sign in sign up
<<
Home , CD154, CD19, CD70, Interferon, Interferon gamma, MHC class II

(2001) 15, 1451–1457  2001 Nature Publishing Group All rights reserved 0887-6924/01 $15.00 www.nature.com/leu T activation following infection of primary follicle center B cells with adenovirus encoding CD154 MJ Cantwell, WG Wierda, IS Lossos, R Levy and TJ Kipps

Immunogenex, Inc., La Jolla, CA; Division of Oncology, Stanford University Medical School, Stanford, CA; and Division of Hematology/Oncology, UCSD School of Medicine, La Jolla, CA, USA

Purified, high-titer adenovirus encoding murine CD154 (Ad- cell cytotoxicity towards nonmodified CLL B cells.13 More- CD154) or human CD154 (Ad-hCD154) was used to infect lymph over, a phase I clinical study using Ad-CD154 modified CLL node cells isolated from patients with follicle center lymphoma. 14 Infection of lymphoma B cells with Ad-CD154 at a multiplicity B cells has shown promising therapeutic results. Similarly, of infection (MOI) ratio of 100 or higher resulted in high-level other groups have shown that modification of various tumor transgene expression. Additionally, upon infection of lym- types to express CD154 can induce effective anti-tumor phoma B cells, only Ad-CD154 resulted in surface expression immune responses in animal tumor models.15–17 of CD154, despite similar, high-level expression of either These encouraging results stimulated us to examine the human or mouse CD154 by HeLa cells infected with Ad-hCD154 feasibility of Ad-CD154 for other or Ad-CD154, respectively. Moreover, infection of lymphoma B cells with Ad-CD154, but not Ad-hCD154 or adenovirus enco- malignancies. B cell lymphoma appears to be an excellent ding Eschericheria coli beta-galactosidase (Ad-LacZ), induced candidate disease for this approach because, similar to CLL, the neoplastic B cells to express higher levels of immune co- lymphoma B cells express CD40 and can be induced to stimulatory molecules that are required for proficient presen- express multiple immune co-stimulatory molecules upon tation of to T cells. Consistent with this, we found that CD40 ligation.18,19 Moreover, since lymphoma cells express Ad-CD154 infected lymphoma B cells could stimulate T cells to clonally restricted immunoglobulins and generally harbor proliferate or produce -gamma in allogeneic or auto- logous mixed interactions. We conclude that cytogenetic abnormalities, these cells should express somati- lymphoma B cells can be infected with Ad-CD154 and that this cally generated that potentially could be targeted by significantly enhances their recognition by allogeneic or auto- the .20 As such, we examined whether logous T cells. As such, Ad-CD154-transduced lymphoma B lymphoma B cells could be infected with recombinant aden- cells may have potential for the active immune therapy of ovirus vectors. Furthermore, we studied the phenotype and patients with follicle center lymphoma. Leukemia (2001) 15, immune function of such cells following infection with adeno- 1451–1457. Keywords: follicle center lymphoma; CD154; adenovirus encoding CD154.

