Apelin is the endogenous ligand for the APJ receptor, and apelin and APJ are expressed in the gastrointestinal (GI) tract. Intestinal inflammation increases intestinal hypoxia-inducible factor (HIF) and apelin expression. Hypoxia and inflammation are closely linked cellular insults. The purpose of these studies was to investigate the influence of hypoxia on enteric apelin expression. Exposure of rat pups to acute hypoxia increased hepatic, stomach-duodenal, and colonic apelin mRNA levels 10-, 2-, and 2-fold, respectively (P < 0.05 vs. controls). Hypoxia also increased colonic APJ mRNA levels, and apelin treatment during hypoxia exposure enhanced colonic APJ mRNA levels further. In vitro hypoxia also increased apelin and APJ mRNA levels. The hypoxia-induced elevation in apelin expression is most likely mediated by HIF, since HIF-activated apelin transcriptional activity is dependent on an intact, putative HIF binding site in the rat apelin promoter. Acute exposure of rat pups to hypoxia lowered gastric and colonic epithelial cell proliferation; hypoxia in combination with apelin treatment increased epithelial proliferation by 50%. In vitro apelin treatment of enteric cells exposed to hypoxia increased cell proliferation. Apelin treatment during normoxia was ineffective. Our studies imply that the elevation in apelin expression during hypoxia and inflammation in the GI tract functions in part to stimulate epithelial cell proliferation.
- gastrointestinal tract
apelin is the endogenous ligand for a G protein-coupled receptor, the APJ receptor (1, 4, 14). Apelin and APJ are expressed widely in the body. They are produced in the brain, kidney, adipose tissue, lung, mammary gland, gastrointestinal (GI) tract, and cardiovascular system (9, 14, 18, 25–27, 35, 38, 39, 41). Like other regulatory peptides, apelin exerts a broad range of activities affecting multiple organ systems, including effects on heart contractility and blood pressure (3, 5), appetite and drinking behavior (22, 36), immune response (20), gastric acid secretion (21), and insulin and cholecystokinin secretion (33, 41). In addition, apelin has been shown to affect cell motility, proliferation, and apoptosis (10, 12, 15, 17, 24, 34, 37, 44, 45).
Our laboratory (10) has reported that apelin expression in the GI tract is activated by inflammation. In mice and rats with experimental colitis, intestinal apelin production is upregulated, and in humans with inflammatory bowel disease, apelin immunostaining is increased in the colonic epithelium (10). In addition, in rodents, lipopolysaccharide (LPS) administration increases intestinal apelin expression levels (Greeley GH, unpublished data). Inflammation and hypoxia are closely associated insults during cellular stress and may exert their influences by activation of common signaling pathways.
The aim of the present study was to examine the influence of acute hypoxia on apelin and APJ expression in the GI tract and liver, as well as in vitro using cell lines derived from the GI tract and liver. In addition, because apelin administration has been shown to enhance angiogenesis as well as proliferation of retinal and colonic epithelial cells (10, 17), we examined the effect of exogenous apelin on enteric cell proliferation during hypoxia. In this report, we demonstrate that hypoxia elevates apelin and APJ expression and that exogenous apelin increases cell proliferation in vivo in the GI tract and in vitro in cells derived from the GI tract and liver. Furthermore, we demonstrate that HIF overexpression stimulates apelin gene transcription. Together, these findings imply that the increased apelin expression during inflammation and hypoxia plays a stimulatory role on cell proliferation.
MATERIALS AND METHODS
All animal experiments were conducted in accordance with mandated standards of humane care and were approved by the University of Texas Medical Branch Institutional Animal Care and Use Committee. Sprague-Dawley rats were maintained in air-conditioned and light-regulated rooms (lights on 0600–1800) and given access to food and water ad libitum. Rats were exposed to hypoxia as reported earlier (23).
