Early reduction in number of T cells producing proinflammatory cytokines, observed after extracorporeal photopheresis, is not linked to apoptosis induction

IMMUNOBIOLOGY, PHYSIOLOGY, AND IMMUNE MONITORING

Early Reduction in Number of T Cells Producing Proinflammatory Cytokines, Observed After Extracorporeal Photopheresis, Is Not Linked to Apoptosis Induction J. Bladon and P.C. Taylor ABSTRACT Immediately following ECP, a significant number of lymphocytes become apoptotic and the number of T cells producing TNF␣ and IFN␥ is reduced. This study sought to determine if the cytokine down-regulation was a direct consequence of apoptosis induction. Methods. Samples were obtained from 6 graft versus host disease (GvHD) and 5 cutaneous T cell lymphoma (CTCL) patients immediately pre-ECP and from the leucocyte collection bag following 8-MOP/UVA exposure, but prior to re-infusion. Separated peripheral blood mononuclear cells (PBMC) were placed in cell culture and stimulated for 6 hours with phorbol myristate acetate (PMA), Ionomycin and Brefeldin A. Using flow cytometry, T cells were identified by CD3 expression and apoptotic T cells sub-selected by Annexin V staining. Both apoptotic and non-apoptotic T cells were evaluated for their intracellular expression of IL2, IL4, IL10, IFN␥ and TNF␣. Results. Neither patient group demonstrated a significant change in IL4 or IL10 expression post ECP. However the number of T cells expressing IL2, IFN␥ and TNF␣ was reduced in both the Annexin V-positive and -negative T cell populations (P ⬍ .05). The nonapoptotic T cells from GvHD patients demonstrated the greatest reduction in cytokine expression. Conclusions. Since proinflammatory cytokines play a major role in the pathology of GvHD, their down-regulation post-ECP may produce a direct clinical benefit. The lowest number of IL2-, IFN␥- and TNF␣-expressing T cells occurred within the apoptotic population; however, Annexin V-negative T cells also demonstrated a marked reduction post-ECP. However, the lack of an increase in IL4 and IL10 expression indicates that this process was not a consequence of skewing toward a Th2 cytokine profile.

0041-1345/03/$–see front matter doi:10.1016/S0041-1345(03)00477-9 1328

© 2003 by Elsevier Inc. All rights reserved. 360 Park Avenue South, New York, NY 10010-1710 Transplantation Proceedings, 35, 1328-1332 (2003)

ECP DOWNREGULATES CYTOKINES

E

XTRACORPOREAL PHOTOPHERESIS (ECP) involves the exposure of white cells to 8-methoxypsoralen (8-MOP) and UVA radiation, following initial isolation by leukopheresis. Treated cells are subsequently reinfused.1 ECP has been reported to be effective treatment for many T-cell–mediated conditions, including graft-versus-host disease (GvHD) and cutaneous T-cell lymphoma (CTCL).1,2 After ECP, lymphocytes immediately begin to express apoptotic markers.3 The majority of treated lymphocytes are apoptotic by 48 hours post-ECP.4 Recently, we observed a reduction in the number of T cells expressing the proinflammatory cytokines IFN-␥ and TNF-␣, when tested immediately after ECP.5 Following successive ECP treatments, the positive clinical outcome of “responders” is believed to involve modulation of the immune system. Antigen-presenting cells (APCs), activated by ECP, process apoptotic lymphocytes stimulating further identification and removal of nontreated cells of similar origin.6 In addition, when proinflammatory cytokines are actively involved in the pathology of the condition, such as GvHD, the immediate downregulation of these mediators by ECP may also be beneficial.5 This study was designed to determine if the reduction in the number T cells expressing proinflammatory cytokines, observed immediately post ECP, represented a direct consequence of apoptosis induction and whether ECP exposure induced an immediate skewing of T cells from a Th1 to a Th2 secretion pattern. PATIENTS AND METHODS Patients Eleven patients receiving ECP treatment at our institution were recruited for this study. Patients included six with chronic GvHD (cGvHD) and five with CTCL. Among the six cGvHD patients, the malignancies requiring transplantation included acute myeloid leukemia (n ⫽ 1), chronic myeloid leukemia (n ⫽ 2), acute lymphoblastic leukemia (n ⫽ 1), chronic lymphoblastic leukemia (n ⫽ 1), and non-hodgkin’s lymphoma (n ⫽ 1). The CTCL cohort of patients consisted of 3 stage III and 2 stage IVa subjects. Diagnosis for each condition was based on established histologic and immunologic criteria. Informed consent was obtained from all patients and local ethics approval was granted for the study.

