|Home | About | Journals | Submit | Contact Us | Français|
Tissue ischemia and ischemia-reperfusion (I/R) remain sources of cell and tissue death. Inability to restore blood flow and limit reperfusion injury represents a challenge in surgical tissue repair and transplantation. A central regulator of blood flow, reperfusion signaling and angiogenesis is nitric oxide (NO). De novo NO synthesis requires oxygen and is limited in ischemic vascular territories. Nitrite (NO2-) has been discovered to convert to NO via heme-based reduction during hypoxia, providing a NO synthase independent and oxygen-independent NO source. Furthermore, blockade of the matrix protein thrombospondin-1 (TSP1) or its receptor CD47 has been shown to promote down-stream NO signaling via soluble guanylate cyclase (sGC) and cGMP-dependant kinase. We hypothesized nitrite would provide an ischemic NO source that could be potentiated by TSP1-CD47 blockade enhancing ischemic tissue survival, blood flow and angiogenesis. Both low dose nitrite and direct blockade of TSP1-CD47 interaction using antibodies or gene silencing increased acute blood flow and late tissue survival in ischemic full thickness flaps. Nitrite and TSP-1 blockade both enhanced in vitro and in vivo angiogenic responses. The nitrite effect could be abolished by inhibition of sGC and cGMP signaling. Potential therapeutic synergy was tested in a more severe ischemic flap model. We found that combined therapy with nitrite and TSP1-CD47 blockade enhanced flap perfusion, survival and angiogenesis to a greater extent than either agent alone, providing approximately 100% flap survival. These data provide a new therapeutic paradigm for hypoxic NO signaling through enhanced cGMP mediated by TSP1-CD47 blockade and nitrite delivery.
The technical aspects of soft tissue surgery and microsurgery invariably induce dramatic changes in regional blood flow and produce some degree of ischemia and ischemia/reperfusion (I/R) injury . Severing of vascular networks and connections during surgery initiates an interval of ischemia, the duration of which and the sensitivity of the affected tissue determines the degree of tissue loss. The results of this process in many instances are tissue necrosis and delayed wound healing [4; 5]. Ischemic tissues experience decreased vessel and capillary caliper, leukocyte retention and activation, endothelial cell dysfunction, and inflammatory mediator production . Restoration of flow to ischemic tissue incurs additional damage through the production of free radicals by xanthine oxidase and NADPH oxidase [7; 8] with the degree of reperfusion injury directly related to the length of the ischemic interval .
Nitric oxide is a bioactive gas produced constitutively by many cell types . Its pro-survival roles include the regulation of vascular tone through relaxation of the smooth muscles cells of arteries, promotion of endothelial cell health, inhibition of inflammatory cell activation, and blocking of thrombus formation. NO thereby maximizes blood flow to ischemic tissues and minimizes I/R injury and tissue loss [12; 13]. Recently it was discovered that the matrix glycoprotein thrombospondin-1 (TSP1), via engaging the cell surface receptor CD47, blocks the pro-survival signaling of NO [14; 15; 16]. Inhibiting TSP1-CD47 signaling maximizes NO signaling and can increase tissue blood flow [17; 18; 19; 20]. While TSP inhibition will enhance cGMP signaling, these effects are downstream of NO production. NO production becomes limited during anoxic conditions since NO synthase requires oxygen as a substrate [21; 22]. At low oxygen tension, however, nitrite (NO2-) can be reduced to NO by several cellular heme proteins including myoglobin and xanthine oxidoreductase [23; 24]. The hypoxic conversion of nitrite to NO promotes potent cytoprotection in part via the dynamic modulation of mitochondrial reactive oxygen species production following reperfusion. This cytoprotective effect of nitrite has been observed in animal models of ischemia-reperfusion of the heart, liver, and brain . Nitrite also increases angiogenesis in murine models of chronic ischemia .
We hypothesized TSP1-CD47 blockade and supplementation of nitrite could have additive activities to prevent tissue loss to ischemia. We examined this using a murine model of dorsal random myocutaneous flaps, which experience predictable and significant amounts of ischemic tissue necrosis. We show that nitrite alone increases ischemic myocutaneous flap survival with enhanced blood flow. Importantly, the combined treatment of nitrite and TSP1-CD47 blockade increases flap survival and flap blood flow significantly beyond either therapy alone.
