|Home | About | Journals | Submit | Contact Us | Français|
Induction of heme oxygenase-1 (HO-1) expression can be achieved by stimulation with cobalt protoporphyrin (CoPPIX) or cobalt chloride (CoCl2). HO-1 has been recently implicated in regulation of angiogenesis and CoCl2 is known to potently activate hypoxia inducible factor-1 (HIF-1) transcription factor, a key regulator of angiogenic response in hypoxia. Here we determined the effect of CoPPIX and CoCl2 on the expression of vascular endothelial growth factor (VEGF) and interleukin-8 (IL-8), the two major angiogenic mediators, in human microvascular endothelial cells (HMEC-1). CoPPIX induced HO-1 expression and strongly enhanced VEGF and IL-8 synthesis, through the activation of VEGF and IL-8 promoters. Inhibition of HO activity by SnPPIX decreased VEGF production, while, interestingly, it did not affect IL-8. CoCl2 activated hypoxia-responsive element (HRE) and consequently VEGF generation via the enhancement of production of reactive oxygen species (ROS). On the other hand, CoCl2 did not influence IL-8 expression, while CoPPIX did not induce ROS elevation neither it affected HRE activity in VEGF promoter. Our data show that although both CoCl2 and CoPPIX induce HO-1, the influence of CoCl2 on VEGF does not involve HO-1 and is HIF-1-dependent, while the effect of CoPPIX does not involve HIF-1 but relies on HO-1.
Vascular endothelial growth factor (VEGF) is the major angiogenic mediator generated by different cell types, including microvascular endothelial cells, smooth muscle cells and fibroblasts (10). It exerts a wide range of actions being a specific mitogen for endothelial cells, increasing their migration and inhibiting apoptosis (35). IL-8, a member of chemokine family, is not only a pro-inflammatory agent but also plays an important role in angiogenesis (23). Both these factors are crucial in various angiogenic-dependent disorders, including tumors and cardiovascular diseases (3).
Heme oxygenases (HO) are the rate limiting enzymes that catalyze the conversion of heme into biliverdin, carbon monoxide (CO) and free iron (24). Two isoforms are known. HO-2 is constitutively expressed in many cells, including endothelial cells and neurons, while HO-1 is the inducible protein. Up-regulation of HO-1 may be an adaptive mechanism protecting cells from stress, in particular during hypoxia or inflammation (25).
Several reports have demonstrated that increased HO-1 expression and activity can upregulate synthesis of VEGF both in vitro (8, 15, 16,26) and in vivo (5,21 ,29). It has been shown that enhancement of VEGF production in response to hemin in rat vascular smooth muscle cells (VSMC) and endothelial cells was dependent on HO-1 activity, as it was completely reversed by HO inhibitor, SnPPIX (8, 15). Also 15-deoxy12, 14 Δ-prostaglandin-J2 (15d-PGJ2) augmented the expression of VEGF in an HO-1 dependent manner in human microvascular endothelial cell line (HMEC-1) (16). Interestingly, in the same cells the production of IL-8 was not affected by hemin or SnPPIX (16), indicating that its synthesis does not involve HO-1 pathway.
However, the stimulation of cells with exogenous heme involves not only the HO-1 induction, but may represent the complex activity of heme as a pro-oxidant and iron-releasing compound. Therefore, in the present study we aimed to validate the role of HO-1 activation in regulation of VEGF and IL-8 synthesis, comparing the effects of two other potent HO-1 inducers, CoPPIX and CoCl2.
Luciferase Assay Reagents, oligo(dT) primers, dNTPs and MMLV reverse transcriptase were purchased from Promega (Madison, WI, USA). The ELISA kit for human VEGF was procured from R&D Systems Europe (Abingdon, UK). BCA-1 protein determination kit, epidermal growth factor (EGF), hydrocortisone and the cell culture medium (MCDB 131) were obtained from Sigma-Aldrich (Poznan, Poland). Tin protoporphyrin-IX (SnPPIX) and cobalt protoporhyrin (CoPPIX) were purchased from Porphyrin Products. The SuperFect transfection reagent was bought from Qiagen. Fetal calf serum was procured from ICN (Irvine, CA, USA). The ELISA-based Trans-Am HIF-1 kit was procured from Active Motif (Belgium).
Human microvascular endothelial cells (HMEC-1) were obtained from Centers for Disease Control and Prevention (Atlanta, GA, USA) and cultured in a MCDB 131 medium containing 10% FCS, L-glutamine (2 mM), EGF (10 ng/ml) and hydrocortisone (1 μg/ml).
