Abstract: Fibroblast culture systems are routinely used to investigate wound contraction under a wide range of experimental conditions. These include the effects of irradiation, inhibition of chronic inflammatory cell mediators, and the biocompatibility of wound management products. In particular, these in-vitro cell systems have been routinely used to assess the cytotoxicity of topical antimicrobial agents and dental materials. L929 cells, derived from an immortalized mouse fibroblast cell line, are internationally recognized cells that are routinely used in in-vitro cytotoxicity assessments. In these studies, it is proposed that equine granulation tissue fibroblasts, cultured from slow healing wounds or from granulating wounds with exuberant granulation tissue removed during normal surgical debridement, may also be used to assess in-vitro cytotoxicity, in particular with respect to topical wound healing products. This model demonstrated that granulation tissue fibroblasts behave similarly to L929 fibroblasts in that they were effective in differentiating the toxicity of a variety of topical iodine-containing formulations. The data presented in this report suggests that currently marketed iodine-containing antiseptic agents show variable toxicity to fibroblasts involved in wound healing. These results suggest that currently available topical iodine antiseptic agents could be detrimental to wound healing if treatment is prolonged.

Disclosure: Financial support for this work was provided by ConvaTec Ltd., United Kingdom.

Introduction

Since the introduction of a fibroblast-populated collagen lattice in the late 1970s, this type of in-vitro model has been extensively used to study fibroblast function.[1,2] Contraction of fibroblast/collagen gels has been used as in-vitro models for investigating the biological mechanisms of wound contraction[3,4] and also the effects of various compounds aimed at stimulating (enhancing wound healing) or reducing (preventing scar formation) the rate of contraction.[5] The benefit of this model is that the fibroblasts are grown in a three-dimensional collagen gel culture, a matrix component native to the wound environment.

Fibroblasts cultured in these conditions have distinct characteristics compared to those cultured on dishes. The fibroblasts cultured in collagen gels acquire a bipolar spindle form, while fibroblasts grown on culture dishes manifest a ruffled membrane that possesses one or more broad-banded pseudopodia.[6,7] In the gels, collagen production is decreased, proteinases and fibronectin production is increased,[8] and fibroblast proliferation is slower.[9] Generally, it appears that the culture of fibroblasts in these gels more closely resembles growth in an in-vivo situation, and this is important when considering cellular interactions and potential cytotoxic compounds.

Fibroblast culture systems have been routinely used to investigate wound contraction in a wide range of experimental conditions, including irradiation effects,[10] inhibition of chronic inflammatory cell mediators,[11] and the biocompatability of wound management products.[12] In particular, these in-vitro cell systems routinely have been used to assess the cytotoxicity of topical antimicrobial agents[13] and dental materials.[14]

In the current studies, the L929 immortalized mouse cell line, a recognized model for wound dressing compatibility studies,[12] was used along with equine fibroblasts, which were harvested and cultured from slow healing wounds or from granulating wounds with exuberant granulation tissue removed during normal surgical debridement.[15]

There were two reasons for choosing equine fibroblasts. First, the chronic wound in the horse is considered to have a similar pathology to the human chronic wound.[2,16] These nonhealing wounds are only present on the lower limbs (i.e., below the knee) of horses and are highly inflamed with high levels of protease activity in wound tissue and wound fluid.[16] The second reason is that equine nonhealing wounds hyper-granulate, and excision of this tissue provides an excellent source of granulation tissue-derived fibroblasts. Equine granulation tissue fibroblasts (EGTF) have been shown to have similar growth rates to those of normal equine fibroblasts yet have a spread-out cuboidal appearance compared to the typical spindle-shaped morphology found in normal fibroblasts.[7]

It was proposed in the first part of these studies to compare the rates of contraction of normal equine fibroblasts with those of EGTF against the well-characterized mouse fibroblast cell line (L929 cells) that is used routinely in cytotoxicity studies.[13–15]

In the second part of the studies, collagen gels containing EGTF and L929 cells were compared by assessing their response to the application of potentially cytotoxic, iodine-containing topical antiseptic agents (Table 1).
There is still much debate regarding the use of antiseptics in wound care,[17,18] although there is a growing consensus that the slower releasing formulations have a role to play.[18]

Materials and Methods

Cell culture. L929 cells were obtained from the European Collection of Cell Cultures, Centre for Applied Microbiology and Research, Porton Down, United Kingdom.

Normal equine fibroblasts were obtained post mortem from horses that had been killed for non-related clinical reasons.

Chronic wound (granulation) tissue fibroblasts were cultured from tissue taken from slow-healing wounds or from granulating wounds with exuberant granulation tissue during normal surgical debridement prior to skin grafting.

