Biochemically Engineered Stromal Cell-Derived Factor 1-alpha Analog Increases Perfusion in the Ischemic Hind Limb

Abstract

Background

Despite promising therapeutic innovation over the last decade, peripheral arterial disease remains a prevalent morbidity, as many patients are still challenged with peripheral ischemia. We hypothesized that delivery of engineered stromal cell-derived factor 1-alpha (ESA) in an ischemic hind limb will yield significant improvement in perfusion.

Methods and Results

Male rats underwent right femoral artery ligation, and animals were randomized to receive a 100μL injection of saline (n = 9) or 6μg/kg dosage of equal volume of ESA (n = 12) into the ipsilateral quadriceps muscle. Both groups of animals were also given an intraperitoneal injection of 40μg/kg of granulocyte macrophage colony-stimulating factor (GMCSF). Perfusion was quantified using a laser Doppler imaging device pre-operatively, and on post-operative day 0, 7, and 14. Immunohistochemistry was performed to quantify angiogenesis on day 14, and an mRNA profile was evaluated for angiogenic and inflammatory markers. Compared with the saline/GMCSF group at day 14, the ESA/GMCSF-injected animals had greater re-perfusion ratios (Saline/GMCSF: 0.600 ± 0.140 vs. ESA/GMCSF: 0.900 ± 0.181, group effect: P = .006, time effect: P < .0001, group*time effect: P < .0001), elevated capillary density (10×, Saline/GMCSF: 6.40 ± 2.01 vs. ESA/GMCSF 18.55 ± 5.30, P < .01), and increased mRNA levels of VEGF-A (Saline/GMCSF: n = 6, 0.298 ± 0.205 vs. ESA/GMCSF: n = 8, 0.456 ± 0.139, P = .03).

Conclusion

Delivery of ESA significantly improves perfusion in a rat model of peripheral arterial disease via improved neovasculogenesis, a finding which may prove beneficial in the treatment strategy for this debilitating disease.

Introduction

Peripheral arterial disease (PAD) remains the cause of significant morbidity and mortality in the US, currently affecting over 8 million individuals with many patients ultimately progressing to lower limb ischemia and, in many cases, amputation.¹ In the last decade, advancements in surgical technique and molecular biology have resulted in an important evolution of therapeutic strategies. However, these therapies often overlook the importance of addressing the microvascular pathways needed to restore perfusion to ischemic tissue, and a need to develop novel molecular therapies to promote micro-revascularization remains.

The bone marrow contains vascular progenitor cells that can be mobilized into the peripheral circulation and directed to ischemic tissues to initiate angiogenesis. These endothelial progenitor cells (EPCs) contribute to angiogenesis and maintenance of endothelial cell layers.² The chemokine stromal cell-derived factor 1-alpha (SDF) has been shown to play an important role in cell-homing via interaction with CXCR4, a transmembrane-specific G protein-coupled receptor (GPCR) expressed in numerous cell lines, including EPCs.³ Vascular endothelial growth factor-A (VEGF-A) has also been shown to induce SDF expression in ischemic tissue, and is an integral protein in the SDF/CXCR4 signaling axis by recruiting CXCR4⁺ cells from the bone marrow.⁴

Initial trials with the delivery of SDF to ischemic muscle have demonstrated angiomyogenesis, neovasculogenesis, and mitigated microvascular endothelial cell apoptosis.^{5, 6} However, recombinant SDF is expensive and has a large and complex tertiary structure, limiting its translational potential and its ability to be utilized with various engineered delivery platforms. Previously, our group designed and synthesized engineered stromal cell-derived factor-1alpha (ESA), which when injected into the myocardium has demonstrated the ability to attract EPCs and promote angiogenesis.⁷ In the present study, we hypothesized that delivery of ESA in a rat peripheral arterial disease model would lead to micro-revascularization of ischemic tissue.

