British \ournu/ of Humutology, 1992. 8 1 , 73-76

Ultraviolet irradiation modulates MHC-alloreactive cytotoxic T-cell precursors involved in the onset of graft-versus-host disease H.c. V A N PH001JEN. M. I . A A R T S - K I E M E NM.S ,A . GRIJZENHOUT*A N D H . V A N WEELDENT Departments o$ lmmunohaeniatology and tDermatology. University Hospital Utrecht, and *National Institute of Public Henlth and Environmental Protection, Bilthoven. The Netherlands Received 27 September 1991: accepted for publication 19 December 1991

Summary. Ultraviolet B (UVB) irradiation of cellular blood components has been proposed as a new technology to prevent HLA sensitization in recipients. Earlier studies have shown that a dose of 2 J/cm2 abrogates the ability of lymphocytes to serve as stimulators in mixed lymphocyte cultures (MLC). In this study we have evaluated the effect of UV energy on T-lymphocytes for the prevention of transfusion-associated graft-versus-host disease (TA-GvHD). The

response of cytotoxic T-lymphocyte precursors against host alloantigens was almost undetectable at a dose of 0.5 J/cmz. T-cell proliferation in MLC or in response to phytohaemagglutinin was inhibited by more than 95% at doses of 1 J/cm2 or higher. The data suggest that UV irradiation can be used to prevent both HLA sensitization and TA-GvHD in recipients.

Transfusion of cellular blood components, contaminated with leucocytes. can give rise to GvHD when donor Tlymphocytes, mismatched with the host for major histocompatibility complex (MHC) antigens become sensitized and attack host tissues. TA-GvHD is primarily seen in patients with a severely compromised cellular immunity following immunosuppressive and myelosuppressive therapy (Anderson et a/. 1991). In these patients donor T-lymphocytes proliferate in response to host cells without being eliminated. TA-GvHD can be prevented by abrogating the proliferative capacity of the T-cells. Gamma irradiation of cellular blood components is the only proven effective technology to prevent TA-GvHD (Thomas et al. 1975). Recently. Deeg et al (1989) reported that the transfusion of UV irradiated blood products prevented the onset of GvHD in a n autologous bone marrow transplant (BMT) model, whereas the transfusion of unirradiated blood products resulted in acute GvHD in 100% of recipients. In this study we have evaluated the effect of UV energy on human cytotoxic T-lymphocytes, which are the most potent cells involved in the development of GvHD. T-cell responses against alloreactive 'host' cells were determined in limiting

dilution assays. The importance of these observations in view of their possible implications in transfusion practice will be discussed. MATERIALS AND METHODS Cell suspensions. Ficoll centrifugation of heparinized blood obtained from healthy donors was used to collect peripheral blood mononuclear cells (PBMCs)for the lymphocyte proliferation and for the limiting dilution assays. All donors selected for the limiting dilution assay were previously typed for HLAA. -B, -C in the standard National Institute of Health lymphocytotoxicity assay and for HLA-DR in the two-colour fluorescence assay using highly selected alloantisera (van Rood et a!, 19 76). U V B and gamma irradiation. PBMCs. isolated after ficoll centrifugation, were washed twice and suspended in plasma. A volume of 2 ml was pipetted in each well of a six-well plate (Costar, Broadway, Cambridge. Mass., U.S.A.) to obtain a layer thickness of 2 mm. The plate was covered with UVtransmittant PVC and placed on a platform rotator set at 60 cycles per minute. The source of UV energy was a bank of eight UVB fluorescent tubes (Philips, TL 2OW/O1) emitting a strong narrow peak around 312 nm (half band width 2 . 5 nm) and a smaller peak around 305 n m (van Prooijen et al. 1990). The distance between the source and the plates was 15 cm and irradiance was 5.4 mW/cm2. Irradiance measure-

Correspondence: Dr H. C. van Prooijen. Department of Immunohaematology, IJniversity Hospital Utrecht. P.O. Box 8 5 5 0 0 . 3 5 0 0 GA Utrecht. The Netherlands.

