- Email: [email protected]
Functional analysis of T lymphocytes infiltrating the dermis and epidermis of post-burn hypertrophic scar tissues
Burns 25 (1999) 43±48
Functional analysis of T lymphocytes in®ltrating the dermis and epidermis of post-burn hypertrophic scar tissues Paola Bernabei a, Laura Rigamonti a, Silvia Ariotti a, Maurizio Stella b, Carlotta Castagnoli c, Francesco Novelli a, * a
Department of Clinical and Biological Sciences, University of Turin, 10043 Orbassano, Italy b Department of Plastic Surgery and Burn Unit, Trauma Centre, Italy c Piedmontese Foundation for Burn Studies and Research, Turin, Italy Accepted 3 August 1998
Abstract The cytokine pro®le of T cell clones (TCC) from the dermis and epidermis of burn patients with hypertrophic scars (HS) in active (AHS) and remission phases (RHS) was determined in this study. We found that AHS tissues are heavily in®ltrated by Type 0±Type 1 polarized CD3 + lymphocytes producing high IFN-g and low IL-4 levels. Analysis of their surface marker phenotype showed that the high IFN-g production was shared equally between the CD4 + TCRa/b and CD8 + TCRa/b clones. The pro®le of TCC from RHS tissues revealed pronounced in®ltration of Type 0±Type 1 polarized lymphocytes with an even more evident Type 1 pro®le. However, the levels of IFN-g produced by RHS-derived TCC were 4±6 times lower than those produced by AHS-derived TCC. These data show that high levels of IFN-g produced by Type 0±Type 1 lymphocytes in®ltrating HS are a feature of AHS, whereas reduction of this ability to produce high levels of IFN-g, though without a shift towards a Type 0±Type 2 phenotype through an increase in IL-4, is characteristic of RHS. # 1999 Elsevier Science Ltd and ISBI. All rights reserved. Keywords: Hypertrophic scars; Th1/Th2; Cytokines; Burns
1. Introduction The pathogenic factors involved in hypertrophic scarring characterized by in¯ammation and hyperproliferation suggest that an altered cellular immune response in epidermis and dermis is primarily responsible [1±5]. This possibility is supported by the detection of abundant in®ltration by activated T lymphocytes [1], and by ectopic expression of HLA-DR [2], ICAM-1 (CD54) [3] and CD36 [4] molecules on keratinocytes and ®broblasts in the tissue involved, while changes in the biosynthesis of some in¯ammatory cytokines, such as IFN-g and TNF-a, in post-burn hypertrophic scar * Corresponding author. Dipartimento di Scienze Cliniche e Biologiche, Ospedale S. Luigi Gonzaga, 10043 Orbassano, Italy. Tel.: +39-11-9038638; fax: +39-11-9038639; E-mail: novelli@pasteur. sluigi.unito.it.
(HS) tissues [5], are evidence of their pivotal role in triggering the sustained in¯ammatory response. A correct immunological response needs the dierentiation of T helper (Th) lymphocytes into eector T cells able to in¯uence the cell-mediated or humoral immune response according to antigen type and characteristics. In the case of parasitic endocellular infections, the adaptive immune response is directed by Type 1 Th cells (Th1), which produce high levels of IFN-g and IL-2, and activate and induce macrophages, natural killer cells (NK) and cytotoxic CD8 + T lymphocytes to trigger the cell-mediated response [6]. In the case of allergens or helminthic infestations, the response is directed by Type 2 Th cells (Th2), which produce IL-4 and IL-5, and induce antibody production (in particular IgE) and eosinophilia to trigger the humoral response [7]. The immune response is also in¯uenced by the functional dierentiation of Th lymphocytes primed in the
0305-4179/99/$19.00 + 0.00 # 1999 Elsevier Science Ltd and ISBI. All rights reserved. PII: S 0 3 0 5 - 4 1 7 9 ( 9 8 ) 0 0 1 2 8 - 4
44
P. Bernabei et al. / Burns 25 (1999) 43±48
epidermis and dermis. Th1 cells are known to play a crucial role in delayed type hypersensitivity and contact hypersensitivity (type IV) [8]. In each case, the antigens make contact with the skin, which is both a mechanical barrier and an important regulator of the immune response. There is some evidence that a wide variety of in¯ammatory cytokines involved in recruitment and activation of the eector cells of an immune response are locally induced by antigenic stimuli or trauma [9]. Immunohistochemical examination of skin-in®ltrating cells indicated an abundant presence of T cells in AHS [1]. Their functional dierentiation and the consequences of their in®ltration were investigated by evaluating the cytokine pro®le of T cell clones (TCC) obtained from the dermis and epidermis of active (AHS) and remission (RHS) HS biopsies. The method described by Moretta et al. [10] enables T cells to be cloned with >80% eciency and was used to form a picture of the entire T cell population in®ltrating HS tissues [1]. Evaluation of the TCC cytokine pro®le provides