Human Alloimmune Responses Proliferating T Lymphocyte

CD30 Expression Identifies the Predominant Proliferating T Lymphocyte Population in Human Alloimmune Responses

This information is current as Keith W. Chan, Corwyn D. Hopke, Sheri M. Krams and of October 5, 2021. Olivia M. Martinez J Immunol 2002; 169:1784-1791; ; doi: 10.4049/jimmunol.169.4.1784 http://www.jimmunol.org/content/169/4/1784 Downloaded from

References This article cites 39 articles, 23 of which you can access for free at: http://www.jimmunol.org/content/169/4/1784.full#ref-list-1 http://www.jimmunol.org/ Why The JI? Submit online.

• Rapid Reviews! 30 days* from submission to initial decision

• No Triage! Every submission reviewed by practicing scientists

• Fast Publication! 4 weeks from acceptance to publication

*average by guest on October 5, 2021

Subscription Information about subscribing to The Journal of Immunology is online at: http://jimmunol.org/subscription Permissions Submit copyright permission requests at: http://www.aai.org/About/Publications/JI/copyright.html Email Alerts Receive free email-alerts when new articles cite this article. Sign up at: http://jimmunol.org/alerts

The Journal of Immunology is published twice each month by The American Association of Immunologists, Inc., 1451 Rockville Pike, Suite 650, Rockville, MD 20852 Copyright © 2002 by The American Association of Immunologists All rights reserved. Print ISSN: 0022-1767 Online ISSN: 1550-6606. The Journal of Immunology

CD30 Expression Identiﬁes the Predominant Proliferating T Lymphocyte Population in Human Alloimmune Responses1

Keith W. Chan,* Corwyn D. Hopke,† Sheri M. Krams,*† and Olivia M. Martinez2*†

CD30 is an inducible member of the TNFR superfamily that is expressed on activated T and B cells and some lymphoid malignancies. We have previously shown that human CD30؉ T cells elicited with allogeneic APC are a major source of IFN-␥ and IL-5 production. In the present study we have used alloantigen, as well as anti-CD3 plus anti-CD28 mAb stimulation, to further -characterize human CD30؉ T cells with respect to function and the expression of other activation-dependent cell surface mole cules, including the related TNFR family members OX-40 and 4-1BB (CD137). Our results indicate that human CD30؉ T cells are a subset of activated T cells that also express CD25 and CD45RO. Moreover, we observed that allogeneic APC consistently induced a greater proportion of CD30؉ cells within the activated T cell population than did stimulation with plate-bound anti-CD3 plus anti-CD28 mAb or stimulation with soluble anti-CD3 plus anti-CD28 and autologous APC. The enhanced induction of CD30 expression by alloantigen was not common to other inducible TNFR family members because anti-CD3 plus anti-CD28 mAbs were Downloaded from far more effective in inducing expression of 4-1BB and OX-40. Furthermore, CD30 expression marked the predominant proliferating T cell population induced by alloantigen as determined by CFSE staining and ﬂow cytometry. These results indicate that CD30, but not 4-1BB or OX-40, is preferentially induced by alloantigen, suggesting that CD30 may be important in human alloimmune responses. The Journal of Immunology, 2002, 169: 1784Ð1791.

D30 is a 120-kDa type 1 transmembrane glycoprotein to induce apoptosis during negative selection in the thymus (12, http://www.jimmunol.org/ and a member of the TNFR superfamily (1). CD30 was 13), although some studies fail to support a role for CD30 in this C originally identified as a molecule expressed on Hodgkin process (14). and Reed Sternberg cells in Hodgkin’s disease but has subse- Other inducible members of the TNFR family expressed on T quently been detected on certain non-Hodgkin’s malignancies as cells, including 4-1BB and OX-40, have also been shown to have well as some virally transformed cells (2). CD30 is also expressed costimulatory function. The interaction between 4-1BB (CD137) on a subset of normal human T lymphocytes following activation and 4-1BB ligand (4-1BBL), expressed on activated B cells, mac- with Ag or mitogen (3, 4). Expression of CD30 has been proposed rophages, and dendritic cells, is particularly important in CD8- to preferentially mark Th2 cells (5), although CD30 has also been mediated responses to alloantigen (15Ð17), viruses (15, 18), and by guest on October 5, 2021 detected on CD4ϩ Th0 and Th1 cells, indicating that CD30 is not tumors (19), although recent studies also suggest a role for 4-1BB/ restricted to Th2 cells (6, 7). We (8) and others (9) have demon- 4-1BBL interactions in CD4ϩ T cell function (20). OX-40, in con- strated that primary stimulation of human T cells with alloantigen trast, primarily promotes cytokine production by CD4ϩ T cells or anti-CD3 mAbs and IL-2 elicits a population of CD30ϩ T cells (21Ð23) and has been implicated in T cell function in animal mod- that is a major source of IFN-␥ and IL-5 production. els of allogeneic graft-vs-host disease (GVHD) (24), experimental Several lines of evidence indicate that CD30 acts as a costimu- allergic encephalomyelitis (25), leishmaniasis (26), inflammatory latory molecule in T cell responses. The human CD30 ligand bowel disease (27), and tumor immunity (28). 4-1BB and OX-40, (CD30L3; CD153) is expressed in the periphery on activated T like CD30, tend to be expressed relatively late following activa- cells, monocytes, B cells, neutrophils, and eosinophils. Engage- tion, suggesting these molecules may be important in sustaining T ment of CD30 by CD30Lϩ cells or agonist anti-CD30 mAbs in- cell responses after the initial CD28/B7-1 and B7-2 interactions. duces IFN-␥ production (9) and proliferation (10) in anti-CD3- Despite the observed similarities in expression and function of stimulated peripheral blood T cells, NF-␬B activation, and viral CD30, OX-40, and 4-1BB in murine systems, relatively little is expression in an HIV-infected T cell line (11), and Ca2ϩ flux in known about the expression patterns of these inducible TNFR fam- Jurkat cells (3). Signaling through CD30 has also been suggested ily members on human T lymphocytes. The purpose of the current study was to examine the relationship of CD30 expression to the expression of other inducible TNFR family members on human T *Department of Surgery and †Program in Immunology, Stanford University School of cells and to further characterize the phenotype and function of Medicine, Stanford, CA 94305 CD30ϩ T cells in alloimmune responses. Received for publication October 11, 2001. Accepted for publication June 13, 2002. The costs of publication of this article were defrayed in part by the payment of page Materials and Methods charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Abs and reagents 1 This work was supported by National Institutes of Health Grant RO1 DK47810 (to The following Abs were purchased from DAKO (Carpinteria, CA): anti- O.M.M.). K.W.C. was supported by a Howard Hughes Medical Institute Fellowship. CD3-FITC, anti-CD3-PECy5, anti-CD4-PE, anti-CD4-PECy5, anti-CD8- 2 Address correspondence and reprint requests to Dr. Olivia M. Martinez, Stanford PECy5, anti-CD30-FITC, anti-CD25-PE, IgG1-FITC, IgG1-PE, and PE- University School of Medicine, 1201 Welch Road, Medical School Lab Surge P312, conjugated rabbit anti-mouse Ig. PE-conjugated mAbs specific for CD25, Stanford, CA 94305-5492. E-mail address: [email protected] OX-40, 4-1BB, CD27, and CD45RO, as well as IgG1-PECy5, were ob- 3 Abbreviations used in this paper: CD30L, CD30 ligand; 4-1BBL, 4-1BB ligand; tained from BD PharMingen (San Diego, CA). Anti-CD28 and anti-CD30 GVHD, graft-vs-host disease; LCL, lymphoblastoid B cell line; NHS, normal human mAbs were purchased from BD PharMingen and anti-CD3 mAb was pur- serum; P/S, penicillin-streptomycin. chased from Coulter (Miami, FL). Human rIL-2 was obtained from the

Biological Resources Branch of the National Cancer Institute (Frederick, anti-CD4-PECy5, or anti-CD8-PECy5 in addition to FITC- and PE-conju- MD). Propidium iodide was purchased from Sigma-Aldrich (St. Louis, gated mAbs, the cells were further gated on the CD3-, CD4- or CD8- MO) and CFSE was obtained from Molecular Probes (Eugene, OR). positive populations to analyze T cell subsets.

