPDH mRNA level were assayed at 0, 1, and 2 h after DRB treatment by Northern blot. Representative images (A) and quantitative graph (B) of COX-2 mRNA levels of 3 independent experiments.

 

   DISCUSSION
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
SEMFs form a network under the epithelium and play a critical role in epithelial restitution, reconstitution, and barrier function (37). Recently, contribution of S1P in intestinal inflammation and carcinogenesis is coming to light, but underlying mechanisms of how S1P exerts its effects remain obscure. In this study, we found that S1P upregulated COX-2 expression and PGE2 production in SEMFs. We show that isolated mouse SEMFs express SphK1 and SphK2 mRNA, which suggests the possibility that SEMFs also contribute to production of S1P in the colon. In normal and colon cancer model animals, it was reported that colon epithelial cells, cancer cells and stromal cells expressed SphK1 (20, 23). However, because SphK1 expression is much higher in epithelial cells than stromal cells, epithelial cells can be considered as the major source of S1P in colon mucosa (20, 23). On the other hand, recent evidence revealed that the site of COX-2 expression early on in polyposis is not the epithelial cells but rather the SEMFs (17). Taken together, we propose an inflammatory cycle, where S1P, produced by epithelial cells, induces COX-2 expression in SEMFs, which results in a generation of PGE2 that contributes to colon inflammation and carcinogenesis.

The mechanism of COX-2 expression induced by S1P is controversial and may be different between cell types. For S1P-induced COX-2 expression in rat aorta smooth muscle cells, although Hsieh et al. (15) reported the involvement of ERK and PI3K/Akt signaling, Nodai et al. (31) suggested the contribution of PKC and Src-family tyrosine kinases, but not ERK and p38 signaling. ERK signaling (but not p38 signaling) is involved in S1P-induced COX-2 expression in amnion-derived WISH cells and human coronary arterial smooth muscle cells (15, 22). Recently, Ki et al. (21) reported involvement of JNK activation in S1P-induced COX-2 expression. Here we observed that ERK, p38, IKK, and PKC activation were all necessary for maximum protein expression of COX-2 in mouse SEMFs. One of the causes of this cell specificity for S1P signaling may be different expression of S1P receptors. In the present study, we observed expression of the S1P receptors, S1P1, S1P2, S1P3, and S1P5. Because it is reported that S1P1 and S1P3 activate ERK and S1P2 activates p38 MAPK (43), these receptor subtypes cooperatively may be involved in the S1P-induced COX-2 expression in SEMFs. Although S1P is known to directly pass through a plasma membrane and may act as an intracellular second messenger with direct targets that are as yet unknown (2, 39), our finding that upregulation of COX-2 by S1P was completely blocked by PTX that inactivates Gi protein indicates that S1P-induced COX-2 expression is mediated through S1P receptors in mouse SEMFs. Although we observed PGE2 release from SEMFs by S1P treatment, the very recent report using RAW264.7 mouse macrophage-like cell line demonstrates that S1P induced COX-2 expression but not PGE2 production, because of the lack of phospholipase A2 activation (18).

In the present study, we examined the involvement of S1P in IL-1β-induced COX-2 expression. As described earlier, S1P contributes to IL-1β-and TNF-{alpha}-induced COX-2 expression in many cells, including the A549 human lung carcinoma cell line, the L929 mouse fibrosarcoma cell line, and the HT-29 human colon cancer cell line (20, 35, 36). It has also been reported that IL-1β and TNF-{alpha} activate SphK in the INS-1 insulinoma cell line, resulting in production of S1P (28). However, in the present study, SphK inhibitors failed to inhibit IL-1β-induced upregulation of COX-2 expression. Moreover, although S1P-induced COX-2 expression was completely inhibited by PTX as indicated above, IL-1β-induced COX-2 expression was not blocked by PTX. These findings indicate that in mouse SEMFs generation of S1P does not play a key role in IL-1β-induced COX-2 expression. Considering the cell types in which S1P is involved in cytokine-induced COX-2 expression, we speculate that the production of S1P is more important in immortalized cells such as cancer cells, because these cells have higher SphK expression. This concept may be supported by the observation that SphK expression is low in normal colon epithelium and highly induced by azoxymethane, an inducer of colon tumor (20).

