The Role of Hemodynamic Shear Stress in Healing Chronic Wounds

Chronic lower extremity wounds, especially those related to diabetes, are a major health care challenge in the United States. Lower extremity wounds affect up to 49 million people worldwide; in the United States alone, the estimated annual cost of chronic wound management is $96.8 billion.^1,2 Problematically, chronic wounds do not always heal, which can lead to an amputation that results in increased debilitation, health care costs, morbidity, and mortality. Up to one-third of the estimated 500 million persons with diabetes worldwide may eventually develop a DFU, which is the greatest risk factor for future DFU development.^3,4 In 2014, the average cost of an amputation among US military veterans was $60 640 per amputation.^3-5 In 2017, overall health care plus lost productivity costs of treating diabetes plus lost productivity in the United States was $237 billion.⁴ Treatment alone costs billions of dollars, and the incidence of diabetes is projected to increase 50% by 2030.^4-6 The mechanisms of a prolonged or nonhealing DFU are addressed in this review, specifically those related to improved blood flow and the associated physiological vascular mechanotransduction and sequelae for healing.⁷ Etiologies of chronic wounds include macro- and microangiopathies, diabetes, venous insufficiency, prolonged vascular closure, and autonomic neuropathy.The pathophysiology of DFUs makes healing difficult for systemic, anatomic, microbiologic, immunologic, molecular, and epigenetic reasons. Nevertheless, compromised blood flow comprehensively exacerbates the above pathophysiologic mechanisms. For example, cellular hypoxia resulting from vascular dysfunction in addition to increased oxygen consumption due to inflammation both affect HIF-1. This transcription factor for approximately 1000 genes responds to SS and regulates oxygen homeostasis by angiogenesis, erythropoiesis, cellular proliferation, migration, and survival, which enhances healing in diabetic wounds. Hypoxia-inducible factor-1α of the heterodimer HIF-1 is both destabilized and inhibited in persons with diabetes.^7-9

Generally, wound healing follows 4 sequential, overlapping phases: hemostasis/coagulation, inflammation, proliferation, and remodeling.^8-10 However, diabetes may complicate, delay, or stall healing. Adequate circulation and normal SSs are vital for healing processes in these 4 phases, but micro- and macroangiopathies in diabetes impair each phase of healing. Diabetic foot ulcers are characterized by prolonged inflammation, persistent infections, bacterial biofilms, impaired dermal and/or epidermal cells in remodeling, and senescent cells (fibroblasts, keratinocytes, ECs, macrophages). Cell populations exhibit impaired proliferative and secretory abilities. Thus, DFU is typically a chronic wound. Cellular senescence is related either to prolonged oxidative stress and DNA-damaged cell cycles or to abnormal metabolism in persons with diabetes.¹⁰ Mesenchymal stem cells that normally engraft into remodeling the microcirculation during healing are deficient and defective in the setting of DFU.^11,12 Healing of DFUs requires not only amelioration of wound causes, but also of systemic and metabolic disturbances, including infection and imbalance of cytokines, growth factors, proteases, mesenchymal stem cells, and other metabolically incompetent cells.^12-16

The authors of this review searched the following databases and source materials for studies on abnormal blood flow with altered endothelial mechanotransduction in diabetic microangiopathy affecting multiple molecular mechanisms of wound healing^16-18: National Library of Medicine, PubMed, ScienceDirect, HighWire Press, and Scopus. This review focuses on wound healing mechanisms that connect physiologic blood flow and SS with chronic wounds such as DFUs during each phase of healing.¹⁸ Healing of DFUs is multifactorial, and healing parameters unrelated to SS are not included in this review. Adequate blood flow and hemodynamic SS at homeostatic levels affect inflammation, coagulation, cytokines, angiogenesis, NO, AR, and MMPs. Each of these biochemical and physiological parameters plays a significant role in chronic wound healing and is reviewed in the context of the sequential stages of wound healing.

Healing of any wound is dependent upon skin perfusion pressure and blood flow to and PO₂ at the wound site; DFU healing mechanisms are not solely dependent on these parameters, however.^20-22 A rheologic component of flowing blood is laminar SS. Normal arterial SS (approximately 15 dyne/cm²) from flowing blood affects healing by reducing the duration of inflammatory responses in both direct and downstream blood vessels.

The mechanotransducive forces of SS on the largest endocrine organ in the body, the vascular endothelium, are critically important to vessel integrity. These forces affect growth factors, differentiation of immature endothelial progenitor cells and embryonic stem cells, autocrine and paracrine healing mechanisms, sequestration of ROS, mitochondrial ATP production, and regulation of approximately 1000 genes. Laminar SS at physiologic levels has antioxidant, antiapoptotic, anti-inflammatory, and antiproliferative effects. Chronic DFUs in patients with diabetes may not heal because of constitutive angiopathy, AGEs, and other factors related to impaired hemodynamics. Lupu et al²³ recently reviewed such forces on ECs in sepsis. The authors of the current study review those forces with respect to DFU healing.

