- Email: [email protected]
Deletion of prolyl hydroxylase domain proteins (PHD1, PHD3) stabilizes hypoxia inducible factor-1 alpha, promotes neovascularization, and improves perfusion in a murine model of hind-limb ischemia
Microvascular Research 97 (2015) 181–188
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Deletion of prolyl hydroxylase domain proteins (PHD1, PHD3) stabilizes hypoxia inducible factor-1 alpha, promotes neovascularization, and improves perfusion in a murine model of hind-limb ischemia Muhammad T. Rishi a,b,1, Vaithinathan Selvaraju a,1, Mahesh Thirunavukkarasu a, Inam A. Shaikh a,b, Kotaro Takeda c, Guo-Hua Fong c, J. Alexander Palesty b, Juan A. Sanchez a, Nilanjana Maulik a,⁎ a b c
Molecular Cardiology and Angiogenesis Laboratory, Department of Surgery, University of Connecticut Health Center, Farmington, CT, USA Stanley J. Dudrick Department of Surgery, Saint Mary's Hospital, Waterbury, CT, USA Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA
a r t i c l e
i n f o
Article history: Accepted 24 October 2014 Available online 3 November 2014 Keywords: Hindlimb ischemia Prolyl hydroxylase HIF VEGF Bcl-2
a b s t r a c t Background: There is an emerging focus on investigating innovative therapeutic molecules that can potentially augment neovascularization in order to treat peripheral arterial disease (PAD). Although prolyl hydroxylase domain proteins 1 and 3 (PHD1 and PHD3) may modulate angiogenesis via regulation of hypoxia inducible factor-1α (HIF-1α), there has been no study directly addressing their roles in ischemia-induced vascular growth. We hypothesize that PHD1−/− or PHD3−/− deficiency might promote angiogenesis in the murine hind-limb ischemia (HLI) model. Study design: Wild type (WT), PHD1−/− and PHD3−/− male mice aged 8–12 weeks underwent right femoral artery ligation. Post-procedurally, motor function assessment and laser Doppler imaging were periodically performed. The mice were euthanized after 28 days and muscles were harvested. Immunohistochemical analysis was performed to determine the extent of angiogenesis by measuring capillary and arteriolar density. VEGF expression was quantified by enzyme-linked immunosorbent assay (ELISA). Bcl-2 and HIF-1α were analyzed by immunofluorescence. Fibrosis was measured by picrosirius red staining. Results: PHD1−/− and PHD3−/− mice showed significantly improved recovery of perfusion and motor function score when compared to WT after femoral artery ligation. These mice also exhibited increased capillary and arteriolar density, capillary/myocyte ratio along with decreased fibrosis compared to WT. VEGF, Bcl-2 and HIF-1α expression increased in PHD1−/− and PHD3−/− mice compared to WT. Conclusions: Taken together these results suggest that PHD1 and PHD3 deletions promote angiogenesis in ischemiainjured tissue, and may present a promising therapeutic strategy in treating PAD. © 2014 Elsevier Inc. All rights reserved.
Introduction Peripheral arterial disease (PAD) is a progressive and debilitating illness that reduces the quality of life in more than 25 million patients in Europe and North America alone (Belch et al., 2003; Norgren et al., 2007; Raval and Losordo, 2013). It is strongly associated with other atherosclerotic pathologies including coronary artery and cerebrovascular disease (Golomb et al., 2006). While increased understanding of PAD has resulted in the development of multiple treatment modalities, amputation rates from PAD still remain high. A recent study looking at the Medicare database from 2000 through 2008 found that among 3
⁎ Corresponding author at: Molecular Cardiology and Angiogenesis Laboratory, Department of Surgery, University of Connecticut Health Center, 263 Farmington Ave., Farmington, CT 06032-1110, USA. E-mail address: [email protected] (N. Maulik). 1 Contributed equally.
http://dx.doi.org/10.1016/j.mvr.2014.10.009 0026-2862/© 2014 Elsevier Inc. All rights reserved.
million Medicare admissions for PAD, a total of 186,338 patients underwent lower extremity amputations, which were 6.8% of all the hospitalized PAD population (Jones et al., 2012). Approximately 50% of the patients with critical limb ischemia (CLI) are unsuitable for any form of revascularization, and those who receive interventional treatment still carry a poor long-term prognosis (Adam et al., 2005; Ouriel, 2001). Additionally, significant morbidity occurs as a result of chronic wound ulceration particularly in diabetics in which an impaired microcirculation results in local tissue ischemia. The management of PAD follows three main approaches: risk factor attenuation, pharmacological intervention, and surgical revascularization. Revascularization can be achieved either by open or endovascular surgery, but neither approach significantly alleviates disease morbidity and mortality (Ouriel, 2001). In 2005, a multicenter, randomized controlled study highlighted the shortfalls of PAD treatment by comparing the effectiveness of bypass versus angioplasty in severe ischemia of the leg (BASIL Trial). In this study, half of the patients
182
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
with CLI were considered unsuitable for any form of revascularization due to advanced disease progression, and those who received treatment presented with very poor prognosis (Adam et al., 2005). It is increasingly apparent that PAD may require other treatment modalities that upregulate the ability of the patient to generate new vasculature. Currently available strategies to promote angiogenesis to treat the consequences of PAD have not resulted in clinically applicable forms of therapy. However, targeted molecular approaches to enhance angiogenesis have considerable theoretical advantages to other forms of treatment. Hypoxia inducible factor-1α (HIF-1α) appears to be one of the key regulators of the downstream effectors of angiogenesis by inhibition of PHDs (Adluri et al., 2011a). Therefore, upregulation of HIF-1α via inhibition of prolyl hydroxylase domain enzymes (PHDs) is a potentially effective option for patients afflicted with PAD. There are three PHD variants in mammalian cells: PHD1 (EglN2), PHD2 (EglN1), and PHD3 (EglN3). There are organ specific differences in the expression of these three isoenzymes. PHDs act as cellular oxygen sensors and play a critical role in triggering angiogenesis. During normoxic conditions, HIF-1α is hydroxylated at two proline residues (P402 and P546) by one of the three identified PHD isoenzymes (Bruick and McKnight, 2001; Epstein et al., 2001). Hydroxylation initiates binding of von Hippel–Lindau tumor suppressor protein and proteasomal degradation of HIF-1α by the E3 ubiquitin ligase complex (Berra et al., 2006). Under hypoxia, PHD enzyme activity is inhibited. The inhibition of PHD enzyme activity allows stabilization and translocation of HIF-1α to the nucleus and dimerization with its β-subunit to form an active transcription factor. This complex then binds to hypoxia-response elements (HRE) in the regulatory regions of target genes to upregulate transcription and translation of angiogenic peptides, including VEGF, eNOS, Notch ligands, platelet-derived growth factor (PDGF) and angiopoietin-2 (Ang-2) (Semenza, 2003). Of these peptides, VEGF plays the most vital role in angiogenesis since it is required for the growth and development of vascular endothelial cells as well as endothelial cell migration and proliferation (Ferrara, 2002; Ferrara et al., 2003; Shizukuda et al., 1999). VEGF, additionally, upregulates Bcl-2 to promote endothelial cell survival (Nor et al., 1999). While a number of studies have explored various ways to alter this angiogenic pathway as a potential treatment for ischemic vascular disease, we still do not have a satisfactory clinically proven answer. Thus, there is an urgent need for more novel approaches. PHDs may constitute promising candidates for angiogenesis related therapies due to their role in regulating HIF-dependent expression of angiogenic factors. While PHD2 has been clearly shown to regulate angiogenesis (Wu et al., 2008), its involvement in a number of other processes such as erythropoiesis suggests that targeting PHD2 may be associated with complications. PHD1 and PHD3, on the other hand, do not substantially affect the blood vessels when targeted individually, and hence may provide less risky angiogenesis targets. Here we have examined whether genetically disabling specific PHDs (PHD1 and PHD3) improves limb perfusion by enhancing angiogenesis following HLI. We show that deletion of PHD1 and PHD3 promotes ischemia-induced angiogenesis and potentiates reperfusion in a murine model of HLI, making pharmacologic inhibition of PHDs a possible useful therapeutic strategy in the treatment of PAD. Materials and methods Experimental animals All animal protocols were approved by the Institutional Animal Care and Committee of the University of Connecticut Health Center, Farmington, Connecticut. PHD1−/− and PHD3−/− mice were kindly provided by Dr. GH Fong and were generated as described previously (Takeda et al., 2006).
Mouse model of hind-limb ischemia Adult, 8–12 week-old male C57BL/6, PHD1−/− and PHD3−/− mice were subjected to femoral artery ligation as described previously (Couffinhal et al., 1998; Limbourg et al., 2009). Briefly, mice were anesthetized with a rodent isoflurane vaporizer set at 2.5% with a constant oxygen supply of 1 L/min. The right femoral artery was surgically isolated and ligated with 8–0 prolene suture both proximal and distal to the origin of the deep femoral artery. After ligation, the intervening segment was excised, and all branches between these two sites were ligated or cauterized. The left femoral artery was used as an internal control without ligation.
Laser Doppler perfusion imaging Laser Doppler perfusion imaging was performed using a PeriScan (PIM 3) System (Perimed, Järfälla-Stockholm, Sweden). Baseline images were acquired before the creation of HLI, and post-operative images were taken on days 0, 3, 7, 14, 21 and 28. Three Doppler-perfusion readings were taken at each time point, and an average was calculated to provide final blood-flow values as the ratio of ischemic to non-ischemic hind-limb perfusion.
