Clinical Ma,rerials 8 (1991) 171-177

Hyaluronate Derivatives and their Application to Wound Healing : Preliminary Observations

Jeffrey M. Davidson, at d Lillian B. Nanney, b$ cs d Kenneth N. Broadley,” Jeffrey S. Whit&t,” Antonio M. Aquino” a Departments of Pathology, ’ Cell Biology and ‘Plastic Surgery, Vanderbilt University School of h/ledicine, Nashville (TN), USA

Mauro Beccaro & Alessandro Rastrelli Fidia/Bioskin Research Laboratories, Via Ponte della Fabbrica 3/A, 35031 Abano Terme (PD), Italy

Abstract: Hyaluronic acid and its derivatives show promise as biomaterials in wound healing applications. Studies of cutaneous wound repair were carried out in two animal models to compare the biological effects of hyaluronic acid and hyaluronic acid ethyl ester, a new semisynthetic derivative. The two compounds were tested in partial-thickness excisional wounds in 40-kg pigs and full-thickness excisional wounds in the rabbit ear as 0.2 % (w/w) formulations in a neutral Na alginate vehicle. All compounds were administered daily under an occlusive, polyurethane dressing. Neither hyaluronic acid nor the hyaluronic acid ethyl ester showed toxic or inflammatory influences over the observation period of about 2 weeks. Morphometric analysis of porcine wounds revealed small differences among treatment groups which may have been masked by the effect of the vehicle. The rabbit ear model data suggested a very slight inhibition of wound closure. Biochemical analysis of ear wounds showed this injury model to be a sensitive system for evaluation of vulnerary agents. The hyaluronate-treated wounds tended to accumulate collagen more slowly, which may reflect the capacity of these compounds to modify the scarring process. Given the ability to fabricate hyaluronate esters into films and fibers, these data suggest that such biomaterials will not, by themselves, exert a negative influence on the repair process and may improve healing, either alone or in combination with other soluble agents.

INTRODUCTION movement of various successive waves of inflam- matory, mesenchymal, and epithelial cells to sites of

There is evidence in the biomedical literature to tissue repair.6 In the fetus, hyaluronate is abundant suggest that hyaluronic acid is intrinsically involved in structures such as skin, and this glycosamino- in the process of wound repair,lv’ and that glycan may account, at least in part, for the unique exogenously-added hyaluronate may serve a bene- reparative properties of fetal skin, which include ficial role in the acceleration or modification of the lack of scarring.‘-’ Hyaluronate has found ap- wound healing process.3’ 4 The biological role of plication in ocular surgery and treatment. Other hyaluronate is best understood in the process of potential applications include orthopedic injury morphogenetic cell movements, where an extra- treatment and burn care. A noteworthy example of cellular matrix abundant in this large, hydrophilic the biological properties of hyaluronate in wound molecull:: becomes permissive for the rapid move- repair is the ability of the GAG to improve the rate ment of cells to new sites in tissue.5 Similarly, it is of experimental tympanic membrane repair with thought that hyaluronate may play a positive role in markedly reduced scarring. lo regenerative and repair processes by facilitating the The physical properties of hyaluronate can be

171 Clinical Materials 0267-6605/91/$03.50 0 1991 Elsevier Science Publishers Ltd, England

172 J. M. Davidson, L. B. Nanney, K. Iv. Broadley, J. S. Whitsett, A. M. Aquino, M. Beccaro, A. ~~~~~~~~~

markedly modified by esterification.ll Various formulations can be fabricated into gels, films, and woven materials.12 The advantages of bio- degradable polysaccharide which can be manu- factured into various forms include low inherent antigenic properties of polysaccharides and po- tential for incorporation of other bioactive sub- stances into the polymer. It was the objective of the present study to determine whether hyaluronic acid ethyl ester (HYAFF@ 7~75; Fidia S.p.A., Abano Terme, Italy) suspended in an alginate-vehicle base, would alter the process of tissue repair in standard- ized wound healing models, and whether its proper- ties differed in a measurable fashion from those of the parent material, hyaluronic acid (Mr 140-160000; Connetivina@‘; Fidia S.p.A.). these purposes, two experimental models of wo healing were used : split-thickness excisional wounds in the pig and a full-thickness punch worm rabbit ear.