Primary Cell Cultures
Colons, ileums, and stomachs of 6-day-old rat pups were harvested and rinsed three times in PBS, minced, and digested by incubation with collagenase A (2 mg/ml, 37°C, 30 min; Roche, Indianapolis, IN). Cells were then washed with Dulbecco's modified Eagle's medium (DMEM), centrifuged at low speed, and resuspended in medium; this procedure was repeated twice before cells were plated onto plates or dishes in DMEM containing 10% fetal bovine serum to facilitate attachment of cells and crypts. Confluent monolayers formed within 3–4 days. Cultured cells were then exposed to hypoxia (0.5% O2, 24 or 48 h). Primary cultured cells consisted of monolayer cultures of a mixed enteric cell population.
Measurement of Epithelial Cell Proliferation by Bromodeoxyuridine Incorporation
Formalin-fixed, paraffin-embedded rat pup stomach and colon sections were deparaffinized, treated with 1% H2O2 for 15 min, and then subjected to antigen retrieval at 100°C for 10 min in DAKO Target Retrieval Solution in a H2800 microwave processor. Slides were treated with 0.1% avidin and 0.01% biotin sequentially, followed by 0.03% casein in 0.05% Tween-PBS before application of a biotinylated mouse anti-bromodeoxyuridine (BrdU) antibody (Molecular Probes, Eugene, OR). Mouse Ig Ready-to-Use (InnoGenex, San Ramon, CA) served as a negative control. Antibody incubations, detection with universal DAKO LSAB2 HRP system, and colorization with diaminobenzidine (DAB) were done with a DAKO Autostainer. Slides were counterstained with Mayer's modified hematoxylin (Poly Scientific, Bay Shore, NY) before they were mounted. The number of BrdU-stained cells added was examined in vehicle- and apelin-treated groups.
Western Blot Analysis
Cells were harvested and sonicated in cell lysis buffer (catalog no. 9803; Cell Signaling, Beverly, MA) to extract total cellular proteins. Supernatants were harvested, and protein concentration was quantitated. Total cellular protein extracts (30 μg) were then boiled for 5 min in the presence of sample loading buffer (NuPAGE LDS sample buffer; Invitrogen, Carlsbad, CA) before separation on a 4–12% SDS-PAGE. After gel separation, proteins were transferred onto nitrocellulose membranes and incubated with a hypoxia-inducible factor-1α (HIF-1α) antibody (NB 100–105H, 1:2,000; Novus, Littleton, CO) for 3 h at room temperature. Membranes were then incubated for 45 min at room temperature with conjugated anti-mouse (1:5,000) IgG-horseradish peroxidase antibody (1030-05; SouthernBiotech, Birmingham, AL) in washing buffer. Detection was done using the enhanced chemiluminescence (ECL) method according to the manufacturer's instructions (NEN Life Science Products, Boston, MA).
Measurement of Apelin and APJ Expression Levels
For measurement of apelin expression levels in cell lines, total cellular RNA was extracted and purified by means of the RNAqueousTM kit (AB-Ambion, Foster City, CA) as described previously (42). RNA was quantified by spectrophotometry. Before real-time PCR analyses were performed, total RNA was used as a template to synthesize first-strand cDNA by the random priming method using the AdvantageTM RT-for-PCR kit (BD Biosciences-Clontech). Briefly, 1 μg of total RNA was incubated with 1 μl of random primer at 70°C for 2 min, chilled on ice, and mixed with a solution containing 4 μl of reaction buffer (5×), 1 μl of deoxynucleotide (10 mM), 0.5 μl of recombinant RNase inhibitor (40 U/μl), and 1.0 μl of Moloney murine leukemia virus reverse transcriptase (200 U/μl). The mixture (20 μl) was incubated for 1 h at 42°C, and the reverse transcriptases were denatured at 94°C for 5 min. Apelin mRNA levels were then measured using real-time RT-PCR assays as described previously (41, 42). Assays were done using an Applied Biosystems 7000 sequence detection system (Foster City, CA). Applied Biosystems Assays-By-Design containing a 20× assay mix of primers and TaqMan MGB probes (6-FAM dye-labeled probe) were used for the target genes: rat apelin (accession no. AF179679), human apelin (accession no. NM_017413), mouse apelin (accession no. NM_013912), rat APJ (accession no. NM_031612), and a predeveloped 18S rRNA (VIC dye-labeled probe). TaqMan assay reagent (P/N 4319413E) was used for the internal control. Primers were designed to span exon-exon junctions. Probe sequences were searched against the Celera database.