Motor function assessment In order to assess motor function after induction of HLI, a preestablished scoring system was used (Webber et al., 2011) preoperatively and on post-operative days 0, 3, 7, 14, 21 and 28. Under this scoring system, hind-limb motor function was graded as unrestricted (grade 5), no active use of toe(s) or spreading (grade 4), restricted foot use (grade 3), no use of foot (grade 2), or no use of limb (grade 1).
Morphometric analysis for capillary and arteriolar density Immunohistochemical analysis for both capillary and arteriolar density was performed with paraffin embedded skeletal muscle sections harvested from the ischemic limb, 28 days after surgery. Sections were deparaffinized, hydrated with serial dilution in ethanol and blocked with 2.5% normal horse serum followed by incubation with primary antibody (Caveolin 1, an endothelial cell specific marker antibody— Santa Cruz Biotechnology, Santa Cruz, CA, USA for capillary density (Messaoudi et al., 2009; Ratajczak et al., 2005) and mouse monoclonal anti-α smooth muscle actin antibody—Abcam Inc., Cambridge, MA, USA for arteriolar density) and respective secondary antibody. Images were obtained at 400× magnification on Olympus BH2 microscope and Zeiss LSM 510 Meta confocal microscope for capillary and arteriolar density respectively. Capillaries and arterioles were counted per millimeter square area using Adobe Photoshop CS4 software (Adluri et al., 2011b).
VEGF quantification by ELISA VEGF quantification was performed via ELISA as previously described (Thirunavukkarasu et al., 2013). Briefly, mice from all three groups (WT, PHD1−/− and PHD3−/− mice) were euthanized 3 days after femoral artery ligation and ischemic hind limb gastrocnemius muscle tissue sections were harvested from each treatment group. Subsequently, these muscle samples were homogenized and suspended (50–75 mg/ml) in sample buffer. Protein was isolated and then the total protein concentration was determined using a BCA (bicinchoninic acid) protein assay kit (Pierce, Rockville, IL). Using a mouse VEGF Duoset (R&D Systems Inc., MN), the expression of VEGF (in pg/ml) was determined according to the manufacturer's instructions.
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
183
Bcl-2 and HIF-1α expression by immunofluorescence analysis Ischemic limb muscles were collected 3 days after HLI and tissue sections were prepared as described previously and used for immunofluorescence staining (Thirunavukkarasu et al., 2013; Limbourg et al., 2009). Briefly, the sections were stained for Bcl-2 and HIF1-α with rabbit polyclonal anti-Bcl-2 and rabbit polyclonal anti-HIF-1α (Santa Cruz Biotechnology, CA, USA), respectively, and incubated overnight at 4 °C. The sections were then treated with respective secondary antibodies and nuclear staining was performed with TO-PRO-3 (Invitrogen, Carlsbad, CA, USA). Finally, the images for Bcl-2 and HIF-1α staining were digitally acquired at 400× magnification using Zeiss LSM Confocor 3 microscope at the same exposure and gain settings for each sample comparison. The corrected total fluorescence was measured as described earlier (Cappell et al., 2012; Duca et al., 2014; Harari-Steinberg et al., 2013). Briefly, area and integrated mean density were quantified using ImageJ software from the acquired images. Five to ten pictures from each samples (n = 3–4) and the corrected total fluorescence (CTF) were calculated by the formula integrated density − (area of selected picture × mean fluorescence of background reading).
Fig. 2. In-vivo comparison of perfusion ratio in WT, PHD1−/− and PHD3−/− mice after right femoral artery ligation. Preoperative images using laser Doppler were taken before induction of ischemia as well as post-operative days 0, 3, 7, 14, 21 and 28. Digital colorcoded images were obtained and then analyzed through PIMsoft (version 1.5.4.8078) to quantify the perfusion in the area of interest. PHD1−/− ( ) and PHD3 −/− ( ) mice show significantly increased perfusion ratio as compared to WT ( ) mice on post-operative days 3, 7, 14, 21 and 28 (*p b 0.05; WT vs. PHD1−/− and #p b 0.05; WT vs. PHD3−/−; n = 9–19).
Results Picro sirius red/collagen staining PHD 1 and 3 deletion enhances blood flow recovery in mice after HLI Picrosirius staining was performed to measure the extent of fibrosis in 5 μm sections of gastrocnemius muscle of ischemic limbs as described earlier (Oriowo et al., 2014). Statistical analysis Statistical analysis was conducted using GraphPad Prism Software to perform one-way ANOVA followed by Newman–Kuels multiple comparison test. The results were considered significant at p b 0.05. Values were expressed as Mean ± S.E.M.
To test if PHD1 or PHD3 deficiency improves blood flow, we performed femoral artery ligation and monitored hind limb perfusion recovery with laser Doppler imaging [Fig. 1]. Baseline hind limb perfusion (Pre-op) in PHD1−/− and PHD3−/− mice showed no significant changes compared to WT mice. Immediately after the surgery (day 0), both PHD1−/− and PHD3−/− mice suffered from equivalent ischemia compared to WT mice. However, PHD1−/− and PHD3−/− mice showed significant improvement in perfusion ratio (laser Doppler imaging) when compared with WT mice on post-operative day 3 until day 28
Fig. 1. Comparison of laser Doppler perfusion images of WT, PHD1−/− and PHD3−/− mice after right femoral artery ligation with representative LDIs from different groups. Laser Doppler was used to take baseline preoperative and post-operative images on days 0, 3, 7, 14, 21 and 28. Red indicates maximum perfusion while black indicates minimum perfusion. PHD1−/− and PHD3−/− mice show significantly increased perfusion ratio as compared to WT mice over the course of a 28 day post-operative period.
184
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
experiment, PHD knockout hastened recovery of motor function, a desirable outcome. Deletion of PHD1 and PHD3 increased capillary density following HLI
Fig. 3. Comparison of motor function score. Assessment of motor function score was performed preoperatively before the induction of hind limb ischemia and post-operatively on days 0, 3, 7, 14, 21 and 28. A pre-established motor function scoring system was used: Unrestricted (grade 5), no active use of toe(s) or spreading (grade 4), restricted foot use ) and (grade 3), no use of foot (grade 2), or no use of limb (grade 1). PHD1−/− ( PHD3−/− ( ) mice show significantly increased motor function score as compared to WT ( ) mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT).
[Fig. 2]. On day 28, blood perfusion in PHD1−/− and PHD3−/− reached 0.82 ± 0.11 and 0.84 ± 0.13, respectively, whereas in WT the extent of perfusion was only 0.43 ± 0.04. Deletion of PHD1 and PHD3 improves motor function after HLI In order to assess the functional benefit of PHD inhibition, the walking patterns of mice from all groups were assessed before and after induction of HLI. PHD1−/− and PHD3−/− mice scored significantly higher on motor function tests compared to WT mice on postoperative day 14 and day 21. PHD1−/−, but not PHD3−/− mice scored significantly higher than WT mice on post-operative day 7 as well. Preoperative and post-operative day 0, day 3 and day 28 motor function scores showed no significant differences between groups [Fig. 3]. While motor function was restored in all groups by the end of the
Direct visualization of angiogenesis was also performed to confirm the functional improvements of PHD knockout that were documented by laser Doppler imaging and assessment of motor function. Significant augmentation of capillary density in ischemic tissue was observed by immunohistochemistry. At the conclusion of the 28-day experiment, the PHD1−/− and PHD3−/− mice showed increased capillary density [1902.00 ± 101.60 (n = 8) and 1980.92 ± 259.50 (n = 6)] when compared to WT group [727.73 ± 63.59 (n = 6) counts/mm2; p b 0.05] [Fig. 4]. This increase in vasculature in the PHD1−/− (2.61 fold) and PHD3−/− (2.72 fold) mice likely explains the attenuation of symptoms from HLI. Deletion of PHD1 and PHD3 increased capillary/myocytes ratio following HLI Capillary/myocyte ratio, another marker for angiogenesis, also was assessed at the conclusion of the 28-day experiment. We observed augmented capillary/myocyte ratio in PHD1−/− and PHD3−/− mice [2.22 ± 0.12 (n = 8) and 1.80 ± 0.10 (n = 6) vs. 1.30 ± 0.11 (n = 6); p b 0.05] when compared to WT mice [Fig. 5]. This increase in microvasculature per myocyte agrees with the preservation of muscle function observed in PHD1−/− and PHD3−/− mice. Deletion of PHD1 and PHD3 results in increased arteriolar density following HLI Adductor muscle samples were harvested for immunohistochemistry to assess arteriolar density at the end of the experiment. We observed increased arteriolar density in PHD1−/− and PHD3−/− mice
Fig. 4. Comparison of capillary density between WT, PHD1−/− and PHD3−/− mice. A) Representative images showing caveolin 1 staining in the gastrocnemius muscle of PHD1−/−, PHD3−/− and Wild Type (WT) mice. Imaging was performed at the end of the 28-day experiment. Images were acquired at 400× magnification using Olympus BH2 microscope and results were expressed as counts per mm2. B) PHD1−/− and PHD3−/− mice showed significant increase in capillary density compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
185
Deletion of PHD1 and PHD3 increased Bcl-2 and HIF-1α expressions following HLI
Fig. 5. Comparison of capillary/myocyte ratio between WT, PHD1−/− and PHD3−/− mice. PHD1−/− and PHD3−/− mice show significantly increased capillary/myocyte ratio compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
[22.31 ± 1.56 (n = 6) and 20.86 ± 1.72 (n = 5) vs. 13.59 ± 0.89 (n = 6) counts/mm2; p b 0.05] when compared to WT animals [Fig. 6].