critical information concerning the eector functions of HS in®ltrating T cells. In this study, we evaluated the production of both IFN-g, a typical proin¯ammatory cytokine [11], and IL-4, a typical anti-in¯ammatory cytokine [12], to evaluate the dierentiation status of the TCC. Their production is, in fact, used to class Th cells as Type 1 (IFN-g producing), Type 2 (IL-4 producing) and Type 0 (IFN-g and IL-4 producing) [13]. The results show that clones from AHS are strongly polarized towards a Type 0±Type 1 phenotype characterized by high IFN-g and low IL-4 production, and that decreased IFN-g production (4±6-fold in this study) is a characteristic of RHS. 2. Materials and methods 2.1. Chemicals and reagents RPMI 1640 medium, fetal calf serum (FCS), L-glutamine, penicillin, streptomycin, gentamycin and pyruvate were from Gibco Laboratories, Grand Island, NY; dispase was from Boehringer Mannheim, Milan, Italy; phytohaemagglutinin (PHA) and ionomycin were from Sigma Chemical Co., St Louis, MO; Terasaki plates and round-bottomed 96-well plates were from Corning Costar Corporation, Cambridge, MA; PBS, BSA, 2-ME, sodium azide, were from Merck Chemicals, Milan, Italy; anti-CD3 OKT3 mAb was from Ortho Pharmaceutical Corp., Raritan, NJ; ¯uoresceinated (FITC) or phycoerythrinated (PE) antiCD3, Leu-2a (CD8)-FITC, Leu-3a (CD4)-PE, TCR a/ b-FITC, TCR g/d-PE, isotype control mouse IgG1 and IgG2 PE or FITC conjugated were from Becton
Dickinson, Milan, Italy; recombinant IL-2 (rIL-2) was from EuroCetus, Milan, Italy. 2.2. Patients Five selected patients with a 10±40% burned surface area (BSA%) discharged from the Turin Burn Centre were followed as outpatients for at least 36 months and classi®ed according to our scheme. AHS were elevated, erythematous, often painful, with variable degrees of skin retraction aecting articular motility: telangectases, small bullae and trophic ulcers were sometimes present despite continuous compressive therapy, massage and physiotherapy, and did not show any sign of regression 18 months after induction. RHS were classi®ed as a function of the progressive decrease of the in¯ammatory symptoms. They were ¯attened and normochromic, and dysaesthesia was decreased. Patients underwent plastic surgery under general anaesthesia for scar correction. Biopsies were taken after informed consent from excised scars. The study was conducted on biopsies from two AHS patients and three RHS patients. 2.3. Media The culture medium was RPMI 1640 containing penicillin, streptomycin, gentamycin, pyruvate, 2 mM ÿ5 L-glutamine and 2 10 M 2-ME, and supplemented with 10% heat-inactivated FCS and 5% heat-inactivated human serum (referred to hereafter as complete medium). All the in vitro cultures were performed at 378C in a humidi®ed 5% CO2 atmosphere. 2.4. Preparation of T cells Dermis and epidermis were dissociated by incubating biopsies with 2.5 U/ml dispase dissolved in complete medium for 12 h at 48C. To obtain dermal, epidermal or total in®ltrating T lymphocytes, dispasetreated or undissociated biopsies were minced and the T cells were allowed to emigrate from the tissue during 3 days of cultivation in complete medium supplemented with 20 U/ml rIL-2. 2.5. TCC and their maintenance T cells from undissociated biopsies or from dermis and epidermis of HS were seeded under limiting dilution conditions (0.1 cells/well) in Terasaki plates in complete medium containing 105 irradiated (50 Gy) peripheral blood mononuclear cells (PBMC) as feeder cells, 2.5 mg/ml PHA and 20 U/ml rIL-2, as reported elsewhere [10]. After 2 weeks, the clones were transferred to round-bottomed 96-well plates and maintained in culture by re-stimulation every 9±12 days
P. Bernabei et al. / Burns 25 (1999) 43±48
with 2.5 mg/ml PHA, 20 U/ml rIL-2 and of irradiated allogeneic PBMC as feeder cells. rIL-2 was replaced every 3 days to allow growth in a state of ongoing activation. These are the standard conditions of our studies. All short-term assays were performed by recovering clones 1 day before re-stimulation to avoid interference with the presence of feeder cells.
2.6. Immunophenotyping of TCC Cell surface marker analysis of ongoing TCC was performed in a FACScan ¯ow cytometer (Becton Dickinson, Milan, Italy) by using ¯uoresceinated (FITC) or phycoerythrinated (PE) anti-CD3, antiCD4, anti-CD8, anti a/b-TCR or anti g/d-TCR mAb.
2.7. Characterization of the cytokine pro®le of TCC After 1 month of culture, the cytokine secretion pro®le of these TCC was evaluated by stimulating 0.2 106 T cell blasts from each clone in 0.200 ml in complete medium with anti-CD3 mAb (5 mg/ml) and ionomycin (500 ng/ml). After 24 h, culture supernatants were collected and stored at ÿ708C until use. IFN-g and IL-4 were determined with a commercial ELISA kit (Genzyme, Cambridge, MA).
