Relevance of Animal Models for Wound Healing

Animal models and in-vitro assays have become indispensable tools for researchers in nearly every scientific discipline. In product development there is a need for translational research to obtain data that can lead to sound clinical trials and ultimately, improved wound care. This process is usually performed in a stepwise fashion starting with in-vitro testing, preclinical, and then clinical evaluations (Figure 1). In-vitro studies help determine which concentrations may be effective in-vivo and determine whether certain products are effective on various cell types (eg, fibroblasts and keratinocytes). The next step is to examine the effect of the product’s use in an animal model(s). This facilitates investigation of the product in the presence of wound fluid, blood, immune cells, proteases, etc., which can have an effect on the activity of the active agent. Many in-vivo animal studies initially investigate the safety and/or irritancy of the product. It is important to be sure that these agents do not have a toxic effect on tissues. Efficacy animal trials are conducted after the safety studies are completed. This eventually allows the product to be evaluated in human trials.
Although definitive studies conducted on human subjects are needed, such studies present several practical, ethical, and moral concerns. For example, in order to examine wounds histologically throughout the entire healing process one must biopsy a human subject at multiple time points, which is impractical. Furthermore, ethical considerations prevent the intentional infection of a wound on a human or the use of an untreated control subject. Some of the practical difficulties lie in obtaining enough subjects with similar or identical situations to conduct well controlled studies. Another complication to factor in with human trials is compliance (eg, subject’s level of cooperation, ability to understand and follow instructions). The above difficulties have led researchers to develop multiple in-vitro and in-vivo models that attempt to mimic or reproduce human conditions.
In-vitro assays. In-vitro assays are great for examining the effect of agents on particular cell types. They are relatively inexpensive, fast, and convenient for the researcher. In addition to providing useful results in a short time, they possess an obvious humane appeal since they usually do not involve the use of animals or humans.¹In-vitro assays are useful in wound healing research for determining the possible effectiveness of various treatments, particularly antimicrobial and healing enhancing agents. Another noteworthy attribute of in-vitro testing is the ability to screen multiple agents or samples simultaneously. Assays can aid in the early detection of antimicrobial resistance among pathogens and determination of minimal inhibitory concentrations (MIC), and allow for highly specific control over the experimental conditions. However, it is difficult to simulate a “real world” application. Although some variables such as pH, salinity, and temperature are easily controlled, in-vitro assays are incapable of completely reproducing biological conditions (eg, immune responses, healing) and diseases, such as diabetes.²
In order to approximate in-vivo experiments, in-vitro assays have been developed that incorporate some variety of cell or tissue system. Wound closure studies have been conducted on single cell monolayer systems. These assays use a cell monolayer culture on the surface of a plastic plate, which can be disrupted much like the skin. Monolayer cells are capable of migrating into the damaged region much like the healing tissue of an animal.^3,4 Similarly, 3-dimensional cell cultures grown on a type I collagen or fibrin framework can be used to more closely approximate normal wound physiology in-vitro.^5,6 Organ cultures such as skin explants, can also be used for in-vitro assays. Although, these skin explants may best approximate the animal models, they can be problematic. It is difficult to maintain viability in organ cultures and the assay conditions vary from donor to donor. Additionally, modification of the matrix or cellular composition is impossible. To overcome these drawbacks, tissue explants from the same donor must be used for any given experimental series. This limits the number of samples and conditions a researcher can modify resulting in limited data.^7,8

The number of studies that use animal models for wound healing has been increasing recently. This is not surprising considering the number of new products that are introduced each year. Interestingly, the number of articles in PubMed pertaining to animal wound healing models has almost doubled every 10 years since 1980 (Figure 2). There are numerous types of in-vivo models available, and each has unique benefits and disadvantages.