Bcl-2 is an important anti-apoptotic factor. Immunofluorescence analysis and quantification showed that both PHD1−/− and PHD3−/− mice displayed increased expression of Bcl-2 in ischemic muscle compared to WT mice. Specifically, PHD1-/- and PHD3-/- mice exhibited 6.70 fold and 5.06 fold, respectively increased Bcl-2 expression as compared to WT mice [Figs. 7B and D]. Similarly, HIF-1α is well characterized as the master regulator of angiogenesis and as a key factor in many cell survival pathways (Semenza, 2003). To evaluate whether knockdown of PHD1 and PHD3 alters the expression of HIF-1α, we performed immunofluorescence analysis and quantification 3 days after induction of HLI. We observed significant augmentation in the expression of HIF-1α in PHD1−/− (3.70 fold) and PHD3−/− (4.08 fold) ischemic mice when compared to WT ischemic mice [Figs. 7C and E]. These results show that deletion of PHD1 and PHD3 increases Bcl-2 and HIF-1α levels in ischemic tissue, which correlates with the upregulation of angiogenesis and the preservation of tissue structure and function.
Deletion of PHD1 and PHD3 increased VEGF level following HLI
Deletion of PHD1 and PHD3 results in decreased fibrosis following HLI
VEGF is well-recognized as the key regulatory peptide of angiogenesis. Quantification of VEGF by ELISA showed increased expression in ischemic hind limb muscles of PHD1−/− and PHD3−/− mice [69.55 ± 12.49 and 66.45 ± 11.55 vs. 23.26 ± 1.55 (n = 3–4) pg/ml; p b 0.05] as compared to WT mice [Fig. 7A].
Fibrosis measurement by picrosirius red staining was done 28 days after HLI in ischemic limb. PHD1−/− and PHD3−/− mice group showed significant reduction in the fibrotic area and collagen accumulation [4.60 ± 0.50 and 5.15 ± 0.36 vs. 12.41 ± 1.01 (n = 6–8); p b 0.05] as compared to WT mice [Fig. 8].
Fig. 6. Comparison of arteriolar density between WT, PHD1−/− and PHD3−/− mice. A) Representative digital micrographs showing α smooth muscle actin immunostaining in the adductor muscle of PHD1−/−, PHD3−/− and Wild Type (WT) mice. Imaging was performed at the conclusion of the 28-day experiment. Zeiss LSM510 Meta confocal microscope used at 400× magnification was used for image acquisition and results were expressed as arterioles per mm2. B) Confocal image of PHD1−/− and PHD3−/− mice showed significant increase in arteriolar density compared to WT (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 5–6).
186
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
Fig. 7. Increased expression of VEGF, Bcl-2 and HIF-1α in the PHD1−/− and PHD3−/− mice. A: Bar graph represents the VEGF levels measured by ELISA. PHD1−/− and PHD3−/− mice showed significantly more VEGF levels than the WT group (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4). B–E: Representative images are shown of immunofluorescence detection of Bcl-2 and HIF-1α staining performed after 3 days of hind limb ischemia. PHD1−/− and PHD3−/− mice group showed increased expression of Bcl-2 and HIF-1α compared to WT. Quantitative analysis for Bcl-2 expression showed that PHD1−/− and PHD3−/− mice exhibited 6.70 fold and 5.06 fold increased Bcl-2 expression as compared to WT mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4). Similar quantitative analysis for HIF-1α expression showed 3.70 fold and 4.08 fold increase in PHD1−/− and PHD3−/− mice as compared to WT mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4).
Discussion Since Folkman first introduced the concept of angiogenesis in tumors, multiple studies have identified important triggers of new vessel formation (Folkman, 1971, 1984; Folkman and Klagsbrun, 1987). Collateral vessel formation from capillaries appears to be part of a spectrum of complementary phenomena that culminates in enhanced perfusion to tissue beds with impaired blood flow (Carmeliet, 2003). Despite gaining this understanding of angiogenesis and showing therapeutic benefit from the administration of pro-angiogenic molecules in animal models, clinical studies attempting to induce angiogenesis have been uniformly disappointing (Lekas et al., 2006; Lunde et al., 2006; Schachinger et al., 2006). For example, a large, TAMARIS Phase III, randomized, double-blinded, and placebo-controlled, multinational trial studying the effect of non-viral 1 FGF (NV1FGF) on amputation and death rates in patients with CLI failed to impact mortality or rate of limb loss (Belch et al., 2011). Our study indicates that PHD1 and PHD3 knockouts are associated with an increase in both perfusion and functional recoveries after induction of HLI. Perfusion ratio was markedly elevated in PHD1−/− and PHD3−/− mice following HLI, indicating increased blood flow to the ischemic limb and faster recovery from ischemic insult when compared to WT animals. A similar trend in motor function scores from the WT and knockout mice corroborates these conclusions. These results suggest a favorable outcome for PHD inhibition as therapy against ischemia. The severity of downstream HLI, resulting in angiogenesis and subsequent recovery is highly dependent upon the operative technique of femoral artery ligation. There are two generally described techniques of murine HLI. The “proximal” method originally described by Couffinhal et al. (1998) involves ligation of proximal end of femoral artery and distal portion of saphenous artery and ligation and/or cauterization of all the intervening side branches. This proximal upstream ligation tends to
produce a severe model of HLI and results in slow return of perfusion to baseline. The second technique of so called “distal” ligation involves obliteration of the femoral artery distal to the profunda femoris artery and proximal to the genu arteries (Helisch et al., 2006). This downstream ligation results in a mild to moderate ischemia with quick recovery to baseline. Our murine model of HLI is in accordance with similar models of “proximal” ligation previously described (Couffinhal et al., 1998). Following right femoral artery ligation of WT mice, we recorded an 86% drop in the hindlimb perfusion ratio as compared to initial preoperative values. PHD1−/− and PHD3−/− mice show a similar drop in the perfusion ratio immediately after ligation; 85% and 84%, respectively. This also verifies that PHD1−/− and PHD3−/− mice have the same baseline vasculature as the WT. Serial laser Doppler imaging of both the ischemic and non-ischemic limb demonstrated a gradual return of the perfusion towards the preoperative value. When compared to immediate postoperative perfusion, a 29% recovery was recorded in the ischemic limb of WT mice at the conclusion of 28 days in contrast to PHD1−/− and PHD3−/− which showed a 67% and 68% recovery in perfusion ratio respectively. This is also evident by the motor score of the PHD1−/− and PHD3−/− mice which showed rapid return to baseline function at postoperative days 14 and 21 as compared to WT mice that showed slow recovery. PHD1−/− has been previously shown to reprogram muscle metabolism and improve tolerance to ischemia. This study also indicated that PHD1 deficiency conferred myocyte hypoxia resistance (Aragones et al., 2008). Our data indicate that PHD1 deletion improved the recovery of hindlimb perfusion with associated angiogenesis. These two lines of evidence are not necessarily mutually exclusive. We propose that highly necrotic muscle tissues are compromised in the secretion of angiogenic factors due to loss of viable cells capable of normal transcription, translation, and protein secretion activities. In contrast, improved
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
187
laboratories have not only confirmed that PHD inhibitors (PHDI) upregulate angiogenic, proliferative, and pro survival peptides downstream of HIF-1α such as VEGF, NFκB, HO-1 and Bcl-2, but it also augments recovery from acute ischemic insult. Most of these trials reported minimal or no significant side effects with PHDI although the risk of tumorigenesis due to neovascularization remains unknown (Selvaraju et al., 2014). Clinical trials will ultimately prove whether the use of novel PHD inhibitor therapies is effective in combating peripheral arterial disease. Since other types of PHDs modulate collagen synthesis and degradation, it would be important to also develop safe, specific inhibitors to the HIF prolyl-hydroxylase-domain isoenzymes which will likely be required before widespread clinical trials are initiated (Myllyharju, 2008). Given the complexities noted in transitioning PHDI to the clinical realm, continued communication between the basic scientists who develop these smallmolecule inhibitors of PHDs and the clinicians who conduct drug trials can help to make this promising therapeutic design a reality. In summary, our study indicates that PHD1 and PHD3 global knockouts are associated with increased vascularity resulting in improved tissue perfusion and functional recovery after induction of HLI. These findings suggest that pharmacologic inhibition of PHD1 and PHD3, or their targets, may offer a novel approach to the treatment of PAD and other atherosclerotic conditions in the future. Fig. 8. Picrosirius red staining was performed to measure the extent of fibrosis at 28 days post-surgery in gastrocnemius muscle of ischemic limbs of the experimental animals. Picrosirius red staining showed significant decrease in the fibrosis in PHD1−/− and PHD3−/− mice compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