We observed inhibition of the ERK, p38 and IKK pathways blocked IL-1β-induced COX-2 expression. Similar to this, Powell's group previously found that ERK, p38, and IKK/NF-{kappa}B are involved in IL-1-induced COX-2 expression in a human SEMF cell line (10, 29). Because IKK inhibition completely blocked the COX-2 expression, IKK/NF-{kappa}B pathway seems crucial for IL-1β-induced COX-2 expression. On the other hand, p38 and ERK inhibitors led to partial reduction in COX-2. Moreover, although IL-1β induced higher COX-2 expression than S1P, phosphorylation level of p38 and ERK are lower than S1P treatment. These results indicate the lesser role of p38 and ERK in IL-1β-induced COX-2 expression. Powell's group also reported the involvement of PKC activation in COX-2 upregulation by IL-1. However, we observed that PKC inhibitor did not block IL-1β-induced COX-2 expression in mouse SEMFs, maybe because of differences of experimental procedures or 18Co cell (human SEMF cell line) vs. primary mouse SEMFs.

Our finding that S1P amplified the effects of IL-1β on COX-2 expression suggests the importance of S1P on COX-2 expression in SEMFs. Both S1P and IL-1β induce phosphorylation of p38, and simultaneous treatment with both of them enhances p38 phosphorylation, suggesting that this amplified COX-2 expression by S1P. Moreover, we demonstrated that PKC inhibition blocked the COX-2 expression induced by combination of S1P and IL-1β to the level of IL-1β alone. So PKC activation is important for the amplification by S1P. We observed increased stability of COX-2 mRNA by S1P treatment. It is well known that p38 and PKC activation leads to COX-2 mRNA stabilization through 3'-UTR recognition by RNA binding protein complex including HuR (8, 12, 13, 25). HuR is a member of Hu family of RNA-binding proteins that recognize AU-rich element-containing mRNAs (7). It is recently reported that HETES enhance IL-1-mediated COX-2 expression via p38 and HuR-dependent augmentation of mRNA stability in 18Co human SEMF cell line (11). Moreover, S1P promotes HuR-dependent COX-2 mRNA stabilization in RAW264.7 mouse macrophage-like cell line (18). Therefore we expect that, in mouse SEMFs, S1P induced activation of p38 and PKC leads to COX-2 mRNA stabilization by HuR-dependent mechanism.

In summary, the present findings indicate that in mouse primary SEMFs S1P enhances COX-2 expression and PGE2 production mainly via p38 and ERK, but IKK and PKC are also involved (Fig. 8). On the other hand, IKK activation is crucial for IL-1β-induced COX-2 expression, and p38 and PKC activation play a lesser role. S1P amplifies IL-1β induced COX-2 protein expression through p38 and PKC activation that lead to COX-2 mRNA stabilization. Although in diverse cell types IL-1β activates SphK and production of S1P is involved in IL-1β-induced COX-2 expression, S1P does not contribute to COX-2 expression stimulated by IL-1β in mouse SEMFs. These observations enhance our understanding of the involvement of S1P in pathological functions of SEMFs in inflammation and carcinogenesis in intestinal mucosa.


Figure 8
View larger version (15K):
[in this window]
[in a new window]

  		Fig. 8. Possible mechanism of S1P- and IL-1β-induced COX-2 expression in mouse SEMFs. S1P binds S1P receptors (S1PR) and induces COX-2 protein expression mainly through ERK and p38 activation and partially through IKK and PKC activation. On the other hand, IKK play a crucial role in IL-1β-induced COX-2 protein expression, and p38 and PKC have lesser roles. It is known that p38 and PKC stabilize COX-2 mRNA through 3'-untranslated region recognition by RNA binding protein complex including human antigen R (HuR); S1P may enhance IL-1β-induced COX-2 protein expression through p38 and PKC dependent upregulation of COX-2 mRNA stability.