Cell preparation CFSE labeling and analysis of cell division by flow cytometry Lymphoblastoid B cell lines (LCL) were maintained in RPMI 1640 with Isolated PBMC were washed, counted, and resuspended at 5 ϫ 107/ml in 10% FCS and 1% penicillin-streptomycin (P/S) in a 5% CO2 atmosphere RPMI 1640 with 1% P/S. A stock solution of CFSE (5 mM) was diluted at 37¡C. Blood was collected from healthy donors by venipuncture in hep- 1/10 with RPMI 1640 and 15 ␮l of the CFSE was added per milliliter of arinized tubes. PBMC were isolated from blood samples by Ficoll- cells. The cell suspension was incubated for 10 min at 37¡C. The cells were Hypaque (Amersham Pharmacia Biotech, Piscataway, NJ) density centrif- then washed and cultured in RPMI 1640 supplemented with 10% NHS and ugation. PBMC were washed, counted, and cultured in RPMI 1640 with 1% P/S on plates precoated with anti-CD3 plus anti-CD28 mAbs or with 10% normal human serum (NHS; Sigma-Aldrich) and 1% P/S at 37¡Cin allogeneic APC in MLR as described above. Cells were collected on days a humidified incubator with 5% CO2. T cells were isolated by negative 3Ð6, washed with PBS/FCS, and stained with unconjugated mouse anti- selection using the MiniMACS (Miltenyi Biotec, Auburn, CA) separation human CD30 mAbs for 20 min at 4¡C. Cells were washed twice and la- protocol. To obtain purified T cells, PBMC were washed and resuspended ϫ 8 beled with PE-conjugated rabbit anti-mouse Ig for 20 min at 4¡C. Cells at a concentration of 1.25 10 /ml. Twenty microliters of Hapten-Ab were washed again and labeled with PECy5-conjugated anti-CD3 mAb for Cocktail (Miltenyi Biotec), containing hapten-conjugated anti-CD11b, anti- 7 20 min at 4¡C. After a final wash, the cells were acquired and analyzed on CD16, anti-CD19, anti-CD36, and anti-CD56 mAbs, was added per 10 a FACScan. Twenty thousand cells were collected and gating was set by cells and the cells were incubated for 10 min at 4¡C. The cells were then forward and side scatter. washed twice and again resuspended at a concentration of 1.25 ϫ 108/ml in PBS containing 1% FCS and 2 mM EDTA (PBS/FCS/EDTA). Twenty microliters of MACS AntiHapten Microbeads (Miltenyi Biotec) was added Measurement of cytokine production 7 per 10 cells and the cell suspension was incubated for 15 min at 4¡C. The Quantification of IFN-␥, IL-2, IL-5, and TNF-␣ was performed as previ- Downloaded from cells were washed and resuspended in 500 ␮l of PBS/FCS/EDTA, loaded ϩ ϩ ously described (29). Briefly, on day 1, a 96-well microtiter plate (Costar) onto a MACS Column Type LS /VS (Miltenyi Biotec), and allowed to was coated with 50 ␮l capture Ab (2 ␮g/ml; BD PharMingen) specific for filter through the column. The column was flushed four times with 3 ml of the cytokine of interest and incubated overnight at 4¡C. On day 2, the plate PBS/FCS/EDTA and the unbound cells were eluted. This negative selec- was washed twice with PBS/0.05% Tween and blocked with PBS contain- Ͼ tion enrichment routinely resulted in a cell population consisting of 98% ing3%BSAfor2hatroom temperature. The wells were washed twice T cells as determined by immunofluorescent staining and flow cytometry. with PBS/Tween and sample supernatants were loaded onto the plate in

