Acellular dermal matrix-based gene therapy augments graft incorporation

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

Available online at www.sciencedirect.com

ScienceDirect journal homepage: www.JournalofSurgicalResearch.com

Acellular dermal matrix-based gene therapy augments graft incorporation Meredith T. Vandegrift, MD, Caroline Szpalski, MD, Denis Knobel, MD, Andrew Weinstein, MD, Maria Ham, MD, Obinna Ezeamuzie, MD, Stephen M. Warren, MD, FACS, and Pierre B. Saadeh, MD, FACS* The Institute of Reconstructive Plastic Surgery, New York University Langone Medical Center, New York, New York

article info

abstract

Article history:

Background: Acellular dermal matrix (ADM) is widely used for structural or dermal

Received 14 April 2014

replacement purposes. Given its innate biocompatibility and its potential to vascularize,

Received in revised form

we explored the possibility of ADM to function as a small interfering RNA (siRNA) delivery

30 October 2014

system. Specifically, we sought to improve ADM vascularization by siRNA-mediated inhi-

Accepted 7 January 2015

bition of prolyl hydroxylase domain-2 (PHD2), a cytoplasmic protein that regulates hypoxia

Available online 13 January 2015

inducible factor-1a, and improve neovascularization. Materials and methods: Fluorescently labeled siRNA was used to rehydrate thin implantable

Keywords:

ADM. Pharmacokinetic release of siRNA was determined. Twelve millimeter sections of

siRNA

ADM reconstituted with PHD2 siRNA (nonsense siRNA as control) and applied to dorsal

Acellular dermal matrix

wounds of 40 FVB mice. Grafts were sewn in, bolstered, and covered with occlusive

Alloderm

dressings. Photographs were taken at 0, 7, and 14 d. Wounds were harvested at 7 and 14 d

Wound healing

and analyzed (messenger RNA, protein, histology, and immunohistochemistry).

Graft incorporation

Results: Release kinetics was first-order with 80% release by 12 h. By day 14, PHD2-containing

Graft vascularity

ADM appeared viable and adherent, whereas controls appeared nonviable and nonadherent.

PHD2

Real-time reverse transcription-polymerase chain reaction demonstrated near-complete knockdown of PHD2, whereas vascular endothelial growth factor and FGF-2 were increased 2.3- and 4.7-fold. On enzyme-linked immunosorbent assay, vascular endothelial growth factor was increased more than fourfold and stromal cell-derived factor doubled. Histology demonstrated improved graft incorporation in treated groups. Immunohistochemical demonstrated increased vascularity measured by CD31 staining and increased new cell proliferation by denser proliferating cell nuclear antigen staining in treated versus controls. Conclusions: We concluded that ADM is an effective matrix for local delivery of siRNA. Strategies to improve the matrix and/or genetically alter the local tissue environment can be envisioned. ª 2015 Elsevier Inc. All rights reserved.

1.

Introduction

Acellular dermal matrix (ADM) is commonly used for breast, abdominal, and head and neck reconstruction [1e7]. A

commonly cited advantage of ADM and other biologic dermal substrates is incorporation into the host tissue, which ultimately confers host tissue-like properties thought to result from vascularization and subsequent recellularization by

All authors approve of this final article. * Corresponding author. The Institute of Reconstructive Plastic Surgery, New York University Langone Medical Center, 305 E33rd Street, New York, NY 10016. Tel.: þ1 212 263 8452; fax: þ1 212 263 8492. E-mail addresses: [email protected], [email protected] (P.B. Saadeh). 0022-4804/$ e see front matter ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jss.2015.01.003

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

361

section of ADM was placed in 40 mL of a solution containing 20-pmol fluorescently labeled siRNA (Qiagen, Valencia, CA, Alexa Fluor 488). Once reconstituted, the samples were formalin fixed, paraffin embedded, and sectioned. Slides were viewed under an epifluorescent microscope and digitized; Sigma Scan software (Systat Software, Inc., San Jose, CA) was used to quantify the relative intensity of fluorescence within the matrix of normal saline-reconstituted ADM versus siRNA reconstituted samples to determine the presence of autofluorescence.