Material and Methods

ESA Polypeptide Design

We have previously reported on the design and synthesis of ESA.^{8, 9} In summation, we minimized the profile of the peptide by deleting the central β-pleated sheet (amino acids 18–54) but preserved the CXCR4 receptor binding N-terminus and the molecular stabilizing C-terminus (Fig 1). Using computational modeling, it was determined that replacing the deleted portion with a two-proline residue linker retained the 3-dimensional (3D) protein configuration similar to that of the native SDF-1α and yielded energetic and conformational advantages. The engineered protein was then synthesized using solid-phase peptide synthesis, where the N α-amino acids were incorporated into the peptide in a stepwise fashion while the C-terminal end was attached to a solid support matrix.

Fig 1.

Crystallographic structural representation of SDF-1α and ESA. The N terminal (green), central region (yellow), and C terminal (magenta) are denoted by their respective colors. The central β- sheet region (yellow) in SDF-1α is replaced by a diproline linker in ESA. The corresponding amino acid sequence of each molecule is also depicted and colored according to region. (From Hiesinger W, Perez-Aguilar JM, Atluri P, et al. Computational protein design to reengineer stromal cell-derived factor-1alpha generates an effective and translatable angiogenic polypeptide analog. Circulation 2011; 124:S18–26.)

Animal Model of Peripheral Arterial Disease

All animals were treated in accordance with the Guide for the Care and Use of Laboratory Animals (National Institutes of Health), and the procedures were approved by the University of Pennsylvania Animal Care and Use Committee. Male Wistar rats weighing 250g (n = 21) and 8 to 10 weeks old underwent right-sided femoral artery ligation using a well-established and highly reproducible model of peripheral vascular disease.¹⁰ The rats were anesthetized in a 2L induction chamber (VetEquip, Pleasantville, Calif.), and 3% isofluorane was continuously delivered. A 16-gauge angiocatheter was used for endotracheal intubation and the rats subsequently were connected to mechanical ventilation (Hallowell; EMC, Pittsfield, Mass.), with 1% isofluorane maintained throughout the operation. A 1 cm incision was made along the right inguinal ligament, allowing for easy access to the iliac and common femoral artery. After ligation, either 6μg/kg of ESA (n = 12) diluted in 100μl of saline or 100μl of saline (n = 9) was injected in the ipsilateral quadriceps muscle. Additionally, 40μg/kg of granulocyte macrophage colony-stimulating factor (GMCSF) was injected intraperitoneally in both groups. Previously, GMCSF was shown to play a critical role in inducing global bone marrow-derived EPC upregulation, but no standalone therapeutic benefit was found without a local chemotactic agent.¹¹ Successful ligations were confirmed by right-foot coloration and laser Doppler imaging.

Laser Doppler Perfusion Images

Rats were anesthetized in a similar manner as presented in the surgical animal model. Each limb was shaved prior to imaging to minimize artifact. Limb perfusion was analyzed using a Moor high-resolution laser Doppler imaging device (LDI2-HR) pre-operatively and post-operatively on days 0, 7, and 14. To quantify perfusion, the LD12-HR perfusion data were expressed as the ratio of the ischemic (right) to normal (left) limb blood flow.

Immunohistochemistry

After 14 days, right-sided quadriceps were explanted, submerged in Tissue Tek OCT compound, and stored in a −80°C freezer. 10μm-thick sections were prepared from each sample (Leica 3050 S). Cryosections of muscle were fixed in acetone at −20 °C for 20 minutes. Sections were blocked with 10% fetal bovine serum (Sigma Aldridge) for 60 minutes at room temperature. After blocking, the sections were incubated with rabbit polyclonal antibody to alpha smooth muscle actin (SMA) (Abcam, 1:50) and sheep polyclonal antibody to von Willebrand Factor (vWF) tagged with Fluorescein isothiocyanate (FITC) (Abcam, 1:100). Bound primary antibodies were detected with Alexa Fluor 555 donkey anti-rabbit IgB (Life Technologies, 1:50). Lastly, sections were stained with 4′,6-diamidino-2-phenylindole (DAPI) (Vector Laboratories). Sections were then visualized using a Leica DFC295 inverted microscope (Leica Microsystems, Wetzlar, Germany) using the 10× objective. Vessel density was quantified by the number of capillaries per high-powered field.