Cells were seeded in 24-well or 6-well plates and grown to full confluence. Next, medium was changed and stimulants were added for 6 or 24 hr. After this time media were collected for determining VEGF and IL-8 protein concentration, while cells were washed twice with cold PBS and then subjected to RNA or protein isolations for RT-PCR analysis or Western blotting, respectively. CoPPIX, SnPPIX and CoCl2 were prepared freshly and added to cells for indicated time. SnPPIX and N-acetylcysteine were applied 1 hr before stimulation. Solvent (0.1 M NaOH for CoPPIX and SnPPIX) at appropriate final dilution was included in control.
RT-PCR was performed on 2 μg of total RNA which was isolated from the cells by acid guanidinium thiocyanate-phenol-chloroform extraction. Reverse transcription was carried out with oligo-dT primers for 1 hr at 42°C using MMLV reverse transcriptase, according to vendor's instruction. PCR amplification with Taq polymerase was performed for 28 cycles (EF2), 30 cycles (HO-1, IL-8) and 35 cycles (VEGF) using the following protocol: 95°C - 40 sec, 58°C - 40 sec and 72°C - 50 sec. The primers specific for VEGF (5'-CAC CGC CTC GGC TTG TCA CAT-3' and 5'-CTG CTG TCT TGG GTG CAT TGG-3'), for HO-1 (5'-GTG GAG ACG CTT TAC GTA GTG C-3' and 5'-CTT TCA GAA GGG TCA GGT GTC C-3'), for IL-8 (5'-CTC TCT TGG CAG CCT TCC TGA-3' and 5'-CCC TCT GCA CCC AGT TTT CCT T-3'), and for housekeeping gene EF2 (5'-GCG GTC AGC ACA ATG GCA TA and 5'-GAC ATC ACC AAG GGT GTG CAG) were used. PCR products were analyzed by electrophoresis in 2% agarose gel. The product length for the VEGF120 was 431 bp, for VEGF164 - 563 bp, for HO-1 - 250 bp, for IL-8 - 240 bp and for EF2 - 218 bp.
Cells were transfected with constructs containing a full-length human VEGF promoter (−2279 to +54) or HRE fragment of this promoter (−1014 to −903) cloned into the luciferase reporter plasmid pGL2 and inserted upstream of the thymidine kinase promoter of pT81luc0 plasmid. Both constructs were kindly provided by Dr. Hideo Kimura (Chiba, Japan). The full-length promoter of the IL-8 gene was cloned upstream of the firefly luciferase coding region in the pGL2-basic vector and was kindly supplied by Dr. Rainer de Martin (Vienna, Austria) (34). HMEC-1 cells growing to 70% confluence in 24-well plates were transfected with 0.25 μg of DNA and 1,25 μl of SuperFect Reagent per well, according to vendor's protocol. After 24 hr cells were treated with indicated stimulants for next 24 hr. The activity of reporter genes was determined in cell lysates at 48 hr after transfection and normalized to the total protein content.
The total cellular protein from HMEC-1 cells was isolated using ice-cold lysis buffer (1 × PBS, 10 mM phenylmethylsulfonyl fluoride (PMSF), 10 mM leupeptine, 10 mM aprotinin and 1% Triton X-100). Samples were centrifuged (20 min, 8000 g, 4°C) and clear supematants were collected. After measurement of protein content with bicinchoninic acid protein assay kit (BCA, Sigma) 10 μg of protein was loaded on 12% SDS-PAGE gel followed by transfer to nitrocellulose membrane Hybond™ECL™ (Amersham Pharmacia Biotech, Buckinghamshire, UK). Membranes were probed with polyclonal antibodies against heme oxygenase-1 (Stressgen Biotech, Canada) followed by biotin-conjugated secondary antibodies at the dilution 1:2500 in TBS with 3% albumin. Alkaline Phosphatase-Conjugated Streptavidin (Dako) at the dilution 1:5000 in TBS was used and the visualisation was performed using the BCIP/NBT Blue liquid substrate for membranes.
HMEC-1 cells were stimulated for 24 hr and DCFH—DA (10 μM) was added for the last hour of incubation. The fluorescence (excitation 485 nm, emission 535 nm) was measured directly from the plate. Data were normalized to the control values.