Fig. 1. Production of IFN-g and IL-4 by T cell clones from the dermis (Q) and epidermis (w) of a patient with an AHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
45
3. Results In the ®rst experiment, 120 TCC from the dermis and 120 TCC from the epidermis of one AHS patient were generated. Their cytokine pro®les showed that both sets produced very high IFN-g levels (mean 529 pg/ml, range 72±2042) and low IL-4 levels (mean 57 pg/ml, range 0±773) (Fig. 1). Analysis of the TCC functional phenotypes revealed a dierent distribution of Type 0 and Type 1 clones between the dermis and the epidermis. TCC from the dermis produced high IFN-g levels and almost all produced little IL-4 (Fig. 1): 99% (119 TCC) were Type 0, 1% (1 TCC) was Type 1 and none was Type 2. In contrast, TCC from the epidermis displayed a clear Type 1 phenotype and the ability to produce IL-4 was lost: 40% (48 TCC) were Type 0, 60% (72 TCC) were Type 1 and none was Type 2. 58 TCC from another AHS patient were analyzed. When their membrane phenotype (CD4, CD8, TCR a/ b, TCR g/d) and IFN-g and IL-4 production was investigated, cyto¯uorimetric analysis revealed that 73% (42 TCC) were CD4 + TCRa/b + , 24% (14 TCC) were CD8 + TCRa/b+ and only 1.5% were CD4 + TCR g/d+ (1 TCC) and CD8 + TCRg/d+ (1 TCC). Characterization of the cytokine production of TCC from AHS patient con®rmed their polarization toward the Type 0±Type 1 phenotype (Fig. 2). While 38 CD4 + TCC produced high levels of both IFN-g (mean 380 pg/ml, range 30±1040) and IL-4 (mean 230 pg/ml, range 0±1393), 13 CD8 + TCC produced high
Fig. 2. Production of IFN-g and IL-4 by CD4 + (w) and CD8 + (.) T cell clones from a patient with an AHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
46
P. Bernabei et al. / Burns 25 (1999) 43±48
Fig. 3. Production of IFN-g and IL-4 by T cell clones (w) from a patient with an RHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
levels of IFN-g (mean 369 pg/ml, range 0±920) and lower levels of IL-4 (mean 6 pg/ml, range 0±57). Phenotyping of each CD4 + or CD8 + TCC con®rmed the Type 0±Type 1 polarization observed in the ®rst experiment. In particular, 47% of 38 CD4 + TCC were Type 0 (18 TCC), 47% were Type 1 (18 TCC) and only 6% were Type 2 (2 TCC). Type 1 polarization became more marked among the 13 CD8 + TCC obtained: 31% produced both IFN-g and IL-4 (4 TCC), while 69% produced only IFN-g (9 TCC) and none produced only IL-4. 76 TCC were obtained from a patient with RHS. Characterization of their cytokine production revealed a marked Type 1 polarization (IFN-g: mean 90 pg/ml, range 24±201; IL-4: mean 5 pg/ml, range 0±342) (Fig. 3): 96% (73 TCC) were Type 1, 3% (2 TCC) were Type 0 and 1% (1 TCC) was Type 2. These TCC, however, produced 4±6-fold less IFN-g than the AHS clone (Fig. 4).
4. Discussion This study indicates that AHS are in®ltrated by strongly Type 0/Type 1 polarized lymphocytes characterized by high IFN-g production in both the dermis and the epidermis. IL-4 production in the dermis was not such as to induce a decrease of the high IFN-g level. IL-4 production was greatly decreased in the epidermis. Analysis of the TCC phenotypes shows that this high IFN-g production is shared equally between the CD4 + TCRa/b and CD8 + TCRa/b clones.
Fig. 4. Production of IFN-g in AHS (w) and RHS (q) derived T cell clones after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
These data ®t in well with those obtained by Castagnoli et al. demonstrating a high production of in¯ammatory cytokines, including IFN-g, in AHS tissues [1]. Moreover, the phenotype data show that RHS are in®ltrated by Type 0±Type 1 polarized lymphocytes with an even more evident Type 1 pro®le. Here, however, IFN-g production is 4±6-times lower than in AHS. Lower IFN-g levels during remission could be determined by a change in the microenvironment in which the eector functions of activated T cells are exerted. One possibility is a diminished availability of costimulatory factors released by the accessory cells (mainly monocytes/macrophages), which, like IL-12, act directly on the IFN-g promoter by raising its transcription levels and thus altering the amount of IFN-g produced [14]. Lower IFN-g levels during healing could also re¯ect a fall in the number of in®ltrating T cells. Remission is, in fact, accompanied by a decreased number of in®ltrating T cells with an activated phenotype [1] that may depend on their diminished ability to produce IFN-g and hence a decrease in their proliferation due to their inability to use it autocrinally [15±18]. Reduction of the number of activated T cells in®ltrating HS could be the result of diminished production in the microenvironment of the chemotactic factors that recruit them and sustain the in¯ammation. Our recent data indicate that healing of HS is correlated with reduced IL-15 production by keratinocytes (Castagnoli C. et al. in preparation). Like IL-2, IL-15 is a growth factor for T cells [19] and its potent chemotactic eect recruits them into the places where it is produced [20]. An enhanced T cell susceptibility to
P. Bernabei et al. / Burns 25 (1999) 43±48
apoptosis during remission could thus account for the reduced in®ltration. The eector response of T cells in peripheral tissues is physiologically governed by mechanisms that trigger their suicide and thus prevent their continuous and consequently dangerous expansion [21]. This apoptosis is mainly controlled both by gene products regulated positively following the antigen stimulus, such as the Fas-Fas-ligand system, and by the availability or otherwise of growth factors, such as IL2, which prevent apoptosis by maintaining a high expression of the genes that provide protection against it [17, 18, 21]. Our data indicating that the remission of HS is associated with a decreased ability of in®ltrating T cells to produce IFN-g apparently con¯ict with those obtained by Harrop et al. showing that IFN-g may have a role in remission of HS through the inhibition of collagen production by ®broblasts [22, 23]. Our data underscore the role of the high amount of IFN-g produced by in®ltrating T cells in sustaining and amplifying the in¯ammatory response observed in AHS. However, the eects of IFN-g on the immune response are often double-edged [15±17]. We have shown that IFN-g may favour both the expansion or the apoptosis of Type 1 lymphocytes depending on the density of the two chains constituting its receptor complex [18]. The kind of signal transduced by IFN-g in activated Type 1 lymphocytes during the remission of AHS could be more important than the levels of IFN-g produced. Interestingly, it has been reported that nitric oxide (NO) production is reduced in HS tissues [24]. Data obtained in our lab indicate that it is a critical environmental factor that up-regulates the expression of IFNg receptor chains and primes T cells for IFN-gmediated apoptosis (Allione A. et al., submitted for publication). When NO production is low, the downregulation of T cells IFN-g receptor chain expression may promote the expansion of Type 1 in®ltrating lymphocytes. In conclusion, these data indicate that study of the in®ltrating T cell response may explain the in¯ammatory events observed in HS. They could also be turned to clinical advantage the enhanced Type 1 response with Type 2 cytokines as IL-4 or IL-12 antagonists. Acknowledgements This work was supported by grants from Fondazione Piemontese per gli Studi e le Ricerche sulle Ustioni, Istituto Superiore di Sanita' (Special Project on AIDS) and Associazione Italiana Sclerosi Multipla to F.N. S.A. was supported by a fellowship from Fondazione Piemontese per gli Studi e le Ricerche sulle Ustioni. We thank Dr J. Ilie for critically reading the manuscript.