Photopheresis Treatment Extracorporeal photopheresis cycles were performed using the XTS system (Therakos, UK). The treatment involves harvesting leukocytes (buffy coat) using a “collect and elutriation” six-cycle apheresis system, which includes a final “concentration” step. This system optimizes the hematocrit (Hct) and increases the white cell/volume ratio in the buffy-coat bag (BCB). The collected cells are exposed to 8-methoxypsoralen (8-MOP; Uvadex; Ben Venue Laboratories, Bedford, Ohio) and a 1.5-J/cm2 UVA radiation source, commencing at the end of the concentration step. The From the Department of Haematology, Rotherham General Hospital, South Yorkshire, UK. Address reprint requests to Dr John Bladon, Department of Haematology, Rotherham General Hospital, South Yorkshire S6O 2UD, UK. E-mail: [email protected]

1329 exposure time of (⬇15 to 60 minutes) depends on the volume and Hct of the buffy coat. After irradiation, the treated cells are reinfused. The process in repeated on the following day; thereafter, the patients return on a monthly basis.

Cell Preparation Venous blood was obtained immediately pre-ECP and from the leukocyte collection bag prior to reinfusion. Peripheral blood mononuclear cells (PBMCs) were separated using a density gradient (Nycomed, Norway), washed with phosphate-buffered saline (PBS), and added to RPMI medium (Bio-Whittaker, UK) containing 10% fetal calf serum, 140 ␮g/mL streptomycin, 50 ␮g/mL vancomycin, and 1% glutamine. The final suspension of 1.0 ⫻ 106 cells/mL was incubated in the dark at 37°C. The cells were immediately stimulated for 6 hours with 30 ng/mL phorbol 12myristate 13-acetate (PMA), 1 ␮g/mL ionomycin, and 10 ␮g/mL brefeldin A (Sigma, UK). After stimulation, the cells were washed with PBS prior to testing.

Cell Staining PBMCs washed with PBS were placed into 490 ␮L of workingstrength “binding buffer,” from an Annexin V identification kit (Immunotech, UK), at a cell count of 2.0 ⫻ 106/mL. Ten microliters of PE-Cy5-conjugated anti-CD3 (Dako, UK) and 1 ␮L of FITC-conjugated Annexin V were then added. The tube was mixed, placed on ice, and incubated for 10 minutes in the dark. After washing, using working strength binding buffer, the cells were treated with a “fix and perm” commercial kit (Harlan Sera-Lab, UK). At the permeabilization stage, 20 ␮L of either PE-conjugated anti–IL-2,–IL-4,– IL-10,– IFN-␥, or –TNF-␣ (R&D Systems, UK) was added. After a further 15-minute room temperature incubation in the dark, the cells were washed in PBS and processed. Appropriate isotype controls were performed.

Flow Cytometry Cells were processed through a Dako Galaxy flow cytometer. Alignment and fluorescence were standardized using alignment beads (Molecular Probes, UK) and flourospheres (Dako, UK). A minimum of 20,000 events were gathered for each test. T cells were identified by their anti-CD3 (PE-Cy5 (FL-III) expression and side scatter (SS), and bitmapped (R1) (Fig 1a). Subselection from R1 identified T cells positive versus those negative for Annexin V (FITC) expression (FL-I versus SS), designated R2 and R3 respectively (Fig 1b). By combining these gates, the percentage of apoptotic T cells (R1 ⫹ R2 ⫽ G1) and nonapoptotic cells (R1 ⫹ R3 ⫽ G2) expressing each cytokine were enumerated. For each bitmap (G1 and G2) isotype controls were used to set “region gates” for each cytokine histogram (PE-FL-II). These region gates were used to determine the percentage of each T-cell subset, expressing IL-2, IL-4, IL-10, IFN-␥, or TNF-␣ (Fig 2).

RESULTS

Neither the CTCL or the cGvHD patient group demonstrated significant changes in IL-4 or IL-10 expression within the apoptotic or nonapoptotic T-cell populations post-ECP. However, the number of T cells expressing IL-2, IFN-␥, and TNF-␣ was reduced in both T-cell populations (P ⬍ .05). The downregulation of IL-2, IFN-␥, and TNF-␣

1330

BLADON AND TAYLOR

Fig 1. T-cell identification. (a) T cells were identified and “bitmapped” (R1) using CD3 (PE-CY-5) expression and side scatter (SS). (b) The T cells contained in R1 were evaluated for Annexin V expression using the FL-I (FITC) channel. The bitmap R2 contained T cells positive for Annexin V, whereas R3 isolated Annexin V–negative T cells. Subsequently, the combined gates of R1 and R2 (G1) were used to assess cytokine expression in the apoptotic T-cell population, whereas R1 and R3 (G2) evaluated the cytokine expression of the nonapoptotic T-cell population.

within the apoptotic T-cell populations of both patient groups is demonstrated in Fig 3a. The changes in the Annexin V–negative T cells of the CTCL and cGvHD patients induced by ECP are illustrated in Fig 3b. Comparisons for each cytokine demonstrated the largest decrease in the apoptotic population. Cytokine expression post-ECP, as a percentage of pre-ECP, fell by between 42% and 78%. Downregulation of each cytokine within the Annexin

V–negative T cells was also profound, ranging from 29% to 57%. GvHD patients demonstrated the largest reduction in expression of all three cytokines within the nonapoptotic T-cell population and the largest decline in the number of apoptotic T cells expressing IL-2. Within the apoptotic T-cell population, the decreases in the IFN-␥– and the TNF-␣– expressing T cells were comparable for both patient groups.

