Skin Wound Healing: Of Players, Patterns, and Processes

Efficient skin wound healing is a mechanism that most people take for granted as a bodily function. Yet it is this seemingly incidental task of the organism that ensures survival from everyday injuries such as cuts and lacerations. Many people realize what constitutes normal wound healing only when complications arise, for example, delayed healing or chronic wounds. This applies not only to the disrupted integrity of the skin but also to healing processes of all other organs of the organism as well. Since the skin is the most important shielding barrier between the sterile inner body and the outside world rich in pathogens, the evolutionary pressure to optimize efficient healing of skin wounds has always been very high. Moreover, wound healing through both regeneration and repair therefore displays a significant advantage in survival of the living [1]. Particularly in surgical specialties, this bodily function is absolutely crucial, both for the patient and the surgeon, in order to lead the surgical intervention to success [2]. However, this long history of evolutionary selection for optimized skin wound healing questions the extent to which the physiological skin wound healing process can be improved at all.

For effective and thorough repair, many different players and processes must work together. Thus, in addition to wound closure, proliferation, migration, and differentiation of skin cells involved in wound healing, the coagulation and immune systems, the regeneration of hair follicles (HFs), sweat and sebaceous glands, and other cellular changes would also be important for successful wound healing. In both cases, the cells of the skin must produce and establish a healthy microenvironment consisting of an extracellular matrix (ECM) and specific peptides and growth factors, all of which are necessary for promoting wound healing.

In this context, it is in the nature of people and especially of physicians to describe and divide bodily functions and their diseases into players, patterns, and processes. On one hand, this division improves the understanding of physiological processes if the overall process appears too complex and confusing at the beginning. On the other hand, it can also be used to categorize bodily dysfunctions and to deduct specific and stage-appropriate therapeutic options. Particularly in skin wound healing, the division into different phases has always played a major role. Although there are very early accounts of possible wound treatments [3], it was in antiquity that the first attempts were made to describe and categorize wound healing theories, treatment techniques, and their interrelationships [4]. Indeed, the clinical focus centered originally on wound treatment and its improvement for a long time. However, a growing interest was also laid on wound healing theories at a very early stage in history.

In that respect, primary and secondary wound healing, the concept of praiseworthy pus, and antisepsis and asepsis are the most important basic concepts in the course of the history of wound healing. While the concepts of antisepsis and asepsis are relatively familiar to the general public, the more precise distinction of primary and secondary wound healing is known predominantly in medical circles. Nowadays, the concept of praiseworthy pus meets with rejection or incomprehension, and the “ubi pus, ibi evacua” concept has been developed from the theory of “pus bonum et laudabile.” Interestingly, both statements probably originated with Hippocrates. However, these wound healing concepts were lost in the meantime and subsequently had to be laboriously rediscovered, elaborated, and modified. Thus, it was not until the introduction of asepsis in the last third of the 19th century that a significant change in the effectiveness of wound treatment was achieved, with a marked reduction in the number of bacteria in wounds. Up to this finding, the gold standard was held to cause even a purposeful suppuration in wounds to bring them to healing. A more differentiated historical view shows that even before the introduction of antisepsis, anti-inflammatory mineral and herbal wound healing agents were used [3]. Unfortunately, this knowledge was lost as well. In today’s world, the scientist or clinician can take advantage of the many sources of information and is faced with the dilemma of separating correct from incorrect knowledge and good from bad research. This is especially true for the interesting and extensive field of wound healing. In this regard, it is quite interesting to state that publications dealing with wound healing issues begin with nearly identical sentences. Authors usually first describe that the skin is the largest organ of the human being [5‒9], whereas a differentiated discussion can already be started with this statement. The question arises as to whether the size specification refers more to the weight or rather the surface area. If one takes the weight, one may be right. The weight of the skin is about 3.5–10 kg, depending on the size of the person. If the fatty tissue is also included, it can even reach a weight of 20 kg. In terms of surface area, however, the intestine beats the skin by far. While the skin is calculated with an area of about 1.8 square meters, the intestine with its villi reaches a total area of 400–500 square meters. This should be given utmost attention when describing the skin as the largest organ.

