Recent in vivo studies demonstrated that aging gastric mucosa has impaired angiogenesis and reduced expression of vascular endothelial growth factor (VEGF). Angiogenesis is triggered by hypoxia and VEGF gene activation, and the latter requires transport of transcription factor(s) into endothelial cell nuclei. We focused on gastric mucosal endothelial cells (GMEC), which are key targets and effectors of gastric angiogenesis, and determined whether and to what extent importin-α, a nuclear transport protein, regulates VEGF gene activation and gastric angiogenesis and the possible role of importin-α in aging gastropathy. GMEC were isolated from rats 3 and 24 mo of age, young (YGEC) and aging (AGEC), respectively. We examined in these cells 1) in vitro angiogenesis, 2) expression of VEGF and importin-α, 3) nuclear transport of hypoxia-inducible factor (HIF)-1α by importin-α, 4) binding of HIF-1α to the VEGF gene promoter, and 5) effects of importin-α silencing in YGEC and its upregulation in AGEC on angiogenesis and VEGF expression. AGEC exhibited significantly impaired in vitro angiogenesis by fourfold and decreased expression of VEGF, importin-α, and nuclear HIF-1α by 1.4-fold, 1.6-fold, and 2.9-fold, respectively, vs. YGEC. Upregulation of importin-α in AGEC significantly reversed all these abnormalities. In YGEC, knockdown of importins-α1 and -α3 significantly reduced in vitro angiogenesis by 93% and 73% and VEGF expression by 48% and 52%, respectively. The above findings demonstrate that importin-α is a novel and critical regulator of gastric angiogenesis. Its reduced expression in AGEC is the key mechanism for impaired angiogenesis and reduced VEGF.
- gastric endothelial cells
- hypoxia-inducible factor-1α
clinical and experimental studies indicate that gastric mucosa of aging individuals (referred to herein as aging gastric mucosa or aging gastropathy) has impaired mucosal defense, e.g., reduced prostaglandin generation, decreased mucus and bicarbonate secretion, and an impaired sensory nerve response to luminal acid (6, 8, 14, 24, 26, 30, 32, 37, 38, 40). Not surprisingly, aging gastric mucosa has increased susceptibility to injury by damaging agents including nonsteroidal anti-inflammatory drugs (NSAIDs) and ethanol (25, 29, 30, 33). Our previous study identified specific cellular and molecular abnormalities in aging gastric mucosa: hypoxia, increased expression of phosphatase and tensin homolog deleted on chromosome 10, activation of caspases -3 and -9, and reduced expression of survivin that increase susceptibility of the gastric mucosa to injury (35). Aging gastric mucosa also exhibits impaired healing of both acute injury and chronic gastric ulcers (12, 26, 32, 36, 37) and reduced therapeutic efficacy of ulcer-healing drugs, including H2 receptor antagonists and proton-pump inhibitors (18).
We recently examined in vivo angiogenesis, new capillary blood vessel formation from preexisting vessels, and the healing of gastric mucosal injury (erosions) in aging and young rats and the underlying mechanisms (2). That study demonstrated that the gastric mucosa of aging rats has severely reduced angiogenesis in response to injury and that this reduction directly correlates with impaired healing of mucosal erosions. One of the main underlying mechanisms of impaired in vivo angiogenesis in aging gastric mucosa is reduced mucosal expression of vascular endothelial growth factor (VEGF), which is a potent and fundamental angiogenic growth factor (9–11). However, reduced expression of VEGF in aging gastric mucosa was unexpected because aging gastric mucosa exhibits a marked hypoxia (35), which usually is a strong stimulus for VEGF gene activation. The mechanism underlying reduced VEGF expression in aging gastric mucosa remains unexplained.