In Vivo Vascular Endothelial Growth Factor-A (VEGF-A) and Inflammatory mRNA Concentration Assay

RNA was isolated from frozen muscle (quadriceps and gastrocnemius) embedded in OCT solution using QuantiGene Sample Processing kit (Affymetrix) according to the manufacturer’s instructions. Briefly, each 5mg tissue was added to 200μl of homogenizing solution with 4μl of proteinase K and was then incubated at 65°C for 1–2 hours while vortexing every 10 minutes. After centrifugation, the supernatant was analyzed immediately for VEGF-A and inflammatory markers (IL-12B, IL-2, IL-1Beta, IL-10, IFN-gamma, IL-1alpha, IL-6) using QuantiGene 2.0 Plex Assay kit (Affymetrix) according to the manufacturer’s protocol. Briefly, 40μl of tissue supernatant was added to each well of the hybridization plate containing 60μl of working bead mix and was incubated for 18–22 hours at 54°C at 600 rpm. 40μl of homogenizing solution was used for assay background. The samples were transferred to the magnetic separation plate followed with pre-amplifier hybridization, amplifier hybridization, probe labeling, and SAPE binding with three washes after each step. Lastly, the plate was analyzed in the Bio-Plex Luminex 200 system (BIO-RAD) and gene expression was calculated using Bio-Plex Manager 5.0 software. Specifically, values were quantified using a fluorescence ratio normalized to Glyceraldehyde 3-phosphate dehydrogenase (GAPDH).

Statistical Analysis

Results are expressed as mean ± standard deviation. Statistically significant differences between groups were compared using the 2-tailed Student t-test. Significance was set at a P value of less than 0.05. For the comparison of hind limb perfusion between the ESA treated and control group across multiple time points, a univariate repeated measures analysis of variance (ANOVA) was performed. Adjusted p-values (Greenhouse-Geisser) are reported for the repeated measures ANOVA analysis to account for slight violation of compound symmetry. Statistical analyses were performed with SAS version 9.4 (Cary, NC).

Results

Capillary Density in the Ischemic Hind Limb

Capillary density in explanted tissue was stained for von Willebrand glycoprotein, a marker for endothelial cells, and alpha smooth muscle actin. The number of capillaries per high-powered field was significantly greater in the ESA-treated hind limbs (18.55 ± 5.30) compared to that of the saline-injected hind limbs (6.40 ± 2.01, P < .01) (Fig 2A), and the morphology of the capillaries was more defined and extensive in the ESA/GMCSF group relative to the saline/GMCSF group (Fig 2B).

Fig 2.

Capillary density stratified by treatment group. A, Ligated hind limb samples were cryosectioned and stained for von Willebrand Factor (vWF), smooth muscle actin (SMA), and DAPI. The ESA/GMCSF group exhibited a significantly higher mean capillary density (n = 12, 18.55 ± 5.30) than that of the saline/GMCSF group (n = 9, 6.40 ± 2.01, *P < .01). Error bars denote SE. B, Representative fluorescent microscopy images of quadriceps sections at 10x magnification. Bar = 100μm.

Increased levels of mRNA expression of VEGF-A in ESA Treated Hind Limbs

The cytokine VEGF-A has been reported to play an important role in the mobilization of EPCs into sites of neovascularization.¹² There were increased levels of mRNA expression of VEGF-A in 14-day muscle sections that were treated with ESA (Quad: n = 8, 0.456 ± 0.139, Calf: n = 8, 0.473 ± 0.106) in comparison to the saline/GMCSF group (Quad: n = 6, 0.298 ± 0.205, Calf: n = 6, 0.285 ± 0.136, P = .03, P = .04). Levels are expressed in VEGF-mRNA fluorescence ratio normalized to GAPDH (Fig 3).

Fig 3.

Graph showing VEGFA-mRNA fluorescence ratio after normalizing to GAPDH. Flourescence was measured in both quadricep and calf samples. ESA/GMCSF-treated group showed significantly higher fluorescence ratios in both the quad and calf (n = 8, Quad: 0.456 ± 0.139, Calf: 0.473 ± 0.106) relative to the saline/GMCSF group (n = 6, Quad: 0.298 ± 0.205, Calf: 0.285 ± 0.136, *P = .03, **P = .04). Error bars denote SE.