Second, as mentioned in the introduction to this article, the statement of the important potential of the skin to heal once injured is a frequently chosen functional description in the introduction [10‒12]. Likewise, wound healing is also readily described as an evolutionary advantage that enables living organisms to survive through regeneration and repair processes [11, 13]. Furthermore, the description of the many functions of the skin as an organ is also popular [5, 7, 9, 10, 13, 14]. To achieve sufficient wound repair, a collaborative and somehow complex interplay of cells and mediators is needed [1, 10]. Some authors start their manuscripts by emphasizing the immense socioeconomic burden of chronic, nonhealing wounds, scars, or burns [9, 15‒17]. Occasionally, this culminates in the brief description of scarring, pointing out that fetal skin and oral mucosa heal without a scar [18‒21]. The most interesting part, though, is that wound healing seems to proceed in different phases, stages, cascades, or sequences [11, 22‒24].

In particular, this article aimed to shed more light on the various classifications of skin wound healing, both in terms of wound healing theory and treatment methods. We present the various underlying concepts of wound repair and discuss the current paradigm. We focus on describing the external, macroscopic processes with the aim to better understand the complex process of skin wound healing and provide sometimes deeper information on individual interesting interactions of cells or factors.

It is crystal clear that the wound healing of the skin requires a complicated synchronization and thus belongs to the most complex processes in the human body. The visualization of the structure of the uninjured skin can already provide some answers at this point. The epidermis is an outer, impermeable layer that protects the body from the rough external threats, containing HFs, sweat and sebaceous glands as sources of epidermal stem cells [25, 26]. It is composed of keratinocytes, Merkel cells, and melanocytes as well as multiple residing hematopoietic cell lines [27]. Resident and engaged cells of the immune system are present in each layer, which constantly monitor the skin for damage and initiate necessary steps to limit pathogenic invasion and defend against it [9, 27]. A survey of the epidermal microenvironment is realized by Langerhans cells, which communicate this information to locoregional lymph node stations to actively generate protective and regulatory host immune responses [28, 29]. CD4⁺- and CD8⁺-T cells, in the form of resident memory cells, remain in the epidermis as cellular remnants of past pathogenic skin challenges, awaiting secondary stimuli [30, 31]. The underlying dermis is a layer rich in ECM, blood vessels, and receptors, delivering the skin with nutrients. Furthermore, it consists of dermal immune cells such as dendritic cells and T-cell subsets, macrophages, mast cells, basophils, eosinophils, and heterogenous populations of dermal fibroblasts. Underneath is the subcutaneous adipose tissue, which serves as an energy source and a constant supplier of growth factors for the dermis to regulate hair growth and mediate antibacterial host defense [32, 33].

Skin wounds in adult mammals usually heal with a fibrotic scar and will not regenerate ectodermal appendages such as HFs or adipose tissue. Surprisingly, new HFs regenerate themselves in the center of large, full-thickness wounds in mice in a process known as wound-induced folliculogenesis [34‒36] or newly called wound-induced hair neogenesis (WIHN) [37]. WIHN is furthermore followed by neogenesis of skin adipose tissue and seems to be mediated by the Wnt-signaling cascade [37]. These wounds show a distinct regeneration pattern, where there is a hairless part from the wound edges and a hairy center with nonpigmented white hairs and underlying regenerated adipocytes and without arrector pili muscles [37‒39]. The critical size defect above which WIHN can proceed appears to be in wounds larger than 1.0 cm² in young mice [39] and larger than 2.25 cm² in adult mice [38]. For wounds that are smaller, the wound area is replaced with pure scar tissue [38]. Regeneration of adipocytes is accomplished by the neogenic HFs entering the anagen phase, which then secrete bone morphogenetic protein to stimulate the conversion of surrounding myofibroblasts into new lipid-containing adipocytes [39, 40]. In the meantime, it has also been shown that repigmentation is possible through the migration of tyrosinase-related protein 2 melanocyte stem cells into neogenic HF [41]. The process further integrates M2 macrophages, which express growth factors important for the regulation of the hair cycle and hair growth, such as fibroblast growth factor (FGF)-2 and insulin-like growth factor (IGF)-2 [42]. However, the WHIN process, and thus to some extent general regeneration, appears to be susceptible to inhibition by mature neutrophils because mice deficient in mature neutrophils have higher WIHN [43].