C57BL/6 male mice aged 14 to 18 weeks were housed under pathogen free conditions and had ad libitum access to filtered water and standard rat chow. Handling and care of animals was in compliance with the guidelines established by the Animal Care and Use Committees of the National Cancer Institute. Male Sprague-Dawley rats (250–500 g; Harlan) were used in accordance with the IACUC of the University of Pittsburgh.
1H-[1,2,4]oxadiazolo[4,3-a]quinoxalin-1-one (ODQ) was obtained from Cayman Chemicals (Ann Arbor, MI). Sodium nitrite, allopurinol and potassium hexacyanoferrate were obtained from Sigma Aldrich (St Louis, MO). A rat monoclonal antibody to murine CD47, Ab 301, was prepared as described. The TSP1 monoclonal antibody clone A6.1 was purchased from Neomarkers/Laboratory Vision (Fremont, CA). An isotype-matched control IgG2 antibody was purchased from Santa Cruz Biotechnology (Santa Cruz, CA). Type I collagen (PureCol) was purchased from Inamed Biomaterial (Fremont, CA). A translation-blocking antisense morpholino oligonucleotide complementary to human and murine CD47 (CGTCACAGGCAGGACCCACTGCCCA) and a 5-base mismatch control (CGTgACAGcCAcGACCgACTGCgCA) were obtained from Gene Tools (Philomath, OR).
Sodium nitrite was dissolved in sterile Millipore filtered water to obtain a 1 mol/L (M) stock which was then serially diluted to a final concentration of 600 µmol/L (µM). Solutions were prepared fresh daily and kept in sealed sterile eppendorfs to minimize atmospheric exposure. Animals received a single bolus dosage of 48 nmols/animal (achieving an approximate blood level of 10 µM) or vehicle via intraperitoneal (i.p.) injection.
C57BL/6 mice were matched for sex and age. Anesthesia was induced by and maintained with inhalation 1.5% isoflurane and a 50:50 mixture of oxygen to room air. Body temperature was monitored via rectal probe and maintained at 37 ° C with a heating pad and lamp during the procedure. The dorsal surface was clipped of hair and depilated with Nair®. The skin was then cleansed with surgical soap and alcohol and the animals draped. Using sterile technique, 1 × 3 cm random myocutaneous flap incorporating the panniculus carnosus and skin were raised. Flap dimensions were marked on the animal skin surface with the aid of a template to insure consistency in dimension. Flaps were orientated parallel to the long axis of the animal with flap bases directed caudal and secured after mobilization with interrupted 5-0 nylon sutures. A minimum number of sutures were employed to limit the effect of suturing on tissue survival. Animals were awakened and returned to individual cages and allowed ad libitum access to food and water. On postoperative day seven, the animals were again anesthetized with inhalation isoflurane and the flaps evaluated.
The necrotic area of the flap was determined by color, refill, eschar, and the pin-prick test. The outlines of viable and nonviable areas were traced using transparent film and the area of flap necrosis versus total flap area determined by comparison with the original flap template. The cut weight of the flap tracing was compared the weight of the original template utilizing the method described by Ueda . Flap survival area was presented as the ratio of flap survival on postoperative day seven divided by the original flap area multiplied by 100%.
A total of 65 animals, 5 per group, underwent flap surgery and received the following: untreated, vehicle alone (sterile saline), nitrite alone, ODQ alone, Ab A6.1 alone, IgG2 control Ab alone, CD47 morpholino alone, control mismatch morpholino alone, nitrite + ODQ, nitrite + Ab A6.1, nitrite + IgG2 control Ab, nitrite + CD47 morpholino, nitrite + control mismatch morpholino. All treatment agents were administered just prior to the initiation of surgery. Nitrite, ODQ and the respective vehicle controls were given via i.p. injection. The TSP1 monoclonal antibody clone A6.1 or a matched isotype control antibody were given by direct injection of the surgical flap. A CD47 targeting morpholino or a mismatched control morpholino were also administered by direct injection of the surgical flap.
