Nitrite anion stimulates ischemic arteriogenesis involving NO metabolism

Abstract

Nitric oxide (NO) is a potential regulator of ischemic vascular remodeling, and as such therapies augmenting its bioavailability may be useful for the treatment of ischemic tissue diseases. Here we examine the effect of administering the NO prodrug sodium nitrite on arteriogenesis activity during established tissue ischemia. Chronic hindlimb ischemia was induced by permanent unilateral femoral artery and vein ligation. Five days postligation; animals were randomized to control PBS or sodium nitrite (165 μg/kg) therapy twice daily. In situ vascular remodeling was measured longitudinally using SPY angiography and Microfil vascular casting. Delayed sodium nitrite therapy rapidly increased ischemic limb arterial vessel diameter and branching in a NO-dependent manner. SPY imaging angiography over time showed that nitrite therapy enhanced ischemic gracillis collateral vessel formation from the profunda femoris to the saphenous artery. Immunofluorescent staining of smooth muscle cell actin also confirmed that sodium nitrite therapy increased arteriogenesis in a NO-dependent manner. The NO prodrug sodium nitrite significantly increases arteriogenesis and reperfusion of established severe chronic tissue ischemia.

peripheral arterial disease (PAD) is a chronic occlusive vascular disorder of the lower extremities, the prevalence of which is rising with increases in geriatric and diabetic populations around the world (1). Moreover, an increasing percentage of patients with PAD will progress to critical limb ischemia, a severe complication of PAD that often requires amputation of the affected limb (21). As such, useful treatments for PAD would invoke repair responses to the affected limb through revascularization of the damaged area involving multiple complex events such as angiogenesis and arteriogenesis (3–5).

It is now appreciated that effective therapeutic revascularization of ischemic tissues requires a combination of arteriogenesis and angiogenesis (3, 8). While numerous studies have investigated mechanisms and effects of angiogenesis, much less information is known regarding important mediators of arteriogenesis activity. Arteriogenesis typically refers to outward remodeling and enlargement of collateral arterioles, allowing increased blood flow to distal vascular beds (3, 4, 8). As such, increased arteriogenesis can significantly augment ischemic tissue blood flow through larger caliber remodeled conduit vessels to restore tissue perfusion. However, experimental studies have generally focused on angiogenesis versus arteriogenesis with a recent review of clinical arteriogenesis trials demonstrating a clear lack of understanding of this important event (2).

Nitric oxide (NO), a potent vasodilator, is present to some degree in all vessels and is a signaling modulator that has been implicated in both arteriogenesis and angiogenesis, with its vasodilatory properties being demonstrated in both immature and mature collaterals (17, 26). NO-producing enzymes endothelial and inducible NO synthase expression and activity are influenced by shear stress levels, such as those observed during chronic ischemic events, and augmented in collateral vessel development (25). In fact, the inhibition of NO synthase-dependent NO production results in decreased collateral flow following arteriogenesis initiated by physical activity and cytokine stimulation (17). Shear stress is likely an important initiator of arteriogenesis, as Dai et al. (7) recently demonstrated that endothelial NO synthase contributes to arteriogenesis through a feed-forward mechanism, increasing shear stress in collateral vasculature.

Therapeutic approaches invoking NO bioavailability in the clinic have not been successful. However, we have previously observed that nitrite therapy through selective reduction back to NO in ischemic tissue clearly augments angiogenesis and possibly arteriogenesis (14). Yet, the physiological and temporal effects of sodium nitrite on arteriogenesis activity during experimental tissue ischemia remain unknown. Here we examine the efficacy and mechanisms of sodium nitrite on arteriogenesis and tissue reperfusion of established severe hindlimb ischemia using traditional approaches and novel SPY angiographic imaging to monitor vascular remodeling activity over time (28).

MATERIALS AND METHODS

Reagents.

Pharmaceutical-grade indocyanine green (IC-Green; ICG) and the use of a SPY imaging device was provided by Novadaq Technologies, based in Toronto, Canada. Anti-CD31 antibody was obtained from BD Biosciences. Vectashield plus 4′,6-diamidine-2′-phenylindole dihydrochloride (DAPI) was obtained from Vector Laboratories. Cy3-conjugated secondary antibodies were purchased from Jackson Immunoresearch, and Alexa Fluor 488-conjugated secondary antibodies were purchased from Molecular Probes. Microfil was purchased from Flow Tech. All other chemical reagents were obtained from Sigma-Aldrich.