Mixed lymphocyte reaction duplicate. A standard curve (0Ð1000 pg/ml) was generated by diluting http://www.jimmunol.org/ recombinant cytokine (BD PharMingen) with PBS containing 1% BSA. Stimulator-responder cell pairs in MLR were selected for HLA disparity The plate was covered and incubated at 4¡C overnight. On day 3, the plate based on previously performed HLA typing. Mismatches were confirmed was washed four times with PBS/Tween, biotinylated anti-human cytokine in preliminary experiments using one-way MLR and [3H]thymidine incor- Ab (1 ␮g/ml) was added, and the plate was incubated for 45 min at room poration on day 6 as a measure of cell proliferation. MLR were done as temperature. The plate was then washed six times with PBS/Tween and previously described (8). Briefly, PBMC were suspended at 1 ϫ 106 was incubated with avidin-HRP (5 ␮g/ml; Sigma-Aldrich) for 30 min at cells/ml in RPMI 1640 with 10% heat-inactivated NHS and 1% P/S and room temperature. The plate was washed six times with PBS/Tween before plated in 24-well cell culture plates (Costar, Cambridge, MA) at 1 ϫ 106 developing with 3,3Ј,5,5Ј-tetramethylbenzidine (Sigma-Aldrich) in phos- cells/well. One million irradiated (5000 rad) HLA-incompatible LCL as phate-citrate buffer. The reaction was stopped with2NH2SO4 and the allogeneic APC were added to each well to yield a final volume of 2 ml. plate was read at 450 nm on an ELISA microplate reader (Molecular De- by guest on October 5, 2021 MLR were incubated at 37¡Cat5%CO2 for 1Ð8 days depending upon the vices, Sunnyvale, CA). specific experiment at which time cells and/or supernatants were collected for flow cytometry and cytokine assays, respectively. A total of four dif- ferent responder-stimulator pairs were used in subsequent experiments on Results CD30 expression and function and each yielded similar findings. CD30 is expressed on a subset of activated human T cells that also express CD25 and CD45RO Primary and secondary in vitro T cell activation We previously demonstrated that CD30 is induced on both CD4 ϫ 6 Purified T cells were washed and cultured at 1 10 /ml in RPMI 1640 and CD8 human T cells following stimulation with alloantigen (8). with 10% heat-inactivated NHS and 1% P/S in 24-well plates (Costar) that ϩ had been precoated with anti-CD3 mAbs (10 ␮g/ml) plus anti-CD28 mAbs Alzona et al. (9) reported that CD30 human T cells generated in (10 ␮g/ml). For stimulation with soluble mAbs, PBMC (1 ϫ 106/ml) were response to anti-CD3 mAbs plus IL-2 express high levels of the cultured with anti-CD3 alone (10 ng/ml) or anti-CD3 (10 ng/ml) plus anti- p55 IL-2R (CD25) and are found within the CD45ROϩ activated/ 6 CD28 (5 ␮g/ml). Parallel cultures were established containing 1 ϫ 10 T memory T cell population. To determine whether a similar pattern ϫ 6 cells and 1 10 irradiated, allogeneic APC. In some experiments cells of CD30 expression was elicited with allostimulation, purified hu- were harvested on days 3Ð8 for phenotypic analysis by immunofluorescent staining and flow cytometry. In other experiments, after 4 days the cultures man T cells were cultured with allogeneic APC. Parallel cultures were supplemented with IL-2 (100 U/ml) and maintained for an additional of purified T cells were activated with plate-bound anti-CD3 plus 7 days. On day 11, cells maintained in IL-2 for 7 days were washed, anti-CD28 mAbs. We previously determined that CD30 is a rela- ϫ 6 counted, and replated at 1 10 /ml in 24-well plates precoated with anti- tively late activation marker with peak expression on days 6Ð8in CD3 and anti-CD28 mAbs, with 1 ϫ 106 of the same allogeneic APC used in the primary stimulation, with soluble anti-CD3 and autologous APC, or MLR (8) and on days 3Ð4 following stimulation with anti-CD3 with soluble anti-CD3 plus anti-CD28 mAbs and autologous APC. Acti- plus anti-CD28 mAbs (K. W. Chan and O. M. Martinez, unpub- vated T cells were harvested on day 14 for flow cytometry analysis. All lished findings). Therefore, cells were recovered on days 7 and 8 cultures were maintained at 37¡C in an atmosphere of 5% CO2. after allogeneic MLR or on days 3 and 4 after anti-CD3 plus anti- Immunofluorescent staining and flow cytometry CD28 mAb stimulation and labeled with fluorochrome-conjugated Abs to CD3, CD30, and CD45RO or CD25 for three-color flow Freshly isolated PBMC, anti-CD3 plus anti-CD28 mAb-stimulated, and cytometric analyses. Gates were established on CD3ϩ cells and T alloactivated mononuclear cells were analyzed for cell surface Ag expression by two- and three-color immunofluorescence and flow cytometry. cells were analyzed for expression of CD30 vs CD25 and CD30 vs Cells were incubated with FITC-, PE-, and PECy5-conjugated mAbs spe- CD45RO. Fig. 1 shows the results of a representative experiment. cific for CD3, CD4, CD30, CD25, CD45RO, OX-40, 4-1BB, and CD27, or On day 7 following allostimulation, all of the T cells that express fluorochrome-conjugated isotype-matched control mAbs for 30 min at 4¡C. CD30 are contained within the CD45ROϩ population (Fig. 1A, The labeled cells were washed with PBS/FCS and acquired on a FACScan right panel). Similarly, activation with anti-CD3 plus anti-CD28 (BD Biosciences, Mountain View, CA) with CellQuest software. Cells ϩ were gated on forward and side scatter and a minimum of 20,000 cells were mAbs elicits a population of CD30 T cells that coexpress collected for each sample. When cells were stained with anti-CD3-PECy5, CD45RO (Fig. 1B, right panel). In allogeneic MLR, 75.1% of 1786 CD30ϩ T CELLS IN ALLOIMMUNE RESPONSES

FIGURE 1. CD30ϩ T cells also express CD25 and CD45RO following primary stimulation with allogeneic APC (A) or anti-CD3 plus anti-CD28 mAbs (B). For CD25 analysis, PBMC were cultured with allogeneic APC for 8 days or plate- bound anti-CD3 and anti-CD28 mAbs for 4 days and then labeled with fluorochrome-conjugated isotype controls or the following mAbs: CD3- Cy5PE, CD25-PE, and CD30-FITC. For CD45R0 analysis, PBMC were cultured with allogeneic APC for 7 days or plate-bound anti-CD3 plus anti- CD28 mAbs for 3 days, and then labeled with fluorochrome-conjugated isotype controls or the following mAbs: CD3-Cy5PE, CD30-FITC, and CD45RO-PE. The data were analyzed by three- color immunofluorescence with gates established on CD3ϩ cells. The percentage of cells found in each quadrant is indicated. Similar results were obtained from three experiments and data shown are from one representative experiment. Downloaded from

CD3ϩ T cells express CD30, whereas only 24.9% of CD3ϩ T cells Activation with alloantigen and anti-CD3 plus anti-CD28 mAbs stimulated with anti-CD3 plus anti-CD28 mAbs express CD30. leads to distinct expression patterns of the TNFR family The expression of CD30 on T cells was compared with another members CD30, 4-1BB, and OX-40 marker of T cell activation, CD25. By day 8 of alloactivation

To determine the effect of allostimulation on the expression of http://www.jimmunol.org/ 89.3% of T cells expressed CD25 (Fig. 1A, middle left panel). All other TNFR family members we analyzed the coexpression of ϩ alloactivated CD30 T cells (58.2% of the total T cell population) CD30 with the activation-dependent molecules 4-1BB and OX-40 coexpress CD25. Similarly, Ͼ99% of T cells activated with anti- and the constitutively expressed molecule CD27. In these experi- CD3 plus anti-CD28 mAbs expressed CD25, and the CD30ϩ T ments T cells were enriched by negative selection which routinely cells were found within this CD25ϩ population (Fig. 1B, middle yielded a population of Ͼ98% CD3ϩ cells as determined by im- left panel). Thus, allostimulation, as well as immobilized anti-CD3 munofluorescence and flow cytometry (data not shown). Purified T plus anti-CD28 mAbs, elicit CD30ϩ T cells that coexpress CD25 cells were stimulated for 4 days with alloantigen or plate-bound and CD45RO. These data are consistent with a previous report (9) anti-CD3 plus anti-CD28 mAbs then analyzed for expression of in which CD30ϩ T cells, generated in response to anti-CD3 plus CD30, CD25, 4-1BB, CD27, and OX-40. As shown in a represen- by guest on October 5, 2021 IL-2, were also found within the CD25ϩCD45ROϩ population and tative experiment in Fig. 3A, Ͼ50% of T cells expressed CD30 indicate that CD30ϩ T cells are a subset of activated cells. following primary alloactivation and, as shown previously, the CD30ϩ T cells were uniformly CD25ϩ (Fig. 3, middle left panels). In contrast, only 16% of T cells expressed CD30 in response to Effect of APC on the induction of CD30 expression on T anti-CD3 plus anti-CD28 mAbs despite the fact that, on the basis lymphocytes of CD25 expression, 98% of the T cells display an activated phenotype. Moreover, the level of CD30 expression was lower on T In the previous experiments we observed that allogeneic APC were more effective than plate-bound anti-CD3 plus anti-CD28 mAbs in inducing the expression of CD30 on T cells. One possible explanation for the difference in the extent of CD30 expression is that the plate-bound anti-CD3 plus anti-CD28 system lacks APC, whereas in the MLR allogeneic APC are present. Alternately, there may be intrinsic differences in T cell activation by alloantigen as compared with anti-CD3 plus anti-CD28 mAbs that account for enhanced CD30 expression in MLR. To distinguish between these possibilities we compared CD30 expression on T cells activated by soluble anti-CD3 plus anti-CD28 mAbs and autologous APC with CD30 expression on T cells activated by plate-bound anti-CD3 FIGURE 2. Effect of APC on the induction of CD30 on T cells. Purified plus anti-CD28 mAbs in the absence of APC (Fig. 2). Stimulation T cells were cultured for 4 days with allogeneic APC (A) or immobilized with soluble anti-CD3 plus anti-CD28 mAbs and autologous APC anti-CD3 plus anti-CD28 (B). PBMC were cultured for 4 days with soluble (Fig. 2D) only modestly increased CD30 expression in comparison anti-CD3 (10 ng/ml) (C) or soluble anti-CD3 (10 ng/ml) plus anti-CD28 (5 to stimulation with plate-bound anti-CD3 plus anti-CD28 mAbs ␮g/ml) (D). As detailed in Materials and Methods, cultures were main- (13 and 5%, respectively) (Fig. 2B). In neither case did the pro- tained in IL-2 for 7 days then restimulated for 3 days under the same portion of T cells expressing CD30 approach the levels induced by conditions as the primary stimulation. Cells were recovered and labeled allogeneic APC (36%) (Fig. 2A). These data indicate that the mode with fluorochrome-conjugated isotype controls (open histograms) or anti- CD3-PE and anti-CD30-FITC (filled histograms). Gates were established of T cell activation (alloantigen vs anti-CD3 plus anti-CD28) ϩ ϩ on CD3 T cells for analysis. The percentage of CD30 cells is shown for rather than the presence or absence of APC accounts for the dif- each histogram. Similar findings were observed after primary stimulation ferential expression of CD30. of T cells. The data shown are representative of two experiments. The Journal of Immunology 1787