The 3 physical forces on blood vessels that affect EC function are as follows: circumferential (tangential) stretch, oscillatory SS (periodic flow reversal with time-averaged SS approaching zero) and low mean SS, and laminar, frictional SS.²³ The most physiologically important and consequential force in DFU healing is laminar SS, that is, the longitudinal force per unit area from flowing blood that is believed to be mechanically transduced through the glycocalyx, caveolae, integrins, ion channels, G proteins, and membrane lipid rafts. Mechanotransduction is the conversion of physical SS forces on the luminal EC surface into chemical activities. The ECs respond mechanically and quantitatively to flow velocity and blood pressure changes. Laminar SS, which is measured in dynes per cm², is homeostatic, stimulating the release of humoral vasoactive substances, changing multiple genes expression patterns, and changing cell metabolism and morphology.²⁴ Shear stress is not only important in wound healing, but it is also greatly involved in normal vascular remodeling.²⁴ Particular vasculature types (arteries and veins) have specific physiologic ranges of SS, above and below which vessels are dysfunctional.^25,26 Thus, ECs play a key role in wound healing, including angiogenesis in the setting of impaired skin perfusion in DFUs.^27,28 In addition to EC dysfunctional mechanotransduction in diabetes and DFUs, the EC function is also related to vessel phenotype, age, collagen, and elastin.^27,28 Many of the effects of SS on ECs were first characterized in in vitro studies using cultured monolayers of ECs in viscometers, controlling for the fluid viscosity and velocity components of SS.^28,29 Subsequent in vivo studies led to the discovery of flow-related EC gene expressions.³⁰ Novel genes such as eNOS, COX-2, and MnSOD produce enzymes resulting from mechanotransduction of ECs that exert strong antithrombotic, anti-inflammatory, antiadhesive, antiproliferative, and antioxidative effects on associative cells such as leukocytes, platelets, and vascular smooth muscle.^26-28

Collectively, normal laminar SS is vasoprotective by way of the endothelium. These multifaceted functions of the ECs contribute substantially to homeostasis in the circulatory system.³¹ These homeostatic functions are summarized in Figure 1. Proinflammatory and proliferative (atherosclerotic) pathways are downregulated when homeostatic flow is restored in diseased vessels or those with disturbed flow. With respect to DFU, maintenance of normal laminar SS to the extent possible appears to be essential to enable impaired vessels to achieve homeostasis. Blood flow homeostasis is a multifactorial interaction of autonomic, humoral, and EC mechanisms that use both negative and positive feedback systems—the bases for physiologic control.³¹ Negative feedback seeks a set point (eg, adequate blood flow to meet the metabolic needs of the tissues) and then responds to the deficit (eg, ischemia) to maintain that set point. Mechanotransduction of the ECs is essential to maintain adequate blood flow and hemodynamic set points in health and diseases. Understanding the effects of disturbed flow in diabetic blood vessels on ECs can provide insights into healing chronic DFUs as well as potentially lead to the development of other therapeutic interventions.^32-34 To the authors’ knowledge to date, the use of nonpharmacologic means such as hemodynamics to address the ischemic and oxidative stress in DFU has not been attempted. For example, physiologic levels of SS have been shown to induce activation and translocation of transcription factor Nrf2 for cytoprotection, as described below.³³ Induced skeletal muscle exercise by electrical stimulation may have therapeutic value as well.³⁵

Wound healing is a dynamic process consisting of 4 partially overlapping sequential phases: phase 1, hemostasis (coagulation); phase 2, inflammation; phase 3, proliferation; and phase 4, remodeling.³⁶ In phase 1, injured blood vessels are constricted or damaged and a fibrin clot ceases blood flow, thus providing the scaffold for inflammatory cell invasion. Wound coagulation and inflammation phases are interactive and connected in a vicious cycle, with inflammation producing coagulation and vice versa.³⁷ The coagulated core of a wound receives no blood flow; thus, SS has no effect except in the surrounding periwound of uncoagulated tissue. In the initiation of a wound, however, platelet aggregation leading to thrombus in phase 1 is highly regulated by SS.³⁸ The normally transitional inflammatory phase (phase 2) may become chronic in the patient with diabetes, with impaired angiogenesis, reduced recruitment of bone marrow–derived endothelial progenitor cells, and impaired fibroblast and keratinocyte proliferation and migration. In chronic, nonhealing wounds such as a DFU, this inflammatory phase is prolonged by bacterial colonization and biofilm. A DFU with a high bacterial load will not heal. In DFUs, the inflammatory phase is additionally prolonged by vasculopathy of diabetes in which the baseline blood flow to the wound is also impaired. Because the coagulated wound core in phase 1 has no blood flow, it is essential that adequate flow to the periwound area be maintained, since the mechanotransduction of SS on functional endothelium in the periwound is essential for potential healing. The coagulated wound has no SS, but the remaining 3 phases of wound healing are highly dependent on the effect of SS on a large number of cell types, growth factors (eg, PDGF), antioxidants, NO vasodilation, cytokines (especially VEGF), extracellular components, and gene expressions.³⁶