Animal model.

Wild-type C57BL/6J mice weighing 23–28 g were used for experiments in this study. In all experiments, drug treatments of either control (PBS twice daily) or nitrite (165 μg/kg twice daily) were started 5 days following ligation. Mice were bred and housed at the Association for Assessment and Accreditation of Laboratory Animal Care, International accredited Louisiana State University Health Science Center-Shreveport animal resource facility, and maintained in accordance with the National Research Council's Guide for Care and Use of Laboratory Animals. All animal studies were approved by the Institutional Animal Care and Use Committee (protocol P-08-041) and conformed to the Guide for the Care and Use of Laboratory Animals, published by the National Institutes of Health.

Novadaq SPY imaging analysis.

The SPY Imaging device (Novadaq Technologies) uses an array of light-emitting diodes at a wavelength of 806 nm to excite a bolus injection of vascular contrast dye. The contrast dye used was ICG (IC-Green, Akorn Pharmaceutical, Lake Forest, IL) injected at a concentration of 165 μM that, upon binding with plasma, fluoresces at an absorption maximum of 800 nm with an emission maximum of 825 nm (9). The bolus injection of ICG (30 μl) was administered retro-orbitally, and angiograms were recorded for 1 min. Angiograms were taken before and after the ligation surgery, as well as days 1, 3, 5, 7, 11, 15, and 23 postligation. All SPY measurements were taken 2 h postadministration of the first daily dose of nitrite. SPY analysis software allowed a region of interest to be designated, in our case the gastrocnemius region. The analysis software package uses a specialized algorithm to provide a cumulative intensity value for the region of interest over the length of the video. This algorithm calculates a running average of pixel intensities and removes intensities >±50% away from the average, specifically selecting for median perfusion of tissue and removing contributions from visible vessels in the selected region of interest. This results in a rapid increase in pixel intensity, eventually reaching a plateau as the tissue is maximally perfused (supplemental Fig. 1). Calculating the slope of ICG relative fluorescence units over time provides a blush rate measurement for both limbs (ischemic and nonischemic). Percent change in blush rates were calculated as A/B × 100, where A is the blush rate of the ischemic limb and B is the blush rate of the nonischemic limb.

In a separate series of experiments, SPY imaging of ischemic limbs was performed on either PBS or sodium nitrite-treated animals up to day 15 with and without the skin to evaluate differences in perfusion images. To maintain imaging modality consistency, all SPY angiograms were performed 2 h postadministration of the first daily dose of nitrite. Cine video of SPY angiographs were captured over 30 s with single frame images obtained at various time points identified in the each figure.

Laser-Doppler tissue blood flow measurement.

The Vasamedics Laserflo BPM2 deep-tissue laser-Doppler device was used to measure hindlimb blood flows as we previously reported (14). The laser probe (0.8 mm in diameter) was placed over six separate regions of the gastrocnemius muscle of both the ischemic and nonischemic hindlimbs of the mice. The six measurements were averaged together for reported muscle blood flows. Readings were taken immediately following SPY imaging, and percent blood flows were calculated as: (a/b) × 100, where a is the ischemic limb average flow and b is the nonischemic limb average flow.

Fluorescent microsphere blood flow analysis.

Fluorescence microsphere bead blood flow measurement and the calculation were performed as described previously (11, 12). Briefly, C57BL/6J mice were anesthetized at 7 and 15 days postligation/therapy to perform the procedure. Abdominal aorta was cannulated with polyethylene catheter (inner diameter, 0.28 mm; and outer diameter, 0.61 mm, Intradermic), and adenosine (1 mg/kg, Sigma) was subsequently injected. Microspheres of different dye colors (orange, blue-green, yellow, green, and crimson; 15 μm; Molecular Probes, Invitrogen) were thoroughly mixed and infused through the cannula. Reference blood was drawn from the distal part of the perfusion system. Hindlimb muscles were harvested, weighed, and digested with SDS/proteinase K solution. Microspheres were counted using flow cytometry (FACSCalibur, BD LSRII). Blood flow calculations from each sample were measured as previously described (11) using the following formula: blood flow (in ml/min) = m_s·IS_rs·W/IS_s·m_rs·t, where m_s are microspheres in sample, IS_rs are the internal standard microspheres in reference sample (blue-green microspheres), W is the weight of each sample, IS_s is internal standard microspheres count in sample (blue-green), m_rs is the microspheres count in the sample, and t = duration when reference sample was withdrawn. Finally, blood flow percent change was expressed as the ratio of flow in ligated versus nonligated hindlimb ×100.