Fig 2. Reduction in IL-2–secreting T cells post-ECP. After isolation of the Annexin V–negative T-cell population (see Fig 1), isotype controls were used to set “region gates” (RN1). (a) Pre-ECP, the majority of nonapoptotic T cells were positive for IL-2 expression. (b) After ECP, the number of IL-2–secreting, Annexin V–negative, T cells was significantly reduced.

ECP DOWNREGULATES CYTOKINES

1331

Fig 3. ECP downregulation of IL-2, IFN-␥, and TNF-␣ expression in all T cells. In both CTCL and GvHD patients, the number of apoptotic T cells (a) and nonapoptotic T cells (b) expressing IL-2, IFN-␥, and TNF-␣ post-ECP (filled bars) was significantly less than pre-ECP (open bars).

DISCUSSION

Immediately after ECP, apoptosis is induced in treated lymphocytes.3,4 At the same time, the number of T cells producing proinflammatory cytokines is reduced.5 Using multicolor flow cytometry, both apoptosis induction and cytokine secretion patterns were determined pre- and postECP exposure. Annexin V identified the apoptotic cells,

based on attachment to the externalized phosphatidylserine residues present on apoptotic T cells, a mechanism that occurs early in the apoptotic process,7 while a “fix and perm” technique, after suitable stimulation, allowed access to cytokine production within the intracellular compartment.8 Although the apoptotic T-cell population, postECP, demonstrated the smallest number of cells expressing

1332

IL-2, IFN-␥, and TNF-␣, the downregulation of cytokine production was unrelated to apoptosis induction. The nonapoptotic T-cell group also showed a marked reduction in T cells secreting IL-2, IFN-␥, and TNF-␣. In addition, the number of T cells showing dual positivity for Annexin V and either IL-2, IFN-␥, and TNF-␣ post-ECP was significantly lower than the comparable groups pre-ECP suggesting that cytokine reduction occurred within an already apoptotic population. Previously, skewing toward a Th2 cytokine profile has been reported after ECP.9 However, the present study failed to observe an increase in IL-4 – and IL-10 – expressing T cells, suggesting that these early changes were not attributable to skewing toward a Th2 cytokine secretion pattern. In GvHD, proinflammatory cytokines play a major pathologic role.10 Pre-ECP, GvHD patients demonstrate higher levels of IL-2–, TNF-␣–, and IFN-␥–secreting T cells than CTCL patients. The present results are consistent with previous findings, indicating a prominent role of these cytokines in the etiology of cGvHD.5 Therefore, the downregulation of IL-2, IFN-␥, and TNF-␣ by ECP may be beneficial to treat this condition.5 Annexin V–positive lymphocytes are processed swiftly and removed after reinfusion.3 If reduction in cytokine expression is limited to the apoptotic population, the beneficial effect may be limited. However, because downregulation proinflammatory cytokine secretion was independent of apoptosis status, the effect may be more pronounced and longer lasting. Previously, we postulated a possible link between cytokine downregulation and apoptosis;5 however, the present simul-

BLADON AND TAYLOR

taneous evaluation of both apoptosis and intracellular cytokine production clearly demonstrates no link. Although no skewing toward a Th2 cytokine profile was detected, no reduction in IL-4 – or IL-10 –secreting T cells was observed. These results indicate a selective process for cytokine downregulation, targeting reactive cells that preferentially secrete proinflammatory cytokines. The downregulation of proinflammatory cytokines after ECP may supplement the immunomodulatory effect induced by apoptosis induction and APC activation. However, these processes are distinct and may thus help to explain why a treatment regimen, associated with modulation of the immune system, is beneficial in conditions associated with a marked elevation in inflammatory mediators.

REFERENCES 1. Edelson R, Berger C, Gasparro F, et al: N Engl J Med 316:297, 1987 2. Owsianowski M, Gollnick H, Siegert W, et al: Bone Marrow Transplant 14:845, 1994 3. Bladon J, Taylor PC: Br J Haematol 107:707, 1999 4. Bladon J, Taylor PC: Br J Dermatol 146:59, 2002 5. Bladon J, Taylor PC: J Clin Apher 17:177, 2002 6. Berger CL, Xu A-L, Hanlon D: Int J Cancer 91:438, 2001 7. Verhoven B, Krahling S, Schleng A: Cell Death Differ 6:262, 1999 8. Prussin C: J Clin Immunol 17:195, 1997 9. Klosner G, Trautinger F, Knobler R, et al: J Invest Dermatol 116:459, 2001 10. Parkman RJ: Curr Op Haematol 5:22, 1998