Inflammatory Markers are not Statistically Significant between ESA and Saline Groups

At 14 days, mRNA concentration assay showed similar levels of interleukins and IFN-gamma levels. Statistical comparison demonstrated no statistically significant difference between ESA/GMCSF (n = 8, IL-12B: 0.0021 ± 0.00067, IL-2: 0.0014 ± 0.00021, IL-1Beta: 0.0018 ± 0.00042, IL-10: 0.0020 ± 0.00064, IFN-gamma: 0.0018 ± 0.00031, IL-1alpha: 0.0014 ± 0.00036, IL-6: 0.0024 ± 0.00061, IL-4: 0.00084 ± 0.00014) and saline/GMCSF groups (n = 6, IL-12B: 0.0020 ± 0.00058, P = .41, IL-2: 0.0014 ± 0.00050, P = .39, IL-1Beta: 0.0021 ± 0.00065, P = .21, IL-10: 0.0020 ± 0.00081, P = .40, IFN-gamma: 0.0015 ± 0.00062, P = .16, IL-1alpha: 0.0020 ± 0.0010, P = .074, IL-6: 0.0019 ± 0.00047, P = .085, IL-4: 0.0011 ± 0.00038, P = .025) (Fig 4).

Fig 4.

mRNA levels of pro-inflammatory markers in quadriceps of ischemic limbs in both treatment groups. There was minimal difference between ESA/GMCSF (n = 8, IL-12B: 0.0021 ± 0.00067, IL-2: 0.0014 ± 0.00021, IL-1Beta: 0.0018 ± 0.00042, IL-10: 0.0020 ± 0.00064, IFN-gamma: 0.0018 ± 0.00031, IL-1alpha: 0.0014 ± 0.00036, IL-6: 0.0024 ± 0.00061, IL-4: 0.00084 ± 0.00014) versus saline/GMCSF groups (n = 6, IL-12B: 0.0020 ± 0.00058, P = .41, IL-2: 0.0014 ± 0.00050, P = .39, IL-1Beta: 0.0021 ± 0.00065, P = .21, IL-10: 0.0020 ± 0.00081, P = .40, IFN-gamma: 0.0015 ± 0.00062, P = .16, IL-1alpha: 0.0020 ± 0.0010, P = .074, IL-6: 0.0019 ± 0.00047, P = .085, IL-4: 0.0011 ± 0.00038, P = .025). Error bars denote SE.

ESA Promotes Perfusion in the Ischemic Hind Limb

Reperfusion of the ischemic hind limb was quantified using laser Doppler perfusion assessment and perfusion data was expressed as the ratio of the ischemic (right) to normal (left) limb blood flow. The rats treated with an injection of ESA into an ischemic hind limb had an enhanced perfusion ratio on day 14 (n = 12, 0.900 ± 0.181) relative to that of the saline/GMCSF-treated group (n = 9, 0.600 ± 0.140, group effect: P = .006, time effect: P < .0001, group*time effect: P < .0001) (Fig 5A and B).

Fig 5.

A, Representative laser Doppler images of the preoperative hind limb and again at the study endpoint for each group. B, Graph depicting the ratio of perfusion in the ischemic hind limb relative to the non-ligated hind limb pre-operatively and at three time points after induced ischemia. The ESA/GMCSF group (n = 12, 0.900 ± 0.181) showed marked perfusion augmentation by Day 14 relative to that of the saline/GMCSF group (n = 9, 0.600 ± 0.140, group effect: P = .006, time effect: P < .0001, group*time effect: P < .0001). Error bars denote SE.

Discussion

The use of cytokines and growth factors to promote revascularization is an important and well-studied therapeutic strategy. Previous studies have confirmed the efficacy of SDF in performing this role in both cardiac and peripheral vascular disease models.^11,13,14 However, there are limitations associated with using recombinant SDF, including its bulkiness and high cost. Our study is novel in its use of a bioengineered analog of SDF to treat hind limb ischemia. This study demonstrates that injection of ESA into an ischemic hind limb is an effective therapy that increases capillary density and enhances perfusion. Herein, we describe an extremely clinically translatable peptide therapy utilizing a smaller and cost-efficient protein analog resulting in effective micro-revascularization of ischemic muscle.