Vascular ligation and casting.

Chronic hindlimb ischemia was induced in C57BL/6J mice by ligating and transecting the left common femoral vein and artery distal to the origin of the profunda artery as well as its associated branches. Mice were anesthetized with an intraperitoneal injection of ketamine (100 mg/kg) and xylazine (8 mg/kg). Aseptic surgery was performed by a linear incision at the left groin, exposing the left common femoral artery and vein. Immediate pallor was observed in the distal hindlimb following ligation of the artery and vein using 5-0 silk.

Arterial vasculature was evaluated postmortem using Microfil vascular casting as we previously reported (27). Nitrite and PBS control-treated mice were anesthetized 7, 15, or 23 days postligation and euthanized, and the vena cava and abdominal aorta immediately distal to the kidneys were ligated with 5-0 silk suture. The aorta was canulated distal to this ligature to isolate the hindlimb for vascular cast perfusion. PBS was perfused through the catheter, and the vena cava distal to the initial suture was severed allowing exit of the perfusate. A papaverine-adenosine (10.6 μM-3.7 μM) mixture was injected, and 10% phosphate-buffered formalin was administered to maximally dilate and fix the vasculature. Blue-pigmented Microfil was perfused at a 7:2 ratio (Microfil:diluent), and the hindlimbs were cleared using graded ethanol solutions of 20, 50, 75, 95, and 100% followed by methyl salicylate each for 24 h. Imaging of these tissues was performed on a Nikon Eclipse AZ100 Multizoom scope using z-axis flattening technology. Vessel diameters and branches at various time points were measured using morphometric analysis tools in Nikon Elements software. Supplementary Fig. 2 illustrates the collateral vessels analyzed in this study. Gracillis collateral arterioles are highlighted by the yellow lines connecting the profunda femoris to the saphenous artery.

Sodium nitrite therapy.

Five days following induction of critical limb ischemia, drug therapy was initiated. Mice were randomized to one of four treatment groups: sodium nitrite (165 μg/kg ip, twice daily), PBS control (injected ip, twice daily), sodium nitrite (165 μg/kg ip, twice daily) + 2-(4-carboxyphenyl)-4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide (cPTIO; 1 mg/kg twice daily), and PBS control (ip, twice daily) + cPTIO (1 mg/kg twice daily). These treatments were maintained until the end of the study.

Tissue NO measurement.

Tissue NO metabolite levels were measured using a chemiluminescent NO analyzer (GE Healthcare) as we previously described (14, 27). Briefly, gastrocnemius tissue (both ischemic and nonischemic) was harvested from animals at the time of euthanasia. This tissue was added to a 300-μl solution containing 800 mM potassium ferricyanide, 17.6 mM N-ethylmaleimide, and 6% Nonidet P-40 in PBS and homogenized. Homogenized tissue was then frozen in liquid nitrogen until chemiluminescent analysis with acidic triodiode.

Immunohistochemical staining of smooth muscle actin and CD31.

Tissue immunohistochemistry was performed as we have previously reported with anti-smooth muscle actin (SMA) and anti-CD31 antibodies on day 15 postligation to observe the progression of arteriogenesis (14, 27). Gastrocnemius muscles from ischemic and nonischemic hindlimbs were dissected and embedded in optimum cutting temperature freezing medium, frozen, and cut into 5-μm sections. The slides were fixed at −20°C in 95% ethanol-5% glacial acetic acid for 1 h. Slides were blocked and stained with anti-CD31 antibody at 1:200 dilution and primary antibody against SMA at 1:500 dilution and incubated at 37°C for 1 h. Slides were then washed and Cy3-conjugated anti-rat secondary antibody, and Alexa Fluor 488-conjugated anti-mouse secondary antibodies were added at 1:500 dilution and incubated at room temperature for 1 h. Slides were mounted using Vectashield-DAPI mounting medium. At least four slides per hindlimb with three sections per slide were made for vascular staining and analysis. A minimum of two fields were acquired per section of muscle. Pictures were acquired with a Hamamatsu digital camera using a Nikon TE-2000 epifluorescence microscope (Nikon, Japan) at ×200 magnification. Simple PCI software version 6.0 (Compix, Sewickley, PA) was used to quantitate the area of CD31 and SMA staining of arterial vessels.