myocyte survival in PHD1−/− mice would maintain more viable myocytes available for expression and secretion of angiogenic factors such as VEGF-A, and thus improve hind limb recovery. On the other hand, improved hind limb perfusion may in turn contribute to further improved myocyte survival, thus initiating a positive feedback cycle leading to accelerated healing. Angiogenic modification through PHD1 has a considerable promise in increasing tissue survivability and recovery following ischemia. We previously have shown that disruption of PHD1 in mice attenuates myocardial ischemia/reperfusion injury ex vivo by increasing HIF-1α transcription-factor activity (Adluri et al., 2011a) and PHD3 gene deleted mice showed an increase in angiogenesis and cardiac function post myocardial infarction (Oriowo et al., 2014). This study not only has successfully transitioned these benefits to a HLI model, but more importantly demonstrates directly a role in angiogenesis which was not investigated in the previous study. Together these studies suggest that PHD inhibition may be a viable form of therapy for ischemia. Our study found histological evidence of increased neovascularization, measured by capillary density and capillary/myocyte ratio in the gastrocnemius muscle in the PHD1−/− and PHD3−/− groups as a putative mechanism for the observed increase in tissue perfusion and functional recovery. Similarly, the experimental groups displayed significantly increased arteriolar density compared to WT animals. This observed increase in angiogenesis from PHD inhibition is associated with stabilization of HIF-1α, which activates a number of angiogenic factors, including VEGF (Semenza, 2003). In our present study we demonstrated that PHD1−/− and PHD3−/− mice have increased VEGF expression compared to WT group. Similarly, Bcl-2, an anti-apoptotic protein, was upregulated in the knockout mice, indicating increased target-cell survival following ischemia. Our murine model of PAD elucidates the potential for PHD inhibition as a novel treatment for acute ischemic insult. As current therapies have failed to significantly decrease the morbidity and mortality of PAD and are limited to patients with large vessel disease, PHD inhibition has enormous translational potential (Criqui et al., 1992). Small-molecule PHD inhibitors have become an even more attractive topic of study after clinical administration of downstream angiogenic factors showed less promising effects (Nagel et al., 2010). Experiments from research
Acknowledgments This study was supported by the Cardiovascular Research Fund. We would also like to thank Joshua W. Goldman, B.S. for proof reading the manuscript. Conflict of interest None. Disclosure This work was presented at American College of Surgeons; Clinical Congress 2012 and in American Heart Association; Scientific Session 2012. References Adam, D.J., et al., 2005. Bypass versus angioplasty in severe ischaemia of the leg (BASIL): multicentre, randomised controlled trial. Lancet 366, 1925–1934. Adluri, R.S., et al., 2011a. Disruption of hypoxia-inducible transcription factor-prolyl hydroxylase domain-1 (PHD-1−/−) attenuates ex vivo myocardial ischemia/reperfusion injury through hypoxia-inducible factor-1alpha transcription factor and its target genes in mice. Antioxid. Redox Signal. 15, 1789–1797. Adluri, R.S., et al., 2011b. Thioredoxin 1 enhances neovascularization and reduces ventricular remodeling during chronic myocardial infarction: a study using thioredoxin 1 transgenic mice. J. Mol. Cell. Cardiol. 50, 239–247. Aragones, J., et al., 2008. Deficiency or inhibition of oxygen sensor Phd1 induces hypoxia tolerance by reprogramming basal metabolism. Nat. Genet. 40, 170–180. Belch, J.J., et al., 2003. Critical issues in peripheral arterial disease detection and management: a call to action. Arch. Intern. Med. 163, 884–892. Belch, J., et al., 2011. Effect of fibroblast growth factor NV1FGF on amputation and death: a randomised placebo-controlled trial of gene therapy in critical limb ischaemia. Lancet 377, 1929–1937. Berra, E., et al., 2006. The hypoxia-inducible-factor hydroxylases bring fresh air into hypoxia signalling. EMBO Rep. 7, 41–45. Bruick, R.K., McKnight, S.L., 2001. A conserved family of prolyl-4-hydroxylases that modify HIF. Science 294, 1337–1340. Cappell, K.M., et al., 2012. Multiple cancer testis antigens function to support tumor cell mitotic fidelity. Mol. Cell. Biol. 32, 4131–4140. Carmeliet, P., 2003. Angiogenesis in health and disease. Nat. Med. 9, 653–660. Couffinhal, T., et al., 1998. Mouse model of angiogenesis. Am. J. Pathol. 152, 1667–1679. Criqui, M.H., et al., 1992. Mortality over a period of 10 years in patients with peripheral arterial disease. N. Engl. J. Med. 326, 381–386. Duca, F.A., et al., 2014. Replication of obesity and associated signaling pathways through transfer of microbiota from obese-prone rats. Diabetes 63, 1624–1636. Epstein, A.C., et al., 2001. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107, 43–54. Ferrara, N., 2002. Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin. Oncol. 29, 10–14. Ferrara, N., et al., 2003. The biology of VEGF and its receptors. Nat. Med. 9, 669–676.
188
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
Folkman, J., 1971. Tumor angiogenesis: therapeutic implications. N. Engl. J. Med. 285, 1182–1186. Folkman, J., 1984. Angiogenesis. In: Jaffe, E.A. (Ed.), Biology of Endothelial Cells. MartinusNijhoff, Boston, pp. 412–428. Folkman, J., Klagsbrun, M., 1987. Angiogenic factors. Science 235, 442–447. Golomb, B.A., et al., 2006. Peripheral arterial disease: morbidity and mortality implications. Circulation 114, 688–699. Harari-Steinberg, O., et al., 2013. Identification of human nephron progenitors capable of generation of kidney structures and functional repair of chronic renal disease. EMBO Mol. Med. 5, 1556–1568. Helisch, A., et al., 2006. Impact of mouse strain differences in innate hindlimb collateral vasculature. Arterioscler. Thromb. Vasc. Biol. 26, 520–526. Jones, W.S., et al., 2012. Temporal trends and geographic variation of lower-extremity amputation in patients with peripheral artery disease: results from U.S. Medicare 2000–2008. J. Am. Coll. Cardiol. 60, 2230–2236. Lekas, M., et al., 2006. Growth factor-induced therapeutic neovascularization for ischaemic vascular disease: time for a re-evaluation? Curr. Opin. Cardiol. 21, 376–384. Limbourg, A., et al., 2009. Evaluation of postnatal arteriogenesis and angiogenesis in a mouse model of hind-limb ischemia. Nat. Protoc. 4, 1737–1746. Lunde, K., et al., 2006. Intracoronary injection of mononuclear bone marrow cells in acute myocardial infarction. N. Engl. J. Med. 355, 1199–1209. Messaoudi, S., et al., 2009. Cardiac aldosterone overexpression prevents harmful effects of diabetes in the mouse heart by preserving capillary density. FASEB J. 23, 2176–2185. Myllyharju, J., 2008. Prolyl 4-hydroxylases, key enzymes in the synthesis of collagens and regulation of the response to hypoxia, and their roles as treatment targets. Ann. Med. 40, 402–417. Nagel, S., et al., 2010. Therapeutic manipulation of the HIF hydroxylases. Antioxid. Redox Signal. 12, 481–501. Nor, J.E., et al., 1999. Vascular endothelial growth factor (VEGF)-mediated angiogenesis is associated with enhanced endothelial cell survival and induction of Bcl-2 expression. Am. J. Pathol. 154, 375–384. Norgren, L., et al., 2007. Inter-society consensus for the management of peripheral arterial disease (TASC II). J. Vasc. Surg. 45 (Suppl. S), S5–S67.
Oriowo, B., et al., 2014. Targeted gene deletion of prolyl hydroxylase domain protein 3 triggers angiogenesis and preserves cardiac function by stabilizing hypoxia inducible factor 1 alpha following myocardial infarction. Curr. Pharm. Des. 20, 1305–1310. Ouriel, K., 2001. Peripheral arterial disease. Lancet 358, 1257–1264. Ratajczak, P., et al., 2005. Expression and localization of caveolins during postnatal development in rat heart: implication of thyroid hormone. J. Appl. Physiol. 99, 244–251 (1985). Raval, Z., Losordo, D.W., 2013. Cell therapy of peripheral arterial disease: from experimental findings to clinical trials. Circ. Res. 112, 1288–1302. Schachinger, V., et al., 2006. Intracoronary bone marrow-derived progenitor cells in acute myocardial infarction. N. Engl. J. Med. 355, 1210–1221. Selvaraju, V., et al., 2014. Molecular mechanisms of action and therapeutic uses of pharmacological inhibitors of HIF-prolyl 4-hydroxylases for treatment of ischemic diseases. Antioxid. Redox Signal. 20, 2631–2665. Semenza, G.L., 2003. Targeting HIF-1 for cancer therapy. Nat. Rev. Cancer 3, 721–732. Shizukuda, Y., et al., 1999. Vascular endothelial growth factor-induced endothelial cell migration and proliferation depend on a nitric oxide-mediated decrease in protein kinase Cdelta activity. Circ. Res. 85, 247–256. Takeda, K., et al., 2006. Placental but not heart defects are associated with elevated hypoxia-inducible factor alpha levels in mice lacking prolyl hydroxylase domain protein 2. Mol. Cell. Biol. 26, 8336–8346. Thirunavukkarasu, M., et al., 2013. Simvastatin treatment inhibits hypoxia inducible factor 1-alpha-(HIF-1alpha)-prolyl-4-hydroxylase 3 (PHD-3) and increases angiogenesis after myocardial infarction in streptozotocin-induced diabetic rat. Int. J. Cardiol. 168, 2474–2480. Webber, M.J., et al., 2011. Supramolecular nanostructures that mimic VEGF as a strategy for ischemic tissue repair. Proc. Natl. Acad. Sci. U. S. A. 108, 13438–13443. Wu, S., et al., 2008. Enhancement of angiogenesis through stabilization of hypoxiainducible factor-1 by silencing prolyl hydroxylase domain-2 gene. Mol. Ther. 16, 1227–1234.