Binding of HIF-1 (hypoxia inducible factor-1) to the hypoxia responsive element (HRE) was measured with the ELISA-based Trans-Am HIF-1 kit using nuclear extracts lysates prepared from HMEC-1 cells. The Trans-Am kit employs 96-well microtier plates coated with an oligonucleotide containing the HRE consensus sequence. Transcription factor binding to HRE sequence is detected by anti-HIF specific antibody followed by colorimetric measurement. Preparation of nuclear extracts was done exactly as recommended by the manufacturer. Specificity was checked by measuring the ability of soluble wild-type or mutated oligonucleotides to inhibit binding.
Luciferase assay was done according to the manufacturer's protocol as described previously (14).
Cell culture media were collected and concentration of VEGF and IL-8 protein was quantified following the manufacturer's protocol.
All experiments were performed in duplicates or triplicates and if not reported otherwise they were repeated at least three times. Data are presented as mean ± SD. Statistical evaluation was done with Student's t test. Differences were accepted as statistically significant at p<0.05.
CoPPIX and CoCl2 can induce HO-1 in many cell types (2,3,30). Our experiments confirm that these compounds are strong activators of HO-1 expression also in microvascular endothelium. We observed a potent upregulation of HO-1 in CoPPIX- and CoCl2-treated HMEC-1 both at the mRNA and protein levels, as demonstrated by RT-PCR (Fig. 1A) and Western-blotting (Fig. 1B), respectively.
Additionally, using DCFH-DA oxidation assay, we found that exposure of HMEC-1 to CoCl2 leads to a strong elevation of reactive oxygen species (ROS). Fluorescence of DCFH-DA oxidation product was inhibited by preincubation of cells with N-acetylcysteine (NAC), confirming the specificity of assay. Of note, CoPPIX did not affect ROS generation in HMEC-1 (Fig. 1C). At the doses used, CoPPIX and CoCl2 did not influence HMEC-1 viability (data not shown). Thus, apart from induction of HO-1 these two cobalt compounds exert distinct effects on the cells.
Basal production of VEGF in HMEC-1 is usually low and ranges around 10-30 pg/105 cells. Along with induction of HO-1, CoPPIX and CoCl2 upregulated by approximately three-fold the release of VEGF into the culture media, as measured by ELISA (Fig. 2A). Augmented synthesis of VEGF protein was paralleled by similar extent of activation of VEGF transcription, as determined by luciferase assay in cells transfected with a reporter plasmid containing a human full-length VEGF promoter (Fig. 2B). It may suggest that regulation of VEGF synthesis by both compounds is exerted at the transcription level.
Despite the similar stimulatory effects of CoPPIX and CoCl2 on VEGF transcription, the underlying pathways appear to be different. Using the cells transfected with a reporter construct regulated by the hypoxia responsive element (HRE) from the VEGF promoter, we demonstrated that CoCl2 strongly activated HRE. In contrast, CoPPIX did not exert any influence (Fig. 2C). Thus, we can suppose that increased transcription of VEGF in response to CoCl2 results mostly from activation of hypoxia inducible factor-1 (HIF-1) and direct stimulation of VEGF promoter through binding of HIF-1 to HRE.
Lack of effect of the second HO-1 activator, CoPPIX, on HRE suggested that HO-1 is not involved in regulation of HIF-1 binding activity. To check this supposition we measured activation of HIF-1 in cells treated with hemin, the most common HO-1 inducer. Analysis done using a TransAm assay confirmed that modulation of HO-1 activity does not affect the HIF-1 binding to the target DNA sequences (data not shown, 101.15 ± 4.03% of control, n=2 independent experiments).
Finally, we evaluated the involvement of HO-1 pathway in CoPPIX- and CoCl2-induced upregulation of VEGF expression. To this aim we pretreated the cells with SnPPIX, a competitive inhibitor of HO-1, and then stimulated them with CoPPIX or CoCl2. As shown at Fig. 3, inhibition of HO-1 significantly reversed the increase in VEGF synthesis induced by CoPPIX. It was demonstrated both at the mRNA (Fig. 3A) and protein (Fig. 3B) level. In contrast, the upregulation of VEGF in response to CoCl2 was not significantly affected by SnPPIX.