In this context, it must be mentioned that different body regions have different wound healing properties of the skin. Thus, the skin in the genital area differs significantly from many other regions and also has to cope with completely different tasks than the rest of the human body [44]. As main differences, there is no subcutaneous adipose tissue, no Scarpa fascia, no firm anchorage of the skin with underlying bone or cartilage, and little or no keratinizing epithelium [45]. The skin of the genitalia is also characterized by reduced hair growth on the skin of the scrotum and labia majora and has a moist environment due to mucosal surfaces by mucus-secreting glands [45]. In addition, there is a permanent microbial contamination due to the proximity to the urogenital and intestinal flora which further challenges the immunological function of this skin [45]. Furthermore, bleeding and swelling are described as two major clinical symptoms for injured genital skin. However, if the hematoma and swelling disappear, the wounds heal almost without visible scarring, even after severe trauma [45]. An example of this is the almost invisible genital scarring after male circumcision [44]. The hormonal influence also differs significantly from normal to genital skin. In genital skin, androgen and estrogen receptors are strongly expressed, from which wound healing in the genital area benefits considerably [45]. Intracrine production and increased aromatase activity in genital skin following injury result in increased intracellular bioavailability of estrogens, which stimulate keratinocyte and fibroblast migration and reduce cellular inflammatory response via downregulation of pro-inflammatory cytokines and signaling [46‒48].

As already described in the introduction, Hippocrates was already able to make a simple form of classification of wound healing, based on the type of wound and its healing, and presented a first concept for wound treatment, which still has its rudimentary correctness today. Hippocrates distinguished between simple cut wounds without tissue debris and contamination on the one hand and complicated contaminated injuries with a lot of dead tissue and possibly already inflamed on the other. The simple wounds could be sutured and healed primarily (i.e., primary wound healing/wound healing by primary intention). The latter had to remain open and close secondarily (i.e., secondary wound healing/wound healing by secondary intention). Under certain circumstances, these wounds could still be transferred to primary wound healing by cleaning the wound with boiled rainwater or disinfection with wine. Hippocrates treated secondary healing wounds with boiled sheep’s wool and was able to achieve wound drainage and promotion of granulation. However, he also described the concept of suppuration of wounds, so that dead tissue debris and contaminants are broken down and removed by pus. In addition, suppuration was also divided into the white, creamy, and thus praiseworthy pus as a sign of an intact body defense and the thin or stinking pus as an indication of a weak defense or particularly dangerous bacteria [49]. This documents that physicians in the era of Hippocrates already had an idea of the importance of pus, although they did not yet understand the underlying mechanisms but were able to assess them by their observations alone. They differentiated very precisely that any other form of pus than the one that looked creamy white was considered prognostically unfavorable. Suppuration was considered a useful, natural, and especially desirable process only when the wound could not heal primarily. If signs of inflammation appeared in primarily healing wounds, anti-inflammatory agents from nature were applied. If these remedies were unsuccessful, Hippocrates had to resort to iron/knife, i.e., surgical intervention. In case of failure, he recommended fire or judged the wound to be incurable. As a result, Celsus was able to describe the signs of inflammation (calor, rubor, dolor, tumor), which are still valid today and are taught as such. In Celsus’ time, qualitatively effective dressing materials were of great importance, especially to draw the bad juices out of the wound. In addition, the miasma theory had to be observed to keep “bad air” away from the wound. The Greek term miasma means impurity or bad smell. The miasma theory assumed that pathogens, the miasmas, are found in bad-smelling air and are transmitted directly to humans through the respiratory tract or even through the skin and make them ill. Bandages were also indispensable for this purpose. To reduce the swelling of the wound area, the dressings were wrapped very tightly [49]. The similarity to this treatment concept is nowadays described in the English-speaking world as the RICE concept [50]. It is recommended by physicians for the early treatment of bone injuries or acute soft tissue injuries and stands as an acronym for rest, ice, compression, and elevation [50].