47
References [1] Castagnoli C, Trombotto C, Ondei S et al. Characterization of T-cell subsets in®ltrating post-burn hypertrophic scar tissues. Burns 1997;23:565±72. [2] Castagnoli C, Stella M, Magliacani G, Alasia ST, Richiardi PM. Anomalous expression of HLA class II molecules on keratinocytes and ®broblasts in hypertrophic scars consequent to thermal injury. Clin Exp Immunol 1990;82:350±4. [3] Castagnoli C, Stella M, Magliacani G, Ferrone S, Richiardi PM. Similar ectopic expression of ICAM-1 and HLA Class II molecules in hypertrophic scars following thermal injury. Burns 1994;20:430±3. [4] Castagnoli C, Stella M, Menegatti E, Calcagni M, Magliacani G, Teich Alasia S, Alessio M. CD36 expression in hypertrophic scars. Annals of Mediterranean Burns Club 1995;8:214±8. [5] Castagnoli C, Stella M, Berthod C, Magliacani G, Richiardi PM. TNF production in hypertrophic scarring. Cell Immunol 1993;147:51±63. [6] Hsieh CS, Macatonia SE, Tripp CS, Wolf SF, O'Garra A, Murphy KM. Development of Th1 CD4 + T cells through IL12 produced by Listeria-induced macrophages. Science 1993;260:547±9. [7] Seder RA, Paul WE, Davis MM, Fazekas de St Groth B. The presence of IL-4 during in vitro priming determines the lymphokine-producing potential of CD4 + T cells from T cell receptor transgenic mice. J Exp Med 1992;176:1091±8. [8] Hauser C. Cultured epidermal Langherans cells activate eector T cells for contact sensitivity. J Invest Dermatol 1990;95:436± 40. [9] Kupper TS, Luger TA, Schwarz T. editors. The Second Decade of Epidermal Cytokines: an Overview. Dekker, New York, 1993. p. 239. [10] Moretta A, Pantaleo G, Moretta L, Cerottini JC, Mingari MC. Direct demonstration of the clonogenic potential of every human peripheral blood T cell. Clonal analysis of HLA-DR expression and cytolytic activity. J Exp Med 1983;157:743±54. [11] Billiau A. Interferon g. Biology and role in pathogenesis. Adv Immunol 1996;62:61±130. [12] Maher DW, Davis I, Boyd AW, Morstyn G. Human interleukin-4: an immunomodulator with potential therapeutic applications. Prog Growth Factor Res 1991;3:43±56. [13] Mosmann TR, Coman RL. Th1 and Th2 cells: dierent patterns of lymphokine secretion lead to dierent functional properties. Annu Rev Immunol 1989;7:145±73. [14] Manetti R, Gerosa F, Giudizi MG et al. Interleukin 12 induces stable priming for interferon gamma (IFN-g) production during dierentiation of human T helper (Th) cells and transient IFN-g production in established Th2 cell clones. J Exp Med 1994;179:1273±83. [15] Novelli F, Giovarelli M, Reber-Liske R, Virgallita G, Garotta G, Forni G. Blockade of physiologically secreted IFN-g inhibits human T lymphocyte and natural killer cell activation. J Immunol 1991;147:1445±52. [16] Novelli F, Di Pierro F, Francia di Celle P, Bertini S, Aaticati P, Garotta G, Forni G. Environmental signals in¯uencing expression of the IFN-g receptor on human T cells control whether IFN-g promotes proliferation or apoptosis. J Immunol 1994;152:496±504. [17] Novelli F, Bernabei P, Ozmen L et al. Switching on of the proliferation or apoptosis of activated T human lymphocytes by IFN-g is correlated with the dierential expression of the alpha and beta chains of its receptor. J Immunol 1996;157:1935±43. [18] Novelli F, D'Elios MM, Bernabei P et al. Expression and role in apoptosis of the alpha and beta chains of the IFN-g receptor on human Th1 and Th2 clones. J Immunol 1997;159:206±13.
48
P. Bernabei et al. / Burns 25 (1999) 43±48
[19] Carson WE, Giri JG, Lindemann MJ et al. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J Exp Med 1994;180:1395±401. [20] Wilkinson PC, Liew FY. Chemoattraction of human blood T lymphocytes by IL-15. J Exp Med 1995;195:1255±62. [21] Lenardo MJ. Fas and the art of lymphocyte maintenance. J Exp Med 1996;183:721±4. [22] Harrop AR, Ghahary A, Scott PG, Forsyth N, Uji-Friedland A, Tredget EE. Regulation of collagen synthesis and mRNA ex-
pression in normal and hypertrophic scars ®broblasts in vitro by interferon-gamma. J Surg Res 1995;58:471±7. [23] Tredget EE, Nedelec B, Scott PG, Ghahary A. Hypertrophic scars, keloids, and contractures. The cellular and molecular basis for therapy. Surg Clin N Am 1997;77:701±30. [24] Wang R, Ghahary A, Shen YJ, Scott PG, Tredget EE. Nitric oxide synthase expression and nitric oxide production are reduced in hypertrophic scar tissues and ®broblasts. J Invest Dermatol 1997;108:438±44.