Flap blood flow was measured using laser Doppler imaging (MoorLD1-2λ, Moor Instruments, Devon, England). Briefly, animals were placed on a heating pad, and anesthesia was provided by 1.5% inhalation isoflurane in a 50:50 mixture of oxygen to room air. Core temperature was maintained via heat lamp at 35.5 °C and monitored by rectal probe. After a 15 minute equilibration to the experimental set-up, analysis of blood flow was performed. The following instrument settings were used: override distance 21 cm; scan time 4 msec/pixel. Results are expressed as the percent change from baseline control of the region of interest (ROI).
On the 7th post-operative day animals were anesthetized with 1% isoflurane in a 50:50 mixture of oxygen to room air. Core temperature was monitored by rectal probe and maintained at 37° C via heat lamp. Flap sutures were removed and flaps gently inverted exposing the vascular network of the deep undersurface of each flap. Image acquisition of flap vascularity was performed with an Olympus C-5500 digital camera (Melville, NY). Analysis of images was performed to quantify the degree of macro-vascularity of each flap using the following grading system – a vessel was defined as the segment of blood vessel between two points of vessel branching. Only vessels filled with red (oxygenated) blood were counted. Results represent the mean ± SD of 5 flaps from each treatment group.
On post-operative day seven dorsal myocutaneous flaps were excised and necrotic tissue removed. The remaining viable portion of each flap was then snap frozen in liquid nitrogen, pulverized and resuspended in lysis buffer. The mixtures were then sonicated in a cold room at 4° C and centrifuged at 13000 rpm for 10 minutes. Aliquots from the cleared supernatant were then used for determination of tissue cGMP via standard immunoassay (Amersham, GE Healthcare, UK). Results were normalized to total protein which was determined for each sample via a BCA protein assay (Pierce, Rockford, IL). Results represent the mean ± SD of 5 flaps from each treatment group.
Male Sprague-Dawley rats underwent surgical mobilization of 2 × 6 cm random myocutaneous flaps. Flaps were excised and immediately homogenized in a sucrose containing buffer. NO generation from tissue homogenates (10 mg/ml) treated with nitrite (10 mM) in the presence or absence of allopurinol (500 µM) and potassium hexacyanoferrate (500 µM) was measured in vessels in which the suspension and headspace were purged with helium and the vessel connected in line to a NO chemiluminescence analyzer.
Muscle biopsies 1 mm3 in size were harvested from the pectoralis major muscle of 10 week old C57BL6 mice and explanted into a three dimensional type I collagen matrix as described . Explants were incubated in the presence of endothelial growth medium (Lonza, Swizterland) with 2% FCS and the indicated treatment agents. Cell migration through the extracellular matrix was measured 96 hours post-explantation. Results represent the mean ± SD of at least three separate experiments.
Tissue units were excised, fixed in 10% buffered paraformaldehyde, paraffin embedded, and sectioned at a thickness of 5 µm. Sections were then stained with hematoxylin and eosin (H&E). Review of each slide was performed by an independent observer unaware of the origin of each tissue slide to quantify the micro-vascularity of each flap. Patent and functional microvessels were defined as those having an identifiable lumen filled with red blood cells. Results represent the mean ± SD of 5 flaps from each treatment group.
Results are presented as the mean ± SD with significance calculated by the Student’s t test (two-tailed) or ANOVA analysis as appropriate using a standard software package (Origin). Significance was assigned a p value ≤ 0.05.
Untreated or vehicle treated random dorsal myocutaneous flaps experienced 60.5 ± 4% (not shown) and 63.6 ± 7.11% necrosis respectively (Fig. 1A). In contrast systemic nitrite supplementation (48 nmols/animal i.p. at the time of surgery) significantly decreased tissue necrosis (47 ± 4%, p = 0.0017). Immediate and significant enhancement of flap blood flow in nitrite treated animals compared to vehicle treated flaps was confirmed by laser Doppler imaging in the first post-operative hour (Fig. 1B). Follow up analysis 72 h after surgery demonstrated persistence of enhanced flap flow in the nitrite treated group (data not shown).