Ischemia, hypoxia, and SS

In acute wounds, the inflammatory phase (phase 2) typically begins immediately after the hemostatic and coagulation stages; this phase lasts approximately 7 days, during which time neutrophils and macrophages are at work.^16,17 Hypoxic and inflammatory wound environments increase the number and type of ROS that also impair ECM proteins, DNA control of cell cycles, and biochemical pathways in persons with diabetes.^12,13 Repeated tissue injury is caused by bacteria and PDGFs such as TGF-β, ECM fragments, while immune cells are constantly being recruited; as such, a proinflammatory cytokine storm persists in DFUs.¹⁵ This inflammatory phase is a self-sustaining cycle that, if not stopped, chronically destroys ECM. It forms a positive feedback loop wherein inflammation promotes coagulation, which promotes inflammation; however, SS breaks this cycle in the periwound tissue.^37,38Figure 2 illustrates this positive feedback loop.

Healing and growth from the periwound serve as the basis for the healing of chronic wounds. Monocytes are recruited within 48 to 96 hours of wounding and transform into tissue-activated macrophages. Chronic hyperglycemia in diabetes, however, (regardless of a current DFU) is associated with low-grade inflammation from excess production of proinflammatory cytokines by macrophages and with ROS from mitochondria and NADPH oxidase. These additional ROS result in impaired antioxidant and autophagic mechanisms.^39,40 Continuous inflammatory recruitment of neutrophils elevates MMPs, especially MMP-8, leading to ECM degradation. Normally, wounds inhibit degradative MMP by protein TIMP. Laminar SS normally upregulates endothelial TIMP-1 by way of TGF-β1 signaling pathways. Therefore, maintenance of physiological blood flow in the periwound would counteract the degrading inflammatory sequelae in DFUs.^40,41 Inflamed diabetic blood vessels impair blood flow to both the DFU and periwound tissues. Diabetic vasculopathy in non-wound tissue with its low SS causes EC dysfunction. Thus, normally functioning ECs subjected to SS would likely enhance wound healing.⁴² Diabetic vasculopathy is associated with enhanced expression of IL-6, VCAM-1, and monocyte chemoattractant protein, as well as decreased production of NO. In persons with diabetes, AGEs sequester existing NO, which also impairs EC function. Furthermore, chronic, ischemic wounds continually release ROS from neutrophils that both damage wound tissues and recruit neutrophils. The conundrum is that inflammation is necessary for healing, but the transition to the proliferation phase (phase 3) cannot occur in the setting of chronic inflammation. Physiological blood flow to the periwound with anti-inflammatory laminar SS (10 dyne/cm²–15 dyne/cm²) and a clean, pathogen-free wound bed must be present to advance from phase 2 to phase 3.

Endothelial SS triggers the transcriptome of hundreds of endogenous, vasculoprotective genes. These include antioxidant genes, phase 2 detoxification enzyme genes, molecular chaperones, and anti-inflammatory costimulating genes.^43-45 Genes for antioxidants include heme oxygenase-1, NAD(P)H:quinone oxidoreductase-1, glutathione S-transferase, catalase, glutathione reductase, thioredoxin reductase, SOD, and others that are all upregulated in EC by physiologic SS. However, such antioxidant expressions are inhibited by inflammatory mediators: cytokines, TNF-α, IL-1β, and PI16.⁴⁶ Catalase, for example, catalyzes the conversion of 2 molecules of hydrogen peroxide into 2 molecules of water and 1 of oxygen. Peptidase inhibitors (eg, PI16) are among the most highly SS-regulated transcripts in humans, discovered during the etiologic investigations of atherosclerosis.⁴⁶ It is important to note that it is laminar SS that upregulates PI16 and stimulates healing, whereas oscillatory SS and cyclic strain on ECs yield an inflammatory phenotype that impairs removal of antioxidants, increases production of ROS, increases permeability of the EC barrier, and favors coagulation. Proteases are essential for normal EC-regulated arterial remodeling, as well as EC attachment, migration, and invasion during the angiogenic phase 3 of wound healing. The SIRT1 protein is encoded by the (so-called anti-aging) SIRT1 gene and is the most conserved mammalian NAD⁺-dependent histone deacetylase. SIRT1 extends life by regulating responses to fasting, caloric restriction, and exercise; in addition, it regulates endocrines, protects against oxidative stress, promotes DNA stability, and decreases age-related disorders such as neurodegeneration, metabolic disorders, and cancer.^47,48 SIRT1 is expressed by ECs in response to normal SS and targets NF-κB and eNOS, reducing oxidative stress and inflammation. SIRT1 is also responsible for autophagy, the cellular self-digestion process that degrades misfolded proteins and damaged organelles and modulates angiogenesis. SIRT1 is reduced in diabetes and cardiovascular disease, whereas its overexpression significantly increases cell viability, decreases cell apoptosis, and reduces proinflammatory cytokines.⁴⁶