FIGURE 3. Phenotypic analysis of CD30ϩ T cells generated by primary stimulation with allogeneic APC (A)or anti-CD3 plus anti-CD28 mAbs (B). PBMC were cultured with allogeneic APC or plate-bound anti-CD3 plus anti- CD28 mAbs for 4 days. The cells were then labeled with fluorochrome-conjugated isotype controls or the following mAbs: CD3-Cy5PE, CD30-FITC, and CD25-PE, 4-1BB-PE, CD27-PE, or OX-40-PE and analyzed by three-color immunofluorescence with gates established on CD3ϩ cells. The percentage of cells in each quadrant is indicated. Similar results were obtained from three independent experiments and representative data are shown. Downloaded from cells activated with anti-CD3 plus anti-CD28 mAbs as indicated shown). Thus, the preferential induction of the TNFR family mem- by the reduced fluorescence intensity of CD30 staining as com- bers according to the mode of stimulation we observed could not pared to CD30ϩ T cells elicited by allogeneic APC. be explained by differences in the kinetics of expression of these Analysis of 4-1BB expression after allostimulation revealed two molecules. populations of CD30ϩ T cells (Fig. 3A, middle panel). The predominant population of CD30ϩ T cells was 4-1BBϪ (36% of all T http://www.jimmunol.org/ cells), while approximately one-third of the CD30ϩ T cells coex- Effect of secondary stimulation on CD30, 4-1BB, CD27, and pressed 4-1BB (17% of all T cells). Conversely, the major popu- OX-40 expression on T cells lation elicited by anti-CD3 plus anti-CD28 mAbs stimulation was To determine whether secondary stimulation affected the pattern of 4-1BBϩCD30Ϫ (30% of all T cells), while a minor population of CD30, 4-1BB, CD27, and OX-40 expression, purified T cells were T cells coexpressed 4-1BB and low levels of CD30 (Fig. 3B, mid- activated with allogeneic APC or anti-CD3 plus anti-CD28 mAbs dle panel). Overall, 19% of T cells activated with alloantigen ex- for 4 days, rested in IL-2 for 7 days, and then restimulated with pressed 4-1BB, whereas 36% of T cells expressed 4-1BB when either allogeneic APC or anti-CD3 plus anti-CD28 mAbs for an

activated with anti-CD3 plus anti-CD28 mAbs. Taken together, additional 3 days before immunofluorescent staining and flow cy- by guest on October 5, 2021 these results indicate that 4-1BB, like CD30, is expressed on a tometry. Fig. 4 shows the results of a representative experiment. subset of activated T cells and that 4-1BB and CD30 are expressed After two rounds of stimulation with alloantigen or anti-CD3 plus independently. anti-CD28 mAbs, as would be expected, essentially all the T cells CD27 is known to be constitutively expressed on T cells. We displayed an activated phenotype on the basis of CD25 expression observed that activation with either alloantigen or anti-CD3 plus (Fig. 4, middle left panels). Whereas Ͻ10% of the CD3ϩCD25ϩ anti-CD28 mAbs did not effect expression of CD27 (Fig. 3, middle cells expressed CD30 in the anti-CD3 plus anti-CD28 mAb-treated right panels). However, in the case of allostimulation, the CD27ϩ cultures, approximately one-half (45%) of the CD3ϩCD25ϩ cells cells were evenly divided between CD30ϩ and CD30Ϫ T cells in the alloactivated cultures expressed CD30. (Fig. 3A, middle right panel). In contrast, 89% of T cells activated With respect to 4-1BB, the general pattern of expression in both with anti-CD3 plus anti-CD28 mAbs were CD27ϩCD30Ϫ, while alloactivated T cells and anti-CD3 plus anti-CD28 activated T cells 10% were CD27ϩCD30low (Fig. 3B, middle right panel), reflecting was similar to that found after primary stimulation. Alloantigen the modest induction of CD30 by anti-CD3 plus anti-CD28 mAbs. induced minimal 4-1BB while anti-CD3 plus anti-CD28 mAbs Unlike CD27, there were marked differences in OX-40 expres- yielded a discrete population of 4-1BBϩCD30Ϫ cells (Fig. 4, A, sion depending on the mode of T cell stimulation. Primary alloa- middle panel,vsB, middle panel). Similarly, restimulation with ctivation resulted in low levels of OX-40 expression (23% of all T alloantigen did not significantly alter the relatively low levels of cells), but most of these OX-40ϩ T cells were found in the CD30ϩ expression of OX-40 that were observed after primary allogeneic population (Fig. 3A, right panel). Stimulation with anti-CD3 plus stimulation (compare Fig. 4A, right panel, and Fig. 3A, right pan- anti-CD28 mAbs induced OX-40 expression on Ͼ80% of T cells, el). However, T cells restimulated with anti-CD3 plus anti-CD28 and these cells were primarily CD30Ϫ (Fig. 3B, right panel). mAbs had notably lower levels of OX-40 expression than did T Taken together, these results suggest that the method of stimu- cells following primary anti-CD3 plus anti-CD28 mAbs lation effects the extent of CD30 expression. Allostimulation stimulation. yields a greater proportion of CD30ϩ T cells than does activation The most significant difference in primary and secondary stim- with anti-CD3 plus anti-CD28 mAbs. In contrast, stimulation with ulation was observed in the expression of CD27. Secondary stim- anti-CD3 plus anti-CD28 mAbs leads to a more pronounced in- ulation with alloantigen induced a more polarized pattern of CD30 duction of OX-40 and, to a lesser extent, 4-1BB than alloantigen. and CD27 expression as compared with primary allogeneic acti- It is important to note that OX-40 and 4-1BB, like CD30, are vation, in that the major T cell populations expressed either CD30 maximally expressed relatively late after activation. However, the or CD27 but not both molecules (Fig. 4A, middle right panel). The predominance of 4-1BB and OX-40 induction in response to anti- CD30ϩCD27ϩ population, which comprised 47% of all T cells CD3 plus anti-CD28 mAbs stimulation vs alloantigen was consis- after primary alloactivation, consisted of only 14% of T cells after tently observed over the course of 8 days of culture (data not secondary stimulation. These results indicate that CD30ϩ T cells 1788 CD30ϩ T CELLS IN ALLOIMMUNE RESPONSES