73

74

H. C. vun Prooijen et ul

ments were performed routinely with a Waldmann UVB meter (Waldmann AG, Schwenningen. Germany), calibrated with a Kipp thermopile type E 11 (Kipp, Delft, The Netherlands). Dose variations were obtained by changing the exposure time while irradiance remained constant. After UV exposure the cells were sedimented and resuspended in RPMI culture medium (Gibco, U.K.) supplemented with 20% heatinactivated pooled human AB serum, 3 p~ glutamine (Gibco. U.K.) and 0.5 pg/ml gentamycin. Gamma irradiations were performed with a Cesium-137 source. Lymphocyte proliferution. PBMCs. irradiated with UVB at doses between 0.5 and 2 J/cm' or gamma at doses between 10 and 40 Gy and subsequently suspended in culture medium, were cultured in quadruplicate (Costar 96-well plates; 150 pl/well; 40 000 cells/well)in the presence of 0 . 1 5 mg/ml phytohaemagglutinin (PHA HA- 1 5 Wellcome Diagnostics, England). After 3 d of incubation (5%COZat 37°C) 1 p~ tritiated thymidine (s.a. 2.1 GBq/mmol) was added to determine proliferation of the lymphocytes after 6-8 h. Lymphocyte proliferation was also evaluated in MLC. Responder cells were irradiated with UVB at doses between 0.1 and 1 J/cmz or with gamma at doses between 1 and 10 Gy. Gamma irradiated ( 2 0 Gy) stimulator cells (40000 cells/ well) were cultured with irradiated responder cells (40 000 cells/well) in quadruplicate in a total volume of 150 pl. After 4 d of incubation 1 p~ tritiated thymidine (s.a. 2.1 GBq/ml) was added to determine lymphoproliferation after 6-8 h. Cytotoxic T-lymphocyte precursor frequencies. Cytotoxic Tlymphocyte precursor (CTLp)frequencieswere determined in limiting dilution assays. The assays were performed in the GvH direction. The responder cells were considered as donor cells and therefore UV irradiated. Limiting numbers of responder PBMCs (20 x lo3to 0 . 3 12 x 10') were cultured in microtitre plates with constant numbers (5 x lo4)of gamma irradiated (20 Gy) stimulator PBMCs suspended in culture medium supplemented with 2 0 U/ml recombinant human interleukin 2. After 7 d of culture the cells were assayed for cytotoxicity against 7 d SICr-labelledPHA blasts from the original stimulator. Supernatants were harvested from the wells and analysed for released radioactivity using a gamma counter. Wells were scored as positive if their radioactive Cr release was greater than the mean plus 3 times the standard deviation (SD) of the control (medium plus irradiated stimulators). The frequency of responding cells and 95% confidence interval were determined as described (Strijbosch et al, 1987). The jack-knife statistical method was taken for each experiment to generate a straight line between responder cell dose and the number of non-responding wells. RESULTS Lymphocyte proliferation Fig 1 shows the effect of UV and gamma irradiation on lymphocyte proliferation when stimulated with PHA or in MLC. PHA experiments were performed with cells obtained from four different donors. A 90% reduction in radioactive thymidine incorporation was seen when responders were gamma irradiation at doses up to 40 Gy. UV irradiation at doses above 1 J/cm2resulted in a 95%or higher reduction of lymphocyte proliferation.

'H thymidine 20 000- incorporation (cprn)

10 000-

2000-

1000-

-l-I

0

10

1-1-1

20

30

40Gy

Fig 1. Etrect of UV and gamma irradiation on lymphocyte incorporation of radioactive thymidine on stimulation with phytohaemagglutinin (PHA: n = 4) and allogeneic cells (MLC: n = 4).

MLC were performed in four separate responder/stimulator lymphocyte combinations and the lymphocyte suspensions were prepared from eight donors. Lymphocyte proliferation was completely abolished after as little as 10 Gy or 1 J/cmL. CTLp frequencies Table I shows that responder and stimulator cells in our studies were completely mismatched for HLA antigens, which explains the high frequencies of MHC alloreactive CTLp in control experiments. Exposure of responders to increasing doses of UVB resulted in a progressive decrease of CTLp frequencies. A dose of 0.5 J/cm2was sufficient to obtain a 30-60-fold decrease of CTLp frequencies (ratio= l/fof UV exposed cells: l/f of control cells). DISCUSSION Transfusion of cellular blood components may lead to TA-GvHD in recipients. The disorder is induced by the T-lymphocytes and is primarily seen in patients with a compromised cellular immunity including BMT recipients, premature newborns and lymphoma patients (Anderson et al, 1991). Recently, TA-GvHD has also been recognized in non-immunocompromised transfusion recipients receiving blood from first-degree family members (Thaler et al, 1989; Arsura et al. 1988; Juji et al, 1989; Otsuka et al, 1989; Sheehan et al, 1987). Reports from Goulmy et al(1983), Tsoi et a l ( l 9 8 3 )and Irle et a l ( l 9 8 5 ) have provided evidence that cytotoxic T-lymphocytes are involved in the aetiology of GvHD. The authors detected anti-host reactive cytotoxic lymphocytes in patients with acute GvHD following allogeneic BMT. Alloreactive cytotoxic lymphocytes were also

UVB Irradiation and Graft-versus-Host Disease

75

Table 1. HLA phenotype of responder/stimulatorpairs and frequencies (I)of responder CTLp before and after W exposure.