Functional analysis of T lymphocytes in®ltrating the dermis and epidermis of post-burn hypertrophic scar tissues Paola Bernabei a, Laura Rigamonti a, Silvia Ariotti a, Maurizio Stella b, Carlotta Castagnoli c, Francesco Novelli a, * a
Department of Clinical and Biological Sciences, University of Turin, 10043 Orbassano, Italy b Department of Plastic Surgery and Burn Unit, Trauma Centre, Italy c Piedmontese Foundation for Burn Studies and Research, Turin, Italy Accepted 3 August 1998
Abstract The cytokine pro®le of T cell clones (TCC) from the dermis and epidermis of burn patients with hypertrophic scars (HS) in active (AHS) and remission phases (RHS) was determined in this study. We found that AHS tissues are heavily in®ltrated by Type 0±Type 1 polarized CD3 + lymphocytes producing high IFN-g and low IL-4 levels. Analysis of their surface marker phenotype showed that the high IFN-g production was shared equally between the CD4 + TCRa/b and CD8 + TCRa/b clones. The pro®le of TCC from RHS tissues revealed pronounced in®ltration of Type 0±Type 1 polarized lymphocytes with an even more evident Type 1 pro®le. However, the levels of IFN-g produced by RHS-derived TCC were 4±6 times lower than those produced by AHS-derived TCC. These data show that high levels of IFN-g produced by Type 0±Type 1 lymphocytes in®ltrating HS are a feature of AHS, whereas reduction of this ability to produce high levels of IFN-g, though without a shift towards a Type 0±Type 2 phenotype through an increase in IL-4, is characteristic of RHS. # 1999 Elsevier Science Ltd and ISBI. All rights reserved. Keywords: Hypertrophic scars; Th1/Th2; Cytokines; Burns
1. Introduction The pathogenic factors involved in hypertrophic scarring characterized by in¯ammation and hyperproliferation suggest that an altered cellular immune response in epidermis and dermis is primarily responsible [1±5]. This possibility is supported by the detection of abundant in®ltration by activated T lymphocytes [1], and by ectopic expression of HLA-DR [2], ICAM-1 (CD54) [3] and CD36 [4] molecules on keratinocytes and ®broblasts in the tissue involved, while changes in the biosynthesis of some in¯ammatory cytokines, such as IFN-g and TNF-a, in post-burn hypertrophic scar * Corresponding author. Dipartimento di Scienze Cliniche e Biologiche, Ospedale S. Luigi Gonzaga, 10043 Orbassano, Italy. Tel.: +39-11-9038638; fax: +39-11-9038639; E-mail: novelli@pasteur. sluigi.unito.it.
(HS) tissues [5], are evidence of their pivotal role in triggering the sustained in¯ammatory response. A correct immunological response needs the dierentiation of T helper (Th) lymphocytes into eector T cells able to in¯uence the cell-mediated or humoral immune response according to antigen type and characteristics. In the case of parasitic endocellular infections, the adaptive immune response is directed by Type 1 Th cells (Th1), which produce high levels of IFN-g and IL-2, and activate and induce macrophages, natural killer cells (NK) and cytotoxic CD8 + T lymphocytes to trigger the cell-mediated response [6]. In the case of allergens or helminthic infestations, the response is directed by Type 2 Th cells (Th2), which produce IL-4 and IL-5, and induce antibody production (in particular IgE) and eosinophilia to trigger the humoral response [7]. The immune response is also in¯uenced by the functional dierentiation of Th lymphocytes primed in the
0305-4179/99/$19.00 + 0.00 # 1999 Elsevier Science Ltd and ISBI. All rights reserved. PII: S 0 3 0 5 - 4 1 7 9 ( 9 8 ) 0 0 1 2 8 - 4
44
P. Bernabei et al. / Burns 25 (1999) 43±48
epidermis and dermis. Th1 cells are known to play a crucial role in delayed type hypersensitivity and contact hypersensitivity (type IV) [8]. In each case, the antigens make contact with the skin, which is both a mechanical barrier and an important regulator of the immune response. There is some evidence that a wide variety of in¯ammatory cytokines involved in recruitment and activation of the eector cells of an immune response are locally induced by antigenic stimuli or trauma [9]. Immunohistochemical examination of skin-in®ltrating cells indicated an abundant presence of T cells in AHS [1]. Their functional dierentiation and the consequences of their in®ltration were investigated by evaluating the cytokine pro®le of T cell clones (TCC) obtained from the dermis and epidermis of active (AHS) and remission (RHS) HS biopsies. The method described by Moretta et al. [10] enables T cells to be cloned with >80% eciency and was used to form a picture of the entire T cell population in®ltrating HS tissues [1]. Evaluation of the TCC cytokine pro®le provides critical information concerning the eector functions of HS in®ltrating T