local tissues [8]. Moreover, rapid ADM vascularization is critical when ADM is used as a dermal replacement to support split-thickness skin grafts. Additionally, although ADM has become widely used in implant-based breast reconstruction in single and two-stage procedures, there are increasing reports of higher seroma, infection, and explantation rates with the use of ADM [9,10]. Interestingly, radiation, which plays a central role in the treatment of breast cancer and which yields characteristic diminishment in tissue vascularity, has been identified as an independent predictor of adverse outcome in the setting of ADM associated implant-based breast reconstruction [11]. Improving ADM vascularization may improve the performance characteristics of ADM and may allow for new and higher volume applications of ADM. In all these circumstances, attempts to vascularize this pocket may prove beneficial in decreasing complications and improving take in complex wound beds. Hypoxia inducible factor (HIF)-1a is the master switch that regulates new blood vessel formation, or angiogenesis, in response to hypoxic conditions [12,13]. In fact, the stabilization of HIF-1a is critical to wound healing [14]. Under homeostatic conditions, that is unwounded skin, prolyl hydroxylase domain (PHD)2 protein binds to HIF-1a and causes its degradation by the von Hippel Lindau protein [15,16], thereby turning off this switch. Several investigators have blocked PHD-2 protein expression, creating constitutive HIF1a expression [15e17]. If one applied this technique to an area with low angiogenic potential, theoretically, angiogenesis could be augmented. The discovery of silent interfering RNA has spawned a wealth of studies ranging from improvement in delivery systems for cellular uptake, the provision of new therapeutics to combat human diseases, the design of anticancer agents, and to applications in agriculture to advance plant science [18]. Studies have shown that small interfering RNAs (siRNAs) can be used to suppress the transcription of specific gene sequences, thereby inhibiting the production of a specific protein product [17]. Moreover, silencing is transient, eliminating the potential for uncontrolled cell proliferation [19]. Subsequently, a growing need has emerged to create a delivery system that can specifically target the subcutaneous tissue for local gene modulation and therapy. Because ADM is already widely used for structural or dermal replacement purposes owing to its innate biocompatibility and potential to vascularize, we explored the possibility of using ADM as a siRNA delivery system. We therefore hypothesized that this mode of gene therapy delivery could modify the wound bed to improve neovascularization. Second, if treated ADM could create new vessels in a wound, it might also increase the vascular cell ingrowth into the matrix itself.

After approval from The New York University Medical Center Animal Care Committee (IACUC #061104-01), 40 adult male FVB mice were obtained from Jackson Laboratories (Bar Harbor, ME) and housed in our animal care facility in accordance with the Division of Laboratory Animal Resources bylaws. Investigative experimental arms included ADM reconstituted with PHD-2 siRNA (n ¼ 20) or reconstituted with nonsense siRNA (n ¼ 20). Ten animals from each group were sacrificed for the various data analyses on days 7 and 14. A novel model of ADM grafting onto the dorsal surface of the mouse was used. Briefly, a singular wound (12-mm) was created on the dorsum of the mouse. Thin graftable ADM was reconstituted in nonsense siRNA (control) or solution containing siPHD2 (treated) as per manufacturer’s protocol [20]. Specifically, 20 pmol of PHD2 siRNA was incorporated into the treated ADM. The ADM was then sewn into the wound bed with interrupted 5-0 nylon sutures, bolstered into place with a petroleum gauze and dry sterile dressing, and covered with an occlusive bandage (3M, St Paul, MN). At 7 and 14 d, the dressing was removed, and the wounds were photographed, assessed, and harvested for further evaluation as described in the following.

2.

Methods

2.4. Messenger RNA and protein quantification of ADM-grafted wound

2.1.

Reconstitution of ADM with siRNA

Preliminary studies to assess the adequacy of siRNA uptake and delivery by ADM were performed. In a petri dish, a 12-mm

2.2.

Measuring siRNA release from ADM

Release kinetics was determined using a 96-well plate with a semipermeable membrane. ADM reconstituted with fluorescently labeled siRNA in saline was placed on the membrane with diffusion of fluorescent compound into the lower well containing saline. Concentrations of the treated wells were measured at various time points (hourly for 1e6 h, then at 12 and 24 h) and compared against standard dilutions by a spectrophotometer at 490 nm. Values were divided by maximum concentration of 2.5 pmol/uL (500 pmol of fluorescent compound per 200 mL of saline) to calculate percent of elution and then graphed against time of elution to determine in vitro kinetics of siRNA release from ADM.