Because treatment of HMEC-1 with CoCl2 elevated the generation of ROS, the effect reversed by NAC (Fig. 1C), we checked whether NAC can modulate the CoCl2-induced production of VEGF. Actually, we found that such a treatment significantly reduced the synthesis of VEGF protein (Fig. 4A) and decreased the activity of HRE part of VEGF promoter (Fig. 4B). Thus, we postulate, that although both CoPPIX and CoCl2 strongly induce HO-1 expression, only in the case of the former compound this induction is an important mediator of VEGF augmentation. In upregulation of VEGF by CoCl2, the concomitant induction of HO-1 does not play a significant role, as much more important appears the activation of HIF-1, a direct inducer of VEGF transcription. Stimulation of HIF-1 in response to CoCl2 may result from the generation of ROS (Fig. 1C).
IL-8 is a pro-angiogenic cytokine able to enhance the proliferation of endothelial cell and promote their survival (23). Basal production of IL-8 by HMEC-1 is relatively high, within the range of ~1000 pg/105 cells. We found that in response to CoPPIX, expression of IL-8 mRNA (Fig. 5A) and synthesis of IL-8 protein (Fig. 5B) were strongly augmented. Similarly as in the case of VEGF, the increased generation of IL-8 was associated with a higher transcription rate from the IL-8 promoter (Fig. 5C).
However, pretreatment of cells with SnPPIX did not modulate the response to CoPPIX (Fig. (Fig.5A,5A, ,5B).5B). Therefore, differently from VEGF expression, the upregulation of IL-8 in HMEC-1 exposed to CoPPIX does not involve HO-1 pathway. This observation was further confirmed by treatment of the cells with CoCl2. In spite of the strong HO-1 induction, CoCl2 did not influence the generation of IL-8 protein (Fig. 5B).
The salient finding of the present study is the demonstration that synthesis of VEGF in HMEC-1 can be enhanced by induction of HO-1, while the synthesis of IL-8 is not reliant on this pathway. However, induction of HO-1 by various chemical stimuli does not necessarily lead to increased VEGF synthesis. Thus, although both CoCl2 and CoPPIX induce HO-1, the effect of CoCl2 on VEGF does not involve HO-1 and is dependent on HIF-1, while the effect of CoPPIX does not involve HIF-1 but relies on HO-1.
HO-1 is an inducible enzyme converting heme to CO, free iron and biliverdin which is reduced to bilirubin by biliverdin reductase (24). The products of this enzymatic reaction have important biological effects, including antioxidant, anti-inflammatory and cytoprotective functions (25). Induction of HO-1 is an adaptive response to several injurious stimuli including heme, heavy metals, nitric oxide (NO) or cytokines. Such induction has been implicated in numerous diseases including transplant rejection, hypertension, atherosclerosis, lung injury, endotoxic shock and others (25).
CoPPIX is used as an inducer of HO-1 activity and its action is confirmed by many studies. It was used to up-regulate HO-1 in the stateotic rat liver model of ex vivo cold ischemia/reperfusion injury, where it was able to improve portal venous blood flow, increase bile production and decrease hepatocyte injury (2). Moreover, it can prevent graft-versus-host disease after experimental allogenic bone marrow transplantation (13) and improve survival after rat renal allografts (32).
CoCl2 has been also reported to up-regulate HO-1 (6,17,30). Additionally, the important mechanism of its action is the stimulation of HIF-1 transcription factor (33). HIF-1 is a heterodimeric protein composed of the HIF-1α and HIF-1β subunits. HIF-1β subunit is constitutively expressed whereas the α subunit is unstable at normal oxygen concentration, while low oxygen tension stabilizes it (for a review see 36). CoCl2 which is often used as a hypoxia mimic can potently induce HIF-1 by stabilization and accumulation of HIF-1α protein (18). After this stabilization active HIF-1 heterodimer is formed and binds to HRE present in the regulatory regions of many genes including VEGF (19).
In our studies pharmacological activators of HO-1, CoPPIX and CoCl2, induced VEGF synthesis. VEGF is a potent angiogenic mediator that plays an important role in both physiological as well as pathological vasculogenesis and angiogenesis (10). It stimulates endothelial cell proliferation and migration in vitro and exerts anti-apoptotic activities by inducing Akt phosphorylation (1). In addition, VEGF acts as a proinflammatory cytokine due to its chemoattractant properties, by increasing endothelial cell pemeability or augmenting the expression of endothelial cell adhesion molecules (20). Taken together, VEGF is likely a key intermediary between cell-mediated immune inflammation and the associated angiogenesis reaction.