Angiogenesis in gastric mucosa is accomplished by endothelial cell migration, proliferation, and tube formation, as demonstrated in our recent study (2). Angiogenesis is initiated and regulated by angiogenic growth factors, e.g., VEGF, a fundamental stimulator of angiogenesis that acts on endothelial cells, which are the key targets and effectors of angiogenesis (9–11). The activation of VEGF gene expression requires nuclear transport and binding of hypoxia-inducible factor (HIF)-1α (the hypoxia sensor and transcription factor) to the VEGF gene promoter (9, 27). In some cells, e.g., HeLa cells, nuclear transport of various proteins such as HIF-1α and other transcription factors is mediated by importin-α (7, 23, 39). We hypothesized that reduced VEGF expression and impaired angiogenesis in aging gastric mucosal endothelial cells are due to downregulation of importin-α that theoretically can transport HIF-1α into the nucleus of endothelial cells. Our overall hypothesis was that importin-α is critical for angiogenesis and VEGF gene expression in gastric mucosal endothelial cells (GMEC) and that its downregulation during aging results in decreased nuclear transport of HIF-1α in these cells, leading to reduced VEGF expression and impaired angiogenesis. We examined this hypothesis in vitro in endothelial cells isolated from young and aging rat gastric mucosae.
The aim of the present study was to dissect the mechanism of impaired angiogenesis and reduced VEGF expression in aging gastric mucosa at the cellular and molecular level. Because GMEC are key targets and effectors of angiogenesis, we focused on these cells and performed in vitro studies using isolated rat GMEC. This in vitro approach enabled us to directly manipulate expression levels of key importin-α isoforms and to dissect the consequences of increased and reduced importin-α expression on the mechanisms that underlie the endothelial cell switch from a quiescent to angiogenic phenotype.
MATERIALS AND METHODS
Isolation and culture of young and aging microvascular GMEC.
All experimental studies in rats were approved by the institutional animal review committee [Subcommittee for Animal Studies (IACUC)] of the VA Long Beach Healthcare System. Microvascular GMEC were isolated from Fisher F-344 rats: 3 mo of age (young, YGEC, n = 24) and 24 mo of age (aging, AGEC, n = 24), purchased from the National Institute on Aging. We used an isolation protocol similar to that used in our previous study (21). Rats were euthanized, and the stomachs were removed and opened along the greater curvature. The gastric mucosa was removed, cut into small pieces, and sequentially incubated with collagenase II and trypsin. The resulting cell suspension was passed through polypropylene meshes. Cells were then selected for platelet endothelial cell adhesion molecule (PECAM)-1 expression using anti-PECAM-1 (anti-CD31) antibody (Chemicon, Temecula, CA) by magnetic bead separation (Miltenyi Biotec, Auburn, CA). The endothelial cells isolated from gastric mucosae of young and aging rats, YGEC and AGEC, respectively, were characterized by positive staining for the endothelial markers, von Willebrand's factor (Factor VIII - related antigen) and PECAM-1 (CD31), and, by absence of staining for the myofibroblast marker, smooth muscle α-actin. Endothelial cells were grown on collagen-coated dishes in endothelial cell growth media containing 10% FBS, heparin, and endothelial cell growth supplements. For some studies, these cells were cultured under hypoxia (5% CO2-94% N2-1% O2) in cell culture dishes inside a modified anaerobic growth chamber, similar as in our previous study (19). In some experiments, AGEC were treated with either 1 mM 5-amino-imidazole-4-carboxamide ribonucleotide (AICAR) (Sigma-Aldrich, St. Louis, MO) to induce importin-α expression (5, 16, 39) or its solvent (PBS; control) for 1, 2, 4, or 6 h. YGEC and AGEC were transfected with either 100 nM of specific importin-α1 and importin-α3 siRNA or nontargeting, nonsilencing control siRNA (Qiagen, Valencia, CA) using HiperFect (Qiagen).
In vitro angiogenesis assay.
Endothelial tube formation on growth factor-reduced Matrigel was determined using the in vitro angiogenesis assay similar to our previous studies (4, 20, 21). YGEC and AGEC were grown in complete growth media in 60-mm tissue culture dishes until they were about 80% confluent. The growth medium was replaced with basal medium supplemented with 1% FBS and antibiotics, and the cells were incubated for another 18 h. The cells were then trypsinized, counted, resuspended in basal medium supplemented with 1% FBS, and seeded onto growth factor-reduced Matrigel in 48-well culture plates for up to 24 h. Six and twenty-four hours later, the seeded cells were photographed using a Nikon inverted-phase contrast photomicroscope (Nikon USA, Garden City, NY) and analyzed using a video image analysis system (MetaMorph 7.0; Molecular Devices, Downington, PA). Tube formation was quantified by measuring the total pixel length of the capillary tubes in five randomly selected standardized fields for each well under ×200 magnification.