cells. In this study, we evaluated the production of both IFN-g, a typical proin¯ammatory cytokine [11], and IL-4, a typical anti-in¯ammatory cytokine [12], to evaluate the dierentiation status of the TCC. Their production is, in fact, used to class Th cells as Type 1 (IFN-g producing), Type 2 (IL-4 producing) and Type 0 (IFN-g and IL-4 producing) [13]. The results show that clones from AHS are strongly polarized towards a Type 0±Type 1 phenotype characterized by high IFN-g and low IL-4 production, and that decreased IFN-g production (4±6-fold in this study) is a characteristic of RHS. 2. Materials and methods 2.1. Chemicals and reagents RPMI 1640 medium, fetal calf serum (FCS), L-glutamine, penicillin, streptomycin, gentamycin and pyruvate were from Gibco Laboratories, Grand Island, NY; dispase was from Boehringer Mannheim, Milan, Italy; phytohaemagglutinin (PHA) and ionomycin were from Sigma Chemical Co., St Louis, MO; Terasaki plates and round-bottomed 96-well plates were from Corning Costar Corporation, Cambridge, MA; PBS, BSA, 2-ME, sodium azide, were from Merck Chemicals, Milan, Italy; anti-CD3 OKT3 mAb was from Ortho Pharmaceutical Corp., Raritan, NJ; ¯uoresceinated (FITC) or phycoerythrinated (PE) antiCD3, Leu-2a (CD8)-FITC, Leu-3a (CD4)-PE, TCR a/ b-FITC, TCR g/d-PE, isotype control mouse IgG1 and IgG2 PE or FITC conjugated were from Becton
Dickinson, Milan, Italy; recombinant IL-2 (rIL-2) was from EuroCetus, Milan, Italy. 2.2. Patients Five selected patients with a 10±40% burned surface area (BSA%) discharged from the Turin Burn Centre were followed as outpatients for at least 36 months and classi®ed according to our scheme. AHS were elevated, erythematous, often painful, with variable degrees of skin retraction aecting articular motility: telangectases, small bullae and trophic ulcers were sometimes present despite continuous compressive therapy, massage and physiotherapy, and did not show any sign of regression 18 months after induction. RHS were classi®ed as a function of the progressive decrease of the in¯ammatory symptoms. They were ¯attened and normochromic, and dysaesthesia was decreased. Patients underwent plastic surgery under general anaesthesia for scar correction. Biopsies were taken after informed consent from excised scars. The study was conducted on biopsies from two AHS patients and three RHS patients. 2.3. Media The culture medium was RPMI 1640 containing penicillin, streptomycin, gentamycin, pyruvate, 2 mM ÿ5 L-glutamine and 2 10 M 2-ME, and supplemented with 10% heat-inactivated FCS and 5% heat-inactivated human serum (referred to hereafter as complete medium). All the in vitro cultures were performed at 378C in a humidi®ed 5% CO2 atmosphere. 2.4. Preparation of T cells Dermis and epidermis were dissociated by incubating biopsies with 2.5 U/ml dispase dissolved in complete medium for 12 h at 48C. To obtain dermal, epidermal or total in®ltrating T lymphocytes, dispasetreated or undissociated biopsies were minced and the T cells were allowed to emigrate from the tissue during 3 days of cultivation in complete medium supplemented with 20 U/ml rIL-2. 2.5. TCC and their maintenance T cells from undissociated biopsies or from dermis and epidermis of HS were seeded under limiting dilution conditions (0.1 cells/well) in Terasaki plates in complete medium containing 105 irradiated (50 Gy) peripheral blood mononuclear cells (PBMC) as feeder cells, 2.5 mg/ml PHA and 20 U/ml rIL-2, as reported elsewhere [10]. After 2 weeks, the clones were transferred to round-bottomed 96-well plates and maintained in culture by re-stimulation every 9±12 days
P. Bernabei et al. / Burns 25 (1999) 43±48
with 2.5 mg/ml PHA, 20 U/ml rIL-2 and of irradiated allogeneic PBMC as feeder cells. rIL-2 was replaced every 3 days to allow growth in a state of ongoing activation. These are the standard conditions of our studies. All short-term assays were performed by recovering clones 1 day before re-stimulation to avoid interference with the presence of feeder cells.
2.6. Immunophenotyping of TCC Cell surface marker analysis of ongoing TCC was performed in a FACScan ¯ow cytometer (Becton Dickinson, Milan, Italy) by using ¯uoresceinated (FITC) or phycoerythrinated (PE) anti-CD3, antiCD4, anti-CD8, anti a/b-TCR or anti g/d-TCR mAb.
2.7. Characterization of the cytokine pro®le of TCC After 1 month of culture, the cytokine secretion pro®le of these TCC was evaluated by stimulating 0.2 106 T cell blasts from each clone in 0.200 ml in complete medium with anti-CD3 mAb (5 mg/ml) and ionomycin (500 ng/ml). After 24 h, culture supernatants were collected and stored at ÿ708C until use. IFN-g and IL-4 were determined with a commercial ELISA kit (Genzyme, Cambridge, MA).