For measurement of apelin expression levels in the rat stomach and ileum, total cellular RNA was extracted and purified from stomach and ileal tissue extracts, and Northern blot analysis was conducted using previously published procedures (8, 41, 42).
The aim of experiment 1 was to examine the effects of acute hypoxia on apelin expression levels in the GI tract and liver and in the colon of 24- and 6-day-old Sprague-Dawley rat pups, respectively. The 24- and 6-day-old rat pups with their mother were exposed to acute respiratory hypoxia (9.5% O2) for 72 and 24 h, respectively. Rat pups were killed immediately after hypoxia exposure. Control rat pups were exposed to ambient air. The liver, stomach, and duodenum were harvested from 6-day-old rat pups, and the colon was harvested from 24-day-old rat pups. Total cellular RNA was prepared for measurement of apelin and APJ expression levels.
The aim of experiment 2 was to examine the effects of acute hypoxia (24 h, 0.5% O2) on apelin expression levels in cultured human liver HepG2 and mouse enteroendocrine (STC-1) cells and in primary cultures of rat stomach, ileal, and colonic cells. Control cells were exposed to 21% oxygen. Cells were harvested, total cellular RNA was prepared, and apelin expression levels were measured. APJ expression levels were measured in primary rat ileal cells exposed to acute hypoxia. Western blotting was used to verify that hypoxia [0.5% O2, 150 μM cobalt chloride (CoCl2)] or HIF overexpression elevated HIF-1α protein levels in liver (HepG2) cells.
The aim of experiment 3 was to examine the influence of hypoxia alone or in combination with apelin treatment on GI cell proliferation in vivo. Stomach and colonic epithelial cell proliferation was measured by incorporation of BrdU. Twenty-four-day-old rat pups were exposed to hypoxia alone for 72 h or to hypoxia plus apelin (50 μg/rat, 2 times per day, intraperitoneally) or vehicle treatment (0.05% BSA in saline). Rat pups were killed 16 h after the last apelin injection, and BrdU (6.4 mg/rat ip) was administered 90 min before harvest of stomach and colon tissues. The hypoxia chamber was opened briefly for apelin and BrdU injections. Extirpated tissues were placed in 10% formalin for subsequent histological processing and examination. Histological sections of the stomach and colon were immunostained for BrdU incorporation, and the numbers of BrdU-immunostained cells per field were recorded for vehicle- and apelin-treated rats. Five separate fields per stomach for each rat and 30 colonic crypts per rat for BrdU incorporation were examined.
The aim of experiment 4 was to examine the influence of apelin treatment on cell number in enteric cells exposed to hypoxia. Liver (HepG2) cells and primary rat stomach cells were plated and grown in medium containing 10% fetal bovine serum for ∼24 h. Medium was then replaced with medium without FBS, and cells were exposed to chemical hypoxia (150 μM CoCl2) or 0.5%O2 for 48 h. During hypoxia exposure, cells were treated with either vehicle (0.05% BSA) or apelin (10−8–10−9M). Cell numbers were measured using a Cell Counting Kit (Dojindo Laboratory, Kumamoto, Japan).