Quantitative real-time RT-PCR.
We examined mRNA levels of VEGF, importin-α1, and importin-α3 by real-time reverse transcription PCR using prevalidated QuantiTect assays (Qiagen) and the iCycler real-time PCR detection system (Bio-Rad, Hercules, CA) and methods described in our previous studies (1, 2). These particular isoforms of importin-α were chosen based on the report of their high-affinity for HIF-1α in HeLa cells (7). Total RNA was isolated from cultured gastric endothelial cells using Trizol reagent (Invitrogen, Carlsbad, CA) and 1 μg of total RNA was treated with deoxyribonuclease I and reverse transcribed using the GeneAmp RNA-PCR kit (Applied Biosystems, Foster City, CA). Quantitative PCR on 2.5 μl cDNA was performed using prevalidated QuantiTect assays (Qiagen) and the iCycler real-time PCR detection system (Bio-Rad). Relative mRNA levels were calculated using the 2−ΔΔCt method, and glyceraldehyde-3-phosphate dehydrogenase was used as a reference.
Western blot analysis.
Total cellular protein or cytoplasmic and nuclear proteins were isolated from YGEC and AGEC using commercial buffers and kits (Pierce, Fisher Scientific, Pittsburg, PA). Expression of VEGF, HIF-1α, importin-α, AMPK, and P-AMPK was determined in these samples by Western blotting with respective specific antibodies using methods described in our previous study (3). β-Actin was used as a reference. The primary antibodies used were against VEGF (1:250; Santa Cruz Biotechnology, Santa Cruz, CA), HIF-1α (1:250; Novus Biologicals, Littleton, CO), AMPK, P-AMPK (1:250; Cell Signaling Technology, Andover, MA), importin-α (1:500; Sigma-Aldrich), or β-actin (1:1,000, Sigma-Aldrich). Importin-α exists in several isoforms, and the antibody used in this study recognizes importins-α1, -α3, -α5, and -α7 isoforms.
The expression and localization of VEGF, HIF-1α, and importin-α in YGEC and AGEC were examined by immunostaining with respective specific antibodies described above using methods described in our previous study (3). The staining signal intensity was quantified using MetaMorph 7.0 (Molecular Devices, Downington, PA) and expressed as arbitrary units.
EMSA to determine the binding of HIF-1α to the VEGF gene promoter.
Electrophoretic mobility shift assay (EMSA) was performed with 50 μg of nuclear protein and a 32P-labeled probe representing the portion of the VEGF promoter containing the HIF-1α consensus binding sequence 5′-TACGTGGG-3′ (Operon Biotechnologies, Huntsville, AL) using methods described in our previous study (3).
Data are presented as means ± SD. Statistical significance was analyzed by either Student's t-test, analysis of variance, or Pearson's correlation. A P value of <0.05 was considered statistically significant.
GMEC isolated from aging rats (AGEC) have impaired angiogenesis and reduced VEGF expression.
We examined in vitro angiogenesis and expression of VEGF in GMEC isolated from young (YGEC) and aging (AGEC) rats at baseline (normoxia). AGEC had a fourfold reduction in capillary-like tube formation on growth factor-reduced Matrigel, which reflects in vitro angiogenesis (Fig. 1A), and 2.6-fold and 2.7-fold decreased VEGF mRNA and protein expression, respectively (Fig. 1B) compared with YGEC (all P < 0.001). Treatment of AGEC with exogenous VEGF increased in vitro angiogenesis by 2.3-fold (P < 0.001) (Fig. 1C). These results suggest that reduced VEGF expression is partly responsible for the impairment of angiogenesis in AGEC.
Detailed analysis of in vitro angiogenesis demonstrated that YGEC form lumina when seeded on Matrigel (Fig. 2, A and B; arrows), whereas AGEC do not form lumina in rudimentary endothelial tubes (Fig. 2, C and D). Thus the angiogenic process is not only reduced quantitatively in AGEC, but the quality of tube and vessel formation is also impaired.