FIGURE 4. Phenotypic analysis of CD30ϩ T cells generated by secondary stimulation with allogeneic APC (A) or anti-CD3 plus anti-CD28 mAbs (B). PBMC were cultured with allogeneic APC or anti-CD3 plus anti-CD28 mAbs for 4 days, maintained in IL-2 (100 U/ml) for 7 days, then restimulated with allogeneic APC or anti-CD3 plus anti-CD28 mAbs for 3 days. The cells were labeled with ﬂuorochrome-conjugated isotype controls or the following mAbs: CD3-Cy5PE, CD30-FITC, and CD25-PE, 4-1BB-PE, CD27-PE, or OX-40-PE and analyzed by three-color immunoﬂuores- cence with gates established on CD3ϩ cells. The percentage of cells in each quadrant is indicated. One representative experiment from three experiments with similar results is shown.

also comprise a significant component of the T cell response dur- IFN-␥, and TNF-␣. IL-4 could not be detected in either type of ing secondary stimulation with alloantigen. However, restimula- culture (data not shown). Thus, differences in cytokine profile ac- Downloaded from tion with anti-CD3 plus anti-CD28 mAbs does not induce signif- company the distinct phenotypes induced by alloantigen and anti- icant CD30 expression. Furthermore, secondary stimulation with CD3 plus anti-CD28 mAbs. Although we did not fractionate alloantigen and anti-CD3 plus anti-CD28 mAbs results in down- CD30ϩ and CD30Ϫ T cells in these experiments, previous studies regulation of CD27 and OX-40, respectively. from our laboratory and others using primary allogeneic stimulation (8) or anti-CD3 plus IL-2 (9) suggest that the IL-5 and IFN-␥ Differential cytokine production in T cell lines generated by detected in supernatants following secondary allogeneic stimula- http://www.jimmunol.org/ alloantigen or anti-CD3 plus anti-CD28 mAbs tion are likely derived from CD30ϩ T cells. The lack of CD30ϩ T We have previously demonstrated that CD30ϩ T cells generated in cells, IL-5, and IFN-␥ following secondary stimulation with anti- response to primary alloantigenic stimulation produce abundant CD3 plus anti-CD28 mAbs is also consistent with this observation. IL-5 and IFN-␥ (8). To begin to address whether the differential Whether or not differential expression of other TNFR family mem- phenotypes induced by stimulation with alloantigen or anti-CD3 bers such as 4-1BB, CD27, or OX-40 can be used to further dis- plus anti-CD28 mAbs were associated with distinct cytokine pro- tinguish functional subsets of T cells activated with alloantigen files, purified T cells were stimulated with two rounds of alloan- remains to be determined. tigen or anti-CD3 plus anti-CD28 mAbs as described previously. On day 3 following the second stimulation the T cells were recov- CD30ϩ T cells are the predominant proliferating cell population by guest on October 5, 2021 ered and cultured in plates containing immobilized anti-CD3 in response to alloantigen mAbs. Supernatants were recovered after 48 h and analyzed for One possible explanation for the relatively large proportion of cytokine production by specific ELISA (Fig. 5). T cells stimulated ϩ CD30 T cells in alloactivated cultures is that these cells are ex- with two rounds of alloantigen produced significant levels of IL-2, panding through cellular proliferation. Therefore, we analyzed the IL-5, and IFN-␥ and modest levels of TNF-␣. Secondary stimu- ϩ Ϫ proliferative capacity of CD30 and CD30 T cells. PBMC were lation with immobilized anti-CD3 plus anti-CD28 mAbs yielded T labeled with CFSE and stimulated with either irradiated, allogeneic cells that produced comparable levels of IL-2 but minimal IL-5, APC or plate-bound anti-CD3 and anti-CD28 mAbs. Normal peripheral blood T cells do not express CD30 and minimal cell division could be detected before day 3 in alloactivated cultures (data not shown). Therefore, cells were harvested on days 3Ð6 and labeled with anti-CD3-PE mAbs to allow gating on T lymphocytes. Nonresponding cells can be observed as a population of CD30ϪCFSEhigh cells (Fig. 6A). On day 3 of alloactivation CD30ϩ T cells are evident and a significant proportion of the CD30ϩ T cells are CFSElow, indicating that they have undergone multiple cell divisions. It is worth noting that CD30ϩCFSEhigh cells are also detectable, indicating that cell division is not required for CD30 expression. Over- all, CD30ϩ T cells constitute the predominant proliferating cell population throughout the time course of the experiment, although some FIGURE 5. Cytokine production by T cells following secondary stim- alloactivated CD30ϪCFSElow T cells are also present. In contrast, ulation with allogeneic APC or anti-CD3 plus anti-CD28 mAbs. PBMC anti-CD3 plus anti-CD28 mAb stimulation elicits a dividing T cell were stimulated with allogeneic APC in MLR or plate-bound anti-CD3 population composed of both CD30ϩ and CD30Ϫ cells (Fig. 6B). plus anti-CD28 mAbs for 4 days, rested for 7 days in IL-2 (100 U/ml), then Quantitative analysis indicates that the proportion of dividing T restimulated in MLR (filled bars) or with immobilized anti-CD3 plus anti- cells that express CD30 remains relatively constant from day 3 to CD28 (open bars) for 3 days as described in Fig. 3. Cells were then har- ϭ vested and cultured in microtiter plates coated with anti-CD3 mAbs (10 day 6 of the MLR (range 65Ð78%) (Fig. 6C). Moreover, the ␮g/ml). Supernatants were harvested after 48 h and analyzed for cytokine persistence of CD30 expression on proliferating T cells in the al- production by specific ELISA. Shown is one representative experiment of loactivated cultures has been observed through day 11 (data not three experiments that yielded similar results. Data represent the mean Ϯ shown). When cells are stimulated with immobilized anti-CD3 ϩ Ϫ SEM of three determinations. plus anti-CD28 mAbs, the CD30 T cells and CD30 T cells The Journal of Immunology 1789