The in-vitro gel contraction model used throughout these studies to evaluate tissue toxicity was in compliance with the ISO Standard for the Biological Evaluation of Medical Devices.[19] Briefly, appropriate samples were taken for fibroblast culture and immediately transferred to a dish, washed in Hank’s balanced salt solution (HBSS) (all cell culture materials were supplied by Gibco, United Kingdom, unless otherwise stated), cut into 5mm2 pieces, and placed into 25cm2 tissue culture flasks containing media. The media was made by the addition of Dulbecco’s Modified Eagle Medium (DMEM), supplemented with 10-percent fetal calf serum (FCS) (Sigma, United Kingdom), 20mM Hepes buffer, 100µg/mL gentamicin, and 0.5µg/mL amphotericin B. Cell cultures were incubated at 37 degrees C in a five-percent CO2/95-percent air environment. Readiness for subculturing was determined by the extent of fibroblast cell outgrowth (5–10 days). Cells were farmed successively in a 1 to 4 split ratio and passages 3 to 8 were used in these experiments.

Fibroblasts were harvested from stock dishes and plated out in 35mm, six-well plates at 1x106 cells/mL in type I collagen (Fred Baker Scientific, United Kingdom) at 2mg/mL. Six wells were set up for each individual formulation and cell type (n=6). Untreated dishes containing fibroblasts in collagen as above (n=6) were established to use as controls. The dishes were incubated in a five-percent CO2/95-percent air environment at 37 degrees C. After a period of one hour when the collagen fibroblast gels had set sufficiently, the gel surface was washed with 1mL of HBSS, and 1mL of media was added. In the first series of studies investigating normal contraction rates, the cells were allowed to contract for a maximum of 144 hours. In the second series of experiments, either 0.5mL or 0.5g of a range of topical antiseptic formulations (Table 1) was separately applied to the surface of the gels. Gel contraction was measured using a set of calipers (mm) at 24-hour intervals and thereafter extending to a maximum period of 96 hours. There was no need to remove the dressings to measure the gels because gel contraction did not exceed surface area of the dressing. In these studies, povidone-iodine solution was used as a positive control at a clinically relevant concentration (i.e., 10% w/v). Gauze was soaked in 0.5mL povidone-iodine solution (10%) prior to application to the surface of the collagen gel. The negative control had no formulation applied to the gel surface.

Table 1

Statistical analysis. A multivariate analysis of variance (Duncan’s Multiple Comparison Test) was used to analyze the data. This test accommodates the experimental design used in these studies. Notably, they were two-factor experiments (i.e., different dressings and time points) with repeated measurements on one factor (i.e., contraction rate). These multifactor experiments have been described in detail by Winer.[20]

Results

The initial study showed that EGTF have greater contractile capacity than normal equine fibroblasts, and these results are in agreement with the results of Germain, et al., who studied human granulation fibroblasts (Figure 1).[4] She suggests that there may be a higher proportion of myofibroblasts present in wound fibroblast populations and that these may enhance wound
contraction.[4]

Figure 1

This graph is a comparison of EGTF with normal equine, L929, and human granulation tissue fibroblasts.

In the second series of experiments, EGTF were compared with L929 cells to evaluate a range of iodine-containing topical antiseptic agents (Table 1). The results are presented in Figures 2 and 3.

Figure 2

This graph shows the gel contraction data with EGTF.

Figure 3

This graph shows the gel contraction data for L929 fibroblasts.

Similar results were observed for both sets of fibroblasts. Only inadine appeared to offer a reduced level of cytotoxicity beyond 24 hours (i.e., greater level of contraction), but even this was seen to be significantly different to the control group (p < 0.05). All the topical antiseptic agents were highly significantly different (p < 0.005) to the control, in both the L929 and EGTF groups.

Discussion

Collagen is the most abundant protein in animals, and its critical role in homeostasis of connective tissue is well documented. The biology of collagen associated with wound healing is not fully understood, and although most of the collagen in adult animals is metabolically stable, some of it is rapidly synthesized and degraded. Fibroblasts, the main collagen-producing cells, are also responsible for remodelling and contraction of collagen. The fibroblast is an important cell type involved in all stages of wound repair, and healthy functioning fibroblasts are necessary for expedient wound healing. The fibroblast-populated collagen gel contraction model, originally described by Bell, et al., has become an accepted model to study wound contraction.[1] This model allows a fast and reliable method for assessing contractile ability in a quantitative manner and has often been used to study the cytotoxicity of topical antimicrobial agents[13] and dental materials.[14]

L929 fibroblasts are commonly used in the measurement of potentially cytotoxic compounds, but they are a transformed cell line and, as such, may not truly be representative of a wound environment. By utilizing available granulation tissue fibroblasts, this may provide a more realistic research screening model in the cytotoxic assessment of topical agents applied to wound tissue. These fibroblasts are actually taken from the wound bed and are, therefore, representative of the wound environment. They have a different rate of contraction to standardized transformed cell lines, such as L929 cells, and may be more sensitive to the application of potentially cytotoxic materials.