In response to hypoxia, AGEC do not exhibit increased angiogenesis, nor do they have increased expression of VEGF or nuclear HIF-1α.
Because hypoxia is a fundamental regulator of VEGF expression (10, 27), we next examined the response of AGEC and YGEC to hypoxia. Culture of YGEC under hypoxia resulted in a 1.5-fold increase in capillary-like tube formation on growth factor-reduced Matrigel (Fig. 3A), a 1.4-fold increase in VEGF protein expression (Fig. 3B), and a 2.9-fold increase in nuclear HIF-1α protein expression (Fig. 3C) compared with normoxia (all P < 0.001). In contrast, in AGEC, hypoxia did not affect either in vitro angiogenesis (Fig. 3A) or VEGF expression (Fig. 3B). AGEC had 2.1-fold reduced levels of nuclear HIF-1α (P < 0.001) vs. YGEC after culture under hypoxia (Fig. 3C), reflecting impaired nuclear transport of HIF-1α. Immunofluorescence staining showed that hypoxia induced the nuclear translocation of HIF-1α in YGEC (Fig. 3C; arrows) but not in AGEC. These results suggest that AGEC have defective responses to hypoxia.
Reduced importin-α levels in AGEC.
To determine whether the levels of nuclear transport protein, importin-α, are changed as a consequence of aging, we next examined expression of importin-α mRNA (isoforms 1 and 3) and total importin-α protein in AGEC and YGEC by real-time RT-PCR and Western blotting, respectively. AGEC had a 1.7-fold and 2.7-fold reduction in expression levels of importin-α1 and importin-α3 mRNA, respectively (both P < 0.01) vs. YGEC (Fig. 4A). AGEC had a 1.6-fold reduction in total importin-α protein levels (P < 0.01) vs. YGEC (Fig. 4B) as demonstrated by Western blotting using antibody, which detects both importin-α1 and importin-α3 isoforms. These results showed that the expression of importin-α in GMEC closely correlates with both nuclear HIF-1α levels (correlation coefficient r = 0.994, P < 0.01) and VEGF levels (correlation coefficient r = 0.993, P < 0.05).
Upregulation of importin-α increases VEGF levels and restores impaired angiogenesis in AGEC.
Because numerous attempts at overexpressing importin-α using transfection of expression vectors into AGEC were not successful, we used AICAR, the pharmacological activator of AMP-activated protein kinase (AMPK) (5, 16, 39), to upregulate and activate importin-α in AGEC. At 2 h, AICAR increased importin-α protein expression in AGEC by 1.6-fold (P < 0.001) vs. vehicle (PBS) (Fig. 5A). AICAR-treated AGEC had a 1.8-fold increase in vitro angiogenesis (P < 0.01) (Fig. 5B) and a 2.1-fold increase in VEGF mRNA expression levels (P < 0.05) (Fig. 5C) vs. vehicle (PBS)-treated AGEC. These results further showed that AICAR-induced importin-α expression in AGEC closely correlated with both increased in vitro angiogenesis (correlation coefficient r = 0.952, P < 0.001) and increased VEGF mRNA expression (correlation coefficient r = 0.968, P < 0.05). AICAR-treated AGEC also had a 2.9-fold increase in nuclear HIF-1α protein levels (P < 0.001) vs. vehicle-treated AGEC (Fig. 5D). The binding of HIF-1α to the VEGF gene promoter determined by EMSA was reduced in AGEC nuclear protein extracts by 1.6-fold (P < 0.001) vs. YGEC, and the treatment of AGEC with AICAR significantly increased HIF-1α binding to VEGF gene promoter by 1.3-fold (P < 0.001) (Fig. 5E). Taken together, these data indicate that dysregulated nuclear transport of HIF-1α, due to reduced levels of importin-α in AGEC, causes reduced VEGF gene expression.
Importin-α is critical for angiogenesis and VEGF gene expression in GMEC.