EXPERIMENTAL PROCEDURES

Wound models

For excisional wounds in the pig, a standardized 2 x 2 cm full-thickness wound of 102-1.4 mm depth was made with a Padgett dermatome (Padgett Dermatome Division of KC Assemblage, Kansas City, Missouri) as previously described.13 To retard the process of wound repair in simulation of potential clinical targets in which healing is com- promised, three 40-kg male, outbred pigs, obtained from a commercial breeder, received an intra- muscular injection of methylprednisolone (40 mg) 2 days prior to surgery. Older animals, while possibly exhibiting altered healing rates, were not practical to handle because of size. Previous studies had indicated that this treatment significantly retarded wound healing in this animal model for at least 1 week after injection (data not shown). Steroid- treated pigs, 2-3 months old, received 67 wounds along each flank, arranged randomly for daily treatment with one of three formulations: alginate vehicle (3 % Na alginate, pH 7*0 in saline), alginate vehicle plus O-2 % hyaluronic acid (Bioskin/Fidia S.p.A., Abano Terme, Italy) or O-2 % hyaluronic acid ethyl ester (Bioskin/Fidia S.p.A.) in vehicle. All wounds, including those receiving no treatment, were covered with a semi-permeable, adhesive, polyurethane dressing (Op-Site@+, Smith and Nephew, Massilon, OH) to prevent contamination, retain the gel formulations, and retard dehydration

site. All treatments were ~~~~~~~~ under general anaes

rabbit earl4 took advantage of two a~~t~~i~~l features of this small ~~irnal : the firm ~~~~~~~~~~ of the dermis to the ear car e, thus ~~~v~~ti~~ wound ~~~~~~~~~~e that t to ~~~~~a~~ the

and the avascular nature of the central ear which necessitates all cell ~~ve~~~~t

chamber model~,~” si the ear cartilage forms the base of an open wou and ~~~~~~~~~~ overgrowth readily occurs from the margins of the wound without contracture. ~~t~~~~~ not performed in this study, the ear model can be further com- promised by partial occlusion of the arterial blood SUpplgr.16 For each treatment, two animals were used at each time point (n = 4). Animals were pre- treated with atropine and ~~~$~~~~~~~ with ketami~~/~yl~~i~~ as a site was treated with

inephrine (1 I BOO 000 ; Astm OUTydB were carefully cut to

with a 6-mm bio Cummins ~~~~~~a~~~~~~a~s~ iami, FL), Punch wounds (6 mm) ‘were treate aily with the three ~orrn~~a~~~~~ ~~~a~~~~~~~ acid, hyaluromc acid ethyl ester, or vehicle; approx. 50 pl) or nothing and covered with polyureth e film. To prevent dis- lodging of the wound essings, ,the pimaa was

gently folded in half along its length ed gauze pad, to maintain mild pressure, a with tape after each daily treatment.

Analysis of wound healing included mor and biochemical parameters of r collagen, protein and

excised wound sit use of a 6mm

tissue were remove

termination was ~~de~e~~e~t of variation in the actual wound area. ~ra~~~a~~o~ tissue and the overlying regenerated epidermis were rea away from the normal, dense ~~~~e~t~v~ . beneath the wound site Tissue was wei

omogenized, and aliquots were taken for

Hyaluronate derivatives and their application to wound healing

III- . 0 VEHICLE

FJ HA-ESTER

0.6- ??HA

??OCCLUSIVE ONLY

0.6

(b) $ zooo-

1 e I%

1000-

5 10 DAYS OF TREATMENT

173

DAYS OF TREATMENT Fig. 1. Effect of hyaluronate formulations on granulation tissue formation in porcine excisional wounds. Morphometric evaluation on days 51 and 10 after injury. (a) Epidermal coverage, expressed as a fraction of total wound surface; (b), Granulation, tissue depth,

expressed in pm. Each value represents the mean of 3 3 measurements on each of 4 wounds in a single animal* SEM.