Immunohistochemistry staining for von Willebrand Factor and smooth muscle actin provides mechanistic support for the finding of ESA-induced perfusion of the ischemic limb. Capillary density was significantly higher in tissues treated with ESA in comparison to the saline group, and capillary morphology was more defined and extensive in the ESA/GMCSF group. Interestingly, many of the ESA/GMCSF sections showed evidence of capillary formation without surrounding smooth muscle. This finding may demonstrate the very early stages of de novo capillary formation. Extension of the experiment to greater time points is needed to confirm this hypothesis to determine whether the capillaries have initiated the maturation process by developing a smooth muscle boundary.

Previous studies have demonstrated that the SDF/CXCR4 and VEGF-A chemokine ligand-receptor axis plays an important role in tissue regeneration and angiogenesis.^{4, 15} Specifically, VEGF has been shown to mobilize CXCR4⁺ cells from the bone morrow and to also induce secretion of SDF from ischemic tissue.^{4, 16, 17} Moreover, Dang and colleagues have shown that loss of the VEGF ligand, Vegfa, in the endothelium of adult vasculature causes significant endothelial abnormalities.¹⁸ Interestingly, we found that mRNA VEGF-A levels in the ischemic hind limb at day 14 were significantly elevated in the ESA/GMCSF group compared to that of the saline/GMCSF group. We suspect that improvements in perfusion and capillary density after injection of ESA can be attributed to the interplay of SDF, VEGF, and CXCR4 signaling pathways. Also important to note is that there is no statistical significance between mRNA levels of inflammatory markers between the ESA/GMCSF and control group. This finding indicates that improvements in perfusion and increased capillary density may arise from the specific activation of VEGF, SDF, and CXCR4 rather than a generalized ESA-induced inflammatory response.

The in vivo results from the laser Doppler imaging are particularly interesting, since it represents the most definitive evidence of reperfusion following ischemia. It is important to note that the rat offers a challenging surgical model for inducing sustainable PAD because of the extensive collateral vessel networks branching from the iliac and common femoral artery.¹⁹ Previous groups have shown the control limb to undergo a moderate amount of reperfusion in a rat model of hind limb ischemia.²⁰ In this two week study of hind limb ischemia, there was a more sustained and accelerated improvement of blood flow in the ischemic limb for the ESA treated group when compared to animals in the saline group (Fig 5B). This data offers promising evidence that significant changes are occurring at the microvascular level that lead to reperfusion.

Although we have demonstrated a statistically significant increase in capillary density and perfusion in ESA-treated hind limbs, this study has some limitations. Ideally, we would expand the study to include an ESA alone group that would not receive an intraperitoneal injection of GMCSF at the time of ligation to examine the effects of the chemokine as a standalone therapy. In addition, further studies are needed to evaluate the role of SDF and VEGF in recruiting endothelial progenitor cells to the site of ischemia. Staining for endothelial progenitor cell markers such as CD34, CD31, and CXCR4 in ischemic quadriceps would allow for examination of the degree of mobilization and recruitment of these cells in response to ESA therapy. Finally, it is recognized that this model does not truly resemble peripheral arterial disease in that acute ischemia occurs; however, it remains an appropriate setting to evaluate the effectiveness of an angiogenic therapy.

In conclusion, ESA administered to ischemic hind limbs resulted in greater VEGF levels, capillary density, and greater reperfusion than rats in the saline control group. These promising results warrant further exploration of ESA therapy for PAD to identify the most ideal delivery method.

Clinical Relevance.

The use of cytokines and growth factors to promote revascularization is a well-studied therapeutic strategy. Previous studies have confirmed the efficacy of SDF in revascularizing both cardiac and peripheral vascular disease models.^11,13,14 However, there are limitations associated with using recombinant SDF, including its bulkiness and high cost. This study demonstrates that injection of a bioengineered analog of SDF, ESA, into an ischemic hind limb is an effective therapy that enhances perfusion. Herein, we describe an extremely clinically translatable peptide therapy utilizing a smaller and cost-efficient protein analog resulting in effective microrevascularization of ischemic muscle.