Fig. 1. Production of IFN-g and IL-4 by T cell clones from the dermis (Q) and epidermis (w) of a patient with an AHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
45
3. Results In the ®rst experiment, 120 TCC from the dermis and 120 TCC from the epidermis of one AHS patient were generated. Their cytokine pro®les showed that both sets produced very high IFN-g levels (mean 529 pg/ml, range 72±2042) and low IL-4 levels (mean 57 pg/ml, range 0±773) (Fig. 1). Analysis of the TCC functional phenotypes revealed a dierent distribution of Type 0 and Type 1 clones between the dermis and the epidermis. TCC from the dermis produced high IFN-g levels and almost all produced little IL-4 (Fig. 1): 99% (119 TCC) were Type 0, 1% (1 TCC) was Type 1 and none was Type 2. In contrast, TCC from the epidermis displayed a clear Type 1 phenotype and the ability to produce IL-4 was lost: 40% (48 TCC) were Type 0, 60% (72 TCC) were Type 1 and none was Type 2. 58 TCC from another AHS patient were analyzed. When their membrane phenotype (CD4, CD8, TCR a/ b, TCR g/d) and IFN-g and IL-4 production was investigated, cyto¯uorimetric analysis revealed that 73% (42 TCC) were CD4 + TCRa/b + , 24% (14 TCC) were CD8 + TCRa/b+ and only 1.5% were CD4 + TCR g/d+ (1 TCC) and CD8 + TCRg/d+ (1 TCC). Characterization of the cytokine production of TCC from AHS patient con®rmed their polarization toward the Type 0±Type 1 phenotype (Fig. 2). While 38 CD4 + TCC produced high levels of both IFN-g (mean 380 pg/ml, range 30±1040) and IL-4 (mean 230 pg/ml, range 0±1393), 13 CD8 + TCC produced high
Fig. 2. Production of IFN-g and IL-4 by CD4 + (w) and CD8 + (.) T cell clones from a patient with an AHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
46
P. Bernabei et al. / Burns 25 (1999) 43±48
Fig. 3. Production of IFN-g and IL-4 by T cell clones (w) from a patient with an RHS after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
levels of IFN-g (mean 369 pg/ml, range 0±920) and lower levels of IL-4 (mean 6 pg/ml, range 0±57). Phenotyping of each CD4 + or CD8 + TCC con®rmed the Type 0±Type 1 polarization observed in the ®rst experiment. In particular, 47% of 38 CD4 + TCC were Type 0 (18 TCC), 47% were Type 1 (18 TCC) and only 6% were Type 2 (2 TCC). Type 1 polarization became more marked among the 13 CD8 + TCC obtained: 31% produced both IFN-g and IL-4 (4 TCC), while 69% produced only IFN-g (9 TCC) and none produced only IL-4. 76 TCC were obtained from a patient with RHS. Characterization of their cytokine production revealed a marked Type 1 polarization (IFN-g: mean 90 pg/ml, range 24±201; IL-4: mean 5 pg/ml, range 0±342) (Fig. 3): 96% (73 TCC) were Type 1, 3% (2 TCC) were Type 0 and 1% (1 TCC) was Type 2. These TCC, however, produced 4±6-fold less IFN-g than the AHS clone (Fig. 4).
4. Discussion This study indicates that AHS are in®ltrated by strongly Type 0/Type 1 polarized lymphocytes characterized by high IFN-g production in both the dermis and the epidermis. IL-4 production in the dermis was not such as to induce a decrease of the high IFN-g level. IL-4 production was greatly decreased in the epidermis. Analysis of the TCC phenotypes shows that this high IFN-g production is shared equally between the CD4 + TCRa/b and CD8 + TCRa/b clones.
Fig. 4. Production of IFN-g in AHS (w) and RHS (q) derived T cell clones after 24 h stimulation with 5 mg/ml anti-CD3 antibody and 500 ng/ml ionomycin.
These data ®t in well with those obtained by Castagnoli et al. demonstrating a high production of in¯ammatory cytokines, including IFN-g, in AHS tissues [1]. Moreover, the phenotype data show that RHS are in®ltrated by Type 0±Type 1 polarized lymphocytes with an even more evident Type 1 pro®le. Here, however, IFN-g production is 4±6-times lower than in AHS. Lower IFN-g levels during remission could be determined by a change in the microenvironment in which the eector functions of activated T cells are exerted. One possibility is a diminished availability of costimulatory factors released by the accessory cells (mainly monocytes/macrophages), which, like IL-12, act directly on the IFN-g promoter by raising its transcription levels and thus altering the amount of IFN-g produced [14]. Lower IFN-g levels during healing could also re¯ect a fall in the number of in®ltrating T cells. Remission is, in fact, accompanied by a decreased number of in®ltrating T cells with an activated phenotype [1] that may depend on their diminished ability to produce IFN-g and hence a decrease in their proliferation due to their inability to use it autocrinally [15±18]. Reduction of the number of activated T cells in®ltrating HS could be the result of diminished production in the microenvironment of the chemotactic factors that recruit them and sustain the in¯ammation. Our recent data indicate that healing of HS is correlated with reduced IL-15 production by keratinocytes (Castagnoli C. et al. in preparation). Like IL-2, IL-15 is a growth factor for T cells [19] and its potent chemotactic eect recruits them into the places where it is produced [20]. An enhanced T cell susceptibility to
P. Bernabei et al. / Burns 25 (1999) 43±48
apoptosis during remission could thus account for the reduced in®ltration. The eector response of T cells in peripheral tissues is physiologically governed by mechanisms that trigger their suicide and thus prevent their continuous and consequently dangerous expansion [21]. This apoptosis is mainly controlled both by gene products regulated positively following the antigen stimulus, such as the Fas-Fas-ligand system, and by the availability or otherwise of growth factors, such as IL2, which prevent apoptosis by maintaining a high expression of the genes that provide protection against it [17, 18, 21]. Our data indicating that the remission of HS is associated with a decreased ability of in®ltrating T cells to produce IFN-g apparently con¯ict with those obtained by Harrop et al. showing that IFN-g may have a role in remission of HS through the inhibition of collagen production by ®broblasts [22, 23]. Our data underscore the role of the high amount of IFN-g produced by in®ltrating T cells in sustaining and amplifying the in¯ammatory response observed in AHS. However, the eects of IFN-g on the immune response are often double-edged [15±17]. We have shown that IFN-g may favour both the expansion or the apoptosis of Type 1 lymphocytes depending on the density of the two chains constituting its receptor complex [18]. The kind of signal transduced by IFN-g in activated Type 1 lymphocytes during the remission of AHS could be more important than the levels of IFN-g produced. Interestingly, it has been reported that nitric oxide (NO) production is reduced in HS tissues [24]. Data obtained in our lab indicate that it is a critical environmental factor that up-regulates the expression of IFNg receptor chains and primes T cells for IFN-gmediated apoptosis (Allione A. et al., submitted for publication). When NO production is low, the downregulation of T cells IFN-g receptor chain expression may promote the expansion of Type 1 in®ltrating lymphocytes. In conclusion, these data indicate that study of the in®ltrating T cell response may explain the in¯ammatory events observed in HS. They could also be turned to clinical advantage the enhanced Type 1 response with Type 2 cytokines as IL-4 or IL-12 antagonists. Acknowledgements This work was supported by grants from Fondazione Piemontese per gli Studi e le Ricerche sulle Ustioni, Istituto Superiore di Sanita' (Special Project on AIDS) and Associazione Italiana Sclerosi Multipla to F.N. S.A. was supported by a fellowship from Fondazione Piemontese per gli Studi e le Ricerche sulle Ustioni. We thank Dr J. Ilie for critically reading the manuscript.