Not only do DFUs exhibit wound coagulation and diabetic angiopathy, they are also ischemic because of increased oxygen consumption as well as reduced or lack of oxygen delivery; thus, they are hypoxic. As a result, DFUs are under more severe oxidative stress than normal wounds.^49,50 Recent evidence suggests hypoxia contributes to extended inflammation and delayed healing or nonhealing in chronic wounds.⁴⁹ Typically, adequate blood flow to any wound results in decreased hypoxia and inflammation. Excessive EC proliferation and inflammation at abnormal SS vascular sites promotes the formation of atherosclerotic lesions.⁵¹ This occurs because both inadequate flow and SS change EC behavior toward increased inflammatory signaling, with upregulation of leukocyte adhesion receptors and HIF-1α with neovascularization of plaque. Diabetic hyperglycemia destabilizes HIF-1α (a key regulator of oxygen homeostasis), is dependent on adequate SS,⁵² and affects hundreds of gene expressions for angiogenesis (eg, VEGF), cell proliferation, migration, and survival (eg, FGF-2).^8,53 Hyperglycemia increases NF-κB and the cytokine TNF-α. Hyperglycemia and TNF-α activate NF-κB, resulting in apoptosis of vascular ECs.⁵⁴ Hyperglycemia with increased levels of ROS in diabetes destabilizes and represses HIF-1α. Increased levels of ROS in DFU, coupled with reduced antioxidant defenses, exacerbate diabetic redox imbalance, leading to tissue damage and necrosis. Levels of the free radical scavenger glutathione are reduced in the wounds of mice and humans with diabetes.⁴⁰ Local overexpression of HIF-1α by gene transfer has resulted in improved wound healing in diabetic mice.^39,40 Endothelial cell function is also important in redox balance for chronic wound healing.

Additionally, regarding redox balance in phase 2 of chronic wounds, transcription factor Nrf2 is translocated into EC and regulated by SS. Normally, Nrf2 is bound with KEAP1 and maintained in its inactivate state; both Nrf2 and KEAP1 are critical for the maintenance of redox, protein, and metabolic balance.⁵⁴ Nuclear factor–erythroid 2–related factor 2 is constitutively expressed in all tissues and has been called “the gatekeeper of species longevity.”⁵⁶ The most widely accepted theory of aging, which was first proposed by Harman⁵⁷ in 1956, is that aging is essentially due to oxidative stress. When subjected to oxidative stress (eg, hydrogen peroxide), activated Nrf2 enters the nucleus to bind with the transcription factor ARE.^39,53,58 Endothelial cell transcription factor KLF2 is the master regulator of endothelial homeostasis, anti-inflammatory and antithrombotic properties, and vascular tone. Transcription factor KLF2 is increased by normal SS and is necessary for the nuclear localization of Nrf2. The combined actions of KLF2 and Nrf2 constitute approximately 70% of SS-induced EC gene expression.⁵⁹ Nuclear factor–erythroid 2–related factor 2 is essential for the ARE-mediated induction of genes such as heme oxygenase-1. Of note, the transcriptional activity of Nrf2 in aging-related oxidative stress is reversible by moderate exercise training.⁶⁰ Oxidants from chronically inflamed DFUs stimulate the production of Nrf2 in ECs that are only subjected to physiologic levels of SS (12 dyne/cm²).⁵³ Transcription factor KLF2 induces both anti-inflammatory and anticoagulant proteins. The combined metabolic balance by the KLF2/Nrf2/ARE pathway is SS-regulated and is the most critical pathway to regulate Nrf2 and the antioxidant defense system in chronic wound healing.^33,58 This KLF2/Nrf2/ARE pathway is summarized in Figure 3.