and total-protein determination, while a portion of the homogenate was lyophilized and hydrolyzed for amino-acid analysis to obtain hydroxyproline values.“7, l8 Dermal thickness and epithelial cover- age” were quantified by computerized morpho- metric analysis of tissue sections. A Planar Morpho- metry Package (Southern Micro Instruments, Atlanta, GA) was used to analyze tissue sections viewed on the stage of a Vanox light microscope interfaced to an IBM/AT computer via video camera and screen. In both models, nine random sections; were measured in each of 24 wounds to obtain a mean 5 SEM for each condition.20 Data were obtained with a computerized interface oper- ated by a blinded observer. Dermal depth was not assessed near the wound edge, nor was it taken immediately adjacent to hair follicles. In the pig, depth was defined as the distance, in microns, from the bottom of the migrating epidermis to the

interface of the granulation tissue with the under- lying, non-wounded dermis. Depth of granulation tissue in the rabbit ear model represented the distance between the upper surface of granulation tissue and the underlying cartilage. Re-epithelial- ization in the pig was scored as the total, integrated length of epidermal coverage (which occurred from both wound margins and epidermal appendages), while the rabbit wounds were scored as the diameter of the base and the top of the wound sites in sections taken from the center of the wound.

RESULTS

Excisional wounds in the pigs were evaluated at 5, 10 and 15 days after surgery.. Morphometric analysis revealed differences among the treatment groups in this model. On day 5, all of the gel formulations showed a trend toward increased

174 .J. M. Davidson, L. B. Nanney, K. IV. Broadley, J.

epidermal coverage, with the HA material showing somewhat higher epithelialization than the alginate vehicle alone (Fig. l(A)). By day 10, when wound coverage had reached nearly X0 %, results with the two HA formulations were not significantly different from control values, with the alginate vehicle producing the lowest epidermal coverage (75 % of control). With respect to granulation tissue formation, the parameter of neodermal depth (Fig. l(B)) showed that all gel formulation treatments stimulated this parameter at day 5 when compared to occlusive dressing alone, while there was a relative reduction in granulation tissue thickness in all treatment groups at day 10. Interestingly, alginate vehicle showed retardation of neodermal thickening relative to occlusive dressing alone, which appeared to be partially reversed by the two hyaluronic acid formulations. Additional, long- term follow-up of treated wounds will be necessary to evaluate qualitative aspects of scar formation and resolution.

Biochemical data showed a regular progression of wound healing parameters. All three gel formu-

somewhat reduced cellularity at day 5 and collagen/DNA ratios at day IO. None of these differences was statistically significant when compared to the occlusive dressing alone (data not shown). At the dose of g vehicle effects tended to effects of the HA, reflecting the a experience in evaluating pstentia agents. In addition, the con 1 material chosen for these studies, a semi-perme e occlusive dressing, can itself markedly affect wound repair.21s 22

The rabbit-ear wound model gave more pro- vocative results with the formulations used. Wounds in 14 animals were followed for 14 days, sampling every other day. Data oints represented the mean values from 4 wounds in two animals. Subsidence of inflammatory cell influx was indi- cated by the falling values between day 2 and Cellularity of wounds was again mildly increased over the controls at 6 and 8 days after treatment (Fig. 2), a time during w ich maximal influx of fibroblasts and microvas lar endothelial cells occurs. This relatively elevated in prolonged, however; the DNA csnte control levels by the end of the experimental period, by which time wound epithelial coverage was nearly complete. Rates of healing were reflected in the progressive increase in collagen : NA ratios (Fig. 3) and in collagen content of the ear wounds (Fig. 4). In this set of wounds, the biochemical response