Recently xanthine oxidoreductase (XOR) has been reported to be capable of reducing nitrite to NO . NO levels were determined in ischemic myocutaneous tissue flaps via chemiluminescence with and without XOR inhibition with allopurinol (Fig. 1C, D). Treatment of flap tissue with nitrite resulted in a rapid increase in NO production which was not significantly suppressed by treatment with allopurinal. Interestingly, a dramatic reduction in nitrite conversion to NO was achieved in ischemic flaps treated with potassium hexacyanoferrate suggesting a possible heme based mechanism.
NO signals in part by stimulating soluble guanylate cyclase (sGC) [30; 31] leading to increased cellular cGMP. cGMP, through activation of the cGMP-dependent kinase, leads to vascular smooth muscle cell relaxation, arterial dilation and increased tissue blood flow. ODQ is a chemical inhibitor of the heme moiety of sGC  and blocks NO-stimulated increases in cellular cGMP. Random flaps were created in C57BL/6 mice. Animals received vehicle, nitrite, vehicle plus ODQ or nitrite plus ODQ (20 µg/g i.p.). Flap survival and blood flow were assessed 7 days following surgery. Flap necrosis increased substantially in animals treated with nitrite plus ODQ (62 ± 3%) versus nitrite alone (42 ± 4.5%, p = 0.0381). Similarly, flaps treated with vehicle plus ODQ showed more necrosis than flaps treated with vehicle alone (72 ± 6 versus 64 ± 4%, p= 0.0073). Importantly, ODQ treatment significantly abrogated the ability of nitrite to restore blood flow in ischemic flaps (Fig. 2B).
Random dorsal 1 × 3 cm myocutaneous flaps were treated with a monoclonal TSP1 antibody (clone A6.1, 2.4 µg in 200 µl of sterile saline injected into the flap) and showed significantly less tissue necrosis (31.5 ± 2.57%) versus flaps treated with a matched isotype control IgG2 antibody (64.4 ± 6%, p < 0.0001) (Fig. 3A, B) or vehicle treated flaps (data not shown since differences in flap necrosis between untreated and vehicle treated animals were not statistically significant).
Flaps treated with a TSP1 blocking monoclonal antibody (clone A6.1) showed enhanced ischemic tissue survival. Likewise flaps treated with systemic nitrite showed enhanced tissue survival. However, ischemic flaps treated with combined therapy had the least degree of tissue necrosis (6.8 ± 6, p < 0.0001) versus untreated flaps, vehicle treated or flaps receiving only single agent treatment (Fig. 3A, B). Further analysis suggested Ab A6.1, in a dose dependent manner, enhanced nitrite effects on tissue survival though significance was not reached until a maximum dose of 2.4 µg per tissue unit (Fig. 3C). Interestingly tissue cGMP levels, which were determined at 7 days post-operatively from the viable portion of each flap, were greatest in tissue units from treated flaps with flaps treated with both a TSP1 antibody and nitrite demonstrating the greatest tissue cGMP levels among all groups (56 ± 6 fmol/µg protein, p = 0.0144) (Fig. 3D).
Morpholino oligonucleotides are stable reagents that allow temporary gene silencing of a target protein with minimal to no non-specific effects. The ability of the morpholino sequence to suppress CD47 protein in cultured cells  and murine and porcine tissue has been recently reported . Random dorsal 1 × 3 cm myocutaneous flaps were injected with an oligonucleotide morpholino targeting CD47 alone (10 µM in 200 µl of sterile saline) or in combination with systemic nitrite (48 nmoles/animal i.p.). Those animals that received the CD47 targeting morpholino alone, though demonstrating increased flap survival, still had residual flap necrosis (32 ± 6%) (Fig. 4A, B). In contrast, the combination of a CD47 suppressing morpholino with systemic nitrite resulted in further enhancement of ischemic tissue survival and blood flow with minimal tissue necrosis (14 ± 3.5%, p = 0.0001). Results achieved with combined therapy were comparable to those previously published in random myocutaneous flaps performed in age and sex matched CD47 null animals  this despite a more stringent ischemic challenge secondary to an increase in flap length in the present series of experiments from a width to length ratio of 1:2 to 1:3. A mismatch control morpholino did not increase ischemic tissue survival above untreated or vehicle treated levels.