Statistical analysis.

One-way ANOVA with Dunnett posttesting was used to analyze differences between nitrite and control groups for data at various time points and between treatments. A P value of <0.05 was required for statistical significance. Statistics were performed with GraphPad Prism 4.0 software. The number of mice used per experiment is reported in the figure legends.

RESULTS

Spy imaging angiography and blood flow.

Most current angiographic analysis techniques using X-ray based imaging that requires postmortem examination thus limiting the ability to make longitudinal observations of vascular remodeling over time in the same animal. We sought to determine whether the SPY angiographic imaging approach using ICG coupled with light-emitting diode excitation and image capture could provide noninvasive imaging of peripheral vasculature. Figure 1 demonstrates that the SPY angiographic imaging technique shows clear changes in limb perfusion of the same animal pre- and postvascular ligation. Figure 1A illustrates preligation lower limb vessel angiography following injection of the perfusion dye ICG. The femoral artery/vein progressing distal to the epigastric artery/vein into the saphenous artery/vein of the hindlimbs can be clearly observed. Figure 1B shows the same animal following unilateral femoral artery/vein ligation of the left limb resulting in diminished blood flow to the affected limb (see supplemental video 1 for angiogram video). Figure 1C shows cumulative blood flow in the same region from Fig. 1A that is reported as an accumulation of ICG fluorescence intensity for the duration of the angiogram. Figure 1D shows accumulated ICG fluorescence illustrating blood flow with only a minimal amount of perfusion noted in the foot of the ligated limb. These images demonstrate the sensitivity and utility of SPY vascular imaging in a multistep manner within the same animal.

Fig. 1.

SPY angiographic imaging of normal and ischemic hindlimbs. A: preligation angiogram with a bilateral perfusion of indocyanine green dye (ICG). B: postligation angiogram showing unilateral perfusion of ICG. C: average blood flow in the form of accumulated ICG intensity of preligation hindlimbs over time (red is most flow; blue is least flow). D: average blood flow in the form of accumulated ICG intensity of postligation hindlimbs over time.

SPY angiographic visualization of sodium nitrite therapy.

The process of vessel development was observed more closely with the aid of a 75-mm lens that allowed for a more detailed view of the ligated limb both in terms of resolution (∼50 μm) as well as magnification. In Fig. 2, the left column of images are from a PBS-treated animal at day 11 postligation (6 days after starting PBS therapy) and the right column represents an animal treated with sodium nitrite at 11 days postligation (6 days after starting sodium nitrite therapy). These series of images illustrate the sensitivity of this imaging technique such that 3 s into the video, the ICG begins to perfuse the lateral marginal artery (yellow arrow) (at 0 s, no dye is visible in the video; image not shown). One second later, perfusion of extended gracillis collateral arterioles connecting the profunda femoris (red arrow) to the saphenous artery (18) is seen in sodium nitrite-treated animals. By 5 s, the majority of the saphenous artery has been perfused in the sodium nitrite-treated mouse, and by 6 s, the distal vasculature from the saphenous artery and lateral marginal artery are well perfused throughout the ischemic limb. Thus it appears in this model that the proximal portion of the gastrocnemius muscle was supplied by the gracillis collateral arterioles connecting from the profunda femoris to saphenous artery and the mid and distal portion of the gastrocnemius muscle supplied by the lateral marginal and saphenous arteries. Monitoring the same time points in a control PBS-treated animal did not result in the identification of similar anatomic vascular features, indicating minimal collateral vessel remodeling (see supplemental videos 2 and 3).

Fig. 2.

Delayed nitrite therapy increases SPY angiogram arteriogenesis. Representative temporal SPY angiogram image stills (3–6 s) are shown at 11 days following ligation and 6 days after beginning therapy (either PBS or sodium nitrite). Left: PBS control angiogram. Right: sodium nitrite angiogram following injection of ICG. n = 5 animals per cohort. Circles identify limb anatomical regions of vascular blush, whereas arrows indicate perfused vessels that progressively occur over time.