The influence of HIF-1α and HIF-1β overexpression on wild-type and mutated rat apelin promoter activity was examined in transient transfection experiments. Liver (HepG2) cells were plated onto 12-well tissue culture plates at optimal densities (2 × 105 cells/well). Approximately 24 h after plating, 5′-upstream rat apelin promoter fragment-luciferase reporter constructs (0.8 μg/well; −3000/−1, −1500/−1, −1000/−1, and −407/−1 bp) were transfected with either HIF-1α (0.8 μg/well) or HIF-1β (0.8 μg/well) alone or HIF-1α (0.4 μg/well) and HIF-1β (0.4 μg/well) in combination. Cells were cotransfected with an internal control expression vector (pRL-SV40, 0.02 μg/well). HIF-1α and HIF-1β expression plasmids were obtained from ATCC (Manassas, VA) and O. Hankinson (13), respectively. Apelin promoter fragment-luciferase reporter constructs were prepared in our laboratory (42).
Sequence analysis identified a putative HIF binding site (31) at −685/−682 bp in the rat apelin promoter. This site was mutated using Quick Change site-directed mutagenesis kits (Stratagene, San Diego, CA). The influence of mutation of the putative HIF binding site on apelin promoter activity was tested in transient transfection experiments using wild-type and mutated −1500/−1 and −1000/−1 bp rat apelin promoter constructs (0.8 μg/well) with HIF-1α (0.4 μg/well) and HIF-1β (0.4 μg/well) overexpression. The mutation reaction was done using Pfu Turbo DNA polymerase at 95°C for 60 s followed by 18 cycles of PCR at 95°C for 50 s, 60°C for 50 s, and 68°C for 6.5 min, and then at 68°C for 10 min. The parental double-stranded DNA was digested with DpnI. Mutant sequences were confirmed by DNA sequencing.
Acute Hypoxia Increases Apelin Expression in the rat GI Tract and Liver
The aim of this experiment was to examine the effect of acute hypoxia on apelin expression (mRNA) levels in the rat GI tract and liver. The 6- and 24-day-old rat pups were exposed to hypoxia (9.5% O2) for 24 and 72 h, respectively. Control pups were exposed to ambient air. In 6-day-old rat pups, acute exposure to hypoxia increased apelin expression levels ∼10-fold in the liver and 2-fold in an extract made from the stomach and proximal duodenum (P < 0.05) (Fig. 1A). In 24-day-old rat pups, acute hypoxia increased colonic apelin expression levels ∼2-fold (Fig. 1B).
Hypoxia Activates Apelin Expression in Cultured Enteric Cells
Liver (HepG2), intestinal enteroendocrine (STC-1) and primary rat stomach, ileal, and colonic cells were exposed to hypoxia (0.5% O2) for 24 h. Control cells were exposed to 21% oxygen. Hypoxia exposure increased apelin mRNA levels ∼25-fold in liver, 3-fold in enteroendocrine, and 2-, 3-, and 4-fold in primary stomach, ileal, and colon cells, respectively (Fig. 2). Figure 3 indicates that hypoxia exposure increased HIF-1α protein levels in liver (HepG2) cells. Chemical hypoxia (CoCl2) and overexpression of HIF-1α also increased HIF-1α protein levels.
Hypoxia and Apelin Treatment Increase APJ Expression in the Rat Colon and in Cultured Ileal Cells
Acute exposure of 24-day-old rat pups to hypoxia (9.5% O2, 72 h) increased APJ expression levels approximately onefold in the rat colon (Fig. 4A). Apelin treatment of rats exposed to hypoxia resulted in a significant increase of colonic APJ expression levels (30%, P < 0.05) compared with APJ expression levels of rats exposed to hypoxia alone. Acute exposure of cultured ileal cells to hypoxia also increased APJ expression levels significantly and approximately onefold (P < 0.05) (Fig. 4B). Although hypoxia exposure in combination with apelin treatment of cultured ileal cells decreased APJ expression levels, the decrease was not statistically significant. In in vivo and in vitro conditions, apelin treatment during normoxia did not affect APJ expression levels.