Nitric oxide, AR, and SS

Nitric oxide is constitutively expressed by ECs and is primarily known for its vasculoprotective and tissue defense roles in vasodilation, coordinating metabolic energy demand, and apoptosis. Nitric oxide is substantially modulated by endothelial SS.^61,62 In healthy vessels, SS controls the amount of NO produced. Increased SS relaxes vascular smooth muscle to accommodate increased blood flow with reduced wall stress and tension. Lesser-known multifaceted roles of NO include its anti-inflammatory, antioxidant, antibacterial, and antithrombotic effects that assist in the healing of chronic wounds.^62,63 Nitric oxide–deficient states, such as in the hemostatic wound and the inflammatory phase in the periwound, are characterized by oxidative stress, EC dysfunction, inflammation, vascular disease, and cellular senescence. Mitochondria, which are the main intracellular sources of ROS and the sites that affect cell physiology and death, are the primary cellular targets for the antioxidant effects of NO.⁶² Additionally, NO is involved in dysfunctional glucose utilization in patients with diabetes. Aldose reductase, the first and rate-limiting enzyme of the polyol pathway, is inhibited by NO (Figure 4).

Aldose reductase in the polyol pathway converts NADPH to reduced NAD, the substrate for producing superoxide anions. Furthermore, NO availability is important for reducing ROS and transitioning to the proliferation phase (phase 3) of periwound healing.

Shear stress serves to delocalize eNOS, which is the basis for tissue remodeling and angiogenesis.⁶⁴ An in vitro study showed a twofold increase in EC migration and ring formation in angiogenesis under SS of 15 dyne/cm².⁶⁴ The same study reported on an in ovo artery ligation model in which SS in the form of one half and complete blood flow block for 30 minutes reduced angiogenesis by 50% and 70%, respectively. Adding an NO donor resulted in a twofold recovery of angiogenesis.

Overall, murine and canine models have shown that beneficial cardiac adaptations from exercise training were mediated by NO, resulting from increased neuronal NO synthase.^65,66 The remodeling phase (phase 4) is mediated by NO at the vascular and cardiac levels.⁶⁷ The findings of this and similar studies led to the concept of exercise-like induced medicine, in which therapeutic alternatives to exercise are identified to treat patients who are physically unable to exercise. For example, patients with a DFU, amputation, sepsis, or other debilitations might benefit from bodily generated NO with its very short in vivo half-life.

Nitric oxide is considered a wound healing agent because of its ability to both regulate inflammation and eradicate bacterial infections.⁶⁸ Once again, shear-induced mechanotransduction is responsible for endothelial homeostasis by regulating molecular, cellular, and vascular control of NO and ROS.⁶⁹

With normoglycemia, glucose is predominantly phosphorylated into glucose 6-phosphate and enters the glycolytic pathway. In diabetes, there is accelerated glucose utilization by AR in the polyol pathway, which is implicated in the pathogenesis of secondary wound complications.^70-72 Aldose reductase, a key enzyme in the polyol pathway, catalyzes NADPH-dependent reduction of glucose to sorbitol and production of ROS, including lipid peroxidation with production of highly toxic, most abundant 4-hydroxynonenal. Depletion of NADPH compromises antioxidative defenses. Hyperglycemia significantly activates AR production, causing dysfunctional ECs that are essential in wound healing. Aldose reductase catalyzes the reduction of glucose to sorbitol, but AR is inhibited by NO. Such inhibition of AR prevents oxidative stress–induced activation of redox-sensitive transcription factors NF-κB and activator protein 1, which control multiple cellular processes of differentiation, proliferation, and apoptosis of ECs.^57,73,74 Diabetic angiopathy is associated with decreased NO bioavailability and thus, reduced inhibition of AR, resulting in hyperglycemia, increased production of inflammatory ROS, and impaired healing in DFUs.⁷⁵ The AR mechanism in inflammatory diseases has been studied for 50 years; currently, researchers are investigating the possibility of pharmacologic inhibition of AR.^76-79 Several AR inhibitors have been developed and clinically tested but have proved inefficient.⁷⁴

Complications of any prospective, homeostatic-therapeutic use of NO in healing DFUs include diabetic vasculopathy and the milliseconds half-life of NO. This vasculopathy includes disruption of cellular integrity with cytoskeletal misalignment and osmotic stress; damage of mechanotransduction elements in EC, resulting in reduced beneficial SS and NO production; and increased utilization of NADPH, with resulting depletion of reducing equivalents, otherwise necessary for the detoxification of increased mitochondrial superoxide production.^63,64 Endothelial cell dysfunction resulting from a glucose level that is either too high or too low contributes to accelerated atherosclerosis, myocardial infarction, stroke, and secondary complications of diabetes such as cataractogenesis, retinopathy, neuropathy, nephropathy, and chronic wounds. Endothelial cells become highly permeable in hyperglycemia because of the osmotic gradient, and solutes leak through the vascular wall and impair transcapillary exchange and tissue healing.⁸⁰

