Up-regulated perforin expression of CD8+ blood lymphocytes in generalized non-anaphylactic drug eruptions and exacerbated psoriasis

ARTICLE

Cytotoxic T cells (CTL) are known to express the CD8 epitope and to recognize their targets through the T cell receptor in an MHC class I-restricted manner. In such a way specific lysis of target cells by CTLs is effected over secretory and non-secretory pathways [1, 2]. In the non-secretory cytolytic mechanism the linkage of the CTL over Fas ligands and Fas receptors to the target cell leads to its programmed death, i.e. apoptosis. On the other hand, the secretory cytolytic mechanism involves the CTL's secretion of granzymes and perforin (syn. cytolysin), which both are released by a Ca⁺-dependent degranulation process [3]. Human perforin is a 70 kD protein which polymerizes to a cylindric structure to be introduced as a pore into the plasma membrane of the target cell [4]. It has been postulated that the penetration of the target cell by granzymes, water and salts through polyperforin plasma membrane pores eventually causes proteolysis and osmotic effects leading to cell death [1, 5-7]. The expression of perforin is known to be substantially up-regulated in CD8⁺ CTL upon activation, but to stay at a relatively stable, constitutively high level in natural killer cells [4].

The perforin expression among CD8⁺ cells in peripheral blood of kidney transplant recipients was found to be increased during rejection crises as compared to patients tolerating the transplant [8]. In dermatology until now, the aspect of CTL activation has been studied predominantly in lesional skin tissue, namely in the context of drug reactions [9] as well as drug- and virus-induced erythema multiforme and Stevens-Johnson syndrome (SJS) [10-12].

The objective of our present study was to study the perforin⁺ proportion of peripheral CD8⁺ lymphocytes as a marker for CTL activity in patients with generalized inflammatory skin diseases. We now report upon increased peripheral blood CTL activity in patients with a) drug-induced macular and papular rashes, b) drug-induced SJS and c) drug-unrelated exacerbated psoriasis. The augmentation of the perforin⁺ proportion of peripheral CD8⁺ lymphocytes proved to be a phenomenon with some degree of disease specificity in reflection of the inflammation activity, because it was missing in drug-independent acute urticaria and abolished due to successful treatment of SJS. The findings seem to be meaningful also with regard to the current controversal discussion arguing for a central role of CD8⁺ versus CD4⁺ lymphocytes in a hypothetical pathogenetic concept explaining psoriasis as a T cell-dependent autoimmune-like disease [13].

Methods

Patients

Perforin expression in CD8⁺ cells of the peripheral blood was studied in a total of n = 50 individuals admitted to our clinic. The cohort comprised subgroups of n = 14 patients with acute macular-papular generalized drug eruptions (ages ranging from 19 to 79 years), n = 10 patients with acute generalized urticaria (26 to 79 years, >= 25% body surface involvement), and n = 11 patients with exacerbated chronic-plaque or eruptive-exanthematic psoriasis (29 to 73 years, PASI scores between 25 and 35). In the latter two subgroups only such patients were included for whom a drug-induction of the skin manifestations could be ruled out by a thorough review of the patient's history. Moreover, those patients being affected by a physical, cholinergic or contact urticaria as well as an urticaria vasculitis were excluded. An additional subgroup consisted of n = 5 patients suffering from drug-induced SJS (27 to 75 years). Controls were n = 10 individuals not affected by any inflammatory skin or systemic disease (21 to 79 years).

Separation of mononuclear cells from peripheral blood

From each patient 20 ml of peripheral venous blood were drawn and simultaneously treated with 2 ml of heparin (Novo Nordisk Pharma, Mainz, Germany). The blood samples were diluted 1:1 (v/v) with a 3.5% polypeptide plasma expander solution (Haemaccel 35, Behring Werke AG, Marburg, Germany) and incubated for 60 minutes at 37° C. Subsequent separation of peripheral blood mononuclear cells (PB-MNC) was performed by a buoyant density centrifugation at 440 xg for 30 minutes on a Ficoll-Paque plus cushion (Ficoll, Uppsala, Sweden) at 4° C [14, 15]. The band of cells at the interface was collected and washed twice with Hank's buffered salt solution (HBSS).

Cell fixation and plasma membrane permeabilization

Preparation of PB-MNC for the subsequent antibody-labeling was performed by using a commercial kit (ICS permeabilization kit, cat. no. 9319, Hölzel Diagnostika, Köln, Germany) according to the manufacturer's instructions. In brief, aliquots of 1 x 10⁶ PB-MNC were fixed with 2% buffered formaldehyde for 10 minutes at room temperature. After two washes with HBSS the plasma membranes were permeabilized for 10 minutes with a buffer containing 0.1% saponin.