In-vitro studies do not always concur with in-vivo studies and consequently should not strictly be used to predict in-vivo toxicity. However, by choosing fibroblasts and, in particular, wound granulation fibroblasts, this model was an attempt to mimic the main type of cell involved in dermal matrix production.[16] While it is appreciated that no in-vitro model can closely mimic the clinical situation due to the complexity of the wound healing process, these fibroblast cells are the primary matrix-producing cells, and fibroblast growth can be monitored in well-controlled experimental conditions.

Iodine is a potent antiseptic agent that is frequently used in the management of wounds (primarily as povidone-iodine or cadexomer iodine) to reduce the risk of infection. Unlike many antibiotics, antiseptics are nonselective chemical agents that, if not used appropriately, may be toxic to host tissue as well as to micro-organisms.

In the povidone-iodine products, the iodine is bound to a synthetic polymer polyvinyl-pyrrolidone. There is minimal available data in the literature for iodine release from these polymeric formulations, but numerous in-vitro studies have shown that these solutions are toxic to granulocytes and monocytes,[21] keratinocytes,[22] and fibroblasts[23] at concentrations of approximately one percent. Only when dilutions were applied did the formulations appear to be nontoxic to host cell lines.[23] Doughty has suggested that at least a 10-fold or preferably a 1000-fold dilution should be considered a “safe” concentration.[24]

Clinical evidence from the literature is divided in opinion with respect to the safety and efficacy of iodine in the wound environment. In a recent meeting of clinicians, scientists, and industrial representatives, under the sponsorship of the European Tissue Repair Society, it was generally agreed that the use of slow-release formulations, which generate low concentrations of iodine, appear to be effective and nontoxic in clinical practice.[18]

However, in the current studies, all the iodine products tested were shown to be toxic beyond 24 hours of application. It has been suggested that for some cadexomer products, iodine is slowly released for up to 72 hours depending on the amount of wound exudate.[25] The cadexomer products are usually comprised of very small hydrophilic biodegradable beads, which incorporate 0.9-percent iodine (w/w) typically in a macrogol ointment base comprised of a mixture of polycondensation products of ethylene and water. Upon application to a wound, wound exudate is absorbed by the cadexomer polymer, which swells and releases the iodine component. The cadexomer can absorb up to six times its weight in wound fluid and organic matter.[25]

Assuming a maximum application time for these products (containing 0.9% iodine w/w) of up to 72 hours as stated above, assuming a linear release profile, and assuming that all iodine is released, this equates to an approximate release rate of 125µg/hr of free iodine per gram of applied formulation. It is clear from the gel contraction plots (Figures 2 and 3) that not all of the iodine in the formulations is available. Even allowing for approximately 50 percent of this iodine to be retained by the dressing matrix, this still results in the availability of approximately 60µg/hr of iodine. It has been suggested that as little as 1 x 10 to 16g of iodine is sufficient to kill one bacterial cell;26 therefore, the amount of iodine would appear to be in excess, and this in turn may contribute to host tissue toxicity.

It is accepted that the in-vitro models described in this report may have limitations with respect to the in-vivo situation. However, it is recognized that the fibroblast is an important cell type involved in all stages of wound repair, and the presence of healthy functioning fibroblasts are necessary for expedient wound healing. Therefore the use of primary EGTF could offer a more meaningful model for wound contraction and cytotoxicity measurements.

The fibroblast-populated collagen gel model is well recognized and offers a controlled environment for studying wound contraction although it is appreciated that the healing process of a chronic wound does not proceed in an orderly manner. This model has previously been used routinely to assess the cytotoxic properties of numerous topical antiseptic agents, but the majority of these studies have utilized normal human fibroblast-embedded collagen gels.[1,11,13,14]

In these studies, it has been shown that EGTF is comparable to the standardized L929 cells under these experimental conditions and may be a more realistic model for screening potentially cytotoxic topical agents that may be applied to wound tissue.

Conclusions

The EGTF gel contraction model has been shown to be comparable to an internationally recognized cell line, L929 fibroblasts, in the in-vitro cytotoxic assessment of a variety of topical iodine-containing formulations.

While iodine is an effective antimicrobial agent, the availability of free iodine is critical. It is, therefore, important to strike a balance between minimizing wound infection while ensuring there is no detrimental effect on host tissue. This balance is a critical factor to enable the wound environment to maintain a host-manageable level of micro-organisms.

The data presented in this report indicate that the currently marketed topical iodine containing antiseptic agents are potentially cytotoxic, and with this in mind, prolonged use may be detrimental to wound healing. This suggestion is in good agreement with recent guidelines issued by the FDA for the use of iodine-based antiseptics, in that they recommend these antiseptics should only be used as nonprescription, short-term treatment products.[27] None of the currently marketed topical iodine products appear to be nontoxic to fibroblasts involved in wound healing.