Consistent with the ability of nitrite to stimulate angiogenesis in a hind limb ischemia model , vascular cell outgrowth and migration through a three dimensional collagen matrix was increased in explants treated with exogenous nitrite (Fig. 5A, B). Nitrite alone enhanced angiogenic response in a dose dependent manner. Antibody blockade of TSP1 or CD47 alone also increased angiogenic response when compared to basal or low dose nitrite treatment (data not shown). Finally, the combination of a CD47 targeting antibody and nitrite or a TSP1 targeting antibody and nitrite resulted in greater vascular cell outgrowth than any single treatment agent alone.
Increased tissue perfusion and wound healing requires both an angiogenic response and functional macro-vessels. Consistent with the significant degree of tissue necrosis untreated random myocutaneous flaps had the lowest vascularity index (Fig. 6A, B). Blockade of TSP1-CD47 signaling with a monoclonal TSP1 antibody resulted in increased flap vascularity. Nitrite therapy alone also tended to increase the flap vascularity index. However, combination therapy increased flap vascularity by more than 3 times over untreated flaps (74.3 ± 8/flap versus 23 ± 6.4/flap, p < 0.0001) (Fig 6B).
Quantification of the functional micro-vessels in ischemic flaps demonstrated greater micro-vessels density after nitrite or TSP1 antibody treatment alone (Fig. 7A, B). Importantly, in terms of tissue perfusion, combined treatment resulted in a 56.3% increase in micro-vessel density. Typical of myocutaneous flaps the functional macro- (see Fig. 6A, B) and micro-vascularity is demonstrated adjacent to the muscle compartment of the tissue unit.
Reconstructive surgery and microsurgery have extended greatly the ability to restore complex wounds both functionally and aesthetically. However, wound healing complications and tissue/flap loss remain significant problems. Tissue necrosis and loss are also found as primary clinical manifestations of several chronic diseases including coronary artery disease, peripheral artery disease, diabetes and stroke.
Nitric oxide is a bioactive gas continuously produced by the endothelial component of the vascular system . Among its several pro-survival roles NO regulates tissue blood flow through modulation of arterial tone. NO activates sGC in the vascular smooth muscle cells (VSMC) of arteries . Activation of sGC leads to uncoupling of the contractile apparatus of VSMC which in turn results in blood vessel dilation and increased tissue blood flow. The matrix cellular protein TSP1 potently blocks NO-driven sGC activation in vascular cells . This process necessarily requires TSP1 engagement of the cell surface receptor CD47 . Animals lacking either TSP1 or CD47 display increased tissue blood flow in response to NO and enhanced tissue survival under ischemic conditions [17; 18].
Nitrite has been identified as a novel reservoir of biologically active NO under conditions of ischemia and hypoxia and provides organ protection to I/R injury in the heart, liver and brain . However, its role in modulating soft tissue responses to ischemia and I/R injury has not been explored. A topical ointment containing nitrite and citric acid applied to cutaneous flaps was reported to modestly enhance flap survival  and to increase wound healing in a diabetic mouse model , the therapeutic effects presumptively being obtained by conversion of nitrite to NO locally through acidic disproportionation. We herein demonstrate that systemic administration of nitrite can enhance myocutaneous flap survival to ischemia. A single bolus of nitrite administered at the time of flap surgery significantly enhanced tissue survival, flap blood flow and tissue cGMP. Pre-treatment with ODQ abolished increases in tissue survival and blood flow obtained with nitrite therapy. ODQ is known to irreversibly inactivate the heme moiety of sGC  suggesting the therapeutic benefit obtained with nitrite includes increased sGC activation and cGMP production. However, ODQ has off target effects and has been reported to also suppress activation of other hemoproteins  some of which could be involved in hypoxic reduction of nitrite to NO. However the authors have recently reported enhanced tissue protection/function and NO production obtained with nitrite therapy that was abolished following administration of either ODQ, the NO scavenger PTIO or Rp-8-Br-cGMP (an inhibitor of cGMP-dependent kinase)[40; 41].