A summary of the inflammatory phase of wound healing (phase 2) and the role of ECs and SS in the production of NO and its effects on AR and the polyol pathway is shown in Figure 4. Chronic wounds (including DFU) are in a state of prolonged inflammation and can remain unresolved indefinitely, affecting approximately 2% of the US population.^81,82 Such wounds cannot progress to phase 3 of healing. Chronic inflammation with ROS production continually kills cells such that MMP-mediated phagocytosis is overwhelmed, and infiltrating neutrophil, macrophage, and lymphocyte production of ROS outweighs their removal. Furthermore, frequent episodes of hyperglycemia lead to increased levels of AGEs, which in turn lead to high concentrations of superoxide anions and reactive nitrogen species, thus exacerbating phase 2 in DFUs, which oxidizes DNA, proteins, and lipids.⁷⁷ A limited inflammatory phase is necessary to progress to the proliferative phase, because the inflammatory phase accelerates the removal of dead tissue and pathogens. Excessive ROS prolongs the inflammatory phase, however. The repetitive trauma in DFUs, which is worsened by hyperglycemia and oxidative stress, damages and enlarges periwound blood vessels. Enlarged veins reduce venous return from the lower extremities, causing increased interstitial pressure and reduced microcirculatory exchange, thus exacerbating the ischemic wound. Inhibition of AR has been shown to prevent inflammatory signals from cytokines, growth factors, endotoxins, high glucose, allergens, and autoimmune reactions both in vitro and in vivo.⁵⁴ Of note, restoration of physiological levels of blood flow and SS has only been indirectly studied by evaluating the conjunctive use of negative pressure wound therapy. Restoring homeostatic levels of venous and arterial SS has not been studied as a possible means of managing DFUs.⁸²

Advanced glycation end products and SS

The health and healing effects of normal SS on hyperglycemia, hypoxia, elevated ROS, and MMP on DFU have been discussed herein. With diabetes, the increased flux of glucose through the polyol pathway increases AGE and RAGE.⁸³ This reaction is believed to be important in the pathophysiology of diabetes, initially for atherosclerosis, and recently for DFU.^84,85 Activation of RAGE in diabetes triggers proinflammatory signaling with generation of ROS; activation of NF-κB; upregulation of VCAM-1, MMP, and/or TIMP; and increased platelet adhesion and chemokines that guide leukocyte recruitment.^86,87 Hyperglycemia-induced ROS increases the expression of RAGE and creates another positive feedback cycle that produces more RAGE, thereby sustaining AGE activity.⁸⁸ Both AGE and pathologic SS synergistically lead to ROS production and inactivation of antioxidants such as SOD, glutathione ester, and α-tocopherol.^57,77 It has been shown that SS normally regulates RAGE expression and inflammatory responses in ECs.⁸⁹ In the study by DeVerse et al,⁸⁴ endothelial RAGE expression was elevated by disturbed blood flow in the aortae of healthy swine. Receptors for AGE were downregulated fourfold in human aortic ECs exposed to normal SS (15 dyne/cm²). Additionally, using a diabetes-induced metabolic stress model stimulated with TNF-α and RAGE ligand, DeVerse et al⁸⁴ demonstrated that the resulting increased ROS, VCAM-1 expression, and NF-κB were reduced under normal SS. This relationship between AGE-SS and NO was tested in vivo, with exercise-trained diabetic rats in a study by Delbin et al,⁸⁵ who showed that SS induced by exercise training fully restored the eNOS protein expression, NO production in tissues, and ROS protection. Advanced glycation end products were not normalized by exercise training.

During the inflammatory phase 2, healing is prevented if continuing inflammation prevents progression to phase 3 (proliferation). Prolonged inflammation allows the onset of opportunistic infection and biofilms that increase inflammation. Diabetic inflammation and its associated vasculopathy prolong phase 2. Sequelae of AGE and RAGE result in EC dysfunction, thereby impairing a key homeostatic mechanism for healing—blood flow and SS. Dysfunctional ECs are a common feature of rheumatoid arthritis, systemic lupus erythematosus, chronic renal disease, and diabetes. Vasculoprotection appears to be a therapeutic avenue for cytoprotection in systemic inflammatory diseases.⁸² Administration of anti-inflammatory biomolecules has not proved to be therapeutically effective; however, the restoration or maintainence of physiologic levels of SS has not been clinically evaluated.