Fig. 2. Cellularity of rabbit ear punch wounds DNA content of granulation tissue excised from punch wounds is expressed as a percentage of control (occlusive covering only). Ail gel. formulations appeared to stimulate this parameter on day 8. Values are the mean of detc~m~~atio~s from 4 wounds in 2

animals i_ SEM

OGCLUSNE ONLY VEMBCLE

??HA-ESTER

B HA

Fig. 3. Collagen/DNA ratios of rabbit ear granuiation tissue. Wound sites were excised with a G-mm biopsy punch, and the wound tissue was carefully teased away from the underlying, firm connective tissue. Collagen was determined from hydroxy- proline content of hydrolyzed tissue homogena content was measured in a fluorimetric assay. the progression of the healing process in this model. Healing

(wound coverage) was nearly complete by day 14.

ents was not remar rent from contrs~s. The A-ester form~latio~ appeared to show the slowes ate of collagen a~~~~~~atio~~~

hyaluronate on s wounds with ~o~y~r~th~~e film only exhibite generally the highest collagen values at each time

mulate early ~~ga~~zat~~~ of the wound site while oderating the excessive ac- ~~rn~~a~~o~ of toll n at the of Scar for-

mation. Future studies will

Hyaluronate derivatives and their application to wound healing 175

“0 5 10 15

t 5 10 15

Oay of Repair

Fig. 4. Relative effects of gel formulations on (a) collagen con- tent, (b) collagen/protein ratios and (c) per cent collagen in the rabbit ear wound model. At early time points (d 6 days) hyaluronic acid esters and HA may have had an enhancing effect relative to vehicle alone. By day 14, gel formulations

appeared to suppress the connective tissue response.

long-term effects of HA and HA-ester on the terms (Fig. 5). In terms of rate of closure and quality and composition of the scar. thickening of the wound site, all three gel

Morphometric analysis of healing in the rabbit formulations showed slight retardlation relative to a wound described wound closure in quantitative simple occlusive dressing.

0 I

0 5 10 15

2000

1000

0

- VEHICLE -O-- HA-ESTER

+ OP-SITE 1

0 6 10 15

A-Y Fig. 5. Morphometric analysis of the rabbit ear wound. Panels depict (a) the lower and (b) the upper diameter of the wound as well as (c) the thickness of granulation tissue measured at the upper margin. Occlusive dressing alone elicited a higher rate of decrease in wound diameter (by formation of granu- lation tissue, not contraction) and wound neodermal thick-

ness.

176 J. M. Davidson, L. B. Nanney, K. N. Buoadley, 9. S. Whitsett, A.

DISCUSSION

Because of the small numbers of wounds tested, and the use of a single dosage, these studies were only indicative of the biological effects of HA or HA derivatives in wound healing. Nevertheless, it was apparent that, at the dosages used, neither HA nor its ethyl ester had an inhibitory or toxic effect on the healing process. Histological analysis did not reveal a marked effect on the appearance of inflammatory cells in the lesions. Thus, both formulations had appropriate properties of biomaterials.

These preliminary data illustrate a number of issues in the design of wound repair studies. There continues to exist a sharp contrast between potential clinical applications for wound healing accelerators or wound dressings and the available animal models. Specifically, there is no true animal parallel to the non-healing wound or ulcer in man. In this report, steroid treatment (pig) or relative avas- cularity of a wound site (rabbit) were used to compromise the repair process. Nevertheless, it must be emphasized that the relatively normal, young animals used in these experiments healed at a very rapid rate; thus only the most potent formulations could be expected to provide sig- nificant acceleration of healing Indeed, the authors’ previous studies with a variety of cytokines have shown the ED50 for animal studies to be usually three orders of magnitude greater than those observed in simplified cell culture systems.” A second issue is the variety of healing models available and the differential responsiveness of experimental wounds to various stimuli. The ex- cisional wound model in the pig has been used, for example, to evaluate vulnerary substances as a model of the burn-graft donor site, since this was thought to be a useful clinical target for growth factor therapy in humans.‘” The model has been modified to simulate retarded healing the depth of the injury and using a long-lasting steroid. As a possible drawback, steroids could mask the effects of exogenous stimuli that require the presence of accessory inflammatory cells for action.