 


   ACKNOWLEDGMENTS
 
We thank David Wotton (University of Virginia) for providing mouse GAPDH probe and Theresa Pizarro (University of Virginia) for useful suggestions regarding this study.


   FOOTNOTES
 

Address for reprint requests and other correspondence: M. Hori, Dept. of Veterinary Pharmacology, Graduate School of Agriculture and Life Sciences, The Univ. of Tokyo, Yayoi 1-1-1, Bunkyo-ku, Tokyo 113-8657, Japan (e-mail: ahori{at}mail.ecc.u-tokyo.ac.jp)

The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


   REFERENCES
TOP
 ABSTRACT
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Adegboyega PA, Ololade O, Saada J, Mifflin R, Di Mari JF, Powell DW. Subepithelial myofibroblasts express cyclooxygenase-2 in colorectal tubular adenomas. Clin Cancer Res 10: 5870–5879, 2004.[Abstract/Free Full Text]
  2. Ammit AJ, Hastie AT, Edsall LC, Hoffman RK, Amrani Y, Krymskaya VP, Kane SA, Peters SP, Penn RB, Spiegel S, Panettieri RA Jr. Sphingosine 1-phosphate modulates human airway smooth muscle cell functions that promote inflammation and airway remodeling in asthma. FASEB J 15: 1212–1214, 2001.[Free Full Text]
  3. Andoh A, Bamba S, Fujiyama Y, Brittan M, Wright NA. Colonic subepithelial myofibroblasts in mucosal inflammation and repair: contribution of bone marrow-derived stem cells to the gut regenerative response. J Gastroenterol 40: 1089–1099, 2005.[CrossRef][Web of Science][Medline]
  4. Ballou LR, Barker SC, Postlethwaite AE, Kang AH. Sphingosine potentiates IL-1-mediated prostaglandin E2 production in human fibroblasts. J Immunol 145: 4245–4251, 1990.[Abstract]
  5. Ballou LR, Chao CP, Holness MA, Barker SC, Raghow R. Interleukin-1-mediated PGE2 production and sphingomyelin metabolism. Evidence for the regulation of cyclooxygenase gene expression by sphingosine and ceramide. J Biol Chem 267: 20044–20050, 1992.[Abstract/Free Full Text]
  6. Billich A, Bornancin F, Mechtcheriakova D, Natt F, Huesken D, Baumruker T. Basal and induced sphingosine kinase 1 activity in A549 carcinoma cells: function in cell survival and IL-1beta and TNF-alpha induced production of inflammatory mediators. Cell Signal 17: 1203–1217, 2005.[CrossRef][Web of Science][Medline]
  7. Brennan CM, Steitz JA. HuR and mRNA stability. Cell Mol Life Sci 58: 266–277, 2001.[CrossRef][Web of Science][Medline]
  8. Cok SJ, Acton SJ, Morrison AR. The proximal region of the 3'-untranslated region of cyclooxygenase-2 is recognized by a multimeric protein complex containing HuR, TIA-1, TIAR, and the heterogeneous nuclear ribonucleoprotein U. J Biol Chem 278: 36157–36162, 2003.[Abstract/Free Full Text]
  9. Dempke W, Rie C, Grothey A, Schmoll HJ. Cyclooxygenase-2: a novel target for cancer chemotherapy? J Cancer Res Clin Oncol 127: 411–417, 2001.[CrossRef][Web of Science][Medline]
  10. Di Mari JF, Mifflin RC, Adegboyega PA, Saada JI, Powell DW. IL-1alpha-induced COX-2 expression in human intestinal myofibroblasts is dependent on a PKCzeta-ROS pathway. Gastroenterology 124: 1855–1865, 2003.[CrossRef][Web of Science][Medline]