In the context of wound healing, the intensive and timely cooperation of cells and factors is of crucial importance to ensure physiological and thus successful wound healing of the skin. The temporal classification of wound healing from seconds after trauma to months- or years-long (Fig. 1) remodeling processes of the scar facilitates the understanding of the processes enormously [51]. Especially in this approach of classification, the description of an orchestra is often used to illustrate that only the precise choreography and the correct type of cells and regulatory molecules allow for successful wound healing [52, 53]. If this is not the case, dysregulation of the wound healing process occurs, leading to a nonhealing or even chronic wound, consecutively [51, 54‒56]. However, pathophysiology, and here as an example, excessive wound healing in terms of hypertrophic or keloid scar, can also be explained by a defective but still collaborative approach of involved cells and factors. Excessive scarring might be caused by skin injuries, which include trauma such as bite wounds, burns, surgery, vaccinations, skin piercings, acne, or infections [12]. After an initially normal phase of hemostasis, the inflammatory process follows, which initiates wound healing. Here, an aberrant form of wound healing with dysregulation seems to occur subsequently, characterized by continuous local inflammation [57, 58]. Thus, all excessive forms of wound healing involve an initial purulent inflammatory process followed by an upregulated fibroblast function with consecutive excessive deposition of ECM [59] and thus the formation of a pathological form of a scar.

Accordingly, it can be stated that the physiological regulation of skin wound healing depends on a highly complicated temporal sequence, interaction of many cell types and mediators, and a timely resolution of the inflammatory phase. For example, the macrophage represents a major player in the transition from the inflammatory to the proliferative phase [10, 60]. Depletion studies demonstrated that the absence of macrophages in the inflammatory or proliferation phase of wound healing resulted in decreased tissue formation or even hemorrhage. Moreover, progression to the next planned phase failed [61]. In contrast, wound repair could also be shown to be unimpaired without inflammatory cells at all [62]. Fortunately, redundancy in the wound healing process is high, and other cells or mediators can take over functions or signaling without major complications and still lead to successful wound healing [11].

The most widespread concept of skin wound healing today is that of consecutive or overlapping phases [63]. Here, it is left to the respective authors to decide whether they describe three to five or even more phases [1, 11, 63, 64]. The respective research hypothesis seems to be particularly relevant here. In the following, the essential phases with their respective most important functions are described (Fig. 1). For a more intensive study, reference is made to relevant literature and corresponding reviews in which these processes are described in detail [1, 9, 11, 22, 24, 53, 54, 63].

It all starts with the initial three-step phase of coagulation, primary and secondary hemostasis immediately after trauma. Main players in this process are the platelet and the matrix component fibrinogen, produced by hepatocytes [65]. Neuronal reflex mechanisms occur because of vessel wall injury, leading to rapid vasoconstriction by vascular smooth muscle cells in the tunica muscularis. This is mainly activated by endothelin from the damaged endothelium but also by circulating catecholamines from other injured cells [66]. Vasoconstriction in combination with the formation of a platelet-rich thrombus by platelet aggregation and plug formation (primary hemostasis) prevents a potential life-threatening blood loss. This process is further triggered by the exposure of collagen in the subendothelial matrix via G-protein receptors [67]. However, vasoconstriction is only effective for a few minutes before tissue hypoxia, and consecutive acidosis in the vessel wall is leading to its passive relaxation with consequent vasodilation. This is accompanied by tissue edema due to increased vascular permeability, fluid extravasation and might resume bleeding.

For platelet aggregation and adhesion with the extracellular environment, platelets produce adhesive glycoproteins and sphingosine-1-phosphate, which further enhance this process [68]. The platelet plug and fibrin mesh, which is formed by the activation of the coagulation cascade (secondary hemostasis), make the thrombus and start the first signaling cascade for wound healing by appropriate release of cytokines at the same time. Growth factors and cytokines such as epidermal growth factor (EGF), transforming growth factor (TGF)-α and TGF-β, insulin growth factor (IGF), interleukin-1 (IL-1), and, of course, platelet derived growth factor (PDGF) are released. These factors play a significant role in mediating and activating other processes in the subsequent phases of wound healing and in particular in the chemotactic recruitment of immune cells. Within the first hour after platelet activation, the release of platelet factors is most intense. However, activated platelets continue to release these factors for up to 7 days, thus exerting a continuous effect on wound healing and the cell types involved [69].

This event promotes the amplification and recruitment of cells for digestion of nonviable tissue, foreign material, and bacteria (neutrophils, macrophages) and initiates the next phase. This phagocytotic activity is critical for the continued progress of wound healing as wounds with bacterial imbalance will not heal.