With respect to the ty;es of formulations tested, the rabbit model was more informative and more encouraging. Data were obtained that were con- sistent with the experimental formulations having the capacity to increase transiently the recruitment of cells and formation of granulation tissue at early time points, while possibly retarding the formation of excess collagen at the latest time point evaluated.

By prolonged follow-~~ in a larger test group, it e possible to observe di

quality of wound repair as well. The of this parameter for comparative pu

ic acid and one of its ological materials with

little apparent toxicity in two models. At higher co~ce~t~at~o~s~

ecome mo apparent,, -esters as

les for other vuln rtant application in t

wounds.

A WLEDGE

This study was sup

1.

2.

3.

4.

5.

6.

7.

8.

9.

Spooner, 8. s. Thompson-Pletscher, Pi. A., Matrix accumulation and the development of form: Proteoglycans and branching morphogenesis. In ~~gu~a~~~~ Matrix Accumulation, ed. R. B. Mecham. Academic s: New York, 1986, pp. 399444. Weigel, P. M., Fuller, 6. M. Bt L euf. W. D., A model for the role of hyaluronic acid and in in the early events during the inflammatory response and wound healing. J. Theor. Ipioi., 119 (1986) 219-34.

N., Arnold, F, & Kumar, S., Angiogenesis induced by degradation products of hya- luronic acid. Science, 228 (1985) 1324-6. Abatangelo, G., Martelli, M. & Vecchia, P., Healing of hyaluronic acid-enriched wounds : histological observa-

RQS., 35 (1983) 41@B6. P., Glycosaminoglycans in morphogenesis. in

0s Exlracellulav Matrix, ed. E. D. Hay. Plenum Press, New York, k 98 I, pp. 259--94. Toole, B. P., nudson, 6. B., Munaim, S. I., Knudson; W., Welles, S. $i Chi-Rosso, G., Hyaluronaie-cell inter‘- actions and regulation of hyaluronate synthesis during embryonic limb development. In Cutnneous Deveiqmerli, Aging and Repair, idia Research Series, vol. 18, ed. C.

Davidson. Liviana Press, Padova,

Krummel, rS. M., Nelson, J. M., Diegehnann, R. F‘., Lindblad, W. J., Salsberg, A. M., Greenheld, L. ji i% Cohen, I. K., Fetal response to injury in the rabbit. J.

22 (19873 640-a. L., Krummel, T. M., Durham. L. A.. ., Thomas, B. E., Nelson, J. M & Diege?-

F., Characterization and quantitation of wound ma&-ix in the fetal rabbit. Matrix, 9 (1989) 224-3 I. Adzick, N. S., Harrison, M. R.., Click, P. E., Beckstead, 3. FT., Villa, R. E., Schwenstuhl, B-I. & Goodson, W. H., Comparison of fetal; newborn and adult wound heahng by histologic, enzyme-histochemical and hydroxyproline determinations. /. Pediatr. Surg., 20 (1985) 315-9.

Hyaluvonate derivatives and their application to wound healing 177

10.

11.