Contents lists available at ScienceDirect
Microvascular Research journal homepage: www.elsevier.com/locate/ymvre
Deletion of prolyl hydroxylase domain proteins (PHD1, PHD3) stabilizes hypoxia inducible factor-1 alpha, promotes neovascularization, and improves perfusion in a murine model of hind-limb ischemia Muhammad T. Rishi a,b,1, Vaithinathan Selvaraju a,1, Mahesh Thirunavukkarasu a, Inam A. Shaikh a,b, Kotaro Takeda c, Guo-Hua Fong c, J. Alexander Palesty b, Juan A. Sanchez a, Nilanjana Maulik a,⁎ a b c
Molecular Cardiology and Angiogenesis Laboratory, Department of Surgery, University of Connecticut Health Center, Farmington, CT, USA Stanley J. Dudrick Department of Surgery, Saint Mary's Hospital, Waterbury, CT, USA Department of Cell Biology, University of Connecticut Health Center, Farmington, CT, USA
a r t i c l e
i n f o
Article history: Accepted 24 October 2014 Available online 3 November 2014 Keywords: Hindlimb ischemia Prolyl hydroxylase HIF VEGF Bcl-2
a b s t r a c t Background: There is an emerging focus on investigating innovative therapeutic molecules that can potentially augment neovascularization in order to treat peripheral arterial disease (PAD). Although prolyl hydroxylase domain proteins 1 and 3 (PHD1 and PHD3) may modulate angiogenesis via regulation of hypoxia inducible factor-1α (HIF-1α), there has been no study directly addressing their roles in ischemia-induced vascular growth. We hypothesize that PHD1−/− or PHD3−/− deficiency might promote angiogenesis in the murine hind-limb ischemia (HLI) model. Study design: Wild type (WT), PHD1−/− and PHD3−/− male mice aged 8–12 weeks underwent right femoral artery ligation. Post-procedurally, motor function assessment and laser Doppler imaging were periodically performed. The mice were euthanized after 28 days and muscles were harvested. Immunohistochemical analysis was performed to determine the extent of angiogenesis by measuring capillary and arteriolar density. VEGF expression was quantified by enzyme-linked immunosorbent assay (ELISA). Bcl-2 and HIF-1α were analyzed by immunofluorescence. Fibrosis was measured by picrosirius red staining. Results: PHD1−/− and PHD3−/− mice showed significantly improved recovery of perfusion and motor function score when compared to WT after femoral artery ligation. These mice also exhibited increased capillary and arteriolar density, capillary/myocyte ratio along with decreased fibrosis compared to WT. VEGF, Bcl-2 and HIF-1α expression increased in PHD1−/− and PHD3−/− mice compared to WT. Conclusions: Taken together these results suggest that PHD1 and PHD3 deletions promote angiogenesis in ischemiainjured tissue, and may present a promising therapeutic strategy in treating PAD. © 2014 Elsevier Inc. All rights reserved.
Introduction Peripheral arterial disease (PAD) is a progressive and debilitating illness that reduces the quality of life in more than 25 million patients in Europe and North America alone (Belch et al., 2003; Norgren et al., 2007; Raval and Losordo, 2013). It is strongly associated with other atherosclerotic pathologies including coronary artery and cerebrovascular disease (Golomb et al., 2006). While increased understanding of PAD has resulted in the development of multiple treatment modalities, amputation rates from PAD still remain high. A recent study looking at the Medicare database from 2000 through 2008 found that among 3
⁎ Corresponding author at: Molecular Cardiology and Angiogenesis Laboratory, Department of Surgery, University of Connecticut Health Center, 263 Farmington Ave., Farmington, CT 06032-1110, USA. E-mail address: [email protected] (N. Maulik). 1 Contributed equally.
http://dx.doi.org/10.1016/j.mvr.2014.10.009 0026-2862/© 2014 Elsevier Inc. All rights reserved.
million Medicare admissions for PAD, a total of 186,338 patients underwent lower extremity amputations, which were 6.8% of all the hospitalized PAD population (Jones et al., 2012). Approximately 50% of the patients with critical limb ischemia (CLI) are unsuitable for any form of revascularization, and those who receive interventional treatment still carry a poor long-term prognosis (Adam et al., 2005; Ouriel, 2001). Additionally, significant morbidity occurs as a result of chronic wound ulceration particularly in diabetics in which an impaired microcirculation results in local tissue ischemia. The management of PAD follows three main approaches: risk factor attenuation, pharmacological intervention, and surgical revascularization. Revascularization can be achieved either by open or endovascular surgery, but neither approach significantly alleviates disease morbidity and mortality (Ouriel, 2001). In 2005, a multicenter, randomized controlled study highlighted the shortfalls of PAD treatment by comparing the effectiveness of bypass versus angioplasty in severe ischemia of the leg (BASIL Trial). In this study, half of the patients
182
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
with CLI were considered unsuitable for any form of revascularization due to advanced disease progression, and those who received treatment presented with very poor prognosis (Adam et al., 2005). It is increasingly apparent that PAD may require other treatment modalities that upregulate the ability of the patient to generate new vasculature. Currently available strategies to promote angiogenesis to treat the consequences of PAD have not resulted in clinically applicable forms of therapy. However, targeted molecular approaches to enhance angiogenesis have considerable theoretical advantages to other forms of treatment. Hypoxia inducible factor-1α (HIF-1α) appears to be one of the key regulators of the downstream effectors of angiogenesis by inhibition of PHDs (Adluri et al., 2011a). Therefore, upregulation of HIF-1α via inhibition of prolyl hydroxylase domain enzymes (PHDs) is a potentially effective option for patients afflicted with PAD. There are three PHD variants in mammalian cells: PHD1 (EglN2), PHD2 (EglN1), and PHD3 (EglN3). There are organ specific differences in the expression of these three isoenzymes. PHDs act as cellular oxygen sensors and play a critical role in triggering angiogenesis. During normoxic conditions, HIF-1α is hydroxylated at two proline residues (P402 and P546) by one of the three identified PHD isoenzymes (Bruick and McKnight, 2001; Epstein et al., 2001). Hydroxylation initiates binding of von Hippel–Lindau tumor suppressor protein and proteasomal degradation of HIF-1α by the E3 ubiquitin ligase complex (Berra et al., 2006). Under hypoxia, PHD enzyme activity is inhibited. The inhibition of PHD enzyme activity allows stabilization and translocation of HIF-1α to the nucleus and dimerization with its β-subunit to form an active transcription factor. This complex then binds to hypoxia-response elements (HRE) in the regulatory regions of target genes to upregulate transcription and translation of angiogenic peptides, including VEGF, eNOS, Notch ligands, platelet-derived growth factor (PDGF) and angiopoietin-2 (Ang-2) (Semenza, 2003). Of these peptides, VEGF plays the most vital role in angiogenesis since it is required for the growth and development of vascular endothelial cells as well as endothelial cell migration and proliferation (Ferrara, 2002; Ferrara et al., 2003; Shizukuda et al., 1999). VEGF, additionally, upregulates Bcl-2 to promote endothelial cell survival (Nor et al., 1999). While a number of studies have explored various ways to alter this angiogenic pathway as a potential treatment for ischemic vascular disease, we still do not have a satisfactory clinically proven answer. Thus, there is an urgent need for more novel approaches. PHDs may constitute promising candidates for angiogenesis related therapies due to their role in regulating HIF-dependent expression of angiogenic factors. While PHD2 has been clearly shown to regulate angiogenesis (Wu et al., 2008), its involvement in a number of other processes such as erythropoiesis suggests that targeting PHD2 may be associated with complications. PHD1 and PHD3, on the other hand, do not substantially affect the blood vessels when targeted individually, and hence may provide less risky angiogenesis targets. Here we have examined whether genetically disabling specific PHDs (PHD1 and PHD3) improves limb perfusion by enhancing angiogenesis following HLI. We show that deletion of PHD1 and PHD3 promotes ischemia-induced angiogenesis and potentiates reperfusion in a murine model of HLI, making pharmacologic inhibition of PHDs a possible useful therapeutic strategy in the treatment of PAD. Materials and methods Experimental animals All animal protocols were approved by the Institutional Animal Care and Committee of the University of Connecticut Health Center, Farmington, Connecticut. PHD1−/− and PHD3−/− mice were kindly provided by Dr. GH Fong and were generated as described previously (Takeda et al., 2006).
Mouse model of hind-limb ischemia Adult, 8–12 week-old male C57BL/6, PHD1−/− and PHD3−/− mice were subjected to femoral artery ligation as described previously (Couffinhal et al., 1998; Limbourg et al., 2009). Briefly, mice were anesthetized with a rodent isoflurane vaporizer set at 2.5% with a constant oxygen supply of 1 L/min. The right femoral artery was surgically isolated and ligated with 8–0 prolene suture both proximal and distal to the origin of the deep femoral artery. After ligation, the intervening segment was excised, and all branches between these two sites were ligated or cauterized. The left femoral artery was used as an internal control without ligation.
Laser Doppler perfusion imaging Laser Doppler perfusion imaging was performed using a PeriScan (PIM 3) System (Perimed, Järfälla-Stockholm, Sweden). Baseline images were acquired before the creation of HLI, and post-operative images were taken on days 0, 3, 7, 14, 21 and 28. Three Doppler-perfusion readings were taken at each time point, and an average was calculated to provide final blood-flow values as the ratio of ischemic to non-ischemic hind-limb perfusion.
Motor function assessment In order to assess motor function after induction of HLI, a preestablished scoring system was used (Webber et al., 2011) preoperatively and on post-operative days 0, 3, 7, 14, 21 and 28. Under this scoring system, hind-limb motor function was graded as unrestricted (grade 5), no active use of toe(s) or spreading (grade 4), restricted foot use (grade 3), no use of foot (grade 2), or no use of limb (grade 1).