Double labeling with anti-perforin and anti-CD8 antibodies

Murine monoclonal antibodies used were directed against human perforin (clone delta G9, cat. no. 16476, Hölzel Diagnostika; final dilution 1:625) and human CD8 (clone UCHT4, Hölzel Diagnostika; final dilution 1:1,000), labeled with FITC and R-Phycoerythrin (R-PE) [16, 17]. As a control for unspecific antibody binding we employed a murine monoclonal IgG2b/FITC derived from the BPC4 plasmocytoma (clone BPC4, Hölzel Diagnostika, final dilution 1: 1,250). For each individual five parallel incubations of the PB-MNC aliquots were performed for 20 minutes at 4° C in the dark, as follows: a) control without antibody, b) single label anti-perforin/FITC control, c) single label anti-CD8/R-PE control, d) IgG2b/FITC control for unspecific binding, e) simultaneous double labeling with anti-perforin/FITC and anti-CD8/R-PE. Subsequently, the cell preparations were treated with 0.1% saponin followed by a final washing with HBSS.

Flow cytometry analysis and statistics

The expression of perforin and CD8 was analyzed on a FACScan^TM flow cytometer (Becton Dickinson, Heidelberg, Germany). Among the total of PB-MNC the lymphocyte subpopulation was identified by its forward/side scattering characteristics and gated. The relative amounts of lymphocytes with positive or negative expressions of perforin and CD8 were detected as explained in Fig. 1. Data are expressed as means ± standard deviation. Statistical analysis was performed by the Mann-Whitney U-test with an overall level of significance set to p ¾ 0.05 [18]. Multiple comparisons required Bonferroni's adjustment.

Results

The overall data of our study comparing the perforin expression in CD8⁺ peripheral blood lymphocytes between selected generalized inflammatory dermatoses are presented in Table I. Representative examples of the flow cytometry data underlying these results are given in Figure 2.

The perforin⁺ proportion in CD8⁺ lymphocytes was found to be significantly increased in generalized drug eruptions (68.8 ± 24.9%) and exacerbated psoriasis (67.2 ± 17.1%) as compared to the control group (43.5 ± 11.6%, p ¾ 0.05; Fig. 3). There was no significant difference for patients with urticaria (58.2 ± 23.1%). Correspondingly, the perforin-negative and CD8⁺ proportion in total lymphocytes was significantly decreased in drug eruptions (7.6 ± 7.1%) as well as in psoriasis (8.6 ± 4.5) in comparison to the control (15.8 ± 5.1), which did not differ from urticaria (10.2 ± 5.2; Table I).

Additionally, a group of n = 5 patients affected by a drug-induced SJS, i.e. the major type of erythema exsudativum multiforme including mucous membrane involvement, was studied. These patients experienced SJS when the following drugs were administrated: allopurinol, gentamicin, acetaminophen (syn. paracetamol), certoparin, penicillin (patient no.1); metamizol (patient no. 2); diclofenac (patient no. 3), acetaminophen, oily extract from various herbs (patient no. 4); lorazepam, lamotrigin, acetaminophen (patient 5). The perforin⁺ proportion in CD8⁺ lymphocytes was determined on the admission day and 7 days later after systemic high-dose steroid medication, and found to decline substantially during this period in each of these patients (Fig. 4). The average of the perforin⁺ proportion in CD8⁺ lymphocytes was 81.6% and 33.0% on the day of admission and day 7, respectively. The relatively wide variation of the parameter, ranging from 66% (patient no. 5) to 100% (patient no. 3) on the admission day (Fig. 4), may be explained by a) the various severities of the skin involvement and b) the fact that some of the patients had been pre-treated with systemic steroids, i.e. for a maximum duration of three days.

Discussion

Activation of CTL is known to occur in viral and bacterial diseases as well as in neoplasia, autoimmune diseases and in the context of allo-transplant rejection [1, 4, 19]. These CD8⁺ effector cells are responsible for the specific destruction of target cells carrying microbial, tumor- or auto-antigens. In the present study we have chosen the perforin⁺ proportion of CD8⁺ peripheral lymphocytes as an indicator for overall CTL activity in generalized inflammatory skin diseases. The detection of the expression of perforin as a cytoplasmic epitope by flow cytometry was rendered possible by using saponin to permeabilize the plasma membrane in order to allow the intracellular penetration of the FITC-labeled anti-perforin antibodies.