12

13

14

15

Hellstrom, S., Laurent, C., Schmidt, S.-H., Spandow, 0. & Fellnius, E., Experimental tympanic membrane perfora- tion-A model for studying wound healing? Effects of hyaluronic acid and hydrocortisone. In Cutaneous De- velopment, Aging and Repair, Fidia Research Series, vol. 18, ed. G. Abatangelo & J. M. Davidson. Liviana Press, Padova, 1989, pp. 179-88. Mazzoleni, F., Chiarelli, A., Siliprandi, L., Cortivo, R., Giro, M. G. & Abatangelo, G., Effects of biological membranes on wound healing. In Cutaneous Development, Aging and Repair, Fidia Research Series, vol. 18, ed. G. Abatangelo & J. M. Davidson. Liviana Press, Padova, 1989, pp. 215-22. Rastrelli, A., Beccaro, M., Biviano, F., Calderini, G. & Pastorello, A., Hyaluronic acid esters, a new class of semisynthetic biopolymers : chemical and physico-chemical properties. In Clinical Implant Materials-Advances in Biomaterials, vol. 9, ed. G. Heimke, U. Soltesz & A. J. C. Lee. Elsevier, Amsterdam, 1990, pp. 199-206. Quaglino Jr., D., Nanney, L. B., Kennedy, R. & Davidson, J. M., Transforming growth factor-/? stimulates wound healing and modulates extracellular matrix gene expression in pig skin. I. Excisional wound model. Lab. Invest., 63 (1990) 307-19. Beck, L. S., Chen, T. L., Hirabayashi, S. E., Deguzman, L., LIee, W. F’., McFatridge, U., Xu, Y., Bates, R. L. & Ammann, A. J., Accelerated healing of ulcer wounds in the rabbit ear by recombinant human transforming growth factor-beta 1. Growth Factors, 2 (1990) 273-82. Clark, E. R., Kirby-Smith, H. T., Rex, R. 0. & Williams, R. G., Recent modifications in the method of studying living cells and tissues in transparent chambers inserted in the rabbit’s e,ar. Anat. Rec., 47 (1930) 187-211.

16.

17.

18.

19.

20.

21.

22.

23.

24.

Ahn, S. T. & Mustoe, T. A., Effects of ischemia on wound healing: a new model in the rabbit ear. Ann. Plast. Surg., 24 (1990) 17-23. Buckley, A., Hill, K. E. & Davidson, J. M., Collagen metabolism. Met/z. Enzymol., 163 (1988) 67494. Davidson, J. M., Klagsbrun, M., Hill, K. E., Buckley, A., Sullivan, R., Brewer, P. S. & Woodward, S. C., Accelerated wound repair, cell proliferation and collagen accumulation are produced by a cartilage-derived growth factor. J. Cell Biol., 100 (1985) 1219-27. Chvapil, M., Gaines, J. A., Benson, D. & Tellez, C., An optimal morphometric method for quantitating wound epithelialization. J. Sung. Res., 44 (1988) 266-76. Nanney, L. B., Epidermal and dermal effects of epidermal growth factor during wound repair. J. Invest. Dermatol., 94 (1990) 624-9. Falanga, V., Occlusive wound dressings. Arch. Dermatol., 124 (1988) 872-7. Dyson, M., Young, S., Pendle, C. L., Webster, D. F. & Lang, S. M., Comparison of the effects of moist and dry conditions on dermal repair. J. Invest. Dermatol., 91 (1988) 434-9. Broadley, K. N., Aquino, A. M., Hicks, B., Ditesheim, J. A., McGee, G. S., Demetriou, A. U., ‘Woodward, S. C. & Davidson, J. M., The diabetic rat as <an impaired wound healing model. Stimulatory effects of transforming growth factor-beta and basic fibroblast growth factor. Biotech. Therap., 1 (1989) 55-68. Brown, G. L., Nanney, L. B., Griffin, J., Cramer, A. B., Yancy, J. M., Curtsinger, L. J., Holtin, L., Schultz, 6. S., Jurkiewicz, M. J. & Lynch, J. B., Enhancement of wound healing by topical treatment with epidermal growth factor. IV. Engl. J. Med., 321 (1989) 76-9.