For the onset of the inflammatory phase, however, another “starting signal” is required. The wound activates transcription-independent pathways, which contain Ca²⁺-waves, reactive oxygen species (ROS) gradients, and purinergic molecules. Within the first minutes after trauma, the concentration of Ca²⁺ at the wound edges already increases maximally and spreads centripetally toward the wound center [70]. Subsequently, injured cells add the damage-associated molecular patterns (DAMPs), hydrogen peroxides (H₂O₂), lipid mediators, and chemokines necessary for neutrophil recruitment (Fig. 2). It is the rapid release of H₂O₂ that minimizes wound infection by further neutrophil recruitment, activates keratinocyte regeneration, and supports new vessel formation [71].

The subsequent inflammatory phase can then be divided into an early and late phase. The early phase of the inflammatory reaction is particularly determined by neutrophil activity, which can be described by the processes of recruitment, rolling, cell adherence, and diapedesis through the tissues, finally reaching the site of skin injury (Fig. 2). As this cell line constitutes about 50% of all cell types in the wound by day 1 after injury, it will be described in more detail [72]. Before neutrophils can help as first responders, they have to react to “find me” signals via DAMPs, H₂O₂, and other chemoattractants (e.g., CXCL4, CXCL8, CXCL10, CXCL12, CCL3-5) [9, 73]. These signals develop a gradient, which is directly leading the neutrophils into the wound site and onto surface. Once there, neutrophils build a provisional barrier to prevent microbes from invading. At the same time, granulocyte colony-stimulating factor (G-CSF) and CXC chemokines are released and delivered to the bone marrow via the blood circulation, allowing more mature neutrophils from the bone marrow to enter the peripheral blood pool and consecutively to the wound sites. At first, this is a CXCR2-mediated chemotaxis alongside the vessel endothelium and a chemokine gradient extravascular. Thereafter, formyl-peptide receptor-dependent signaling prevails, and neutrophils migrate into necrotic zones through necrotaxis via a formyl-peptide gradient (Fig. 3). Neutrophils can counteract the infectious threat by releasing toxic granules containing, for example, lysozyme, cathepsin G, elastase, and proteases [74], generating an oxidative burst, initiating phagocytosis, and producing neutrophil extracellular traps (NETs) [75, 76]. NETs are chromatin filaments coated with proteases (especially histones) that extend into the extracellular space and are capable of scavenging and degrading exogenous pathogens through the process called NETosis [75, 76]. NETosis can proceed in two different ways (Fig. 4). In suicidal late NETosis, NADPH-dependent ROS are responsible for neutrophil death [77]. Nuclear and granular membrane disruption occurs, and chromatin is decondensed and dispersed into the cytoplasm, where it is mixed with cytoplasmic proteins. Rupture of the plasma membrane finally releases the NETs into the extracellular space [77, 78]. Early vital NETosis occurs without the death of neutrophils, allowing them to continue to participate in subsequent processes such as phagocytosis or chemotaxis [76]. Vital NETosis is an oxidant-independent process that does not release mitochondrial DNA and is activated by a toll-like receptor (TLR)-2-dependent mechanism stimulated by Staphylococcus aureus or TLR-4 by, for example, platelets and DAMPs. However, NET production is not always accompanied by a positive effect. It has been shown that impaired NET production leads to increased WIHN, indicating that neutrophils may even hinder regeneration in the later course of wound healing [43].

In the following, neutrophil activity gradually changes within several days once all bacteria have been removed. To initiate the next phase of wound healing, neutrophils must be eliminated by extrusion and apoptosis processes. Extrusion occurs at the wound surface and is recognized as wound slough. In the course of wound healing, macrophages represent a very crucial cell type, as their depletion leads to significant disturbances in the wound healing process, at least in rodents [79, 80], whereas the increase of monocyte-macrophage numbers resulted in an acceleration of normal and disturbed diabetic skin wound healing in mice [81]. Macrophages phagocytose the remaining cellular debris, thus initiating the late inflammatory phase after tissue injury (Fig. 5). Their number peaks around day three after wounding and decreases to baseline levels by day ten [82]. The appearance, respectively, the differentiation of macrophages from monocytes in the wound healing process is also accompanied by a variety of chemo-attractive agents (e.g., stromal-derived factor 1 [SDF1], CXCL12).