Cells involved in phase 3 of wound healing include neutrophils, monocytes, macrophages, keratinocytes, fibroblasts, T cells and B cells, mast cells, and ECs that are involved in the production and circulation of cytokines and growth factors.^88,89 This proliferation phase of  healing is characterized by granulation, reepithelialization, and vascularization, and represents the most important repair stage in DFU healing.^90-92 Excessive levels of proteases, cytokines, ROS, and senescent cells prolong or prevent progression to the proliferation phase. Elevated proteases, for example, break down the ECM and attract more inflammatory cells. Monocytes transform into macrophages that produce proinflammatory cytokines IL-1β, TNF-α, TGF-β, VEGF, and insulin-like growth factor. Microvascular ECs are the primary parenchymal cells that participate in wound angiogenesis; in addition, they signal more than 200 genes related to SS.^2,25,93 Vascular endothelial growth factor (upregulated by SS) is the master regulator of angiogenesis during phase 3 both in the development and repair of vessels and cytokine interactions (eg, PDGF and SS co-regulate splitting [branching] angiogenesis in skeletal muscle by limiting VEGF-induced endothelial proliferation).^94-96 If inflammation is prolonged, proliferation will not occur, angiogenesis will be insufficient, and wounds will become chronic.⁹⁷ Local ischemia due to microvascular complications in diabetic vasculopathy exacerbates wound healing by decreasing microRNAs that affect epithelialization, inflammation, fibroblast and keratinocyte migration, and angiogenesis.⁹³ Per the US Wound Registry, of the 6.5 million patients annually with chronic wounds, more than one-third will not heal and cost $96.8 billion annually.² Of those wounds that healed, healing required 105 to 230 days, at a cost of $25 billion annually.⁹⁴

Several signaling pathways (VEGF-A, Notch, BMP, PDGF, angiopoietin-1 and -2, S1P, and SS) are known to regulate angiogenesis.^97,98 Periwound ECs escape microvessel basal laminae and migrate toward the angiogenic stimuli in the wound, then sprout to form new vessels. Shear stress modulates this EC invasion into 3D collagen matrices, and this appears to be regulated by intracellular protein P311 and the PO₂ gradient between the wound and periwound.^96-98 The most understood mechanism of vessel growth into wounds is sprouting angiogenesis, in which special endothelial tip cells sense a VEGF and oxygen gradient and then migrate toward its source to proliferate into the adjacent wound; after which, proliferating stalk cells form new vessel lumina.^98,99

Proliferation is dependent on declining inflammation with adequate blood flow in the periwound area. Normally, proliferation begins approximately 3 days after acute wounding with fibroblasts, collagen, and ground substance forming a tissue scaffold that lasts for 2 to 3 weeks. Angiogenesis occurs within the granulation scaffolding by EC proliferation, migration, and branching off from existing vessels in the periwound into the wound proper.^100,101 In addition to circulation of local cells, other cells, including mesenchymal progenitor cells from bone marrow (ie, BMP), collectively support new vessel formation.^95,102 Angiogenesis is highly dependent upon laminar SS for sprouting in both genesis of replacement vessels (angiogenesis) and new vessel development (vasculogenesis).¹⁰³ Sphingosine 1-phosphate is a biologically active lysosphingolipid, deposited by activated platelets, that acts on ECs to strongly promote angiogenesis under normal and pathologic conditions.^104,105 Sphingosine 1-phosphate acts synergistically with SS, MMP, and growth factors to cause angiogenesis. More recently, in a full-thickness wound defect model in mice, Zhou et al⁹⁶ proposed that as ECs secrete VEGF and P311, it is the P311 that causes VEGF to alter EC responses and angiogenesis. The P311-VEGF/VEGFR-ECs response is likely a mechanism of angiogenesis. The TGF-β pathways for vasculogenesis and angiogenesis are also regulated by P311 in wound healing. Thus, it may be that P311 regulates angiogenesis by an SS–P311–TGF-β–VEGF/VEGFR signaling pathway. Finally, S1P regulates the quiescence of lymphatic vessels by enhancing VEGFR signaling, leading to sprouting lymphangiogenesis. Both oscillatory and laminar SS are also necessary for complementary lymphatic vessel development.⁹⁶ New tissue growth is also dependent on tissue oxygen tensions of 20 mm Hg, which confirms the dependence on angiogenesis and blood flow from the periwound into the wound. Blood flow in the periwound and SS are clearly necessary for phase 3 healing of chronic wounds.