Morphometric analysis for capillary and arteriolar density Immunohistochemical analysis for both capillary and arteriolar density was performed with paraffin embedded skeletal muscle sections harvested from the ischemic limb, 28 days after surgery. Sections were deparaffinized, hydrated with serial dilution in ethanol and blocked with 2.5% normal horse serum followed by incubation with primary antibody (Caveolin 1, an endothelial cell specific marker antibody— Santa Cruz Biotechnology, Santa Cruz, CA, USA for capillary density (Messaoudi et al., 2009; Ratajczak et al., 2005) and mouse monoclonal anti-α smooth muscle actin antibody—Abcam Inc., Cambridge, MA, USA for arteriolar density) and respective secondary antibody. Images were obtained at 400× magnification on Olympus BH2 microscope and Zeiss LSM 510 Meta confocal microscope for capillary and arteriolar density respectively. Capillaries and arterioles were counted per millimeter square area using Adobe Photoshop CS4 software (Adluri et al., 2011b).
VEGF quantification by ELISA VEGF quantification was performed via ELISA as previously described (Thirunavukkarasu et al., 2013). Briefly, mice from all three groups (WT, PHD1−/− and PHD3−/− mice) were euthanized 3 days after femoral artery ligation and ischemic hind limb gastrocnemius muscle tissue sections were harvested from each treatment group. Subsequently, these muscle samples were homogenized and suspended (50–75 mg/ml) in sample buffer. Protein was isolated and then the total protein concentration was determined using a BCA (bicinchoninic acid) protein assay kit (Pierce, Rockville, IL). Using a mouse VEGF Duoset (R&D Systems Inc., MN), the expression of VEGF (in pg/ml) was determined according to the manufacturer's instructions.
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
183
Bcl-2 and HIF-1α expression by immunofluorescence analysis Ischemic limb muscles were collected 3 days after HLI and tissue sections were prepared as described previously and used for immunofluorescence staining (Thirunavukkarasu et al., 2013; Limbourg et al., 2009). Briefly, the sections were stained for Bcl-2 and HIF1-α with rabbit polyclonal anti-Bcl-2 and rabbit polyclonal anti-HIF-1α (Santa Cruz Biotechnology, CA, USA), respectively, and incubated overnight at 4 °C. The sections were then treated with respective secondary antibodies and nuclear staining was performed with TO-PRO-3 (Invitrogen, Carlsbad, CA, USA). Finally, the images for Bcl-2 and HIF-1α staining were digitally acquired at 400× magnification using Zeiss LSM Confocor 3 microscope at the same exposure and gain settings for each sample comparison. The corrected total fluorescence was measured as described earlier (Cappell et al., 2012; Duca et al., 2014; Harari-Steinberg et al., 2013). Briefly, area and integrated mean density were quantified using ImageJ software from the acquired images. Five to ten pictures from each samples (n = 3–4) and the corrected total fluorescence (CTF) were calculated by the formula integrated density − (area of selected picture × mean fluorescence of background reading).
Fig. 2. In-vivo comparison of perfusion ratio in WT, PHD1−/− and PHD3−/− mice after right femoral artery ligation. Preoperative images using laser Doppler were taken before induction of ischemia as well as post-operative days 0, 3, 7, 14, 21 and 28. Digital colorcoded images were obtained and then analyzed through PIMsoft (version 1.5.4.8078) to quantify the perfusion in the area of interest. PHD1−/− ( ) and PHD3 −/− ( ) mice show significantly increased perfusion ratio as compared to WT ( ) mice on post-operative days 3, 7, 14, 21 and 28 (*p b 0.05; WT vs. PHD1−/− and #p b 0.05; WT vs. PHD3−/−; n = 9–19).
Results Picro sirius red/collagen staining PHD 1 and 3 deletion enhances blood flow recovery in mice after HLI Picrosirius staining was performed to measure the extent of fibrosis in 5 μm sections of gastrocnemius muscle of ischemic limbs as described earlier (Oriowo et al., 2014). Statistical analysis Statistical analysis was conducted using GraphPad Prism Software to perform one-way ANOVA followed by Newman–Kuels multiple comparison test. The results were considered significant at p b 0.05. Values were expressed as Mean ± S.E.M.
To test if PHD1 or PHD3 deficiency improves blood flow, we performed femoral artery ligation and monitored hind limb perfusion recovery with laser Doppler imaging [Fig. 1]. Baseline hind limb perfusion (Pre-op) in PHD1−/− and PHD3−/− mice showed no significant changes compared to WT mice. Immediately after the surgery (day 0), both PHD1−/− and PHD3−/− mice suffered from equivalent ischemia compared to WT mice. However, PHD1−/− and PHD3−/− mice showed significant improvement in perfusion ratio (laser Doppler imaging) when compared with WT mice on post-operative day 3 until day 28
Fig. 1. Comparison of laser Doppler perfusion images of WT, PHD1−/− and PHD3−/− mice after right femoral artery ligation with representative LDIs from different groups. Laser Doppler was used to take baseline preoperative and post-operative images on days 0, 3, 7, 14, 21 and 28. Red indicates maximum perfusion while black indicates minimum perfusion. PHD1−/− and PHD3−/− mice show significantly increased perfusion ratio as compared to WT mice over the course of a 28 day post-operative period.
184
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
experiment, PHD knockout hastened recovery of motor function, a desirable outcome. Deletion of PHD1 and PHD3 increased capillary density following HLI
Fig. 3. Comparison of motor function score. Assessment of motor function score was performed preoperatively before the induction of hind limb ischemia and post-operatively on days 0, 3, 7, 14, 21 and 28. A pre-established motor function scoring system was used: Unrestricted (grade 5), no active use of toe(s) or spreading (grade 4), restricted foot use ) and (grade 3), no use of foot (grade 2), or no use of limb (grade 1). PHD1−/− ( PHD3−/− ( ) mice show significantly increased motor function score as compared to WT ( ) mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT).
[Fig. 2]. On day 28, blood perfusion in PHD1−/− and PHD3−/− reached 0.82 ± 0.11 and 0.84 ± 0.13, respectively, whereas in WT the extent of perfusion was only 0.43 ± 0.04. Deletion of PHD1 and PHD3 improves motor function after HLI In order to assess the functional benefit of PHD inhibition, the walking patterns of mice from all groups were assessed before and after induction of HLI. PHD1−/− and PHD3−/− mice scored significantly higher on motor function tests compared to WT mice on postoperative day 14 and day 21. PHD1−/−, but not PHD3−/− mice scored significantly higher than WT mice on post-operative day 7 as well. Preoperative and post-operative day 0, day 3 and day 28 motor function scores showed no significant differences between groups [Fig. 3]. While motor function was restored in all groups by the end of the
Direct visualization of angiogenesis was also performed to confirm the functional improvements of PHD knockout that were documented by laser Doppler imaging and assessment of motor function. Significant augmentation of capillary density in ischemic tissue was observed by immunohistochemistry. At the conclusion of the 28-day experiment, the PHD1−/− and PHD3−/− mice showed increased capillary density [1902.00 ± 101.60 (n = 8) and 1980.92 ± 259.50 (n = 6)] when compared to WT group [727.73 ± 63.59 (n = 6) counts/mm2; p b 0.05] [Fig. 4]. This increase in vasculature in the PHD1−/− (2.61 fold) and PHD3−/− (2.72 fold) mice likely explains the attenuation of symptoms from HLI. Deletion of PHD1 and PHD3 increased capillary/myocytes ratio following HLI Capillary/myocyte ratio, another marker for angiogenesis, also was assessed at the conclusion of the 28-day experiment. We observed augmented capillary/myocyte ratio in PHD1−/− and PHD3−/− mice [2.22 ± 0.12 (n = 8) and 1.80 ± 0.10 (n = 6) vs. 1.30 ± 0.11 (n = 6); p b 0.05] when compared to WT mice [Fig. 5]. This increase in microvasculature per myocyte agrees with the preservation of muscle function observed in PHD1−/− and PHD3−/− mice. Deletion of PHD1 and PHD3 results in increased arteriolar density following HLI Adductor muscle samples were harvested for immunohistochemistry to assess arteriolar density at the end of the experiment. We observed increased arteriolar density in PHD1−/− and PHD3−/− mice
Fig. 4. Comparison of capillary density between WT, PHD1−/− and PHD3−/− mice. A) Representative images showing caveolin 1 staining in the gastrocnemius muscle of PHD1−/−, PHD3−/− and Wild Type (WT) mice. Imaging was performed at the end of the 28-day experiment. Images were acquired at 400× magnification using Olympus BH2 microscope and results were expressed as counts per mm2. B) PHD1−/− and PHD3−/− mice showed significant increase in capillary density compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
185
Deletion of PHD1 and PHD3 increased Bcl-2 and HIF-1α expressions following HLI
Fig. 5. Comparison of capillary/myocyte ratio between WT, PHD1−/− and PHD3−/− mice. PHD1−/− and PHD3−/− mice show significantly increased capillary/myocyte ratio compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
[22.31 ± 1.56 (n = 6) and 20.86 ± 1.72 (n = 5) vs. 13.59 ± 0.89 (n = 6) counts/mm2; p b 0.05] when compared to WT animals [Fig. 6].