Kxpt

I

HLA phenotype

Before

Responder

Stimulator

l/f

9 5% confidence interval

I/f

9 5% confidence interval

1/f

9 5% confidence interval

A1.2

A9.~33

2506

1876-3788

8217

6918-9976

62500

55780-76390

B8.12

B17.40 DR2,w6 1300

1042-1805

5 13 3

42 17-9044

4381 7

35688-54322

754

593-1032

9091

7407-11905

58824

38462-125000

DR3.4 I1

111

A1 R8.w70 DR 3

A3.32 B16.35 DR1.5

A11.32 H17.18

B16.35

DR 3

DR1.5

A3.32

0.25 J/cm2

cloned from skin biopsies of BMT patients with GvHD (Reinsmoen et al, 1984). Gamma irradiation of cellular blood components is now the only proven effective method to prevent TA-GvHD. This paper provides evidence that the use of UV energy is also very effective. Lindall-Kiessling & Safwenberg ( 19 7 1) reported that IJV irradiated white cells were not able to proliferate but also unable to stimulate allogeneic responder cells in MLC. The latter observation stimulated several investigators to develop UV irradiators for treatment of platelet concentrates (Kahn ef al. 1985: Pamphilon et al. 1989: Andreu et al. 1990). For a variety of reasons we advocated the use of a more specific narrow band UVB source and reported that a dose of 2 J/cmz was sufficient to abrogate lymphocyte function almost completely (van Prooijen at al. 1990). We report here that under similar conditions T-cell proliferation in MLC or in response to PHA was inhibited by more than 95% at doses of 1 J/cm2 or higher. T-cell responses against MHC-alloantigens were found to be more sensitive for UV energy: the frequency of UV irradiated cytotoxic T-cell precursors against host alloantigens was almost undetectable at a dose of 0 . 5 J/cm2.The effect of UV irradiation on the prevention of GvHD in vivo has been evaluated in animal studies. Hardy et a1 (1986) showed in a murine BMT model that the addition of UV irradiated donor splenocytes to T-cell depleted histo-incompatible bone marrow did not result in GvHD, whereas lethal GvHD resulted in all control animals. Deeg et a/ (1989) found no GvHD after transfusion of UV irradiated blood products in an autologous BMT canine model, whereas acute GvHD developed in all control animals. In the clinical setting Kaminski et a1 (1989) found a close correlation between a decreased T-cell response and a low grade of GvHD in BMT recipients after transplantation of HLA-identical marrow from an unrelated donor. The application of UV irradiation for the inactivation of leucocytes in cellular blood components is rather attractive. At this moment, cellular blood components for bone marrow transplant recipients are first filtered and subsequently gamma irradiated for the prevention ofHLA sensitization and

0.50 J/cm2

TA-GvHD respectively. Filtration alone is not effective for the prevention of TA-GvHD and gamma irradiation alone is not effective for the prevention of HLA sensitization. UV irradiation seems to be the only procedure effective for the prevention of both HLA sensitization and TA-GvHD but the results of ongoing clinical studies remain to be seen. ACKNOWLEDGMENTS We are grateful for the cooperation of Ms M. de Haan from our Department of Immunohaematology in the selection of unrelated donors and thank Ms S. van Bree from the Department of Immunohaematology and Bloodbank Leiden for all laboratory and technical assistance. REFERENCES Anderson, K.C.. Goodnough. L.T., Sayers, M.. Pisciotto, P.T.. Kurtz. S.R.. Lane, T.A., Anderson. C.S. & Silberstein. L.E. (1991) Variation of blood component irradiation practice:implicationsfor prevention of transfusion-associated graft-versus-host disease. Blood. 77, 2096-2102. Andreu, G.. Boccaccio. C.. Lecrubier. C.. Fretault. J., Coursaget. J., Leguen. J.P.. Aleggini, M.. Fournel, J.J. & Samama, M. (1990) Ultraviolet irradiation of platelet concentrates: feasibility in transfusion practice. Transfusion, 30, 401-406. Arsura, E.L.. Bertelle, A.. Minkowitz,S., Cunningham.J.N. &Grob.D. (1988) Transfusion-associatedgraft-vs-hostdisease in a presumed immunocompetent patient. Archives of Internal Medicine, 148, 1941-1 944. Deeg. H.J., Graham, T.C.. Gerhard-Miller. L.. Appelbaum. F.R., Schuening. F. & Storb. R. (1989) Prevention of transfusioninduced graft-versus-host disease in dogs by ultraviolet irradiation. Blood, 74, 2592-2595. Goulmy, E.. Gratama, J.W.. Blokland. E.. Zwaan. F.E. & Rood, J.J. van (1983) A minor transplantation antigen detected by MHC restricted cytotoxic T-lymphocytes during graft-versus-host disease. Nature. 302, 159-1 61. Hardy, M.A.. Chabot. J.. Tannenbaum. G. & Lau. H.T. (1986) Immunomodulation by ultraviolet irradiation. Transplantation:

76

H. C. van Prooijen et a1

Approaches to Graft Rejection (ed. by H. T. Meryman), p. 119. Liss, New York. Irle. C.. Beatty. P.G.. Mickelson. E.. Thomas, E.D. & Hansen. J.A. ( 1985) Alloreactive T-cell responses between HLA identical siblings. Transplantation. 40, 329-334. Juji. T.. Takahashi. K., Shibata, Y.. Ide. H.. Sakakibara, T.. Ino, T. & Mori. S. (1989) Post-transfusion graft versus host disease in immunocompetent patients after cardiac surgery in Japan. (Letter). New England journal ofMedicine, 321, 56. Kahn. R.A., Duffy, B.F. & Rodey, G.C. (1 985) Ultraviolet irradiation of platelet concentrate abrogates lymphocyte activation without affecting platelet function in vitro. Transfusion, 25, 547-550. Kaminsky. E.. Hows. J., Man, S., Brookes. P.. Mackinnon, S., Hughes, T.. Avakian. 0..Goldman, J.M. & Batchelor. J.R. (1989)Prediction of graft versus host disease by frequency analysis of cytotoxic T cells after unrelated donor bone marrow transplantation. Transplantation, 48, 608-6 1 3 . Lindahl-Kiessling, K. & Safwenberg, J. (1971) Inability of UVirradiated lymphocytes to stimulate allogeneic cells in mixed lymphocyte culture. lnternational Archives of Allergy and Applied Immunology. 41. 670-678. Otsuka. S.. Kunieda, K., Hirose. M.. Takeuchi. H.. Mizutani. Y.. Nagaya, M.. Sato. G.. Kasuya. S.. Matsutomo. K.. Noma. A. & Saji. S. ( 1 989) Fatal erythroderma (suspected graft versus host disease) after cholecystectomy: retrospective analysis. Transfusion. 29, 544-548. Pamphilon, D.H.. Corbin. S.A.. Saunders, J. & Tandy. N.P. (1989) Application of ultraviolet light in the preparation of platelet concentrates. Transfusion. 29, 3 79- 38 3. Prooijen. H.C. van. Marwijk Kooy, M. van, Weelden. H. van, Aarts

Riemens, M.I.. Borghuis, L. & Akkennan, J.W.N. (1990) Evaluation of a new UVB source for irradiation of platelet concentrates. British journal of Haematology, 75, 573-577. Reinsmoen, N.L., Kersey. J.H. & Bach. F.H. (1984) Detection of HLA restricted antiminor histocompatibility antigen reactive cells from skin GVHD lesions. Human Immunology. 11, 249-257. Rood, J.J. van, Leeuwen. A. van & Ploem. J.S. (1976) Simultaneous detection of two cell populations by two-colour fluorescence and application to the recognition of B-cell determinants. Nature, 262, 795-797. Sheehan. T., McLaren. K.M., Brettle. R. & Parker, A.C. (1987) Transfusion-induced graft versus host disease in pregnancy. Clinical and Laboratory Haematology, 9, 205-207. Strijbosch. L.W.G.. Buurman. W.A.B., Does, R.J.M.M., Zinken. P.H.& Groenewegen. G. ( 198 7) Limiting dilution assays experimental design and statistical analysis. journal of lmmunological Methods, 97, 133-136. Thaler, M.. Shamiss, A.. Orgad. S., Huszar. M..Nussinovitch. N.. Meisel. S., Gazit, E.. Lavee. J. & Smolinsky. A. (1989) The role of blood from HLA-homozygous donors in fatal transfusion-associated graft versus host disease after open-heart surgery. New England journal of Medicine. 321, 25-28. Thomas, E.D.. Storb. R.. Clift. R.A., Fefer, A., Johnson, F.L.. Neiman. P.E.. Lerner. K.G.. Glucksberg, H. & Buckner. C.D. (1975) Bone marrow transplantation. New England journal of Medicine. 292, 832-844. Tsoi, M.S.. Storb. R.. Santos. E. & Thomas, E.D. (1983) Anti-host cytotoxic cells in patients with acute graft versus host disease after HLA identical marrow grafting. Transplantation Proceedings, 15, 1484-1487.