There are no fully proven mechanisms controlling the transition from the inflammatory to the proliferative phase of wound healing.¹⁰⁴ Extracellular matrix formation and infiltrating angiogenesis characterize the portion of the proliferative phase that involves blood vessel formation. Inflammation remains the limiting factor, and its persistence is why some wounds do not heal.⁹⁷ Although healing phases overlap and many cell types and signaling molecules are involved, the presence of both angiogenesis by vessels containing functional ECs emanating from the periwound and vasculogenesis is the key indicator of phase 3 healing.¹⁰⁵ It has been suggested that ECM and MMP activity are directly regulated by SS.¹⁰⁶ Reepithelialization in wound closure is also stimulated by multiple signals (including NO, VEGF, PDGF, and FGF) activated by ECs in existing vessels by SS. Additionally, BMP signals marrow-derived endothelial progenitor cells that orchestrate vasculogenesis and vessel calcification responses to SS.^104,107

The remodeling (ie, maturation) phase (phase 4) is never attained in wounds with uncontrolled inflammation.⁹² Shear stress is highly involved in remodeling because the robust angiogenesis in phase 3 tapers off in phase 4.¹⁰⁸ With sufficient blood supply restored, the microenvironment can support epidermal and dermal cell migration while fibroblasts proliferate within the wound to rebuild the ECM.¹⁰⁹ Remodeling, which occurs over months or years, consists of regression of neovasculature, periodic deposition into the ECM, epithelialization, and conversion of granulation into scar tissue.⁹¹ Matrix metalloproteinases degrade and reconstitute the ECM. Also, TIMP eventually blocks the MMP. Macrophages become fibrinolytic and remove excess ECM, debris, and apoptotic cells. It has been suggested that TGF-β, TNF-α, and FGF-2 stimulate apoptosis when epithelialization is complete and the wound is closed.¹¹⁰ Angiogenic vessels from phase 3 remain leaky, with loose cell-to-cell adhesions and proliferations into granulation tissue; however, the SS-related control of growth factors provided by these vessels is vital in phase 4.¹¹¹ During remodeling, neovessels are pruned (caused in part by SS) into stable, mature, well-perfused, normal homeostatic vasculature.^108,109 Meanwhile, SS on ECs resumes its normal mechanotransductive role in regulating blood flow with homeostatic transcriptional effect on hundreds of genes. Multiple negative feedback control mechanisms related to normalized SS are resumed, such as: (a) hypoxia response pathway; (b) inhibition of Sprouty-2 (an intracellular protein that downregulates the SS effects of VEGF on EC proliferation); and (c) production of vasohibin that impairs VEGF.^88,110 The wound is no longer hypoxic, and the core returns to normoxia with normal oxygen demands. This signals a maintenance stage, a return to pre-wound quiescence, and the end of angiogenesis. Oxygen levels are tightly controlled in all stages of wound healing; this is an important role of SS in normally functioning vessels.¹¹¹

The aforementioned description relates to normal wound remodeling, which is highly dependent on angiogenesis followed by pruning into normalized vasculature. Diabetes impairs progenitor cell recruitment, proliferation, and growth factor release, whereas SS increases all these processes. Diabetic vasculopathy and hyperglycemia cause an altered angiogenic state, with dysfunctional EC and impaired SS mechanisms. The delicate balance of vessel growth, proliferation, maturation, and quiescence is impaired, with disturbances in vascular integrity as well as the presence of apoptotic and detached ECs flowing in the bloodstream.¹⁰⁹ Diabetic micro- and macrovascular disease with dysfunctional SS is the basis for nonhealing wounds stalling and not achieving phase 4. Additionally, small resistance arteries have a reduced capacity to adapt in response to chronic changes in blood flow, and SS-mediated remodeling of small resistance arteries is impaired.¹¹¹ For example, diabetic retinopathy involves excessive, uncontrolled angiogenesis leading to microaneurysms, hemorrhages, and vascular edema.

Laminar SS on vascular ECs is involved in both normal cardiovascular homeostasis and 3 of the 4 phases of wound healing. Diabetes, which is the cause of most chronic wounds in the United States, causes multifactorial DFUs that resist healing because wound debris–related inflammation persists within ischemic periwound tissues, with vasculopathy and dysfunctional ECs. Chronic DFUs colonize skin microbes that reproduce in wounds, thus prolonging inflammation. As a result, current therapeutic efforts to heal DFUs by applications of 1 or 2 missing healing components (eg, growth factors) have not been successful. It seems as though a broad array of the healing mechanisms upregulated by normalized SS in periwound tissues would enhance healing in chronic wounds such as DFUs, venous leg ulcers, and pressure injuries. Appropriately designed studies in which normal SS is restored may aid in the development of therapies for managing DFUs and other chronic wounds.