To determine whether and to what extent importin-α is critical for in vitro angiogenesis and VEGF gene expression in YGEC and AGEC, we knocked down importin-α protein expression using specific siRNA and examined in vitro angiogenesis and VEGF gene expression in these cells cultured under hypoxia. The transfection efficiency determined using incorporation of fluorescence-labeled Alexa Fluor 488-labeled siRNA was 85–90% for both YGEC and AGEC. Downregulation of importin-α1 and importin-α3 in YGEC significantly reduced in vitro angiogenesis by 93% and 73% (both P < 0.001), respectively (Fig. 6A), and reduced VEGF mRNA expression levels by 52% and 48% (both P < 0.05), respectively, resulting in levels comparable to the AGEC controls (Fig. 6B). Downregulation of importin-α1 and importin-α3 in AGEC further reduced the already impaired in vitro angiogenesis by 68% (P < 0.001) and 47% (P < 0.01), respectively (Fig. 6A). In addition, treatment of AGEC with importin-α1- and importin-α3-specific siRNA further reduced VEGF mRNA expression by 19% and 27% (both P < 0.05; Fig. 6B), respectively. These findings demonstrate that importin-α is critical for angiogenesis and VEGF gene expression in gastric endothelial cells.
Angiogenesis in aging gastric mucosa has not been examined before except our recent in vivo study in rats, which demonstrated the extent of its inhibition and the close relationship between reduced VEGF expression, reduced angiogenesis, and impaired gastric mucosal healing (2). However, that study was not able to either dissect specific cellular targets and mechanisms or provide information regarding specific abnormalities or their causes in microvascular endothelial cells of aging gastric mucosa, which are the key targets and effectors of gastric angiogenesis. In the present study, we used an in vitro approach to examine the mechanisms of impaired angiogenesis and reduced VEGF expression in aging gastric mucosa at the cellular and molecular level. We directly manipulated expression levels of importin-α in GMEC and examined the consequences of increased and reduced importin-α expression on angiogenesis and VEGF expression in these cells. We recognize that the isolation and subsequent culture of YGEC and AGEC cells can inadvertently select a subpopulation of cells that may not fully represent the overall microvascular endothelial cell population within the gastric mucosa in vivo. However, we used identical approaches for the isolation and expansion of YGEC and AGEC. Additionally, for culturing YGEC and AGEC, we used a rich growth medium containing, in addition to serum, several endothelial growth supplements including VEGF. Thus it is unlikely that there existed a selective pressure favoring a “more fit” subpopulation over less fit/more fastidious cells.
Our present study demonstrates for the first time that importin-α is critical for angiogenesis and VEGF gene expression in gastric mucosal endothelial cells and that an aging-related reduction in importin-α expression results in impaired angiogenesis and reduced VEGF. Our present study demonstrates that, in contrast to YGEC, AGEC do not increase either in vitro angiogenesis or VEGF expression in response to hypoxic conditions, likely attributable to a defect in the transport of HIF-1α (the hypoxia sensor) into the cell nuclei. Furthermore, our present study demonstrated that, in contrast to YGEC, AGEC do not form lumina during in vitro angiogenesis. Thus, not only is angiogenesis reduced quantitatively, but also the quality of tube and vessel formation is also severely impaired. These in vitro findings provide direct relevance to in vivo angiogenesis in gastric mucosa, where endothelial lumen formation is an important feature of angiogenesis, as shown in our recent paper (2). Another study also demonstrated that endothelial cell tubes do in fact form lumina during in vivo angiogenesis, confirming that in vitro angiogenesis has a relevance to the capillary blood vessel formation in vivo (22). However, the lack of lumen formation in AGEC during in vitro angiogenesis is an entirely novel finding not reported previously.
Our study uncovered a novel, previously unrecognized role for importin-α in angiogenesis by demonstrating for the first time that siRNA silencing of importin-α significantly abolishes in vitro gastric angiogenesis and significantly reduces VEGF expression. In this study, we showed that pharmacologically induced upregulation of importin-α expression in AGEC significantly restored angiogenesis and VEGF gene expression. Thus our findings with a focus on gastric endothelial cells provide an entirely new insight into the mechanism regulating gastric angiogenesis in general and also the key mechanism underlying the impairment of angiogenesis in aging gastric mucosa. We acknowledge however, that the apparent rescue of angiogenesis by pharmacological modulation of AMPK activity could reflect, not only importin-α overexpression, but also other effects not related to importin-α.