Apelin Treatment During Hypoxia Increases Enteric Epithelial Cell Proliferation
Previous reports indicate that either hypoxia exposure or apelin treatment alone is associated with changes in cell proliferation and apoptosis (10, 15, 17, 24, 34, 37, 44). The aim of this experiment was to examine the effect of acute hypoxia exposure alone or in combination with apelin treatment on GI cell proliferation in vivo. Proliferation was monitored by measurement of BrdU incorporation into the rat gastric and colonic epithelium. Acute exposure of rat pups to hypoxia (9.5% O2, 72 h) decreased colonic and gastric epithelial cell proliferation by ∼50% (Table 1, Fig. 5). Apelin treatment in combination with hypoxia increased gastric and colonic epithelial proliferation by 50% compared with hypoxia alone. In rats exposed to normoxia, apelin treatment did not affect epithelial cell proliferation.
In a second experiment, the effect of apelin treatment during hypoxia on enteric cell number was examined in vitro. During chemical hypoxia, the number of cultured liver HepG2 and stomach cells decreased significantly (Fig. 6, B and C). Apelin treatment (10−8M) of liver cells exposed to hypoxia significantly increased cell number by 40% (P < 0.05) (Fig. 5A). Apelin treatment (10−9M) of either liver or primary rat stomach cells exposed to chemical hypoxia increased cell number by 40% (P < 0.05) (Fig. 5, B and C). In contrast, apelin treatment of liver and primary stomach cells exposed to normoxia did not affect cell number.
HIF Overexpression Increases Apelin Promoter Activity
Sequence analysis of the rat apelin 5′-upstream region identified two putative HIF binding sites (−2987/−2984 and −685/−682 bp) (Fig. 7, top). Transient transfection experiments using an apelin promoter-luciferase reporter construct (−3000/−1 bp) in liver (HepG2) cells showed that overexpression of HIF-1α plus HIF-1β increased apelin promoter activity ∼150% (Fig. 7, bottom). Overexpression of HIF-1α but not of HIF-1β alone stimulated apelin promoter activity marginally.
Transient transfection experiments were done in liver (HepG2) cells using apelin promoter deletion-luciferase reporter constructs that span either two or one putative HIF binding sites (−3000/−1, −1500/−1, or −1000/−1 bp). Cells were transfected with either HIF-1α plus HIF-1β expression vectors or empty control vectors. Apelin promoter activity increased 1.5- to 2.5-fold in response to overexpression of HIF-1α plus HIF-1β compared with control transfections (Fig. 8). Promoter activity of the −407/−1-bp apelin promoter-reporter construct was not influenced by HIF overexpression; this apelin promoter fragment does not contain a putative HIF binding site.
Transient transfection experiments in liver (HepG2) cells using wild-type and mutated rat apelin promoter-reporter constructs (−1500/−1 and −1000/−1 bp) showed that mutation of a putative HIF binding site in the apelin promoter (−684/−682 bp) abolished HIF-induced promoter activity (Fig. 9).
The apelin-APJ signaling system is expressed in the GI tract and pancreas (33, 35, 41). Our laboratory has shown that colonic apelin expression increases during colonic inflammation (10). The underlying mechanisms for this elevation in apelin production during inflammation are not known. Hypoxia and inflammation are closely linked responses to cellular insults (6, 40), and hypoxia occurs during colitis (2, 16, 28). The influence of hypoxia independent of inflammation on enteric apelin expression has not been investigated previously.
In the present study, we showed that hypoxia activates apelin expression in the GI tract in vivo and in cultured enteric cells. In addition, we showed in in vivo and in in vitro experiments that hypoxia increases expression of APJ, the apelin receptor. The most intriguing findings are that exogenous apelin stimulates a modest elevation in epithelial proliferation in vivo and in vitro during hypoxia but not in normoxia. Furthermore, this study showed that hypoxia increases apelin expression levels by activation of apelin transcriptional activity, since overexpression of the hypoxia-responsive transcription factor HIF increases apelin promoter activity. This elevation is mediated by binding of HIF to a putative HIF binding site in the apelin promoter, since mutation of this putative HIF binding site abolishes HIF-induced promoter activity.