Bcl-2 is an important anti-apoptotic factor. Immunofluorescence analysis and quantification showed that both PHD1−/− and PHD3−/− mice displayed increased expression of Bcl-2 in ischemic muscle compared to WT mice. Specifically, PHD1-/- and PHD3-/- mice exhibited 6.70 fold and 5.06 fold, respectively increased Bcl-2 expression as compared to WT mice [Figs. 7B and D]. Similarly, HIF-1α is well characterized as the master regulator of angiogenesis and as a key factor in many cell survival pathways (Semenza, 2003). To evaluate whether knockdown of PHD1 and PHD3 alters the expression of HIF-1α, we performed immunofluorescence analysis and quantification 3 days after induction of HLI. We observed significant augmentation in the expression of HIF-1α in PHD1−/− (3.70 fold) and PHD3−/− (4.08 fold) ischemic mice when compared to WT ischemic mice [Figs. 7C and E]. These results show that deletion of PHD1 and PHD3 increases Bcl-2 and HIF-1α levels in ischemic tissue, which correlates with the upregulation of angiogenesis and the preservation of tissue structure and function.
Deletion of PHD1 and PHD3 increased VEGF level following HLI
Deletion of PHD1 and PHD3 results in decreased fibrosis following HLI
VEGF is well-recognized as the key regulatory peptide of angiogenesis. Quantification of VEGF by ELISA showed increased expression in ischemic hind limb muscles of PHD1−/− and PHD3−/− mice [69.55 ± 12.49 and 66.45 ± 11.55 vs. 23.26 ± 1.55 (n = 3–4) pg/ml; p b 0.05] as compared to WT mice [Fig. 7A].
Fibrosis measurement by picrosirius red staining was done 28 days after HLI in ischemic limb. PHD1−/− and PHD3−/− mice group showed significant reduction in the fibrotic area and collagen accumulation [4.60 ± 0.50 and 5.15 ± 0.36 vs. 12.41 ± 1.01 (n = 6–8); p b 0.05] as compared to WT mice [Fig. 8].
Fig. 6. Comparison of arteriolar density between WT, PHD1−/− and PHD3−/− mice. A) Representative digital micrographs showing α smooth muscle actin immunostaining in the adductor muscle of PHD1−/−, PHD3−/− and Wild Type (WT) mice. Imaging was performed at the conclusion of the 28-day experiment. Zeiss LSM510 Meta confocal microscope used at 400× magnification was used for image acquisition and results were expressed as arterioles per mm2. B) Confocal image of PHD1−/− and PHD3−/− mice showed significant increase in arteriolar density compared to WT (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 5–6).
186
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
Fig. 7. Increased expression of VEGF, Bcl-2 and HIF-1α in the PHD1−/− and PHD3−/− mice. A: Bar graph represents the VEGF levels measured by ELISA. PHD1−/− and PHD3−/− mice showed significantly more VEGF levels than the WT group (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4). B–E: Representative images are shown of immunofluorescence detection of Bcl-2 and HIF-1α staining performed after 3 days of hind limb ischemia. PHD1−/− and PHD3−/− mice group showed increased expression of Bcl-2 and HIF-1α compared to WT. Quantitative analysis for Bcl-2 expression showed that PHD1−/− and PHD3−/− mice exhibited 6.70 fold and 5.06 fold increased Bcl-2 expression as compared to WT mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4). Similar quantitative analysis for HIF-1α expression showed 3.70 fold and 4.08 fold increase in PHD1−/− and PHD3−/− mice as compared to WT mice (*p b 0.05; PHD1−/− vs. WT and #p b 0.05; PHD3−/− vs. WT; n = 3–4).
Discussion Since Folkman first introduced the concept of angiogenesis in tumors, multiple studies have identified important triggers of new vessel formation (Folkman, 1971, 1984; Folkman and Klagsbrun, 1987). Collateral vessel formation from capillaries appears to be part of a spectrum of complementary phenomena that culminates in enhanced perfusion to tissue beds with impaired blood flow (Carmeliet, 2003). Despite gaining this understanding of angiogenesis and showing therapeutic benefit from the administration of pro-angiogenic molecules in animal models, clinical studies attempting to induce angiogenesis have been uniformly disappointing (Lekas et al., 2006; Lunde et al., 2006; Schachinger et al., 2006). For example, a large, TAMARIS Phase III, randomized, double-blinded, and placebo-controlled, multinational trial studying the effect of non-viral 1 FGF (NV1FGF) on amputation and death rates in patients with CLI failed to impact mortality or rate of limb loss (Belch et al., 2011). Our study indicates that PHD1 and PHD3 knockouts are associated with an increase in both perfusion and functional recoveries after induction of HLI. Perfusion ratio was markedly elevated in PHD1−/− and PHD3−/− mice following HLI, indicating increased blood flow to the ischemic limb and faster recovery from ischemic insult when compared to WT animals. A similar trend in motor function scores from the WT and knockout mice corroborates these conclusions. These results suggest a favorable outcome for PHD inhibition as therapy against ischemia. The severity of downstream HLI, resulting in angiogenesis and subsequent recovery is highly dependent upon the operative technique of femoral artery ligation. There are two generally described techniques of murine HLI. The “proximal” method originally described by Couffinhal et al. (1998) involves ligation of proximal end of femoral artery and distal portion of saphenous artery and ligation and/or cauterization of all the intervening side branches. This proximal upstream ligation tends to
produce a severe model of HLI and results in slow return of perfusion to baseline. The second technique of so called “distal” ligation involves obliteration of the femoral artery distal to the profunda femoris artery and proximal to the genu arteries (Helisch et al., 2006). This downstream ligation results in a mild to moderate ischemia with quick recovery to baseline. Our murine model of HLI is in accordance with similar models of “proximal” ligation previously described (Couffinhal et al., 1998). Following right femoral artery ligation of WT mice, we recorded an 86% drop in the hindlimb perfusion ratio as compared to initial preoperative values. PHD1−/− and PHD3−/− mice show a similar drop in the perfusion ratio immediately after ligation; 85% and 84%, respectively. This also verifies that PHD1−/− and PHD3−/− mice have the same baseline vasculature as the WT. Serial laser Doppler imaging of both the ischemic and non-ischemic limb demonstrated a gradual return of the perfusion towards the preoperative value. When compared to immediate postoperative perfusion, a 29% recovery was recorded in the ischemic limb of WT mice at the conclusion of 28 days in contrast to PHD1−/− and PHD3−/− which showed a 67% and 68% recovery in perfusion ratio respectively. This is also evident by the motor score of the PHD1−/− and PHD3−/− mice which showed rapid return to baseline function at postoperative days 14 and 21 as compared to WT mice that showed slow recovery. PHD1−/− has been previously shown to reprogram muscle metabolism and improve tolerance to ischemia. This study also indicated that PHD1 deficiency conferred myocyte hypoxia resistance (Aragones et al., 2008). Our data indicate that PHD1 deletion improved the recovery of hindlimb perfusion with associated angiogenesis. These two lines of evidence are not necessarily mutually exclusive. We propose that highly necrotic muscle tissues are compromised in the secretion of angiogenic factors due to loss of viable cells capable of normal transcription, translation, and protein secretion activities. In contrast, improved
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
187
laboratories have not only confirmed that PHD inhibitors (PHDI) upregulate angiogenic, proliferative, and pro survival peptides downstream of HIF-1α such as VEGF, NFκB, HO-1 and Bcl-2, but it also augments recovery from acute ischemic insult. Most of these trials reported minimal or no significant side effects with PHDI although the risk of tumorigenesis due to neovascularization remains unknown (Selvaraju et al., 2014). Clinical trials will ultimately prove whether the use of novel PHD inhibitor therapies is effective in combating peripheral arterial disease. Since other types of PHDs modulate collagen synthesis and degradation, it would be important to also develop safe, specific inhibitors to the HIF prolyl-hydroxylase-domain isoenzymes which will likely be required before widespread clinical trials are initiated (Myllyharju, 2008). Given the complexities noted in transitioning PHDI to the clinical realm, continued communication between the basic scientists who develop these smallmolecule inhibitors of PHDs and the clinicians who conduct drug trials can help to make this promising therapeutic design a reality. In summary, our study indicates that PHD1 and PHD3 global knockouts are associated with increased vascularity resulting in improved tissue perfusion and functional recovery after induction of HLI. These findings suggest that pharmacologic inhibition of PHD1 and PHD3, or their targets, may offer a novel approach to the treatment of PAD and other atherosclerotic conditions in the future. Fig. 8. Picrosirius red staining was performed to measure the extent of fibrosis at 28 days post-surgery in gastrocnemius muscle of ischemic limbs of the experimental animals. Picrosirius red staining showed significant decrease in the fibrosis in PHD1−/− and PHD3−/− mice compared to WT (*p b 0.05; ischemic PHD1−/− vs. ischemic WT; n = 6–8 and #p b 0.05; ischemic PHD3−/− vs. ischemic WT; n = 6).