Macrophages themselves also produce important signaling molecules to subsequently activate keratinocytes, fibroblasts, and endothelial cells (ECs). In the beginning, the M1 macrophage is a pro-inflammatory and microbicidal cell type that recognizes and ingests pathogens in intracellular phagosomes, which have a high ROS content and can quickly kill most pathogens [83]. By secreting pro-angiogenic cytokines and VEGF, M1 macrophages further stimulate the initiation of sprouting angiogenesis [84]. As the inflammatory response subsides, the M1 macrophage converts to its anti-inflammatory M2 form, contributing especially to new vessel formation as well as subsequent vessel regression and remodeling [85, 86]. Through a process known as vascular mimicry and expression of the endothelial surface marker Tie2, M2 macrophages participate in vascular anastomosis by fusing branched endothelial vessels and connecting them to the systemic vasculature [87, 88].

Macrophages are further prone to granulocyte-macrophage colony-stimulating factor (GM-CSF) stimulation. In a model of macrophage depletion, it could be recently shown that these wounds demonstrated a significantly prolonged regeneration process. The GM-CSF injection, however, resulted in rising number of macrophages in the blood as well as in the injured skin tissue, accelerating the wound healing process with increased VEGF levels for improved angiogenesis and less scar formation [89].

As the last cell type, T lymphocytes, specifically attracted by IL-1, migrate into the wound. Especially dendritic epidermal T cells (DETCs) play an important role to start into the epithelialization process. Twenty-four to 48 h upon injury, DETCs change their morphology from dendritic to round and get activated by this to release keratinocyte growth factor (KGF)-1 and -2 and IGF-1. This leads to proliferation of keratinocytes in the wound. The importance of DETCs was demonstrated by depletion studies, in which there was a significant delay in wound closure [90, 91].

In the proliferation phase, the wound starts with tissue repair. Due to the formation of a new stroma by fibroblasts, the sprouting of new blood vessels, and the deposition of the ECM as well as collagen synthesis, this phase is also called the granulation phase or the tissue as granulation tissue. The name derives from the granular pattern of the newly formed tissue in histology specimen and in the macroscopic view, which was first described by the British surgeon John Hunter in the 18th century and later characterized in more detail by the French surgeon Alexis Carrel. The formation of granulation tissue consists mainly of activated fibroblasts, which synthesize a new ECM and contract the wound.

A central role in the proliferation phase is played by the micro- and macrovascular system itself and its restoration [92]. Thus, as described earlier, it not only contributes to the initial hemostasis, reduces blood loss, and establishes a provisional wound matrix but it also provides blood clot-derived cytokines and growth factors. The provisional wound microenvironment with its crucial cytokines and growth factors represents the starting point for new vessel formation, thus ensuring the restoration of the blood flow, necessary to provide oxygen and nutrients and to sustain cell metabolism and maintain the healing process. For a better understanding, the phase of angiogenesis can also be subdivided into two phases. While in the pro-angiogenic phase there is an excessive production of blood vessels, the vascular network experiences a maturation process in the subsequent anti-angiogenic phase, which finally leads to a significant reduction in the number of vessels [92].

Local microvascular ECs and, to a much lesser extent, pericytes are mainly involved in the angiogenic processes (Fig. 6). The sophisticated process of new vessel formation by sprouting requires the detection of low oxygen levels in the tissue to activate ECs by hypoxia-responsive growth factors such as vascular endothelial growth factor (VEGF) or PDGF. First, the capillary basement membrane is degraded by enzymes. After this, the ECs begin to proliferate. The foremost ECs, called tip cells, extend filopodia along a pro-angiogenic gradient that indicates the precise orientation of the developing capillary sprout through the ECM (Fig. 6) [93, 94]. The filopodia secrete large amounts of proteolytic enzymes to provide a pathway for the newly formed capillary sprout through the ECM [95]. Tip cells are equipped with many VEGF-A receptors that allow them to sense the precise VEGF concentrations in the ECM and to then align toward the highest gradient [96]. The tip cell is directly followed by a stalk cell (Fig. 6), a proliferating EC that elongates a sprout forming the trunk of a new capillary. If sprouting tip cells converge with each other, they fuse and are subsequently remodeled into tubules that can connect to other tubules or existing vessels, recreating oxygenated blood flow. These newly formed vessels, of course, are still leaky. This allows easier infiltration of immune cells into the wound area until blood flow is fully restored and the blood vessel is sealed by pericytes (Fig. 6). Consequently, if the surrounding tissues receive appropriate amounts of oxygen, the VEGF concentration is reduced and the vascular maturation process is initiated. The maturation and stabilization of the newly formed capillaries requirets the recruitment and attachment of pericytes to the outer vascular endothelium and the deposition of ECM accompanied by shear stress (Fig. 6) [97].