In the present study the perforin⁺ proportion of CD8⁺ peripheral lymphocytes was found to be significantly increased in patients with generalized cutaneous macular-papular drug eruptions (average of 69%, n = 14) as compared to controls not affected by any inflammatory skin disease (44%, n = 10). A maximum average value of the parameter, i.e. 82%, was determined for a group of patients affected by drug-induced SJS, i.e. immediately after admission to our clinic. These data are in good agreement with results from immuno-histological studies which demonstrated an intense perforin-positive dermal infiltrate in lesional skin biopsies from drug-induced SJS [11]. The generally pivotal pathogenic role of drug-reactive, CD8⁺ lymphocytes in drug-related bullous skin eruptions had been demonstrated in earlier studies [10]. Similar results of an increased lesional expression of perforin were reported for cases of virally induced erythema multiforme [12].

In the patients with drug-induced SJS syndrome we monitored the perforin⁺ proportion of CD8⁺ peripheral lymphocytes in the course of the disease. For all five patients the values of the parameter substantially decreased after seven days of systemic steroid administration, i.e. the average dropped from 82 to 33%. Given these data, the perforin⁺ proportion of CD8⁺ peripheral lymphocytes seems to reflect directly disease activity in SJS.

An abundance of lesional perforin⁺ lymphocytes in skin biopsies has also been described for the transfusion-associated graft-versus-host-disease (GVHD) and some cases of acute GVHD after bone marrow transplantation [20, 21]. Under such conditions of acute GVHD the occurrence of perforin⁺ and CD8⁺ lymphocytes was observed especially as an epidermotropic invasion phenomenon [22, 23]. In a patient afflicted with a Shulman syndrome-like scleroderma due to chronic GVHD following allogeneic hematopoietic stem cell transplantation because of acute myeloic leukemia, we observed an extremely high perforin⁺ proportion of peripheral CD8⁺ lymphocytes, i.e.~ 100% [C. Behrendt, H. Gollnick, B. Bonnekoh: unpublished data].

Noteworthy, the establishing of CD8⁺ T cell lines from skin lesions of Behcet's disease expressing perforin mRNA has been described [24]. Tumor cells of three patients with angiocentric lymphoma and granulomatous panniculitis were shown to express perforin as well [25].

That an increased perforin⁺ proportion of CD8⁺ peripheral lymphocytes is however not a general feature of generalized inflammatory skin diseases, is pointed out by our observation in patients with drug-unrelated acute urticaria. When studying the parameters in these patients we could not detect any significant difference as compared to the controls. Obviously, activation of perforin expression does not represent a decisive step in the pathomechanism of this type of acute urticaria.

However, in patients with drug-unrelated exacerbation of psoriasis we found a significantly increased perforin⁺ proportion of CD8⁺ peripheral lymphocytes (average of 67%) as compared to the controls (44%). At the present time the pathogenetic meaning of this finding is not really clear, and it provides a point of discussion, to what extent possibly autoantigen-directed actions of cytotoxic CD8⁺ lymphocytes might be involved in psoriasis. Currently, it is a major undecided debate, if the CD4⁺ or the CD8⁺ lymphocyte may represent the pivotal cell type in the pathogenetic cascade of immunological events in psoriasis. Our data are in line with recent findings demonstrating intraepidermal lymphocytes in psoriatic lesions to be CD8⁺ and to express GMP-17 (syn. TIA-1), known as a cytotoxic granule protein found in activated CTL [26].

Thus, we want to propose the perforin⁺ proportion of CD8⁺ lymphocytes as a useful, although not highly disease-specific routine parameter, to be conveniently measured by flow cytometry in peripheral blood preparations, for the evaluation of the CTL activity in drug-induced non-urticarial skin reactions, especially with regard to the monitoring under therapeutic interventions. Moreover, our data point to a possible involvement of activated peripheral CD8⁺ lymphocytes in the pathogenesis of the drug-unrelated exacerbation process of psoriasis. The pathogenetic meaning of an increased perforin expression is obvious in SJS by its probable contribution to the well known epidermal necrolysis phenomena, but will have to be elucidated in the future for macular-papular drug rashes and psoriasis where skin cell necroses are only rare events.

Article accepted on 26/4/00

CONCLUSION

Acknowledgements

The study was supported in part by a grant FKZ 2587A/0027H from the Kultusministerium des Landes Sachsen-Anhalt awarded to Prof. Dr. B. Bonnekoh.

REFERENCES

1. Berke G. The CTL's kiss of death. Cell 1995; 81: 9-12.

2. Liu CC, Persechini PM, Young JD. Perforin and lymphocyte-mediated cytolysis. Immunol Rev 1995; 146: 145-75.

3. Kraut RP, Bose D, Cragoe EJ, Greenberg AH. The influence of calcium, sodium, and the Na⁺/Ca²⁺ antiport on susceptibility to cytolysin/perforin-mediated cytolysis. J Immunol 1990; 144: 3498-503.

4. Liu CC, Walsh CM, Young JDE. Perforin: structure and function. Immunol Today 1995; 16: 194-201.

5. Krahenbuhl O, Tschopp J. Involvement of granule proteins in T-cell-mediated cytolysis. Nat Immun Cell Growth Regul 1990; 9: 274-82.