2.3. Application of siRNA-containing ADM to cutaneous wounds

Wounds were harvested on days 7 and 14 after grafting. Real-time reverse transcription-polymerase chain reaction (RT-PCR) and enzyme-linked immunosorbent assay (ELISA) of the wound homogenate were performed. Specific genes

362

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

Fig. 1 e siRNA distribution within ADM. Sections of ADM were reconstituted with normal saline (control) versus saline solution containing 20 pmol of fluorescently labeled siRNA. Saline-reconstituted ADM (A) had some evidence of autofluorescence. However, there was increased intensity of fluorescence noted in the ADM reconstituted in fluorescently labeled siRNA with homogenous distribution of the siRNA throughout (B). (Color version of figure is available online.)

investigated in this study focused on those relevant to angiogenesis in a wounded bed, particularly PHD-2 and HIF1a. Additional angiogenic markers studied included vascular endothelial growth factor (VEGF)-A and fibroblast growth factor (FGF-2) as well as CD31 levels as a measurement of vasculogenesis. PCR primers were custom designed to PHD-2 (CGTCACGTTGATAACCCAAA), HIF-1a (TGATGACCAGCAACT TGAGG), VEGF-A (TAACGATGAAGCCCTGGAGT), FGF-2 (CAAG GGAGTGTGTGCCAAC), and normalized to 18S (GTAACC CGTTGAACCCCATT). Gene expression of PHD-2, HIF-1a, VEGF-A, and FGF-2 was examined using messenger RNA (mRNA) quantification via RT-PCR and expressed as fold change of treated ADM-grafted wound beds relative to nonsense-treated controls. Gene expression of VEGF-A was also performed using protein quantification via ELISA to further confirm effects of PHD-2 siRNA on vasculogenesis.

2.5.

Immunohistochemistry of ADM-grafted wound

Wounds were harvested at days 7 and 14 after grafting. The histopathologic sections were transferred to 10% neutral-

buffered formalin for 24 h and then transferred to 70% ethanol in preparation for paraffin embedding. Formalinfixed, paraffin-embedded tissue samples were cut, sectioned, and stained with purified antiemouse CD31 primary antibody in a 1:100 dilution (Abcam, Cambridge, MA) and PNCA in a 1:200 dilution (Abcam), and viewed on an Olympus BX51 (Olympus America Inc., Center Valley, PA) microscope. The slides were then digitized, and Sigma Scan software was used to segment and quantify the number and cumulative cross-sectional area of CD31 and Proliferating cell nuclear antigen (PCNA) þ cells across five nonconsecutive tissue sections for each wound.

2.6.

Statistical analysis

All data are expressed as mean  standard deviation. To compare mRNA levels of PHD2, HIF-1a, VEGF, FGF2, and CD31 as a fold change relative to the control, a one-sample t-test was used. To compare VEGF protein levels, a two-sample independent measures t-test was used. For all tests, a P value <0.05 was considered statistically significant.

Fig. 2 e Release kinetics of siRNA delivery from ADM. ADM reconstituted in fluorescently labeled siRNA was placed in a 96well plate with a semipermeable membrane, and concentrations of siRNA in the wells were measured at various time points. Percent of elution over time was calculated to observe siRNA release from ADM, which appeared to follow first-order kinetics with a peak of 80% of siRNA delivery at 6 h, plateau after 12 h, and essentially no change at 24 h.

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

363

Fig. 3 e Gross evaluation of ADM in situ. Thin graftable ADM was reconstituted with PHD-2 siRNA (treated) versus only saline (control) then sutured on the wounded dorsum of FVB mice and bolstered in place. Photographs taken at day 7 demonstrated clinically softer and more adherent ADM with better incorporation into the surrounding tissue in treated grafts (A) as compared with that of controls (B). (Color version of figure is available online.)

3.

Results

3.1.

Analysis of ADM in vitro

80% delivery after 6 h plateau after 12 h, and essentially no change at 24 h (Fig. 2).

3.2. ADM reconstituted in normal saline demonstrated some evidence of autofluorescence (Fig. 1A). More significantly, ADM reconstituted in saline solution containing fluorescently labeled siRNA revealed increased intensity with homogenous distribution of the siRNA throughout the matrix (Fig. 1B), suggesting adequate and uniform incorporation of siRNA into ADM as a delivery vehicle. Relative intensity of fluorescence was 4.2  0.38 in the treated sample versus 2.03  0.27 in the control sample, P < 0.001. Elution experiments demonstrated first-order kinetics with

Analysis of ADM in vivo

We developed an adherence scale to quantitatively evaluate the gross properties of the grafts. Wounds were evaluated at days 7 and 14, and points were assigned to measure graft “take” as adherence to the wound bed as follows: 0 ¼ nonadherent, 1 ¼ adherent requiring elevation, and 2 ¼ adherent requiring excision. Gross analysis demonstrated that ADM treated with PHD2 siRNA was softer, moister, and more adherent with better incorporation into the surrounding tissue (Fig. 3A), with more pronounced

Fig. 4 e mRNA analysis of PHD2 expression in the wound bed of control versus treated ADM at days 7 and 14. RT-PCR analysis of PHD2 expression demonstrates a significant reduction in the wound beds at both days 7 and 14. Asterisk denotes statistical significance. (Color version of figure is available online.)