Interestingly, VEGF acts not only on ECs but also on fibroblasts, which in turn can also promote the angiogenesis process in wounds. For example, a recent study demonstrated that overexpression of VEGF in fibroblasts leads to increased angiogenesis and the formation of granulation tissue in the early phase of the healing process [98].

The entire process of the formation as well as the maturation of the vessels does not seem to follow a recognizable pattern at first hand. Though a process could be described in which a ring of circularly arranged vessels is initially formed directly at the wound edge, whose organization is irregular and whose blood flow is inconsistent. As the process continues, this vascular ring at the wound edge contracts toward the center of the wound, leaving radially shaped vessels that supply the interior of the wound and establish a connection with the uninjured skin (Fig. 7). These vessels are more uniform in their degree of organization, and the blood flow can be described as almost physiological [99]. If disturbances occur in the neovascularization process, this might lead to a vicious circle and to wound healing disorders or chronic ulcers, as they typically occur in venous insufficiency, arteriosclerotic diseases, or diabetic foot ulcers.

In the context of improving vascularization, transplantation of three-dimensional microvascular fragments derived from adipose tissue is currently under intense investigation [100, 101]. Microvascular fragments have high angiogenic potential because, based on a randomized mixture of functional arterioles, capillaries, and venules, they can rapidly assemble into new microvascular networks after their transplantation into tissue defects [102‒104]. Therefore, this might also be a new strategy to improve skin wound healing by accelerated vascularization, blood vessel maturation as well as epithelialization [105].

The epidermal layers contain keratinocytes in the form of an epithelium connected by cell-cell junctions, mainly desmosomes. A special ECM, the basement membrane, connects the basal layer of the epidermis downward toward the dermis through hemidesmosomes and focal adhesions. For the epithelization process, cell motility through the mechanisms of protrusion, adhesion, and traction is particularly important. This process already activates a few hours after injury, in which there is a transformation of cobblestone stationary keratinocytes into flat migrating keratinocytes. The change in keratinocyte phenotype is also called epithelial-mesenchymal transition (EMT), which is classified as a type II EMT, occurring during tissue repair [106]. This process describes the transformation of an adherent epithelial cell morphology to a motile mesenchymal cell [106]. Through the process of lamellipodial crawling [107] and shuffling, the first keratinocytes move over the wound site [108, 109].

For this, they must loosen their cell-cell and cell-substratum contacts, which are maintained through desmosomes and hemidesmosomes [108, 109], change their apical-basal polarity, and rearrange their actin cytoskeleton and adhesive structures [110]. The apical-basal polarity of the cells is changed in order to allow the leading-edge keratinocytes to migrate laterally across the wound as preparation to restore the epidermal layer [110‒112]. Especially E-cadherin plays a pivotal role in the process of maintaining lateral cell contacts, adhesion, and immobility of cells [112, 113]. Via the upregulation of vimentin, E-cadherin is downregulated and the phenotype changes to a mesenchymal one, with respective markers and skills [111]. The temporally upregulated expression of integrins increases cell motility even further [114]. Interestingly, EMT does not always have to be entirely complete. It could be shown that along a gradient, an incomplete transition of one and the same cell can take place, with both epithelial and mesenchymal characteristics [106, 110].

Keratinocytes in the “second row” behind are getting activated and will proliferate to secure the supply of cells. The leading row of activated keratinocytes drags themselves over the blood clot-derived fibrin, fibronectin, and vitronectin (lamellipodial crawling) and forward across the newly formed wound matrix. Remarkably, the cells do not migrate centripetally to the wound center but change their shape, break their cell-cell contacts, rearrange themselves, and leave the anterior margin (shuffling) [108]. Once in the wound center, contact inhibition stops the migration process of keratinocytes, and wound coverage is completed. Of interest, this described process holds mainly true for human wound healing. In rodents, wound contraction plays the main mechanism for wound closure and must be considered for respective experimental models and their consecutive results [115].

