Thermal damage to skin collagen

Bums (1991) 17, (3), 209-212

Printed in Great Britain

209

Thermal damage to skin collagen J. Hambleton and P. G. Shakespeare Laing Laboratory

for Burn Injury Investigation,

Odstock Hospital, Salisbury, UK

Y&effectof temperatureupon the solubility offrozen skin collagen in vitro and its susceptibility to digestion by proteolytic enzymes has been studied Both ofthese parametersare increased with temperature. Above 52°C there is a sudden increase in both the solubilify of collagen and its swceptibility to digestion, suggesting that this temperature is associated with changes in the structure of the skin collagen. This increase in susceptibility to digestion may have an influence upon the nature of the healing process in the bum wound.

Introduction Thermal damage to the skin has effects on both the cellular activities and its structural matrix. The thermal damage to serum proteins and the cellular components of skin caused by relatively mild heating is well documented (Davies, 1982). The effects of heat upon the structure of the skin are less well characterized. Knowledge of such effects will help with the understanding of the process of wound healing since the demarcation between ‘viable’ and ‘non-viable’ injured tissue may not be simply a reflection of damage to the cellular activities of the skin. Increased susceptibility of collagen to digestion by proteolytic enzymes has been reported (Axelrod and Martin, 1953). One consequence of thermal injury is the infiltration of the damaged tissue with proteolytic enzymes. The change in susceptibility to digestion may allow the heated collagen to be digested and reformed rather than be used as a matrix in the healing burn wound. Whether any remaining matrix would subsequently crosslink and form into its original collagen structure or build up into the fibre arrangement of scar tissue is not certain (McPherson and Piez, 1988). The effect of heat on the susceptibility of skin collagen to digestion by collagenase has been investigated. Samples of normal skin and hypertrophic scars were studied. The results indicated that there is a transition temperature above which collagen becomes more susceptible to digestion.

Materials and methods Skin tissue which had been stored deep frozen immediately after excision was used. Normal skin was obtained from breast reduction surgery and hypertrophic scar tissue from scar revision surgery on bums patients. Portions (100mg) were rinsed with isotonic saline, blotted and coarsely minced with scissors. A portion was added to I ml of isotonic saline and maintained for 20min at the test temperature. After centrifugation the supematant was 0 1991 Butterworth-Heinemann Ltd 0305-4179/91/030209-04

reserved for assay and the heat-treated tissue rinsed, blotted and added to I ml of an appropriate enzyme reaction mixture. The amount of soluble collagen produced was estimated by hydroxyproline analysis (Bergman and Loxley, 1963). Samples of supematant and untreated tissue were subjected to acid hydrolysis. This was achieved by overnight heating at 120°C in 6 N hydrochloric acid in sealed glass ampoules. The hydrolysates were then neutralized with sodium carbonate (5 per cent solution) prior to the hydroxyproline assay. The values were expressed as a percentage of the initial total collagen present in the tissue. The soluble proteins present in the supematant after heating tissue were investigated by two-dimensional immunoelectrophoresis (2DIE) (Davies et al., 1971) and polyacrylamide gel electrophoresis (PAGE) (Laemmli, 1970). Bacterial collagenases Type IV (product no. C5138) and the purer Type VII (product no. CO773), trypsin (T8253) and pepsin (P7012) were obtained from the Sigma Chemical Co. (Poole, Dorset, UK), as were the chemicals required for buffers and for the hydroxyproline assay. Leech collagenase, ‘Calonase’, was obtained from Biophann (UK) Ltd (Hendy, Dyfed, UK). The conditions for the enzyme reactions were based on those recommended by the manufacturer and the units referred to are those defined in the manufacturer’s product information. For example, 1 unit of Calonase is equivalent to 1 mU of bacterial collagenase activity.

Results Normal skin collagen was not soluble to any appreciable extent in neutral solution below SO-52°C. Thereafter the solubility increased with increase in temperature (Figttre I). A time dependence for collagen dissolution was observed over the 2O-min heating period for samples maintained at temperatures above 52°C. A standard incubation time was used to eliminate this variable. Samples of tissue in saline were rapidly cooled at the end of 5, 10 and 20 min heating periods and incubated at 4°C overnight before centrifuging. These were compared with samples separated immediately after heating. Similar levels of soluble collagen were found in both sets of supematant solutions, suggesting that cooling halted the dissolution process. Hypertrophic scars, excised 2 years after bum injury, showed a similar, though more pronounced, increase in solubility with increase in temperature which only occurred above 50°C (Figttre 2). Heating at 70°C solubilized approxi-

Bums (1991) Vol. 17/No. 3

210

60’

50.

$40. 6 =8 8 0) 30.

20

-

IO -

I o,o.

??

I

*

30

40

,

I

50 60 70 Temperature OC

O/ 30

I

80

90

Figure 1. The percentages of total collagen solubilized from pieces of normal skin maintained for 2Omin in saline at various temperatures. 50’

40,

40

50 60 Temperature Oc

70

80

Figure2. Comparison of the percentages of total collagen solubilized by heating normal skin (0) and hypertrophic scars excised 2 years after burn injury (0) (A). The percentages of the initial total collagen solubilized from the heat-treated tissues by collagenase IV digestion (290 units/ml, 50 min incubation at 37°C) are shown by the corresponding closed symbols: 0, normal skin; ?? , A, hypertrophic scars.