In the present report ischemic soft tissue NO levels were enhanced by ex vivo treatment with nitrite and this process was not suppressed with allopurinol suggesting that hypoxic reduction of nitrite to NO in ischemic myocutaneous flaps is not dependent upon XOR activity. In contrast, a near 80% reduction in NO signal was achieved in similar ischemic tissue flaps treated with nitrite in the presence of ferricyanide suggesting a possible role for heme in the conversion of nitrite to NO under these conditions. These results are in contrast with the recent report that demonstrates XOR mediated changes in nitrite stimulated cardiovascular effects . However this report did not differentiate between NO-dependent and NO-independent effects arising from nitrite treatment.
No complications were noted to arise from systemic nitrite administration. This is not unexpected since nitrite reduction to NO occurs along a pH and oxygen gradient . Likewise reductase conversion of systemically administered nitrite is localized to regions of low blood flow, and deceased pH and oxygen, conditions specifically found in ischemic tissues thus minimizing systemic effects upon blood pressure and cardiac response. However, it is possible that systemic nitrite administration might produce an alteration in blood pressure in this model. In contrast, the local administration of therapeutic agents that block TSP1-CD47 signaling, through enhancing regional blood flow in soft tissue flaps, would be unlikely to alter blood pressure.
NO supplementation has been applied to several models of tissue ischemia/wound healing both through administration of NO-releasing agents such as isosorbide dinitrate [18; 43; 44; 45] or through administration of the nitric oxide synthase substrate L-arginine with enhanced tissue survival [46; 47] and wound breaking strength  reported. However, ischemic tissue flaps treated with either primary NO donors or NOS substrate demonstrate residual flap necrosis  suggesting that TSP1-CD47 inhibition of NO signaling remains a barrier to complete tissue survival through exogenous NO supplementation. In the present report we found that concurrent blockade of TSP1-CD47 signaling and nitrite supplementation increased, in an additive manner, ischemic tissue survival beyond what any single agent achieved. We also found in an ex vivo assay that simulates the angiogenic sprouting and cell migration of wound healing increased response to both nitrite and monoclonal antibodies targeting TSP1 and CD47. That combined therapy demonstrated additive rather than synergistic increases in ischemic tissue survival may reflect the robust improvement each agent (either nitrite or a TSP1 monoclonal antibody) provided alone. At the doses employed each individual agent achieved close to 75% tissue survival. When combined ischemic tissue survival approached 100%. The murine dorsal surface reasonably accommodates a soft tissue flap 3 cm in length and limits our flap width to length ratio to 1:3. A greater flap width to length ratio in a larger rodent would perhaps allow for demonstration of synergistic rather than mere additive increases in tissue survival with combination therapy.
Both CD47 Ab/nitrite and TSP1 Ab/nitrite in combination proved to be in the present model of myocutaneous flap ischemia nearly equal in effectiveness. However, targeting CD47 may be theoretically more beneficial since platelets represent a constant potential reservoir of preformed TSP1 . Of great clinical relevance combined therapy with nitrite and a TSP1-CD47 blocking antibody produced significantly more functional micro- and macro-vascularity in ischemic soft tissue flaps than either treatment alone. The results of the present work highlight that a multi-directional approach of delivery of the NO pro-drug nitrite and blocking TSP1-CD47 inhibition of NO signaling achieves maximum tissue protection from ischemic injury with enhancement of angiogenesis and macro- and micro-vascularity (Fig. 8).
We thank Dr. David D. Roberts (NCI, NIH), Dr. William A. Frazier (Washington University School of Medicine), and Dr. Jack Lawler (Harvard Medical School) for providing reagents.
Sources of Funding
This work was supported by NIH grant K22 CA128616 (J.S.I.).
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures: Dr. Jeff Isenberg is Chair of the Scientific Advisory Board of Vasculox, Inc. (St. Louis, MO).