In-vivo models: small mammal wound healing models. Rodent and small mammal models of wound healing have emerged as the model of choice for many researchers. This type of study is beneficial to wound research for multiple reasons. Small animals are inexpensive, easily obtainable, and require less space, food, and water.⁹ Additionally, they often have multiple offspring, which develop quickly allowing experiments to proceed through multiple generations. Small animals usually have accelerated modes of healing compared to humans, thus experiment duration lasts for days, as opposed to weeks or months in human experiments. Some small mammals can easily be altered genetically and provide a wound model capable of approximating defective human conditions such as diabetes, immunological deficiencies, and obesity.¹⁰ Another advantage of small mammal models is their ability to serve in experiments where death is an endpoint, as is some cases of bacterial or viral infection.
Small animals provide a multitude of model choices for various human wound conditions. Some models have been developed to investigate the mechanistic particulars of certain aspects of healing. For example, the rabbit ear chamber^11,12 and the Algire chamber (a transparent plastic window placed in the dorsal subcutaneous tissue of a mouse)¹³ have been employed for visualizing vascularization and measuring angiogenesis. Likewise, various superficial wound models use techniques that separate the different layers of the skin in order to evaluate epidermal regeneration and matrix production. Superficial wounds are often made on the back of the animal using either an electrokeratome or a blister technique. Partial- or full-thickness wounds can be used to recreate more accurately a clinical situation where tissue damage occurs in conjunction with tissue loss. Either excising the tissue or obtaining a punch biopsy can be used to generate these wounds. Wounds of this nature usually heal by re-epithelialization, dermal reconstitution, and wound contraction allowing an analysis of any phase of healing or of the entire healing process. This model is frequently used on the backs or ears of animals and results in reproducible wounds, even on different types of animals. The ear models provide cartilage as the wound bed that prevents healing by contraction. The healing process in these models can easily be followed by histological analysis of the wounds.⁸
Despite the convenience and frequent use of small mammals in wound healing research, there are some limitations. The use of small animals limits the number of wounds that can be investigated and requires a greater number of animals. In some situations, such numbers lead to loss of the economic benefit of using small animals. Additionally, securing bandages or treatments to small animals may be problematic due to the nature of their skin and fur.
Small mammal models also have several anatomical and physiological differences. Small mammals have a follicular pattern and hair growth cycle, which is different in comparison to humans and tend to be covered in dense fur. They have relatively thin epidermal and dermal layers and are considered “loose-skinned.” Although some hairless and tight-skinned mouse models exist, skin thickness and healing mechanisms differ from those in humans. The epidermis is extremely thin and consists of only a few cell layers where the hair follicles are empty, and the subepithelial layer is rich with bundles of skeletal muscle fibers.¹⁴ Particular problems arise in animals with a subcutaneous panniculus carnosus muscle, which aids wound repair by contraction.¹⁵
In-vivo model: porcine wound healing models. Although rat, mouse, rabbit, and guinea pig wound models exist, swine skin is the most similar to humans and has been shown to be an excellent tool to evaluate wound healing therapies.^9,16–18 Porcine skin is structurally similar to human skin with similar epidermal thickness and dermal-epidermal thickness ratios.¹⁹ Pig skin and human skin share similar patterns of hair follicles and blood vessels. Biochemically, pigs contain dermal collagen and a dermal elastic content that is more similar to humans than other commonly used mammals.²⁰ Additionally, pigs and humans have similar physical and molecular responses to various growth factors.²¹ Unfortunately, pigs have a significant cost disadvantage to smaller animals and because the amount of wound contraction varies depending on where the wound is made, strict standardized procedures must be used.^7,22 Pigs also grow quickly and can become difficult to handle should the study continue past a few months.