In our previous papers we have demonstrated that HO-1 expression and activity can enhance synthesis of VEGF in vitro (8,15,16). We have demonstrated earlier that SnPPIX, an inhibitor of HO activity, completely blocked cytokine-induced VEGF expression, whereas stimulation of HO-1 with hemin increased its expression. Similar effect was obtained by overexpressing ho-1 gene in VSMC (8) and HMEC-1 (16). Also in vivo studies clearly showed that VEGF synthesis can be dependent on HO-1 activity (21,29). Here we have obtained analogous results using CoPPIX as inducer of both HO-1 and VEGF. Accordingly, inhibition of HO-1 activity by SnPPIX reversed CoPPIX-induced VEGF production. On the other hand, CoCl2 which also potently induces HO-1 upregulates production of VEGF in an HO-1 independent manner.
Our results clearly demonstrated that CoCl2 induces VEGF synthesis by increased VEGF transcription through HRE activation and this mechanism is completely different from the action of CoPPIX. We showed that although full VEGF promoter was activated by CoPPIX, HRE element was not affected by this compound. In turn, lack of involvement of HO-1 in CoCl2-induced VEGF expression may result from the very fast stabilization of HIF-1 and immediate induction of VEGF, which may preclude the significant influence of HO-1 pathway, requiring more time for activation.
Recent studies suggest that changes in the level of ROS provide a redox signal for HIF-1 induction by hypoxia (for a review see 36). Also under normoxic conditions ROS can mediate HIF-1α protein accumulation and HIF-1 -dependent transcription by hormones, growth factors, and transition metals (9, 14, 18,28). Moreover, ROS mediated induction of HIF-1 has been reported to regulate VEGF expression in different cell types like smooth muscle cells (4,28), human ovarian cancer cells (9) or fibroblasts (14). Accordingly, data presented here indicate that CoCl2-induced HIF-1α activation is dependent on ROS generation.
In this study we have also examined the effect of CoPPIX and CoCl2 on IL-8 synthesis in HMEC-1. IL-8, the CXC chemokine is not only a potent chemoattractant for neutrophils and T lymphocytes and regulator of leukocyte survival but is also involved in the regulation of angiogenesis (23). It is produced by many cell types, including endothelial cells, after stimulation with the inflammatory cytokines, such as tumor necrosis factor-α and interleukin-1β (12). Same papers show that this production can be also connected with the activities of VEGF, which can induce IL-8 synthesis (22,27).
Our results show that two agents containing cobalt ion, namely CoPPIX and CoCl2, regulate IL-8 expression in a completely different way. CoPPIX induces IL-8 mRNA, promoter activity and protein synthesis, whereas CoCl2 does not influence IL-8 generation. Importantly, the action of CoPPIX on IL-8 synthesis seems to be HO-1-independent. This mechanism is different from VEGF regulation, which is reliant on HO-1 pathway. Our data are partially opposite to those obtained by Pae and coworkers (26). They have proposed that HO-1 induced VEGF acts as an IL-8 regulator. In their experiments induction in IL-8 was dependent on HO-1 as they have observed a decrease in IL-8 protein synthesis after transfection of siRNA against HO-1 (26). Those differences might be caused by cell specific effect as HUVEC were used in that study (26) while HMEC-1 in ours.
It is well established that NO can induce both HO-1 (11,31) and VEGF (for a review see: 7) expression in various cell types, including endothelial cells, probably via induction of oxidative stress but in our experiments we did not observe upregulation of nitric oxide synthase expression after CoPPIX treatment (data not shown). Moreover, we have checked if CoPPIX can cause oxidative stress and in that way induce IL-8 expression. However, we did not observe increase in free radical formation.
Taken together, CoPPIX induces both VEGF and IL-8 productions but the mechanisms of these inductions seem to be completely different; VEGF expression is augmented in HO-1 dependent manner, whereas IL-8 production is HO-1 independent. In contrast, CoCl2 induces VEGF through HIF-1 activation, regardless the HO-1 expression. Moreover, it does not up-regulate IL-8 synthesis. Our data strongly suggest that although both CoPPIX and CoCl2 induce HO-1 they have also other specific activities. Further studies are underway to elucidate the mechanism(s) mediating the induction of VEGF and IL-8 by HO-1 activators.
This work was partially supported by grants from Ministry for Science and Informatics Technology (PBZ-KBN-107/P04/2004, 2 PO4B 016 26). We thank Mr. Slawomir Golda for help in measuring ROS generation. Dr Jacek Miedzobrodzki is acknowledged for access to the EG Berthold chemoluminometer. A. Jozkowicz is the recipient of the Wellcome Trust International Senior Research Fellowship.