The mechanism of HIF-1α transport into the nuclei of gastric endothelial cells has not been examined before. In some cells, e.g., HeLa cells, nuclear transport of some proteins is mediated by importin-α (7, 23, 39). Our previous study in aging myocardial microvascular endothelial cells demonstrated reductions in importin-α protein levels, VEGF protein levels, and in vitro angiogenesis vs. young myocardial microvascular endothelial cells. That study, however, did not examine the causal mechanistic relationship between the downregulation of importin-α, reduced VEGF expression, and impaired angiogenesis. In addition, that study did not examine the effect of downregulating or upregulating importin-α protein levels on angiogenesis or VEGF expression in young and aging microvascular endothelial cells. Our present study in gastric endothelial cells clearly establishes that reduced importin-α expression in AGEC is a key mechanism underlying the aging-related reduction in VEGF expression and impaired angiogenesis. This conclusion is further supported by our demonstration that VEGF expression and in vitro angiogenesis in gastric endothelial cells can be modulated by regulating importin-α levels. We showed that upregulation of importin-α in AGEC using AICAR increases VEGF gene activation, VEGF gene expression, and in vitro angiogenesis. This is the first demonstration in AGEC that AICAR, not only activates, but also induces the expression of importin-α and stimulates angiogenesis. Importantly, we also showed that the downregulation of importin-α in YGEC using specific importin-α siRNA significantly decreases VEGF expression and dramatically reduces in vitro angiogenesis in these cells. Thus our study indicates that importin-α is a critical requirement for in vitro angiogenesis and that decreased importin-α levels in AGEC impair the nuclear transport of HIF-1α and result in reduced VEGF expression and impaired angiogenesis.
Our findings that importin-α silencing almost completely abolished in vitro angiogenesis while only partly blocking VEGF mRNA expression and that exogenous VEGF only partly restored in vitro angiogenesis indicate that importin-α regulates angiogenesis by both VEGF-dependent and VEGF-independent mechanisms. The VEGF-independent mechanism by which importin-α regulates angiogenesis could involve notch1 signaling. This contention is supported by the demonstration that angiogenesis requires notch1 signaling (28, 34) and that importin-α mediates nuclear transport of notch1 (17). Therefore, the incomplete reversal of inhibited angiogenesis by exogenous VEGF could have been due to attenuation of notch1 signaling following importin-α silencing in GMEC. Further studies are necessary to investigate the role of importin-α in nuclear transport of notch1 and other possible angiogenic effectors.
It is estimated that, by the year 2020, about one in six people in the United States will be over 65 yr of age (13). In aging individuals, the widespread chronic use of NSAIDs, which inhibit angiogenesis and gastric injury healing (15, 20), and the significantly increased complications, such as gastric mucosal injury caused by NSAIDs (13, 31, 32), make aging gastropathy an important clinical issue. Our study significantly advances and defines a key target and mechanism underlying the impairment of angiogenesis in aging gastropathy, which affects the growing aging population increasingly using aspirin, other NSAIDs, and ulcerogenic drugs. In addition to implications for basic science, these findings have potential therapeutic implications for stimulating and/or inhibiting gastric angiogenesis. Compounds that increase importin-α levels, such as AICAR, or importin-α gene therapy can potentially be used to increase VEGF expression and improve angiogenesis and injury healing in gastrointestinal tissues of aging individuals. Conversely, inhibitors of importin-α can be potentially used for the inhibition of pathological angiogenesis induced by tumors of the gastrointestinal tract.
This work was supported by the VA Merit Review (to A. Tarnawski) and the American Heart Association Beginning Grant in Aid (to A. Ahluwalia).
No conflicts of interest, financial or otherwise, are declared by the authors.
Author contributions: A.A. and A.S.T. conception and design of research; A.A. and A.S.T. performed experiments; A.A., M.K.J., and A.S.T. analyzed data; A.A. and A.S.T. interpreted results of experiments; A.A. and A.S.T. prepared figures; A.A. and A.S.T. drafted manuscript; A.A., M.K.J., and A.S.T. edited and revised manuscript; A.A., M.K.J., and A.S.T. approved final version of manuscript.
- Copyright © 2014 the American Physiological Society