The healing of wounds, and in particular the formation of the new epidermis, occurs much more slowly with increasing age. This seems to be due to the cohesion of keratinocytes from eccrine sweat glands in aged skin. Although the density of eccrine sweat glands is the same in aged compared to young skin, their activity, however, to form epithelial outgrowths during skin repair is reduced by 50% [116]. This subsequently leads to reduced cell-cell contacts during the repair process, with intercellular gaps becoming larger, and the number of desmosomes being reduced, thinner epidermal repair, and overall delayed wound closure in aged skin [116].

The end point of healing in most clinical settings is considered to be the full epithelialization of the wound. Below, however, the healing continues for months or even years. Interestingly, many studies have shown that especially in this phase of wound healing, many disturbances might lead to an unpleasant ending of wound healing resulting in chronic wounds, hypertrophic or even keloid scar formation [117‒122].

In the final phase, remodeling occurs with the formation of a scar in infant and adult wound healing, whereas only fetal and mucosal wound healing heal without scarring. In genital skin, as already mentioned above, only minimal scarring is apparent [44]. Furthermore, the endometrium tissue also regenerates without scarring [123]. During menstruation, the tissue in the luminal area ruptures and exposes the underlying stroma. This is accompanied by a strong inflammatory reaction. The re-epithelialization and structural re-organization of the tissue is restored within a few days without scarring and is then again able to respond appropriately to endocrine signals [123‒125]. In parallel, regression of the neovasculature, deposition of ECM, and remodeling and degradation of granulation tissue also take place. Collagen type III, which is mainly present in granulation tissue, is partially replaced by the stronger collagen type I by remodeling processes in the ECM. As part of these processes, myofibroblasts synthesize matrix metalloproteinases and their inhibitors, the tissue inhibitors of metalloproteinases, whose very precise balance of action is of crucial importance for the remodeling of the ECM and granulation tissue [126, 127]. If there is a disbalance between matrix metalloproteinases and tissue inhibitors of metalloproteinases, the ECM cannot be modified in a phase-appropriate pattern and a chronic wound may develop [118, 119]. Myofibroblasts then undergo apoptosis, a process that in turn also affects other cells of the granulation tissue and, if absent, may lead to potentially hypertrophic scar formation [128‒130]. Macrophages then subsequently clear the cellular debris and apoptotic cells by phagocytosis, so that at the end of this phase, the wound appears poor of cells and avascular.

The severity of the inflammatory response during wound healing appears to be directly related to the extent of scar formation presented here because an excessive or persistent inflammatory response increases the incidence of hypertrophic scar or keloid formation [131]. IL-6 in particular is involved in the regulation of the fibrotic network of fibroblasts, macrophages, keratinocytes, and vascular ECs and the formation of hypertrophic scars. However, its role is controversial. On the one hand, the activated IL-6 signaling pathway is directly related to keloid formation [132], while other studies showed that fibroblasts express anti-fibrotic genes in the presence of IL-6 [133] and that anti-IL-6 agents promote the decline of excessive collagen deposition and genes involved in pathological fibrosis and scarring [134].

In order to gain an initial overview of (human) skin wound healing, simple classifications are advantageous for understanding and thus deserve to exist. In this context, it also seems important to consider and appreciate the historical development. Despite significant progress in the study of wound healing processes, Hippocrates developed at his time a similar, albeit simple, concept of the processes of wound healing about 2,500 years ago, which in part still has its significance and is based on similar approaches for therapy today. However, the complexity of the underlying biology of skin wound healing takes on a multidimensional configuration upon closer examination, in which new actors are constantly being identified, making the events more precise and comprehensible but also significantly confusing when viewed as a whole. From this point of view, the healing process must be categorized so that the observer does not get lost in the multitude of interacting processes. In view of the steadily increasing knowledge, which includes in parallel the physiological as well as the pathophysiological processes of wound healing, the classification according to function in the sense of consecutive and overlapping phases seems the most convenient and considers the corresponding processes more precisely. Despite that many mechanisms and specific cellular functions in wound healing have been identified, many underlying (patho-)physiological processes still remain unknown. Currently, a substantial part of research activities in medicine is limited to molecular levels, while evidence for therapies currently in use is lacking or newly gained knowledge is quite far from clinical applicability and reality.

The authors have no conflicts of interest to declare.

There is no funding to declare relevant to this article.

Heiko Sorg and Christian G.G. Sorg contributed to the review article conception, design, and data collection and analysis. The first draft of the manuscript was written by Heiko Sorg. All authors read and approved the final manuscript.