Our findings showing that hypoxia increases enteric apelin and APJ expression extend earlier reports that apelin expression is activated by hypoxia in vivo in the heart, lung, skeletal muscle, adipose tissue, and tumor vascular tissue and in vitro in cells derived from these tissues (7, 29, 32, 34). Hypoxia will also increase APJ expression in the heart and lung in vivo (32). This upregulation of APJ expression may be one mechanism underlying the stimulatory action of exogenous apelin on epithelial proliferation during hypoxia.
We showed that apelin treatment stimulates a modest proliferation of enteric cells in in vitro and in vivo hypoxia experiments. These findings agree with and extend an earlier report from our laboratory (10) showing that apelin stimulates proliferation of the colonic epithelium in mice with experimentally (sodium dextran sulfate)-induced colonic inflammation. In this earlier report, apelin administration caused a 50% increase in proliferation. Because hypoxia and inflammation are intimately linked cellular insults (2, 16, 28), our findings imply that apelin acts as a mitogenic agent in the GI tract during conditions of stress. This notion extends a report made by another laboratory indicating that the apelin-APJ system is a mediator of oxidative stress in vascular tissue (11).
It should be pointed out that despite the stimulatory activity of exogenous apelin on epithelial proliferation during hypoxia, proliferation is not restored to a level comparable to that measured in animals exposed to normoxia. This finding agrees with and extends data showing a reduced colonic epithelial proliferation in animals exposed to hypoxia or with experimental colitis and having increased expression of colonic apelin. Together, these findings suggest that the hypoxia-induced elevation in endogenous apelin expression and the exogenous apelin treatment during hypoxia are inadequate to maintain or restore epithelial proliferation completely. Other growth factors are likely required.
In addition, the present data demonstrate that epithelial proliferation in the GI tract continues during hypoxia but at a reduced level. Although speculative, this low level of proliferation may be supported by the elevation in intestinal apelin expression.
The present study also demonstrates that the hypoxia-sensitive transcription factor HIF mediates the elevation in apelin expression in the GI tract during hypoxia. In our work, exposure of enteric cells to either hypoxia or the chemical HIF activator, CoCl2 increased apelin expression levels. Hypoxia in vivo also elevated enteric apelin expression. It is well known that cellular hypoxia will increase expression of numerous genes involved in enhancing cell proliferation, metabolism, and oxygenation (19, 30, 43, 46). Putative HIF binding sites have been identified in the mouse and human apelin promoters (32), and in this study we identified two putative HIF binding sites in the rat apelin promoter. We also showed that the proximal HIF binding site is functional in hypoxia-induced apelin expression, since mutation of this site abolished HIF-induced apelin promoter activity. Interestingly, our findings are in agreement with a previous report showing that HIF-1α overexpression increases apelin expression moderately in mouse fibroblasts that is abolished in fibroblasts derived from a HIF-1α gene knockout mouse (7).
Perspectives and Significance
The important findings of the current study are that hypoxia will increase apelin and APJ expression in the GI tract. This upregulation of apelin and APJ by hypoxia may be clinically relevant, since hypoxia occurs during many, if not all, enteric diseases. Furthermore, hypoxia-induced elevations in enteric apelin and APJ expression may be an adaptive mechanism that functions partly to promote epithelial proliferation, albeit at a reduced level. The extent to which endogenous apelin exerts this proposed activity can be better defined by measuring proliferation in an animal model deficient in either apelin or APJ.
This work is supported by Crohn's & Colitis Foundation of America Grant CON 16034 and National Institute of Diabetes and Digestive and Kidney Diseases Grant P01 DK035608.
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.
- Copyright © 2008 the American Physiological Society