  11. Di Mari JF, Saada JI, Mifflin RC, Valentich JD, Powell DW. HETEs enhance IL-1-mediated COX-2 expression via augmentation of message stability in human colonic myofibroblasts. Am J Physiol Gastrointest Liver Physiol 293: G719–G728, 2007.[Abstract/Free Full Text]
  12. Doller A, Akool ES, Huwiler A, Muller R, Radeke HH, Pfeilschifter J, Eberhardt W. Posttranslational modification of the AU-rich element binding protein HuR by protein kinase Cdelta elicits angiotensin II-induced stabilization and nuclear export of cyclooxygenase 2 mRNA. Mol Cell Biol 28: 2608–2625, 2008.[Abstract/Free Full Text]
  13. Doller A, Huwiler A, Muller R, Radeke HH, Pfeilschifter J, Eberhardt W. Protein kinase C alpha-dependent phosphorylation of the mRNA-stabilizing factor HuR: implications for posttranscriptional regulation of cyclooxygenase-2. Mol Biol Cell 18: 2137–2148, 2007.[Abstract/Free Full Text]
  14. Edwards RA, Smock AZ. Defective arachidonate release and PGE2 production in Gi alpha2-deficient intestinal and colonic subepithelial myofibroblasts. Inflamm Bowel Dis 12: 153–165, 2006.[CrossRef][Web of Science][Medline]
  15. Hsieh HL, Wu CB, Sun CC, Liao CH, Lau YT, Yang CM. Sphingosine-1-phosphate induces COX-2 expression via PI3K/Akt and p42/p44 MAPK pathways in rat vascular smooth muscle cells. J Cell Physiol 207: 757–766, 2006.[CrossRef][Web of Science][Medline]
  16. Hu W, Mahavadi S, Huang J, Li F, Murthy KS. Characterization of S1P1 and S1P2 receptor function in smooth muscle by receptor silencing and receptor protection. Am J Physiol Gastrointest Liver Physiol 291: G605–G610, 2006.[Abstract/Free Full Text]
  17. Husoy T, Knutsen HK, Cruciani V, Olstorn HB, Mikalsen SO, Loberg EM, Alexander J. Connexin43 is overexpressed in Apc(Min/+)-mice adenomas and colocalises with COX-2 in myofibroblasts. Int J Cancer 116: 351–358, 2005.[CrossRef][Web of Science][Medline]
  18. Johann AM, Weigert A, Eberhardt W, Kuhn AM, Barra V, von Knethen A, Pfeilschifter JM, Brune B. Apoptotic cell-derived sphingosine-1-phosphate promotes HuR-dependent cyclooxygenase-2 mRNA stabilization and protein expression. J Immunol 180: 1239–1248, 2008.[Abstract/Free Full Text]
  19. Kabashima K, Saji T, Murata T, Nagamachi M, Matsuoka T, Segi E, Tsuboi K, Sugimoto Y, Kobayashi T, Miyachi Y, Ichikawa A, Narumiya S. The prostaglandin receptor EP4 suppresses colitis, mucosal damage and CD4 cell activation in the gut. J Clin Invest 109: 883–893, 2002.[CrossRef][Web of Science][Medline]
  20. Kawamori T, Osta W, Johnson KR, Pettus BJ, Bielawski J, Tanaka T, Wargovich MJ, Reddy BS, Hannun YA, Obeid LM, Zhou D. Sphingosine kinase 1 is up-regulated in colon carcinogenesis. FASEB J 20: 386–388, 2006.[Abstract/Free Full Text]
  21. Ki SH, Choi MJ, Lee CH, Kim SG. Galpha12 specifically regulates COX-2 induction by sphingosine 1-phosphate. Role for JNK-dependent ubiquitination and degradation of IkappaBalpha. J Biol Chem 282: 1938–1947, 2007.[Abstract/Free Full Text]