A separate series of SPY imaging studies were performed using day 15 ischemic hindlimbs to determine the effect of skin on the resolution of vascular imaging. Figure 3 shows vascular angiograph images from ischemic limb adductor and gastrocnemius muscle from PBS or sodium nitrite treatments 5 s after injection of ICG. Figure 3, A and B, shows PBS and sodium nitrite angiograms with skin attached, whereas Fig. 3, C and D, shows the same ischemic limbs without skin. Likewise, Fig. 3, E–H, shows SPY images from the same limbs under identical conditions at 15 s into the video. These pictures show that SPY angiographic imaging is most clear and sharp in limbs without skin and that nitrite therapy appears to selectively augment the amount of collateral vessel remodeling compared with that of PBS.

Fig. 3.

Skin effects on high-magnification SPY imaging. Mice with unilateral hindlimb ischemia were treated with PBS or sodium nitrite beginning at day 5 postligation and subsequently imaged at day 15. A and B: representative SPY angiographs at 5 s in PBS and sodium nitrite-treated animals with the skin intact, respectively. C and D: SPY angiograms 5 s after ICG injection from the same animals after removal of the skin. E and F: representative SPY angiographs at 15 s in PBS and sodium nitrite-treated animals with the skin intact, respectively. G and H: SPY angiograms 15 s after ICG injection from the same animals after removal of the skin.

Figure 4 illustrates representative longitudinal SPY images from PBS or sodium nitrite-treated mice over time. All of these images contain limbs with skin to determine the potential utility of repeated use of this imaging modality. Figure 4, A–F, shows baseline and postligation changes up until day 3 postligation. Respective therapy was started at day 5 postligation with subsequent SPY angiographic images taken at days 7 and 15. Evaluation of perfusion differences in the ischemic leg reveals that robust vascular perfusion occurs in sodium nitrite-treated ischemic limbs (arrows) along with enhanced perfusion of the foot (compare circles among all images). Together, these data demonstrate that the SPY angiographic imaging approach allows for longitudinal evaluation of sodium nitrite therapy on the ischemic vascular remodeling process over time.

Fig. 4.

Longitudinal SPY imaging of ischemic vascular remodeling responses. Longitudinal SPY angiographs were made for PBS and sodium nitrite-treated animals. A and B: preligation (Pre) images for both experimental cohorts. C and D: representative postligation (Post) images for both experimental cohorts. E and F: representative day 3 SPY images for both cohorts before experimental treatments were began. G and H: representative day 7 SPY images. I and J: representative day 15 SPY images between PBS and sodium nitrite therapy, respectively. Circles highlight dye perfusion of the foot, whereas arrows point to distal perfusion of the popliteal and tibial arteries.

Changes in ischemic tissue blood flow as measured by SPY imaging, laser-Doppler flowmetry, and fluorescent microsphere bead infusion techniques.

Increasing blood perfusion of ischemic tissues is an ultimate objective of ischemic vascular remodeling. Thus we next determined ischemic and nonischemic limb blood flow in PBS- or sodium nitrite-treated animals using three different methods. Figure 5A reports the calculated blush rate by SPY imaging that was preferentially increased in nitrite-treated animals at different time points of the study. Laser-Doppler flowmetry data (Fig. 5B) show a progressive increase of ischemic limb blood flow by sodium nitrite that was significant by day 11. Lastly, changes in tissue blood flow were examined using the fluorescence microsphere perfusion technique at days 7 and 15 after ligation and therapy. Figure 5, C and D, shows that nitrite therapy increased blood flow compared with PBS treatment. Importantly, the restoration of blood flow by nitrite therapy was significantly blunted by cPTIO (Fig. 5, A–D) using all methods, suggesting the role of nitrite mediated NO formation in recovery of ischemic hindlimb blood flow.

Fig. 5.

SPY imaging, laser Doppler, and fluorescent microsphere-based blood flow measurements. A: blush rate perfusion ratio between ischemic versus nonischemic tissue by SPY imaging. B: blood flow perfusion measured using laser Doppler. C and D: blood flow perfusion data using fluorescent microspheres at 7 and 15 days postligation/therapeutic cohorts, respectively. Therapeutic cohorts include PBS, nitrite alone, and nitrite + 2-(4-carboxyphenyl)-4,4,5,5-tetramethyl-imidazoline-1-oxyl-3-oxide (cPTIO). n = 5 animals per cohort. *P < 0.05 increase over PBS ischemia; #P < 0.05 decrease from nitrite ischemia.

Morphometry parameters of collateral gracillis artery remodeling in response to nitrite.