In 1979, Mertz and Eaglstein²³ published a novel porcine wound model that evaluated the effect of a polyurethane film and a topical steroid on partial-thickness wounds. They were able to observe that the polyurethane films significantly enhanced epithelization while the steroid retarded healing. They used a salt-split technique that separates the epidermis from the dermis (Figure 3). Since the initial study, they have examined hundreds of compounds using this technique. One advantage of the salt-split technique is that it takes into account the entire wound surface. In order to withstand the separation technique, the specimen needs to have a mature epidermis (eg, more than 5–7 cell layers thick). Histologically, the wound may have 1 cell layer covering the entire surface and would be considered healed (completely epithelized), whereas with the salt split technique, the wound is considered not healed. An advantage of histological analysis is that it allows the researcher to observe dermal effects (eg, white cell infiltrate) and measure the thickness of epidermis. However, it should be noted that during histological evaluation of partial-thickness wounds that the number of hair follicles present may give the impression of a wound being more epithelized when compared to a section of the same wound where no follicles are present. This is because the epidermal cells migrate not only from the wound edges but also from the epidermal appendages.
A porcine model can be used to study various wound types, such as partial- and full-thickness wounds, second-degree burns, ischemic wounds, incisional wounds, and various types of laser induced wounds or treatments. Partial-thickness wounds are made with a specialized dermatome to create a uniform rectangular wound. Burn wounds are made using a specially designed brass rod that is heated in boiling water and placed on the skin of the animal. Our ischemic wound model is a modified flap model where a flap of tissue is separated from the animal using a scalpel. The underlying subcutaneous tissue is covered with a polyurethane film and the flap folded over the barrier since the tissue will revascularize if allowed to encounter the dermal base.¹⁰ The tissue is then separated from nutrients and becomes necrotic. These wound types are used to study the effect on healing by various agents, wound dressings, or debridement techniques. In addition to the sodium bromide salt split technique, punch biopsies of the wounds can be collected and genetic material (usually RNA) purified. Using isolated RNA in combination with RT-PCR allows for analysis of the expression of various factors known to be indicative of the 3 phases of wound healing (inflammation, proliferation, and remodeling).²¹
Inoculating the wounds with known amounts of pathogenic bacteria can be used to test for antimicrobial agents. The various treatments are applied with different frequencies and wound bacteria recovered in a time dependent manner. In wounds, bacteria can exist in a free-swimming state or can be encapsulated within an extracellular polysaccharide (EPS) matrix.²⁴ These two states are known as planktonic and biofilm associated, respectively. Bacterial biofilms have been shown to be an important source of antimicrobial resistance and virulence for pathogens.²⁵ It is sometimes useful to analyze them separately because of the bacteria’s different susceptibly to agents in these states. When separation is necessary, wounds are first flushed to recover the planktonic bacteria (Figure 4) and then scrubbed to recover the biofilm bacteria (Figure 5). The two recoveries are then cultured on solid medium and colonies counted to determine viable colony forming units (CFU). When a total bacterial count is more desirable, the wounds are only scrubbed, and both states recovered together. This technique allows researchers to analyze the effects nondiscriminating antimicrobials, biofilm disrupting agents, and agents capable of specifically targeting bacteria in either state.²⁶ Wound isolated bacteria can also be preserved or extracted for genetic material in order to conduct molecular analysis. Bacterial RNA can be used do determine which virulence factors, if any, that the various treatments are affecting.
 

Conclusion
When attempting to use or develop a model, the benefits and disadvantages must be considered carefully, since a model attempts to imitate some other situation. In-vitro assays can be useful since they are quick, relatively inexpensive, and can be used to screen a wide variety of conditions or samples simultaneously. Unfortunately, in-vitro assays are incapable of replicating all of the factors involved in complex processes, such as wound healing. Generally, animal models approximate human wound healing studies better than in-vitro assays. However, the choice of animal can be difficult since its ability to model human wound healing must be coupled with its practicality. Small mammals are inexpensive, easy to maintain, can be genetically modified, and are easy to handle. Swine are more expensive and more difficult to handle; however, their anatomical and physiological similarities to humans make them a preferred model choice.
Although in-vitro assays and animal models have several unique advantages and disadvantages, no one model is capable of completely representing human wound healing. In-vitro assays and animal models are stepping stones to well controlled clinical trials.