  22. Kim JI, Jo EJ, Lee HY, Cha MS, Min JK, Choi CH, Lee YM, Choi YA, Baek SH, Ryu SH, Lee KS, Kwak JY, Bae YS. Sphingosine 1-phosphate in amniotic fluid modulates cyclooxygenase-2 expression in human amnion-derived WISH cells. J Biol Chem 278: 31731–31736, 2003.[Abstract/Free Full Text]
  23. Kohno M, Momoi M, Oo ML, Paik JH, Lee YM, Venkataraman K, Ai Y, Ristimaki AP, Fyrst H, Sano H, Rosenberg D, Saba JD, Proia RL, Hla T. Intracellular role for sphingosine kinase 1 in intestinal adenoma cell proliferation. Mol Cell Biol 26: 7211–7223, 2006.[Abstract/Free Full Text]
  24. Kuehl FA Jr, Egan RW. Prostaglandins, arachidonic acid, and inflammation. Science 210: 978–984, 1980.[Abstract/Free Full Text]
  25. Lasa M, Mahtani KR, Finch A, Brewer G, Saklatvala J, Clark AR. Regulation of cyclooxygenase 2 mRNA stability by the mitogen-activated protein kinase p38 signaling cascade. Mol Cell Biol 20: 4265–4274, 2000.[Abstract/Free Full Text]
  26. Mahida YR, Beltinger J, Makh S, Goke M, Gray T, Podolsky DK, Hawkey CJ. Adult human colonic subepithelial myofibroblasts express extracellular matrix proteins and cyclooxygenase-1 and -2. Am J Physiol Gastrointest Liver Physiol 273: G1341–G1348, 1997.[Abstract/Free Full Text]
  27. Maines LW, Fitzpatrick LR, French KJ, Zhuang Y, Xia Z, Keller SN, Upson JJ, Smith CD. Suppression of ulcerative colitis in mice by orally available inhibitors of sphingosine kinase. Dig Dis Sci 53: 997–1012, 2008.[CrossRef][Web of Science][Medline]
  28. Mastrandrea LD, Sessanna SM, Laychock SG. Sphingosine kinase activity and sphingosine-1 phosphate production in rat pancreatic islets and INS-1 cells: response to cytokines. Diabetes 54: 1429–1436, 2005.[Abstract/Free Full Text]
  29. Mifflin RC, Saada JI, Di Mari JF, Adegboyega PA, Valentich JD, Powell DW. Regulation of COX-2 expression in human intestinal myofibroblasts: mechanisms of IL-1-mediated induction. Am J Physiol Cell Physiol 282: C824–C834, 2002.[Abstract/Free Full Text]
  30. Mutoh M, Takahashi M, Wakabayashi K. Roles of prostanoids in colon carcinogenesis and their potential targeting for cancer chemoprevention. Curr Pharm Des 12: 2375–2382, 2006.[CrossRef][Medline]
  31. Nodai A, Machida T, Izumi S, Hamaya Y, Kohno T, Igarashi Y, Iizuka K, Minami M, Hirafuji M. Sphingosine 1-phosphate induces cyclooxygenase-2 via Ca2+-dependent, but MAPK-independent mechanism in rat vascular smooth muscle cells. Life Sci 80: 1768–1776, 2007.[CrossRef][Web of Science][Medline]
  32. Ohama T, Hori M, Momotani E, Elorza M, Gerthoffer WT, Ozaki H. IL-1β inhibits intestinal smooth muscle proliferation in an organ culture system: involvement of COX-2 and iNOS induction in muscularis resident macrophages. Am J Physiol Gastrointest Liver Physiol 292: G1315–G1322, 2007.[Abstract/Free Full Text]
  33. Ohama T, Hori M, Momotani E, Iwakura Y, Guo F, Kishi H, Kobayashi S, Ozaki H. Intestinal inflammation downregulates smooth muscle CPI-17 through induction of TNF-alpha and causes motility disorders. Am J Physiol Gastrointest Liver Physiol 292: G1429–G1438, 2007.[Abstract/Free Full Text]