FIGURE 6. Cell division of CD30ϩ T cells and CD30Ϫ T cells activated by alloantigen or anti-CD3 plus anti-CD28 mAbs. PBMC were labeled with CFSE, then cultured with allogeneic APC (A)or anti-CD3 plus anti-CD28 mAbs (B). Cells were recovered on days 3Ð6 and labeled with anti-CD30 FITC and anti- CD3-Cy5PE mAbs. Gating was established on CD3ϩ cells for analysis. Data are representative of three independent experiments. Quantitative analysis of T cell Downloaded from proliferation in response to allogeneic APC (C) or anti-CD3 plus anti-CD28 mAb (D) was also performed. Data shown are the proportion of total dividing T cells that are CD30ϩ (ﬁlled bars) or CD30Ϫ (open bars). Data represent the mean Ϯ SEM of three http://www.jimmunol.org/ experiments. by guest on October 5, 2021

constitute similar proportions of the total cell population that has that can promote proliferation, cytokine production, and apoptosis undergone cell division on days 3 and 4 (Fig. 6D). However, by depending on the cell type and its state of differentiation and day 5 the majority (60.5%) of dividing cells induced by anti-CD3 whether other cell surface receptors are concomitantly engaged. plus anti-CD28 mAbs are CD30Ϫ and on day 6 CD30Ϫ T cells Furthermore, CD30ϩ T cells have been proposed to function as represent 72.5% of the cells that have undergone cell division. regulatory cells through their ability to produce cytokines. How- These results indicate that CD30ϩ T cells constitute the primary ever, little is known about the potential role of CD30ϩ T cells in proliferating cell population elicited by alloantigen, whereas plate- alloimmune responses and, more generally, about the relative ex- bound anti-CD3 plus anti-CD28 mAbs stimulate proliferation of pression of inducible TNFR family members on human T cells. both CD30ϩ and CD30Ϫ T cells. We have taken the approach of characterizing the phenotype and function of the human CD30 cells elicited by alloantigen as a Discussion means to begin to understand their contribution to alloreactivity. Numerous studies have demonstrated the important role of the T We previously reported that alloantigen induces the robust ex- cell costimulatory molecules CD28 and CD154 (CD40L) in allo- pression of CD30 on human T cells and that these cells are the immune responses and graft rejection (30Ð32). Recently, however, primary source of IL-5 and IFN-␥ (8). We now demonstrate that several members of the TNFR family have also been implicated as CD30ϩ T cells generated by allostimulation express high levels of T cell costimulatory molecules that participate in graft rejection. In CD25 and are found within the CD45RO population. A similar particular, 4-1BB costimulation has been shown to augment pro- pattern of CD30, CD25, and CD45RO coexpression was also liferation of alloreactive CD8 T cells in vitro (17), enhance the found in T cells stimulated with plate-bound anti-CD3 plus anti- generation of allospecific CTL in a murine model of acute GVHD, CD28 mAbs in the current study and with anti-CD3 plus IL-2 in and accelerate the rejection of cardiac and skin grafts (16). Mice previous studies by others (3). deficient in both 4-1BBL and CD28 exhibit a delay in skin allo- Moreover, using CFSE labeling to identify dividing cells, we graft rejection (15). In addition, in a rat model of GVHD, OX-40ϩ found that CD30 expression marks the predominant proliferating T donor-derived T cells were shown to be alloreactive and may con- cell population elicited by alloantigen. In contrast, the T cell pop- stitute effector cells (24) while CD27-CD27L interactions promote ulation that proliferates in response to immobilized anti-CD3 plus proliferation and generation of allospecific CTL (33). anti-CD28 mAbs is comprised of similar proportions of CD30ϩ Another member of the TNFR family, CD30, also has costimu- and CD30Ϫ cells. This may, at least in part, explain why allo- latory function and has been characterized as a signaling molecule stimulation consistently yielded a greater proportion of CD30ϩ T 1790 CD30ϩ T CELLS IN ALLOIMMUNE RESPONSES cells as compared with stimulation with anti-CD3 plus anti-CD28 In any case, the factors that account for the differential induction mAbs, despite the fact that only a subset of naive T cells (estimated and regulation of these TNFR family members on human T cells to be 1Ð10%) are capable of responding to alloantigen, whereas are unknown. An obvious difference in the two activation systems essentially all T cells should be activated by anti-CD3 plus anti- used in the current study is that the MLR contains purified re- CD28 mAbs. In the initial phase of the response to anti-CD3 plus sponder T cells and allogeneic APC, whereas the anti-CD3 plus ϩ Ϫ anti-CD28 mAbs, similar proportions of CD30 and CD30 T anti-CD28 mAbs system contains only purified T cells and no cells underwent cell division, whereas in the latter phase of the APC. The LCL used as allogeneic APC in the present study are response the predominant proliferating T cell population was known to express a variety of molecules including B7-1, B7-2, Ϫ CD30 . This shift is likely attributable to loss of CD30 expression ICAM-1, CD30L, and CD40, which could influence the subse- ϩ ϩ on CD30 T cells, rather than death of CD30 T cells, because we quent expression of inducible molecules on T cells. It should be ϩ were unable to detect significant apoptosis in the CD30 fraction noted that the use of allogeneic PBMC as APC in place of allo- (K. W. Chan and O. M. Martinez, unpublished data). geneic LCL yielded similar results with respect to CD30 expres- We also observed marked differences with respect to the ex- sion (K. W. Chan and O. M. Martinez, unpublished data). Further- pression of other TNFR family members on T cells, depending on more, our experiments comparing CD30 induction by plate-bound the mode of cell activation. Primary and secondary stimulation anti-CD3 plus anti-CD28 mAbs with CD30 induction by soluble with anti-CD3 plus anti-CD28 mAbs led to induction of a discrete Ϫ anti-CD3 plus anti-CD28 mAbs and autologous APC indicate that population of activated CD30 T cells that express moderate to it is the mode of T cell activation, rather than the presence or high levels of 4-1BB. In contrast, only low levels of 4-1BB ex- absence of APC, that dictates the extent of CD30 expression. pression were induced by alloantigen. Kim et al. (34) also analyzed CD30 expression has also been reported to be influenced by Downloaded from expression of CD30 and 4-1BB on human T cells after stimulation cytokines and CD28. Some studies have suggested that IL-4 is with anti-CD3 plus anti-CD28 mAbs. These authors concluded essential for CD30 expression on human T cells (36), but other that 4-1BB and CD30 were expressed on mutually exclusive T cell studies found that IL-4 inhibits expression of CD30 on human T subsets. However, their analysis was restricted to CD4 T cells that cells (3). Experiments in murine models indicate that IL-4 is im- had undergone secondary stimulation with anti-CD3 plus anti- portant in regulation of CD30 expression (37). However, IL-4 does CD28 mAbs. Indeed, when we analyzed CD3ϩ T cells after sec-