364

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

Fig. 5 e mRNA analysis of vasculogenic markers in the wound bed of control versus treated ADM at days 7 and 14. RT-PCR analysis of HIF-1a, VEGF, and FGF-2 expression in the wound beds demonstrate that PHD2 silencing augments vasculogenic marker expression most significantly at day 7 but also at day 14. CD31, an endothelial cell marker in blood vessels, demonstrated increased expression at both time points in treated wound beds relative to controls. Asterisk denotes statistical significance. (Color version of figure is available online.)

adherence at day 7; however, the median adherence score for siRNA-treated wounds (1.0; interquartile range [IQR] [0e1.25]) was not significantly different from that of controls (0.5; IQR [0e1.0]). ADM with nonsense siRNA only was clinically more desiccated and easily removed (Fig. 3B), but the median adherence score for siRNA (1.5; IQR [1.0e2.0]) was not statistically different from that of the control (1.0; IQR [0.75e2]).

3.3. mRNA and protein quantification of ADM-grafted wound At day 7, the mRNA level of siPHD2-treated animals (0.28  0.067) was 3.6-fold lower relative to that of nonsense siRNA, P ¼ 0.041. At day 14, the mRNA level of siPHD2-treated

animals (0.069  0.015) was 14.5-fold lower relative to that of nonsense siRNA, P ¼ 0.007. The mRNA level of siPHD2treated animals at day 7 and day 14 was not significantly different from each other (Fig. 4). The treated ADM-grafted wounds had a 1.87  0.28 fold increase in HIF-1 a expression, P < 0.001. FGF-2 expression was increased by a fold change of 3.788  0.96, P < 0.001, whereas CD31 expression was increased by a fold change of 2.01  0.04, P < 0.001 (Fig. 5). Wounds harvested at day 14 demonstrated near-complete PHD2 suppression, with fold change of 0.07  0.05. In contrast, wound beds from nonsense siRNA-reconstituted ADM showed a 5.76  0.52 fold increase (P < 0.05). Although HIF-1a was only mildly still increased over controls, VEGF expression was remarkably elevated in the siPHD2-treated wound bed samples with a 2.03  0.64 fold induction. FGF-2 mRNA expression

Fig. 6 e Protein analysis of VEGF in the wound beds at days 7 and 14. ELISA evaluated translation of VEGF protein in the wound bed. Expression is increased in the treated group versus controls at both time points. Asterisk denotes statistical significance. (Color version of figure is available online.)

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

365

Fig. 7 e Histologic analysis of ADM in situ and the wound beds treated with PHD-2 siRNA versus control groups from day 7 (day 14 images not shown here). Hematoxylin and eosin staining with improvement in cellularity of the treated matrix and

366

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

demonstrated significantly remarkable elevations in the PHD2 siRNA-treated samples compared with the controls, at 1.87 þ 1.06 and 0.10  0.01, respectively. The treated wound had a 1.07  0.06 fold induction, whereas the CD31 mRNA in the control wound remained low at 0.07  0.01 (P < 0.03). ELISA was performed to evaluate protein expression of VEGF in siPHD2-treated grafts versus controls. At day 7, the VEGF protein level of siRNA-treated animals (220.75  0.52) was significantly greater than that of nonsense siRNA (168.09  8.56), P ¼ 0.001. At day 14, the VEGF protein level of siPHD2-treated animals (220.75  0.52) was significantly greater than that of nonsense siRNA (129.  8.56), P ¼ 0.001; Fig. 6).

3.4. Evaluation of cellularity of the ADM and wound bed, PHD-2 knockdown, and matrix neovascularization The following immunohistochemical staining results are shown from day 7 only; stains from day 14 not shown here. Hematoxylin and eosin staining of the ADM revealed an improvement in cellularity of the PHD-2 siRNA-treated matrix and confirmation of better incorporation into the wound bed (Fig. 7A) as compared with the control (Fig. 7B). PCNA staining of the wound bed demonstrated greater DNA replication and cellular proliferation of the wound bed treated with ADM-containing PHD-2 siRNA as compared with the control (Fig. 7C and D). Evaluation with PHD-2 staining showed knockdown in the treated wound beds versus controls (Fig. 7E and F). CD31 staining exhibited denser CD31 profiles and consequently increased neovascularization of the ADM wound beds treated with PHD-2 siRNA in contrast to ADM reconstituted in nonsense siRNA (Fig. 7G and H).