47
References [1] Castagnoli C, Trombotto C, Ondei S et al. Characterization of T-cell subsets in®ltrating post-burn hypertrophic scar tissues. Burns 1997;23:565±72. [2] Castagnoli C, Stella M, Magliacani G, Alasia ST, Richiardi PM. Anomalous expression of HLA class II molecules on keratinocytes and ®broblasts in hypertrophic scars consequent to thermal injury. Clin Exp Immunol 1990;82:350±4. [3] Castagnoli C, Stella M, Magliacani G, Ferrone S, Richiardi PM. Similar ectopic expression of ICAM-1 and HLA Class II molecules in hypertrophic scars following thermal injury. Burns 1994;20:430±3. [4] Castagnoli C, Stella M, Menegatti E, Calcagni M, Magliacani G, Teich Alasia S, Alessio M. CD36 expression in hypertrophic scars. Annals of Mediterranean Burns Club 1995;8:214±8. [5] Castagnoli C, Stella M, Berthod C, Magliacani G, Richiardi PM. TNF production in hypertrophic scarring. Cell Immunol 1993;147:51±63. [6] Hsieh CS, Macatonia SE, Tripp CS, Wolf SF, O'Garra A, Murphy KM. Development of Th1 CD4 + T cells through IL12 produced by Listeria-induced macrophages. Science 1993;260:547±9. [7] Seder RA, Paul WE, Davis MM, Fazekas de St Groth B. The presence of IL-4 during in vitro priming determines the lymphokine-producing potential of CD4 + T cells from T cell receptor transgenic mice. J Exp Med 1992;176:1091±8. [8] Hauser C. Cultured epidermal Langherans cells activate eector T cells for contact sensitivity. J Invest Dermatol 1990;95:436± 40. [9] Kupper TS, Luger TA, Schwarz T. editors. The Second Decade of Epidermal Cytokines: an Overview. Dekker, New York, 1993. p. 239. [10] Moretta A, Pantaleo G, Moretta L, Cerottini JC, Mingari MC. Direct demonstration of the clonogenic potential of every human peripheral blood T cell. Clonal analysis of HLA-DR expression and cytolytic activity. J Exp Med 1983;157:743±54. [11] Billiau A. Interferon g. Biology and role in pathogenesis. Adv Immunol 1996;62:61±130. [12] Maher DW, Davis I, Boyd AW, Morstyn G. Human interleukin-4: an immunomodulator with potential therapeutic applications. Prog Growth Factor Res 1991;3:43±56. [13] Mosmann TR, Coman RL. Th1 and Th2 cells: dierent patterns of lymphokine secretion lead to dierent functional properties. Annu Rev Immunol 1989;7:145±73. [14] Manetti R, Gerosa F, Giudizi MG et al. Interleukin 12 induces stable priming for interferon gamma (IFN-g) production during dierentiation of human T helper (Th) cells and transient IFN-g production in established Th2 cell clones. J Exp Med 1994;179:1273±83. [15] Novelli F, Giovarelli M, Reber-Liske R, Virgallita G, Garotta G, Forni G. Blockade of physiologically secreted IFN-g inhibits human T lymphocyte and natural killer cell activation. J Immunol 1991;147:1445±52. [16] Novelli F, Di Pierro F, Francia di Celle P, Bertini S, Aaticati P, Garotta G, Forni G. Environmental signals in¯uencing expression of the IFN-g receptor on human T cells control whether IFN-g promotes proliferation or apoptosis. J Immunol 1994;152:496±504. [17] Novelli F, Bernabei P, Ozmen L et al. Switching on of the proliferation or apoptosis of activated T human lymphocytes by IFN-g is correlated with the dierential expression of the alpha and beta chains of its receptor. J Immunol 1996;157:1935±43. [18] Novelli F, D'Elios MM, Bernabei P et al. Expression and role in apoptosis of the alpha and beta chains of the IFN-g receptor on human Th1 and Th2 clones. J Immunol 1997;159:206±13.
48
P. Bernabei et al. / Burns 25 (1999) 43±48
[19] Carson WE, Giri JG, Lindemann MJ et al. Interleukin (IL) 15 is a novel cytokine that activates human natural killer cells via components of the IL-2 receptor. J Exp Med 1994;180:1395±401. [20] Wilkinson PC, Liew FY. Chemoattraction of human blood T lymphocytes by IL-15. J Exp Med 1995;195:1255±62. [21] Lenardo MJ. Fas and the art of lymphocyte maintenance. J Exp Med 1996;183:721±4. [22] Harrop AR, Ghahary A, Scott PG, Forsyth N, Uji-Friedland A, Tredget EE. Regulation of collagen synthesis and mRNA ex-
pression in normal and hypertrophic scars ®broblasts in vitro by interferon-gamma. J Surg Res 1995;58:471±7. [23] Tredget EE, Nedelec B, Scott PG, Ghahary A. Hypertrophic scars, keloids, and contractures. The cellular and molecular basis for therapy. Surg Clin N Am 1997;77:701±30. [24] Wang R, Ghahary A, Shen YJ, Scott PG, Tredget EE. Nitric oxide synthase expression and nitric oxide production are reduced in hypertrophic scar tissues and ®broblasts. J Invest Dermatol 1997;108:438±44.