Animals from different treatment cohorts at days 7, 15, and 23 postligation were used for Microfil vascular casting to confirm vascular changes observed by SPY image analysis. Figure 6, A and B, reveals no discernible difference between control or nitrite therapies at day 7 (2 days after beginning nitrite therapy). However, Fig. 6D illustrates that sodium nitrite therapy increased gracillis artery collateral vessel remodeling at day 15 compared with PBS treatment (Fig. 6C). Figure 6F shows that increased collateralization is still observed with sodium nitrite treatment at day 23 along with the appearance of corkscrew arteries that are absent in PBS-treated animals (Fig. 6E). These vascular Microfil images corroborate SPY angiography data, demonstrating increased arteriogenesis following sodium nitrite treatment over time.

Fig. 6.

Microfil vascular casting of gracillis arteriole remodeling in response to PBS control or sodium nitrite. A: representative image of PBS control therapy 7 days following ligation and 2 days following the beginning of treatment. B: corresponding image of nitrite therapy (165 μg/kg twice daily) 7 days following ligation and 2 days after the beginning of treatment. C and D: PBS- and nitrite-treated vascular casts at day 15 (10 days posttreatment initiation). E and F: PBS- and nitrite-treated arterial vascular casts at 23 days following ligation (18 days posttreatment initiation). n = 5 animals per cohort.

Figure 7 reports arterial vessel measurements from Microfil vascular casts with measurements of gracillis collateral arterioles connecting the profunda femoris to saphenous artery. Figure 7A reports the number of gracillis collateral arterioles counted over a 90 mm² area. Sodium nitrite therapy significantly increased the number of gracillis collateral arterioles on day 15 postligation (day 10 posttreatment) in the ischemic limb without affecting arteriolar count in the nonischemic leg. Figure 7B shows that sodium nitrite significantly increases gracillis collateral arteriolar diameter at all time points. Overall, sodium nitrite therapy increased arteriole numbers and resulted in larger diameter arterioles in the ischemic limb over time without significantly affecting arteriole diameters in nonischemic limbs. These data demonstrate that nitrite therapy in established ischemic tissue stimulates robust remodeling of the gracillis collateral arterial vasculature. We next confirmed whether the effects of sodium nitrite were due to NO formation by using the NO scavenger cPTIO coadministered with sodium nitrite therapy. Figure 7, C and D, illustrates the effect of cPTIO on sodium nitrite-dependent changes in gracillis collateral arteriole number as well as diameter. In both cases, cPTIO significantly inhibited the effect of sodium nitrite on collateral vascular remodeling responses.

Fig. 7.

Sodium nitrite therapy increases gracillis collateral artery remodeling. A: numbers of gracillis collateral arterioles among the various treatment groups. B: gracillis collateral arteriole diameter changes among various treatment cohorts. C: number of gracillis collateral arterioles at day 15 with nitrite ± cPTIO (1 mg/kg twice daily) treatment. D: gracillis collateral arteriole diameter changes at day 15 with nitrite ± cPTIO (1 mg/kg twice daily) treatment. n = 5 animals per cohort. *P < 0.05 increase over PBS ischemia (Isch); #P < 0.05 decrease from nitrite ischemia (NI).

Sodium nitrite therapy augments tissue NO bioavailability.

To confirm that sodium nitrite therapy augmented tissue NO bioavailability, we analyzed gastrocnemius tissue from treated groups at days 7 and 15 using chemiluminescent detection of tissue NO, nitrate, and nitrite (NOx) levels. Figure 8A reports tissue data from animals 7 days postligation, corresponding to 2 days after starting sodium nitrite therapy, showing no statistical significance between the treatment groups. Figure 8B is data from animals 15 days postligation and 10 days after beginning sodium nitrite therapy. Nitrite therapy resulted in significant increases in tissue NOx levels in both ischemic and nonischemic tissue. Elevation of NOx in the nonischemic limb of animals receiving sodium nitrite therapy supports the hypothesis that NO exerts its effects in an endocrine-type pathway proposed by Elrod et al. and supported by us (10, 27). Moreover, cPTIO treatment prevented increased tissue NOx levels, suggesting that nitrite reduction to NO is involved in augmenting ischemic tissue NO metabolites.

Fig. 8.