myocyte survival in PHD1−/− mice would maintain more viable myocytes available for expression and secretion of angiogenic factors such as VEGF-A, and thus improve hind limb recovery. On the other hand, improved hind limb perfusion may in turn contribute to further improved myocyte survival, thus initiating a positive feedback cycle leading to accelerated healing. Angiogenic modification through PHD1 has a considerable promise in increasing tissue survivability and recovery following ischemia. We previously have shown that disruption of PHD1 in mice attenuates myocardial ischemia/reperfusion injury ex vivo by increasing HIF-1α transcription-factor activity (Adluri et al., 2011a) and PHD3 gene deleted mice showed an increase in angiogenesis and cardiac function post myocardial infarction (Oriowo et al., 2014). This study not only has successfully transitioned these benefits to a HLI model, but more importantly demonstrates directly a role in angiogenesis which was not investigated in the previous study. Together these studies suggest that PHD inhibition may be a viable form of therapy for ischemia. Our study found histological evidence of increased neovascularization, measured by capillary density and capillary/myocyte ratio in the gastrocnemius muscle in the PHD1−/− and PHD3−/− groups as a putative mechanism for the observed increase in tissue perfusion and functional recovery. Similarly, the experimental groups displayed significantly increased arteriolar density compared to WT animals. This observed increase in angiogenesis from PHD inhibition is associated with stabilization of HIF-1α, which activates a number of angiogenic factors, including VEGF (Semenza, 2003). In our present study we demonstrated that PHD1−/− and PHD3−/− mice have increased VEGF expression compared to WT group. Similarly, Bcl-2, an anti-apoptotic protein, was upregulated in the knockout mice, indicating increased target-cell survival following ischemia. Our murine model of PAD elucidates the potential for PHD inhibition as a novel treatment for acute ischemic insult. As current therapies have failed to significantly decrease the morbidity and mortality of PAD and are limited to patients with large vessel disease, PHD inhibition has enormous translational potential (Criqui et al., 1992). Small-molecule PHD inhibitors have become an even more attractive topic of study after clinical administration of downstream angiogenic factors showed less promising effects (Nagel et al., 2010). Experiments from research
Acknowledgments This study was supported by the Cardiovascular Research Fund. We would also like to thank Joshua W. Goldman, B.S. for proof reading the manuscript. Conflict of interest None. Disclosure This work was presented at American College of Surgeons; Clinical Congress 2012 and in American Heart Association; Scientific Session 2012. References Adam, D.J., et al., 2005. Bypass versus angioplasty in severe ischaemia of the leg (BASIL): multicentre, randomised controlled trial. Lancet 366, 1925–1934. Adluri, R.S., et al., 2011a. Disruption of hypoxia-inducible transcription factor-prolyl hydroxylase domain-1 (PHD-1−/−) attenuates ex vivo myocardial ischemia/reperfusion injury through hypoxia-inducible factor-1alpha transcription factor and its target genes in mice. Antioxid. Redox Signal. 15, 1789–1797. Adluri, R.S., et al., 2011b. Thioredoxin 1 enhances neovascularization and reduces ventricular remodeling during chronic myocardial infarction: a study using thioredoxin 1 transgenic mice. J. Mol. Cell. Cardiol. 50, 239–247. Aragones, J., et al., 2008. Deficiency or inhibition of oxygen sensor Phd1 induces hypoxia tolerance by reprogramming basal metabolism. Nat. Genet. 40, 170–180. Belch, J.J., et al., 2003. Critical issues in peripheral arterial disease detection and management: a call to action. Arch. Intern. Med. 163, 884–892. Belch, J., et al., 2011. Effect of fibroblast growth factor NV1FGF on amputation and death: a randomised placebo-controlled trial of gene therapy in critical limb ischaemia. Lancet 377, 1929–1937. Berra, E., et al., 2006. The hypoxia-inducible-factor hydroxylases bring fresh air into hypoxia signalling. EMBO Rep. 7, 41–45. Bruick, R.K., McKnight, S.L., 2001. A conserved family of prolyl-4-hydroxylases that modify HIF. Science 294, 1337–1340. Cappell, K.M., et al., 2012. Multiple cancer testis antigens function to support tumor cell mitotic fidelity. Mol. Cell. Biol. 32, 4131–4140. Carmeliet, P., 2003. Angiogenesis in health and disease. Nat. Med. 9, 653–660. Couffinhal, T., et al., 1998. Mouse model of angiogenesis. Am. J. Pathol. 152, 1667–1679. Criqui, M.H., et al., 1992. Mortality over a period of 10 years in patients with peripheral arterial disease. N. Engl. J. Med. 326, 381–386. Duca, F.A., et al., 2014. Replication of obesity and associated signaling pathways through transfer of microbiota from obese-prone rats. Diabetes 63, 1624–1636. Epstein, A.C., et al., 2001. C. elegans EGL-9 and mammalian homologs define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell 107, 43–54. Ferrara, N., 2002. Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin. Oncol. 29, 10–14. Ferrara, N., et al., 2003. The biology of VEGF and its receptors. Nat. Med. 9, 669–676.
188
M.T. Rishi et al. / Microvascular Research 97 (2015) 181–188
Folkman, J., 1971. Tumor angiogenesis: therapeutic implications. N. Engl. J. Med. 285, 1182–1186. Folkman, J., 1984. Angiogenesis. In: Jaffe, E.A. (Ed.), Biology of Endothelial Cells. MartinusNijhoff, Boston, pp. 412–428. Folkman, J., Klagsbrun, M., 1987. Angiogenic factors. Science 235, 442–447. Golomb, B.A., et al., 2006. Peripheral arterial disease: morbidity and mortality implications. Circulation 114, 688–699. Harari-Steinberg, O., et al., 2013. Identification of human nephron progenitors capable of generation of kidney structures and functional repair of chronic renal disease. EMBO Mol. Med. 5, 1556–1568. Helisch, A., et al., 2006. Impact of mouse strain differences in innate hindlimb collateral vasculature. Arterioscler. Thromb. Vasc. Biol. 26, 520–526. Jones, W.S., et al., 2012. Temporal trends and geographic variation of lower-extremity amputation in patients with peripheral artery disease: results from U.S. Medicare 2000–2008. J. Am. Coll. Cardiol. 60, 2230–2236. Lekas, M., et al., 2006. Growth factor-induced therapeutic neovascularization for ischaemic vascular disease: time for a re-evaluation? Curr. Opin. Cardiol. 21, 376–384. Limbourg, A., et al., 2009. Evaluation of postnatal arteriogenesis and angiogenesis in a mouse model of hind-limb ischemia. Nat. Protoc. 4, 1737–1746. Lunde, K., et al., 2006. Intracoronary injection of mononuclear bone marrow cells in acute myocardial infarction. N. Engl. J. Med. 355, 1199–1209. Messaoudi, S., et al., 2009. Cardiac aldosterone overexpression prevents harmful effects of diabetes in the mouse heart by preserving capillary density. FASEB J. 23, 2176–2185. Myllyharju, J., 2008. Prolyl 4-hydroxylases, key enzymes in the synthesis of collagens and regulation of the response to hypoxia, and their roles as treatment targets. Ann. Med. 40, 402–417. Nagel, S., et al., 2010. Therapeutic manipulation of the HIF hydroxylases. Antioxid. Redox Signal. 12, 481–501. Nor, J.E., et al., 1999. Vascular endothelial growth factor (VEGF)-mediated angiogenesis is associated with enhanced endothelial cell survival and induction of Bcl-2 expression. Am. J. Pathol. 154, 375–384. Norgren, L., et al., 2007. Inter-society consensus for the management of peripheral arterial disease (TASC II). J. Vasc. Surg. 45 (Suppl. S), S5–S67.
Oriowo, B., et al., 2014. Targeted gene deletion of prolyl hydroxylase domain protein 3 triggers angiogenesis and preserves cardiac function by stabilizing hypoxia inducible factor 1 alpha following myocardial infarction. Curr. Pharm. Des. 20, 1305–1310. Ouriel, K., 2001. Peripheral arterial disease. Lancet 358, 1257–1264. Ratajczak, P., et al., 2005. Expression and localization of caveolins during postnatal development in rat heart: implication of thyroid hormone. J. Appl. Physiol. 99, 244–251 (1985). Raval, Z., Losordo, D.W., 2013. Cell therapy of peripheral arterial disease: from experimental findings to clinical trials. Circ. Res. 112, 1288–1302. Schachinger, V., et al., 2006. Intracoronary bone marrow-derived progenitor cells in acute myocardial infarction. N. Engl. J. Med. 355, 1210–1221. Selvaraju, V., et al., 2014. Molecular mechanisms of action and therapeutic uses of pharmacological inhibitors of HIF-prolyl 4-hydroxylases for treatment of ischemic diseases. Antioxid. Redox Signal. 20, 2631–2665. Semenza, G.L., 2003. Targeting HIF-1 for cancer therapy. Nat. Rev. Cancer 3, 721–732. Shizukuda, Y., et al., 1999. Vascular endothelial growth factor-induced endothelial cell migration and proliferation depend on a nitric oxide-mediated decrease in protein kinase Cdelta activity. Circ. Res. 85, 247–256. Takeda, K., et al., 2006. Placental but not heart defects are associated with elevated hypoxia-inducible factor alpha levels in mice lacking prolyl hydroxylase domain protein 2. Mol. Cell. Biol. 26, 8336–8346. Thirunavukkarasu, M., et al., 2013. Simvastatin treatment inhibits hypoxia inducible factor 1-alpha-(HIF-1alpha)-prolyl-4-hydroxylase 3 (PHD-3) and increases angiogenesis after myocardial infarction in streptozotocin-induced diabetic rat. Int. J. Cardiol. 168, 2474–2480. Webber, M.J., et al., 2011. Supramolecular nanostructures that mimic VEGF as a strategy for ischemic tissue repair. Proc. Natl. Acad. Sci. U. S. A. 108, 13438–13443. Wu, S., et al., 2008. Enhancement of angiogenesis through stabilization of hypoxiainducible factor-1 by silencing prolyl hydroxylase domain-2 gene. Mol. Ther. 16, 1227–1234.