6. Berke G. Killing mechanisms of cytotoxic lymphocytes. Curr Opin Hematol 1997; 4: 32-40.

7. Shi L, Mai S, Israels S. Granzyme B autonomously crosses the cell membrane and perforin initiates apoptosis and Gra B nuclear localization. J Exp Med 1997; 185: 855-66.

8. Rukavina D, Balen-Marunic S, Rubesa G. Perforin expression in peripheral blood lymphocytes in rejecting and tolerant kidney transplant recipients. Transplantation 1996; 61: 285-91.

9. Hertl M, Jugert F, Merk HF. CD8⁺ dermal T cells from a sulphamethoxazole-induced bullous exanthema proliferate in response to drug-modified liver microsomes. Br J Dermatol 1995; 132: 215-20.

10. Hertl M, Bohlen H, Jugert F, Boecker C, Knaup R, Merk HF. Predominance of epidermal CD8⁺ T-lymphocytes in bullous cutaneous reactions caused by beta-lactam antibiotics. J Invest Dermatol 1993; 101: 794-9.

11. Inachi S, Hitoshi M, Masayuki S. Epidermal apoptotic cell death in erythema multiforme and Stevens-Johnson Syndrome. Arch Dermtol 1997; 133: 845-8.

12. Sayanna K, Watanabe M, Miki Y. Localization of perforin in viral vesicles and erythema multiforme. Dermatology 1994; 188: 305-9.

13. Griffiths CEM. Immunological pathways in psoriasis. In: Roenigk HHJr, Maibach HI (eds): Psoriasis, 3rd ed, Dekker, New York, 1998: 341-8.

14. Boyum A. Separation of white blood cells. Nature 1964; 204: 793-4.

15. Boyum A. Isolation of mononuclear cells and granulocytes from human blood. Scand J Clin Lab Invest 1968; 97: 77-89.

16. Hameed A, Olsen KJ, Cheng L, Fox WM, Hruban RH, Podack R. Immunohistochemical identification of cytotoxic lymphocytes using human perforin monoclonal antibody. Am J Pathol 1992; 140: 1025-30.

17. Kataoka K, Naomoto Y, Kojima K, Horiki S, Hizuta A, Tanaka N, Yahagi N, Miyaazaki Y, Okumura K, Orita K. Flow cytometric analysis on perforin induction in peripheral blood mononuclear cells with interleukin-2 or OK-432. J Immunother 1992; 11: 249-56.

18. Mann HB, Whitney DR. On a test of whether one of two random variables is stochastically larger than the other. Ann Math Stat 1947; 18: 50-60.

19. Smyth MJ, Ortaldo JR, Shinkai YI, Yagita H, Nakata M, Okumura K, Young HA. Interleukin 2 induction of pore-forming protein gene expression in human peripheral blood CD8⁺ T cells. J Exp Med 1990; 171: 1269-281.

20. Clement MV, Soulie A, Legros-Maida S, Guillet J, Gluckmann E, Sigaux N, Sasportes M. Perforin and granzyme B: predictive markers for acute GvHD or cardiac rejection after bone marrow or heart transplantation. Nouv Rev Fr Hematol 1991; 33: 465-70.

21. Takata M. Immunhistochemical identification of perforin-positive cytotoxic lymphocytes in graft-versus-host disease. Am J Clin Pathol 1995; 104: 705-6.

22. Sale GE, Anderson P, Browne M, Myerson D. Evidence of cytotoxic T-cell destruction of epidermal cells in human graft-vs-host disease. Immunhistology with monoclonal antibody TIA-1. Arch Pathol Lab Med 1992; 116: 622-5.

23. Baker MB, Altman NH, Podack ER, Levy RB. The role of cell-mediated cytotoxicity in acute GVHD after MHC-matched allogeneic bone marrow transplantation in mice. J Exp Med 1996; 183: 2645-56.

24. Mochizuki M. Characteristics of T cell lines established from skin lesions of Behcet's disease. J Dermatol Sci 1997; 15: 9-13.

25. Takeshita M, Kimura N, Suzumiya J, Ohshima K, Kikuchi M, Watanabe R, Okamura T, Goto H. Angiocentric lymphoma with granulomatous panniculitis in the skin expressing natural killer cell and large granular T-cell phenotypes. Vir Arch 1994; 425: 499-504.

26. Austin LM, Coven TR, Bhardwaj N, Steinman R, Krueger JG. Intraepidermal lymphocytes in psoriatic lesions are activated GMP-17(TIA-1) +CD8+CD3+ CTLs as determined by phenotypic analysis. J Cutan Pathol 1998; 25: 79-88.