Exogenous nitrite administration augments tissue nitric oxide, nitrate, and nitrite (NOx) levels. A: tissue NOx levels at day 7 postligation, 2 days following the beginning of sodium nitrite therapy (165 μg/kg twice daily). B: tissue NOx levels at day 15 postligation, 10 days following the beginning of sodium nitrite therapy (165 μg/kg twice daily). In B, treatment groups that received cPTIO (1 mg/kg twice daily) are included. n = 5 animals per cohort, *P < 0.05.

Sodium nitrite increases arterial SMA staining.

Lastly, we used SMA and CD31 costaining in frozen tissue sections to verify that changes of the vasculature involved histological signs of arteriogenesis. Figure 9A illustrates that sodium nitrite therapy increases SMA staining (green positive areas) that demarcates arterioles (indicated by arrows). Figure 9, B and C, shows results from treatment with cPTIO + sodium nitrite or cPTIO alone, respectively. Coadministration of cPTIO with nitrite significantly decreased the amount of SMA-CD31 staining. Figure 9D reports the amount of colocalization of SMA-positive stain (green color) with CD31-positive stain (red color). Sodium nitrite therapy increased SMA staining of arteries within ischemic tissue in a NO-dependent manner.

Fig. 9.

Nitrite therapy increases arterial smooth muscle actin (SMA) staining. A: SMA staining of sodium nitrite-treated ischemic limb tissue (green areas indicate SMA stain; arrows) 15 days after ligation and 10 days following beginning of sodium nitrite therapy. B: SMA staining of sodium nitrite plus cPTIO treatment in ischemic tissue (arrows indicate staining) 15 days after ligation and 10 days following beginning of sodium nitrite therapy. C: SMA staining of cPTIO-treated ischemic limb tissue (indicated with arrows) 15 days after ligation and 10 days following beginning of sodium nitrite therapy. D: colocalization data of SMA staining (green) with CD31 vascular staining (red) from ischemic tissues of different treatment cohorts 15 days after ligation and 10 days following beginning of sodium nitrite therapy. n = 5 animals per cohort, *P < 0.05 compared with PBS; #P < 0.05 compared with nitrite ischemia.

DISCUSSION

Physiological differences between arteriogenesis and angiogenesis suggest that in a model of chronic limb ischemia, arterial occlusion would stimulate arteriogenesis activity to restore bulk blood flow to ischemic tissue, whereas angiogenesis augments microvascular density to restore tissue oxygen levels in ischemic regions. This paradigm has been substantiated by several previous studies (6, 15, 23). The present study details the role of exogenous sodium nitrite administration on arteriogenesis over time in a severe model of unilateral permanent hindlimb ischemia. An important feature separating our current study from other previous reports is the use of the combined artery and vein ligation model resulting in severe ischemia and delayed therapeutic intervention with sodium nitrite after establishing longer-term tissue ischemia.

We have previously reported that immediate sodium nitrite therapy following femoral artery ligation enhances the restoration of ischemic limb reperfusion and increases microvascular density (14). We further reported that sodium nitrite therapy serves a protective role early on by day 3 postligation in this model through differential expression of genes enhancing muscle and vascular growth and function with later gene expression responses at day 7 affecting tissue repair and extracellular matrix development (22). Results from these studies suggested that day 5 would be an ideal time to investigate whether delayed therapy to examine whether sodium nitrite therapy could augment vascular remodeling responses independent of its rapid protective effects at day 3 using an immediate therapy model. Here we find that using a model of delayed sodium nitrite therapeutic intervention beginning at day 5 postligation, ischemic vascular remodeling rapidly occurs involving robust ischemic arteriogenesis leading to ischemic limb reperfusion. In this study, we chose to use a more severe model of hindlimb ischemia involving ligation and excision of both the artery and vein, which is not typically observed in clinical peripheral vascular disease and therefore may be difficult to extrapolate to a clinical setting compared with historical evaluation of arterial ligation and subsequent remodeling responses (18, 19). However, the choice of this model stemmed from our desire to determine whether sodium nitrite therapy could augment vascular remodeling in a more severe model and whether novel vascular imaging modalities could enable noninvasive image angiographic analysis. Our data clearly illustrate that sodium nitrite therapy is efficacious in a severe model of hindlimb ligation even when given in a delayed manner.