not appear to play a role in our studies because we have been http://www.jimmunol.org/ ondary stimulation with anti-CD3 plus anti-CD28 mAbs we found unable to detect IL-4 in supernatants from MLR or anti-CD3 plus only a small population (5%) of cells that coexpress CD30 and anti-CD28 mAb-stimulated T cell cultures. CD28 costimulation 4-1BB. In contrast, our data indicate that primary allogeneic stim- has also been implicated as a requirement for CD30 expression ulation can induce a CD30ϩ4-1BBϩ T cell population. (38). In the current study CD28 signaling was provided by anti- We also found that primary stimulation with anti-CD3 plus anti- CD28 mAbs in the plate-bound Ab system or by B7-1 and B7-2 on CD28 mAbs induced OX-40 on the majority of T cells, whereas LCL stimulators in the MLR system. Because CD28 engagement Ͻ25% of activated T cells stimulated with alloantigen expressed occurs in both systems it is not likely that CD28 signaling accounts OX-40. Primary stimulation with either alloantigen or anti-CD3 Ͼ for the difference in expression of CD30; however, we cannot rule plus anti-CD28 mAbs resulted in CD27 on 90% of activated T by guest on October 5, 2021 out the possibility that the nature of CD28 signaling differs when cells. However, secondary stimulation with alloantigen resulted in ϩ agonist anti-CD28 mAbs are used instead of the natural ligands down-regulation of CD27 on the CD30 subset. Interestingly, ϩ ϩ B7-1 and B7-2. there was a slightly higher frequency of CD8 CD30 cells and CD8ϩ4-1BBϩ cells compared with CD4ϩCD30ϩ and CD4ϩ4- Another obvious difference in the two modes of stimulation is 1BBϩ cells, whereas there was no difference in the expression of the strength of the TCR signal. Anti-CD3 mAbs have a higher OX-40 and CD27 in T cell subsets (K. W. Chan and O. M. Mar- affinity for CD3 than MHC does for the TCR. However, we have tinez, unpublished data). Wen et al. (35) also found that a greater seen a similar pattern of CD30 expression on T cells in the plate- proportion of CD8 cells express 4-1BB compared with CD4 cells bound anti-CD3 plus anti-CD28 system even when suboptimal when human T cells were stimulated with immobilized anti-CD3. doses of anti-CD3 mAbs are used (C. D. Hopke and O. M. Mar- Taken together our data suggest that the mode of stimulation can tinez, unpublished data). These results suggest that the strength of influence the expression of the TNFR family members CD30, the TCR signal alone does not determine the pattern of CD30 4-1BB, CD27, and OX-40. expression. Although numerous studies in murine models support a role for The function of the CD30 molecule on alloactivated T cells remains unclear. Several reports have suggested that CD30 may 4-1BB and OX-40 in allogeneic responses (15Ð17, 24) our studies ϩ ϩ indicate that a relatively small subset of alloreactive human T cells play a role in delivering an apoptotic signal to CD4 CD8 thy- express these molecules. Whether there are differences in the role mocytes in negative selection (12) and in elimination of activated, of these molecules in allogeneic responses in human and mouse peripheral T cells (13, 39). However, we did not detect enhanced ϩ systems or whether the small subset of alloactivated human T cells apoptosis in the CD30 activated T cell population generated by which express 4-1BB or OX-40 contributes to alloimmunity re- either alloantigen or anti-CD3 plus anti-CD28 mAbs (K. W. Chan mains to be determined. and O. M. Martinez, unpublished data). Alternately, it is plausible Recent studies in murine systems using DO11.10 transgenic T that CD30 costimulation in alloactivated T cells influences cellular cells as well as transgenic T cells specific for an LCMV-derived survival, proliferation, or cytokine production. Experiments are peptide indicate that individual T cells can simultaneously express under way to test these possibilities. CD30, OX-40, and 4-1BB (20). Our results indicate a more re- In summary, we have determined that alloimmune stimulation is stricted, less overlapping pattern of expression on human T cells characterized by induction of CD30 in a discrete subset of activated T among these TNFR family members. Several factors may account cells. Importantly, while CD30 expression is limited to a subset of for the observed discrepancy, including differences in murine and activated cells, its expression marks the predominant proliferating cell human systems, the use of transgenic T cells vs normal T cells, or population in response to alloantigen. Thus, it will be important to the use of peptide Ag vs alloantigen and anti-CD3 plus anti-CD28 address the function of the CD30 molecule on alloactivated T cells stimulation. and the role of CD30ϩ T cells in transplantation. The Journal of Immunology 1791

References 21. Gramaglia, I., A. D. Weinberg, M. Lemon, and M. Croft. 1998. Ox-40 ligand: a potent costimulatory molecule for sustaining primary CD4 T cell responses. 1. Durkop, H., U. Latza, M. Hummel, F. Eitelbach, B. Seed, and H. Stein. 1992. J. Immunol. 161:6510. Molecular cloning and expression of a new member of the nerve growth factor 22. Kopf, M., C. Ruedi, N. Schmitz, A. Gallimore, K. Lefrang, B. Ecbert, receptor family that is characteristic for Hodgkin’s disease. Cell 68:421. B. Odermatt, and M. F. Bachmann. 1999. Ox40-deficient mice are defective in Th 2. Schwab, U., H. Stein, J. Gerdes, H. Lemke, H. H. Kirchner, M. Schaadt, and cell proliferation but are competent in generating B cell and CTL responses after V. Diehl. 1982. Production of a monoclonal antibody specific for Hodgkin and viral infection. Immunity 11:699. Sternberg Reed cells of Hodgkin’s disease and a subset of normal lymphoid cells. 23. Flynn, S., K.-M. Toellner, C. Raykundalia, M. Goodall, and P. Lane. 1998. CD4 Nature 299:65. T cell cytokine differentiation: the B cell activation molecule, OX40 ligand, in- 3. Ellis, T. M., P. E. Simms, D. J. Slivnick, H.-M. Jack, and R. I. Fisher. 1993. structs CD4 T cells to express interleukin 4 and upregulates expression of the CD30 is a signal-transducing molecule that defines a subset of human activated chemokine receptor, Blr-1. J. Exp. Med. 188:297. CD45ROϩ T cells. J. Immunol. 151:2380. ϩ 24. Tittle, T. V., A. D. Weinberg, C. N. Steinkeler, and R. T. Maziarz. 1997. Ex- 4. Martinez, O. M. 2001. Immunobiology of CD30 T lymphocytes. Graft 4:180. pression of the T cell activation antigen, OX-40, identifies alloreactive T cells in 5. Romagnani, S., G. Del Prete, E. Maggi, M. Chilosi, F. Caligaris-Cappio, and acute graft-versus-host disease. Blood 89:4652. G. Pizzolo. 1995. CD30 and type 2 T helper (Th2) responses. J. Leukocyte Biol. 25. Weinberg, A. D., K. W. Wegmann, C. Funatake, and R. H. Whitham. 1999. 57:726. Blocking Ox-40/OX-40 ligand interaction in vitro and in vivo leads to decreased 6. Hamann, D., C. M. U. Hilkens, J. L. Grogan, S. M. A. Lens, M. L. Kapsenberg, T cell function and amelioration of experimental allergic encephalomyelitis. M. Yazdanbakhsh, and R. A. W. van Lier. 1996. CD30 expression does not J. Immunol. 162:1818. discriminate between human Th1- and Th2-type T cells. J. Immunol. 156:1387. 26. Akiba, H., Y. Miyahira, M. Atsuta, K. Taked, C. Nohara, T. Futagawa, 7. Bengtsson, A., C. Johansson, M. T. Linder, G. Hallden, I. Van der Ploeg, and H. Matsuda, T. Aoki, H. Yagita, and K. Okumura. 2000. Critical contribution of A. Scheynius. 1995. Not only Th2 cells but also Th1 and Th0 cells express CD30 OX40 ligand to T helper cell type 2 differentiation in experimental leishmaniasis. after activation. J. Leukocyte Biol. 58:683. J. Exp. Med. 191:375. 8. Martinez, O. M, J. C. Villanueva, S. Abtahi, P. R. Beatty, C. O. Esquivel, and S. M. Krams. 1998. CD30 expression identifies a functional allospecific T lym- 27. Higgins, L. M., S. A. C. McDonald, N. Whittle, N. Crockett, J. G. Shields, and phocyte population. Transplantation 65:1240. T. T. MacDonald. 1999. Regulation of T cell activation in vitro and in vivo by 9. Alzona, M., H.-M. Jack, R. Fisher, and T. M. Ellis. 1994. CD30 defines a subset targeting the OX40-OX40 ligand interaction: amelioration of ongoing inflamma- of activated human T cells that produce IFN-␥ and IL-5 and exhibit enhanced B tory bowel disease with an OX40-IgG fusion protein, but not with an OX40 Downloaded from cell helper activity. J. Immunol. 153:2861. ligand-IgG fusion protein. J. Immunol. 162:486. 10. Smith, C. A., J.-J. Gruss, T. Davis, D. Anderson, T. Farrah, E. Baker, 28. Weinberg, A. D., M.-M. Rivera, P. Prell, A. Morris, T. Ramstad, J. T. Vetto, G. R. Sutherland, C. I. Brannan, N. G. Copeland, N. A. Jenkins, et al. 1993. CD30 W. J. Urba, G. Alvord, C. Bunce, and J. Shields. 2000. Engagement of the OX-40 antigen, a marker for Hodgkin’s lymphoma, is a receptor whose ligand defines an receptor in vivo enhances antitumor immunity. J. Immunol. 164:2160. emerging family of cytokines with homology to TNF. Cell 73:1349. 29. Martinez, O. M., J. C. Villanueva, L. Lawrence-Miyasaki, M. B. Quinn, K. Cox, 11. Biswas, P., C. A. Smith, D. Goletti, E. C. Hardy, R. W. Jackson, and A. S. Fauci. and S. M. Krams. 1995. Viral and immunologic aspects of Epstein-Barr virus 1995. Cross-linking of CD30 induces HIV expression in chronically infected T infection in pediatric liver transplant recipients. Transplant 59:519. 30. Turka, L. A., P. S. Linsley, H. Lin, W. Brady, J. M. Leiden, R. Q. Wei,