  34. Ohama T, Hori M, Sato K, Ozaki H, Karaki H. Chronic treatment with interleukin-1beta attenuates contractions by decreasing the activities of CPI-17 and MYPT-1 in intestinal smooth muscle. J Biol Chem 278: 48794–48804, 2003.[Abstract/Free Full Text]
  35. Pettus BJ, Bielawski J, Porcelli AM, Reames DL, Johnson KR, Morrow J, Chalfant CE, Obeid LM, Hannun YA. The sphingosine kinase 1/sphingosine-1-phosphate pathway mediates COX-2 induction and PGE2 production in response to TNF-alpha. FASEB J 17: 1411–1421, 2003.[Abstract/Free Full Text]
  36. Pettus BJ, Kitatani K, Chalfant CE, Taha TA, Kawamori T, Bielawski J, Obeid LM, Hannun YA. The coordination of prostaglandin E2 production by sphingosine-1-phosphate and ceramide-1-phosphate. Mol Pharmacol 68: 330–335, 2005.[Abstract/Free Full Text]
  37. Powell DW, Adegboyega PA, Di Mari JF, Mifflin RC. Epithelial cells and their neighbors I. Role of intestinal myofibroblasts in development, repair, and cancer. Am J Physiol Gastrointest Liver Physiol 289: G2–G7, 2005.[Abstract/Free Full Text]
  38. Spiegel S, Merrill AH Jr. Sphingolipid metabolism and cell growth regulation. FASEB J 10: 1388–1397, 1996.[Abstract]
  39. Spiegel S, Milstien S. Exogenous and intracellularly generated sphingosine 1-phosphate can regulate cellular processes by divergent pathways. Biochem Soc Trans 31: 1216–1219, 2003.[Web of Science][Medline]
  40. Turini ME, DuBois RN. Cyclooxygenase-2: a therapeutic target. Annu Rev Med 53: 35–57, 2002.[CrossRef][Web of Science][Medline]
  41. Valentich JD, Popov V, Saada JI, Powell DW. Phenotypic characterization of an intestinal subepithelial myofibroblast cell line. Am J Physiol Cell Physiol 272: C1513–C1524, 1997.[Abstract/Free Full Text]
  42. Wallace JL, Devchand PR. Emerging roles for cyclooxygenase-2 in gastrointestinal mucosal defense. Br J Pharmacol 145: 275–282, 2005.[CrossRef][Web of Science][Medline]
  43. Watterson KR, Ratz PH, Spiegel S. The role of sphingosine-1-phosphate in smooth muscle contraction. Cell Signal 17: 289–298, 2005.[CrossRef][Web of Science][Medline]
  44. Zhang Z, Andoh A, Inatomi O, Bamba S, Takayanagi A, Shimizu N, Fujiyama Y. Interleukin-17 and lipopolysaccharides synergistically induce cyclooxygenase-2 expression in human intestinal myofibroblasts. J Gastroenterol Hepatol 20: 619–627, 2005.[CrossRef][Web of Science][Medline]
  45. Zhou H, Murthy KS. Distinctive G protein-dependent signaling in smooth muscle by sphingosine 1-phosphate receptors S1P1 and S1P2. Am J Physiol Cell Physiol 286: C1130–C1138, 2004.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
295/4/G766    most recent
ajpgi.90423.2008v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Ohama, T.
Right arrow Articles by Ozaki, H.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Ohama, T.
Right arrow Articles by Ozaki, H.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Visit Other APS Journals Online
Copyright © 2008 by the American Physiological Society.