Use of ICG angiographic imaging has been previously examined for monitoring vascular perfusion in animals and humans, thus forming the foundation of the SPY Imager device that is primarily used to monitor vascular reanastamoses during different surgical procedures (13, 16, 20, 24, 29). However, in this study we found that the SPY Imager was very capable and useful for performing serial measurements of vascular perfusion of ICG that allowed for enhanced imaging of vascular remodeling responses in situ over time. Our data revealed that skin can limit the resolution of the acquired images, suggesting that detailed vascular morphometry be performed using a window system that does limit the ability to repetitively make such high-resolution images. However, it is clear from our study that longitudinal imaging of the limbs over time reveals distinct differences in perfusion as observed between PBS control and sodium nitrite treatments. One clear anatomical difference that we routinely observed was that ischemic foot perfusion was consistently greater in sodium nitrite-treated animals. Refinement of ischemic foot analysis would be an ideal longitudinal measurement that we are currently developing. Nonetheless, the SPY Imager modality was found to be very easy to operate, to provide reliable and consistent angiograms, and to generate useful anatomic vascular images, highlighting its potential usefulness in experimental studies of vascular remodeling.

We also found that the SPY imaging method could measure changes in tissue perfusion in a sensitive manner compared with other traditional methods of measuring blood flow. Interestingly, blush rates calculated from the SPY imager analytical software revealed significant changes in blood flow at earlier time points compared with laser-Doppler measurements, which is most often used for hindlimb ischemia studies (19). Having observed such distinct differences between the methods, we also employed the fluorescent microbead perfusion technique that measures blood flow throughout the tissue in a very sensitive and accurate manner. Fluorescent microsphere perfusion data at days 7 and 15 correlated very well with the SPY imaging technique versus laser-Doppler measurements. This is likely due, in part, to the fact that laser-Doppler flowmetry is unable to penetrate much beyond 0.5–0.8 mm in depth. Together, blood flow data from the various modalities highlight the potential utility of the SPY imaging technique to sensitively measure tissue blood flow in ischemic tissues.

A second goal of this study was to determine the efficacy of delayed sodium nitrite therapy in a severe model of hindlimb ischemia with the intent of uncovering any biological or physiological limitations. Interestingly, we found that nitrite-mediated NO formation was important for ischemic vascular remodeling in this severe hindlimb ischemia model such that NO scavenging with cPTIO significantly prevented nitrite-mediated effects on arteriogenesis and collateral remodeling. This is likely due to the fact that sodium nitrite therapy elicits rapid differential changes in ischemic tissue gene expression. For example, sodium nitrite therapy quickly downregulates the serotonin receptor 2B, which could augment vasodilation responses that may influence shear-mediated vascular remodeling responses (22). Under the same conditions, purinergic receptor P2Y purinoceptor 2 and vascular endothelial-cadherin were also found to be significantly upregulated that could further augment tissue blood flow and vascular remodeling (22). Future studies antagonizing these and other mediators will be needed to precisely understand which of these mechanisms modulates sodium nitrite-mediated ischemic vascular remodeling.

In conclusion, our study has examined the effect of delayed sodium nitrite in a severe model of hindlimb ischemia and expanded our understanding of potential therapeutic effects. We have shown that the SPY Imager angiograph approach effectively allows noninvasive vascular perfusion image analysis in both detailed resolution and longitudinal formats, indicating the utility of this modality for future study. Studies are needed to refine various measurement protocols (e.g., blood flow and longitudinal perfusion analysis) to obtain additional data besides anatomical images of vascular remodeling responses.

GRANTS

This work was supported by National Heart, Lung, and Blood Institute Grants HL-80482 and HL-94021.

DISCLOSURES

J. Docherty, D. Goyette, and P. Dvorsky are employed by Novadaq Technologies, the company that produces the angiography machine used in the study. C. G. Kevil is a participant on a pending U.S. patent (no. 61/003150) regarding the use of nitrite salts in chronic ischemia and has a commercial interest in TheraVasc.

AUTHOR CONTRIBUTIONS

S.C.B., C.B.P., J.D., D.G., P.D., and C.G.K. conception and design of research; S.C.B., C.B.P., S.P., G.K.K., and C.G.K. performed experiments; S.C.B., C.B.P., S.P., G.K.K., and C.G.K. analyzed data; S.C.B., C.B.P., S.P., G.K.K., J.D., D.G., P.D., and C.G.K. interpreted results of experiments; S.C.B., C.B.P., and C.G.K. prepared figures; S.C.B. and C.B.P. drafted manuscript; J.D., D.G., P.D., and C.G.K. edited and revised manuscript; C.G.K. approved final version of manuscript.