cells. Cell 2:587. http://www.jimmunol.org/ 12. Amakawa, R., A. Hakem, T. M. Kundig, T. Matsuyama, J. J. L. Simard, M. L. Gibson, X. G. Zheng, S. Myrdal, D. Gordon, et al. 1992. T cell activation E. Timms, A. Wakeham, H.-W. Mittruecker, H. Griesser, H. Takimoto, et al. by the CD28 ligand B7 is required for cardiac allograft rejection in vivo. Proc. 1996. Impaired negative selection of T cells in Hodgkin’s disease antigen CD30- Natl. Acad. Sci. USA 89:11102. deficient mice. Cell 84:551. 31. Parker D. C., D. L. Greiner, N. E. Phillips, M. C. Appel, A. W. Steele, 13. Chiarle, R., A. Podda, G. Prolla, E. R. Podack, G. J. Thorbecke, and G. Inghirami. F. H. Durie, R. J. Noelle, J. P. Mordes, and A. A. Rossini. 1995. Survival of 1999. CD30 overexpression enhances negative selection in the thymus and me- mouse pancreatic islet allografts in recipients treated with allogeneic small lym- diates programmed cell death via a Bcl-2 sensitive pathway. J. Immunol. 163: phocytes and antibody to CD40 ligand. Proc. Natl. Acad. Sci. USA 92:9560. 194. 32. Larsen, C. P., E. T. Elwood, D. Z. Alexander, S. C. Ritchie, R. Hendrix, 14. DeYoung, A. L., O. Duramad, and A. Winoto. 2000. The TNF receptor family C. Tucker-Burden, H. R. Cho, A. Aruffo, D. Hollenbaugh, P. S. Linsley, et al. member CD30 is not essential for negative selection. J. Immunol. 165:6170. 1996. Long-term acceptance of skin and cardiac allografts after blocking CD40 15. DeBenedette, M. A., T. Wen, M. F. Bachmann, P. S. Ohashi, B. H Barber, and CD28 pathways. Nature 381:434. 33. Brown, G. R., K. Meek, Y. Nishioka, and D. L. Thiele. 1995. CD27-CD27ligand/ K. L. Stocking, J. J. Peschon, and T. H. Watts. 1999. Analysis of 4-1BB ligand by guest on October 5, 2021 CD70 interactions enhance alloantigen-induced proliferation and cytolytic activ- (4-1BBL)-deficient mice and of mice lacking both 4-1BBL and CD28 reveal a ϩ role for 4-1BBL in skin allograft rejection and in the cytotoxic T cell response to ity in CD8 T lymphocytes. J. Immunol. 154:3686. influenza virus. J. Immunol. 163:4833. 34. Kim, Y.-J., S. H. Kim, P. Mantel, and B. S. Kwon. 1998. Human 4-1BB regulates 16. Shuford, W. W., K. Klussman, D. D. Tritchler, K. T. Loo, J. Chalupny, CD28 co-stimulation to promote T cell responses. Eur. J. Immunol. 28:881. A. W. Siadak, T. J. Brown, J. Emswiler, H. Raecho, C. P. Larsen, et al. 1997. 35. Wen, T., J. Bukczynski, and T. H. Watts. 2002. 4-1BB ligand-mediated costimu- 4-1BB costimulatory signals preferentially induce CD8ϩ T cell proliferation and lation of human T cells induces CD4 and CD8 T cell expansion, cytokine pro- lead to amplification in vivo of cytotoxic T cell responses. J. Exp. Med. 186:47. duction, and the development of cytolytic effector function. J. Immunol. 168: 17. Tan, J. T., J. Ha, H. R. Cho, C. Tucker-Burden, R. C. Hendrix, R. S. Mittler, 4897. T. C. Pearson, and C. P. Larsen. 2000. Analysis of expression and function of the 36. Annunziato, F., R. Manetti, L. Cosmi, G. Galli, C. H. Heusser, S. Romagnani, and costimulatory molecule 4-1BB in alloimmune responses. Transplant 70:153. E. Maggi. 1997. Opposite role for interleukin 4 and interferon-␥ on CD30 and 18. Tan, J. T., J. K. Whitmire, R. Ahmed, T. C. Pearson, and C. P. Larsen. 1999. lymphocyte activation-gene (LAG-3) expression by activated naive T cells. Eur. 4-1BB ligand, a member of the TNF family, is important for the generation of J. Immunol. 27:2239. antiviral CD8 T cell responses. J. Immunol. 163:4859. 37. Nakamura, T., R. K. Lee, S. Y. Nam, B. K. Al-Ramadi, P. A. Koni, K. Bottomly, 19. Melero, I., W. W. Shuford, S. A. Newby, A. Aruffo, J. A. Ledbetter, E. R. Podack, and R. A. Flavell. 1997. Reciprocal regulation of CD30 expression ϩ K. E. Hellstrom, R. S. Mittler, and L. Chen. 1997. Monoclonal antibodies against on CD4 T cells by IL-4 and IFN-␥. J. Immunol. 158:2090. the 4-1BB T cell activation molecule eradicate established tumors. Nat. Med. 38. Gilfillan, M. C., P. J. Noel, E. R. Podack, S. L. Reiner, and C. B. Thompson. 3:682. 1998. Expression of the costimulatory receptor CD30 is regulated by both CD28 20. Cannons, J. L., P. Lau, B. Ghumman, M. A. DeBenedette, H. Yagita, and cytokines. J. Immunol. 160:2180. K. Okumura and T. H. Watts. 2001. 4-1BB ligand induces cell division, sustains 39. Lee, S. Y., C. G. Park, and Y. Choi. 1996. T cell receptor-dependent cell death survival, and enhances effector function of CD4 and CD8 T cells with similar of T cell hybridomas mediated by the CD30 cytoplasmic domain in association efficacy. J. Immunol. 167:1313. with tumor necrosis factor receptor-associated factors. J. Exp. Med. 183:669.