4.

Discussion

Our experiments demonstrated that ADM is an effective vehicle for siRNA delivery. Uptake was homogenous throughout the matrix as seen on immunofluorescence, and release kinetics demonstrated good delivery (over 80%). Grossly, the grafts were more viable as demonstrated by softer texture and stronger adherence to the wound bed in dressing removal as well as during wound bed harvesting. Clinically, we felt that there was a difference in extraction of the grafts; unfortunately our adherence scores did not reach statistical significance with this small data set. Perhaps a larger sample size would achieve significance. RT-PCR demonstrated almost complete knockdown of PHD2 in the wound bed with consequent increases in downstream angiogenic factors from the HIF-1a pathway such as VEGF and FGF-2. This finding is consistent with earlier work performed in this laboratory, which demonstrated that PHD2 silencing led to improved diabetic wound healing, presumably

through a vasculogenic mechanism via HIF-1a stabilization (unpublished data). Additionally, Kelly et al. [21] showed that increasing or stabilizing the expression of HIF-1a proved sufficient to induce angiogenesis, even in normoxia. This evidence supports selectively targeting PHD2 with siRNA to promote vascular ingrowth of an ADM. After HIF-1a stabilization, upregulation of downstream vasculogenic effectors was observed, as measured by RT-PCR and ELISA. This upregulation could in part explain the observed denser CD31 staining, correlating with increased new blood vessel growth into the treated AlloDerm compared with control. Additionally, PCNA demonstrated increased cellular proliferation overall in the treated ADM. This correlated with improved quality of the graft as well as graft take, presumably due to increased neovascularization. Other investigators have observed the ability for ADM to vascularize. Menon et al. [22] demonstrated that ADM implanted into the rabbit abdominal wall served as a durable substitute to mesh, in part due to its ability to vascularize. Likewise, in a similar study looking at ventral hernia repair in the rabbit in an infected wound bed, the ability of the dermal matrix to do well despite the presence of Staphylococcus aureus was attributed to ADM’s capacity to vascularize and subsequently clear the bacteria [23]. The potential advantages of a vascularized matrix could fill a great void in the prevention of infection, seroma formation, and explantation, especially where the wound bed is largely avascular. This study demonstrates a novel use of ADM: a cutaneous delivery system for siRNA to modulate local gene expression. Additionally, we showed that treatment with PHD-2 improved graft vascularity. When implanted into a murine wound bed, we demonstrated the ability of ADM to promote durable cutaneous protein suppression of PHD2, with secondary upregulation of proangiogenic factors to augment neovascularization of the dermal matrix. Future directions in ADM-based gene therapy may include examining the role of ADM coverage for chronic hypoxic diabetic wounds. siRNA delivery via this system could not only provide the needed platform for transcutaneous gene modulation of the dysfunctional wound bed but also modulate the qualities of the acellular scaffold to maximize therapeutic intervention while minimizing complications.

Acknowledgments Authors’ contributions: M.T.V., D.K., S.M.W., and P.B.S. contributed to the conception and design. M.T.V., C.S., D.K., A.W., M.H., and O.E. did the data collection. M.T.V., C.S., D.K., A.W., M.H., O.E., S.M.W., and P.B.S. did the writing of the article. M.T.V., C.S., S.M.W., and P.B.S. did the critical

= confirmation of better incorporation into wound bed (A) versus control (B). PCNA staining with greater DNA replication and increased cellular proliferation of wound bed with treated matrix (C) versus control (D). PHD-2 staining with significant knockdown and suppression of PHD-2 expression in treated wound bed (E) versus control (F). CD31 staining with denser CD31 profiles and increased neovascularization in treated wound bed (G) versus control (H). (Color version of figure is available online.)

j o u r n a l o f s u r g i c a l r e s e a r c h 1 9 5 ( 2 0 1 5 ) 3 6 0 e3 6 7

revisions. D.K., A.W., M.H., and O.E. did the analysis and interpretation. P.B.S. did the data analysis.