,\” 5 g 30. uo d $ 20. ln 10.

0

20

30

40

50 60 Temperature OC

70

Figure 3. The percentage of total collagen solubilized from heat-treated normal skin by digestion with various enzymes (I.5 h incubation at 37°C): collagenase IV (180 units/ml, 0); collagenase VII (196units/ml, ?? ); trypsin (1404 units/ml, 0); pepsin (807 units/ml, a). The percentage of collagen solubilized by the heat treatment alone is also shown (0).

mately 4 per cent of the total of normal skin compared with 20 per cent of the total collagen of scar tissue. Preheating normal tissue at temperatures above 53°C greatly increased its susceptibility to bacterial collagenase Type IV (Figures2, 3). Digestion with trypsin and pure bacterial collagenase Type VII showed similar patterns of change in susceptibility (Figtrre3). Pepsin appeared not to show such a marked threshold. Digestion of collagen in hypertrophic scars showed a similar pattern. The susceptibility to collagenase Type IV in tissue pretreated at temperatures below. 50°C was consistently higher than that obtained in normal tissue ‘(Figure2). Leech collagenase (Calonase) which acts in a similar manner to human collagenase was tested for its ability to digest collagen in tissue (Figure4). There was a marked increase in susceptibility in tissue preheated above 50°C and

0120

30

40

50

60

70 80 90 100 OC Figure 4. The percentage of total collagen solubilized by heating normal skin (0) and the corresponding percentage of collagen solubilized by digestion of the heat-treated skin with leech collagenase (1000 units/ml, 6.5 h incubation at 37”C, 0). Temperature

no activity at all observed in the presence of tissues preheated below 50 “C. Once over the threshold there was only a very modest increase with increase in pretreatment temperature. The low percentage of collagen solubilized by Calonase compared with other collagenases may reflect the low enzyme concentration present in the reaction mixture. 2DIE and PAGE confirmed that serum proteins normally present in the skin were progressively destroyed by the heat treatment of the neutral salt extract. PAGE in the presence of sodium dodecyl sulphate (SDS) showed a collagen triplet, three bands of similar molecular weight (approximately 110 000 Da), in the supematant from tissues heated at higher temperatures (Figure 5~). Collagen was seen as a single band,

Hambleton and Shakespeare: Thermal damage to skin collagen

12345678 j-0

I

-COL

14~13

I- F

a 12345678

b Figure 5. a, SDS-PAGE

of the supematants from heat-treated skin. Tracks I, 2 and 3 from hypertrophic scars and 4 from normal skin heated to 75°C. 5,6 and 7 from hypertrophic scars and 8 from normal skin heated to 37°C. b, SDS-PAGE of the supematants following collagenase IV treatment of preheated skin. Tracks I, 2 and 3 from hypertrophic scars and 4 from normal skin heated to 75°C. 5,6 and i’from hypertrophic scars and 8 from normal skin heated to 37°C. The origin (0), the solvent front (F) and the positions of albumin (ALB) and collagen (COL) are shown.

molecular weight approximately 110 000 Da, in the reaction medium after collagenase IV digestion of preheated tissue (Fipre 5b).

Discussion The results of this study imply that heat damage causes changes in the collagen structure of frozen skin that render it more susceptible to the action of digestive enzymes than in normal physiological conditions. At a temperature of 50-52°C a change in the structure of collagen in skin must occur. This change could be the weakening of either the

211

strong covalent crosslinks between fibres or the hydrogen bonding between adjacent collagen chains. The extensive polymerization of collagen is possible because of the high hydroxyproline content and this confers stability by links involving the hydroxyl groups (Traub and Piez, 1971). On denaturation by heat, collagen, unlike most other proteins, becomes more soluble. The denaturation temperatures of various collagens, i.e. their change in solubility and susceptiblity to enzyme digestion, have been found to relate to the hydroxyproline content (Fessler, 1974). In studies with synthetic polypeptides the hydroxyl groups have been found to increase thermal stability (Sakakibara et al., 1973). It is likely therefore that the resistance to denaturing below 50°C that was observed in the present study is a function of the high content of hydroxyl groups in skin collagen. The action of Calonase, a very pure enzyme preparation having a similar action to human collagenase, suggests that the transition temperature 50-52°C is very critical to the denaturing process. The slightly different profile observed with less pure enzyme preparations may be due partly to the action of non-specific proteases present. The pattern of change in susceptibility to digestion by pepsin was not so marked. This may be because pepsin digests collagen in a different manner from that of collagenase, digesting the extension peptides of the collagen which are not bound in the helical structure of the main molecule. Scar tissue when first formed has a less crosslinked structure with a looser matrix than normal skin (Bailey et al., 1975). Further crosslinks form over several years. This would account for the higher initial susceptibility of collagen in hypertrophic scar to enzyme action compared with normal skin. Collagen crosslinking density has been found to correlate with resistance to bacterial collagenase in reconstructed collagen fibrils (Siegel et al., 1978). Certain enzymes involved in energy production in skin show a reduction in activity at temperatures as low as 45”C, the amount of impairment gradually increasing with increase in skin temperature (Hershey et al., 1965). Cells are dependent on the production of energy for the maintainance of their integrity and survival. Since damage to cells can be initiated at lower temperatures than damage to collagen, a situation could exist where ‘dead’ skin might be present with a ‘viable’ skeleton. This is of interest in healing of bum injuries where ‘skeletal’ components of the skin, if not damaged by heat, might resist degradation by infiltrating proteolytic enzymes and in time the original matrix might be repopulated with viable cells. Digestion of collagen in the heat-damaged matrix would presumably continue for as long as the extracellular space is infiltrated with active proteolytic enzymes and may account for the high urine content of hydroxyproline during the healing of bum wounds. Abnormal levels may still be observed 2 months after extensive bums (Smith et al., 1974). From the present study it would seem unlikely that any skin tissue heated above 52”c, a temperature likely to be reached throughout the skin in a severe scald injury (Moncrief, 1979), would maintain its original structure entirely. Indeed it may not necessarily reform or be remodelled during healing into a matrix identical with that present in normal tissue. Irnmediate cooling after injury is obviously of benefit in limiting the amount of heat which is conducted to neighbouring tissue as well as restricting the depth of injury in the visibly traumatized area. It may also be of benefit in limiting heat damage to the connective tissue matrix of dermis. The present study deals only with the effect of heat upon the gross susceptibility of collagen to digestion. The

Bums (1991)Vol. 17/No.3

212 susceptibility of individual types, particularly the minor collagens making up basement membrane structures, has not been studied. A useful extension of the study would be to determine the characteristics of the collagen remaining in the damaged tissue excised from bum wounds prior to grafting. It is possible that the changes in collagen susceptibility described here may in part determine the balance between healing of the wound to give a normal structure and appearance to the healed skin, and the production of unacceptable scar tissue at the site of injury.

References Axelrod A. E. and Martin C. J. (1953) Biochemical changes in thermally-injured cutaneous tissue. Proc. Sac. Erp. Biol. Med. 83, 463. Bailey A. J., Bazin S., Sims T. J. et al. (1975) Characterisation of the collagen of human hypertrophic and normal scars. Biochim. Biophys. Acta 405,412. Bergman I. and Loxley R. (1963) Two improved and simplified methods for the spectrophotometric determination of hydroxyproline. Analyf. Chem. 35,1961. Davies D. R., Spurr E. D. and Versey J. B. (1971) Modifications to the technique of two dimensional immunoelectrophoresis. Clin. Sci. 40,411. Davies J. W. L. (1982) Effects of burning skin and ofher tissues. In: Physiological Responses to Burning lnjuy. London: Academic, pp. 9-36. Fessler J. H. (1974) Self-assembly of collagen. 1. Supramolec. Strucf. 2,103.

Hershey F. B., Lewis C., Johnson G. et al. (1965) Effects of heat on pyridine nucleotide dehydrogenase, aldolase and fumarase of human epidermis. 1. Invest. Dewnatol. 44,~. Laemmli U. K. (1970) Cleavage of structural proteins during the assembly of the head of bacteriophage T4. Nature 22 7,680. McPherson J. M. and Piez K. A. (1988) Collagen in dermal wound repair In: Clark R. A. F. and Henson P. M. (eds), The Mokmdar and Cellular Biology 4 Wowld Repair. New York: Plenum. pp. 471-491. Moncrief J. A. (1979) The body’s response to heat. In: Artz C. P., Moncrief J. A. and Pruitt B. A. (eds), Burns: A Team Approach. Philadelphia: W. B. Saunders, pp. 23-44. Sakakibara S., Inouye K., Shudo K. et al. (1973) Synthesis of (Pro-Hyp-Gly), of defined molecular weights. Evidence for the stabilization of collagen triple helix by hydroxyproline. Biochim. Biophys. Acta 303, 198.

Siegel R. C., Vater A. and Harris E. D. Jr (1978) The effect of cross-linking in the collagen fibril on degradation by rheumatoid synovial collagenase. Arthritis Rheum. 21,591. Smith J. G., Wehr R. F., Badger N. L. et al. (1974) Urinary hydroxyproline: source of increase after thermal bums. I+OC.%c. Exp. Biol. Med. 145, 897. Traub W. and Piez K. A. (1971) The chemistry and structure of collagen. Adv. Rot. Chem. 25, 243.

Paper accepted 5 Feburary 1991.

Correspondenceshouti be addressed to: Dr P. G. Shakespeare, Laing Laboratory for Burn Injury Investigation, Odstock Hospital, Salisbury SP2 8BJ, Wiltshire, UK.

4th European Bums Congress 23-26

September 1991, Barcelona, Spain

There wilJ be Panel Sessions, State of the Art Lectures, Free papers and a pre-congress course in Basic Bum Care on 23 September. For further information contact: Professor J. A. Banu Roda, c/ Pau Claris 138, 7"4a, Edifice Layetana, 08009 Barcelona, Spain. Fax: (3) 215 72 87