Disclosure The authors certify that, to the best of their knowledge, no financial support or benefits have been received by any co-author, by any member of their immediate family, or any individual or entity with whom or with, which we have a significant relationship from any commercial source, which is related directly or indirectly to the work, and which is reported in the article.

references

[1] Leon-Villapalos J, Eldardiri M, Dziewulski P. The use of human deceased donor skin allograft in burn care. Cell Tissue Bank 2010;11:99. [2] Helgeson MD, Potter BK, Evans KN, Shawen SB. Bioartificial dermal substitute: a preliminary report on its use for the management of complex combat-related soft tissue wounds. J Orthop Trauma 2007;21:394. [3] Langer A, Rogowski W. Systematic review of economic evaluations of human cell-derived wound care products for the treatment of venous leg and diabetic foot ulcers. BMC Health Serv Res 2009;9:115. [4] Maurice SM, Skeete DA. Use of human acellular dermal matrix for abdominal wall reconstructions. J Surg 2009;197:35. [5] Breuing KM, Warren SM. Immediate bilateral breast reconstruction implants and Inferolateral AlloDerm slings. Ann Plast Surg 2005;55:232. [6] Jung SN, Chung JW, Yim YM, Kwon H. One stage skin grafting of the exposed skull with acellular human dermis (AlloDerm). J Craniofac Surg 2008;19:1660. [7] Pushpoth S, Tambe K, Sandramouli S. The use of AlloDerm in the reconstruction of full thickness eyelid defects. Orbit 2008; 27:337. [8] Wong AK, Schonmeyer BH, Singh P, Carlson DL, Li S, Mehrara BJ. Histologic analysis of angiogenesis and lymphangiogenesis in acellular human dermis. Plast Reconstr Surg 2008;121:1144. [9] Chun YS, Verma K, Rosen H, et al. Implant-based breast reconstruction using acellular dermal matrix and the risk of postoperative complications. Plast Reconstr Surg 2010;125:429.

367

[10] Antony AK, McCarthy CM, Cordeiro PG, et al. Acellular human dermis implantation in 153 immediate two-stage tissue expander breast reconstructions: determining the incidence and significant predictors of complications. Plast Reconstr Surg 2010;125:1606. [11] Rawlani V, Buck DW, Johnson SA, Heyer K, Kim JYS. Tissue expander breast reconstruction using pre-hydrated human acellular dermis. Ann Plast Surg; 2010. In Press. [12] Bardos J, Ashcroft M. Negative and positive regulation of HIF-1: a complex network. Biochemica Biophys Acta 2005; 1775:107. [13] Takeda K, Cowan A, Fong GH. Essential role of prolyl hydroxylase domain protein-2 in oxygen homeostasis of the adult vascular system. Circulation 2007;116:774. [14] Botusan IR, Sunkari VG, Savu O, et al. Stabilization of HIF1alpha is critical to improve wound healing in diabetic mice. Proc Natl Acad Sci U S A 2008;105:19426. [15] Wu S, Nishiyama N, Kano M, et al. Enhancement of angiogenesis through stabilization of hypoxia-inducible factor-1 by silencing prolyl hydroxylase domain-2 gene. Mol Ther 2008;16:1227. [16] Loinard C, Ginouve`s A, Vilar J, et al. Inhibition of prolyl hydroxylase domain proteins promotes therapeutic revascularization. Circulation 2009;120:50. [17] Hefferon KL. Innovations in siRNA research: a technology comes of age. Recent Pat Antiinfect Drug Discov; 2010 [Epub ahead of print]. [18] Wadhwa R, Kaul SC, Miyagishi M, Taira K. Know-how of RNA interference and its applications in research and therapy. Mutat Res 2004;567:71. [19] Thanik VD, Greives MR, Lerman OZ, et al. Topical matrixbased siRNA silences local gene expression in a murine wound model. Gene Ther 2007;14:1305. [20] http://www.lifecell.com/health-care-professionals/lifecellproducts/allodermr-regenerative-tissue-matrix/. [21] Kelly BD, Hackett SF, Hirota K, et al. Cell type-specific regulation of angiogenic growth factor gene expression and induction of angiogenesis in nonischemic tissue by a constitutively active form of hypoxia-inducible factor-1. Circ Res 2003;93:1074. [22] Menon NG, Rodriguez ED, Byrnes CK, et al. Revascularization of human acellular dermis in full thickness abdominal wall reconstruction in the rabbit model. Ann Plast Surg 2003; 50:523. [23] Milburn ML, Holton LH, Chung TL, et al. Acellular dermal matrix compared with synthetic implant material for repair of ventral hernia in the setting of peri-operative Staphylococcus aureus implant contamination: a rabbit model. Surg Infect (Larchmt) 2008;9:433.