Materials and methods Introduction Patient samples CD40–CD154 (CD40-) interactions are critical for immune recognition.1,2 CD154 is transiently expressed on + After obtaining informed consent, lymph node were CD4 T cells following (TCR) engagement by performed on patients with B cell follicle center lymphoma antigen-presenting cells through major and prepared into single cell suspensions, as described.21 The complex (MHC) class II molecules.3–6 This, in turn, can cause cells then were subjected to density-gradient centrifugation activation of CD40-expressing antigen presenting cells (APCs), over cushions of Ficoll–Paque PLUS (Amersham Pharmacia including B cells, dendritic cells, , and macro- Biotech, Uppsala, Sweden) to isolate mononuclear cells, as phages.7,8 Such CD40-activated cells can set off a cascade of per the manufacturer’s instructions. The cells were suspended immune-activating events that lead to a specific and effective in fetal calf serum containing 10% dimethylsulfoxide (Sigma, against foreign antigens, such as or St Louis, MO, USA) for storage in liquid nitrogen until use. tumors.9,10 The importance of CD40–CD154 interactions is Blood mononuclear cells were similarly prepared from either underscored by the finding that individuals who have normal donors or follicle center lymphoma patients as indi- inherited defects in the ligand for CD40 have a profound cated. Cell cultures were maintained in serum free AIM-V immune deficiency.11,12 Such patients have an immune media (Gibco-BRL, Grand Island, NY, USA). deficiency syndrome associated with impaired formation, defective isotype switching, and marked susceptibility to various bacterial and viral pathogens. Adenovirus vector constructs In of the role that this receptor–ligand pair plays in immune regulation, the potential for using CD154 for immun- Replication-incompetent adenovirus vectors encoding murine otherapy is under active investigation. For example, chronic CD154 (Ad-CD154), human CD154 (Ad-hCD154), or ␤- lymphocytic leukemia (CLL) B cells modified to express galactosidase (Ad-LacZ) cDNA were made as described.13,22 CD154 using a replication defective adenovirus vector can Each cDNA was flanked by the (CMV) enhance CLL and induce autologous T /enhancer on the 5Ј end and the bovine growth hor- mone polyadenylation signal sequence on the 3Ј end, allowing for high-level expression of the transgene mRNA in Correspondence: TJ Kipps, Division of Hematology/Oncology, Ј Department of Medicine, University of Californai-San Diego, 9500 eukaryotic cells without 3 non-translated regulatory 10 Gilman Dr., La Jolla, CA 92093-0663, USA; Fax: 858 534 5620 sequences. Purified, high-titer (eg 0.8–2 × 10 p.f.u./ml), Received 27 March 2001; accepted 13 May 2001 recombinant adenovirus was produced by Molecular Medi- T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1452 cine, LLC (San Diego, CA, USA; www.molecularmed.com) by each well of a 96-well U-bottom plate. Subsequently, 105 pur- anion-exchange chromatography, as described.23–25 Adeno- ified blood T cells from allogeneic normal donors or 105 auto- virus purity and titers were determined by high performance logous blood T cells were added to each well. In each case, liquid chromatography (Figure 1) and by a plaque forming the CD3+ T cells were isolated to Ͼ93% purity using the T assay using 293 cells, as described.26 Cell Negative Isolation Kit from Dynal, Inc (Lake Success, NY, USA), by which CD19, CD14, CD16, CD56, and HLA-DR- positive cells were depleted. For autologous MLR reactions, Flow cytometry we added recombinant human -2 (IL-2; Biosource International, Cambridge, MA, USA) to each well to a final Cells were washed and suspended in staining buffer, com- concentration of 25ng/ml. For some samples, blocking-mAb posed of phosphate buffered saline (PBS, pH = 7.4) containing specific for MHC-class I (clone W6/32) was added at 3 ␮g/ml 3% fetal calf serum, 0.05% sodium azide, and 10 ␮g/ml pro- final concentration. Following 2 days’ incubation, aliquots of pidium iodide. Cells then were incubated with saturating the media were collected to measure IFN-␥ production by amounts of fluorochrome-conjugated monoclonal -linked immunosorbent assay (ELISA) using two IFN- (mAb) for 30 min at 4°C. Stained cells were washed twice ␥-specific mAbs (Pharmingen) and the manufacturer’s proto- with staining buffer and analyzed using a FACSCaliber flow col. To assess cell proliferation, we added 3H-thymidine to a cytometer (Becton Dickinson, San Jose, CA, USA). Dead cells final concentration of 0.5 ␮Ci/well 5days after the initiation and debris were excluded from the analysis by light scatter of the MLR and then harvested the cells 12 to 16 h later to and by gating out cells that failed to exclude propidium iod- measure incorporated 3H-thymidine using a cell harvester ide. We determined the percentages of cells that expressed (Tomtec, Hamden, CT, USA) and a beta-scintillation counter the transgene product and the mean fluorescence intensity (Wallac, Turku, Finland). ratio (MFIR) of transgene-expressing cells to assess the relative expression levels of the transgene. The MFIR is calculated by dividing the mean fluorescence intensity (MFI) of cells that Results were stained with the fluorochrome-conjugated antigen-spe- cific mAb by the MFI of cells that were stained with a control Adenovirus infection of lymphoma cells fluorochrome-conjugated mAb of the same isotype, but of irrelevant specificity. All were obtained from We initially determined if replication-incompetent adenovirus Pharmingen, Inc. (La Jolla, CA, USA). These included fluor- encoding either murine CD154 (Ad-CD154) or human CD154 escein-conjugated mAb specific for CD3, CD27, CD70, and (Ad-hCD154) could infect lymphoma B cells and result in sur- CD80, and PE-conjugated mAbs specific for mouse CD154, face expression of CD154. Each vector was prepared to high- human CD154, CD86, CD95, and CD71. In addition, we used titer and purity, as assessed via anion-exchange chromato- CyChrome-conjugated mAb specific for HLA-DR and APC- graphy (Figure 1). Following infection with Ad-CD154 or Ad- conjugated mAb specific for CD19. hCD154 at increasing MOI ratios, HeLa cells expressed sur- face mouse or human CD154, respectively (Figure 2). More- over, the levels of surface CD154 expression on HeLa cells Mixed lymphocyte reactions were similar for either vector at the same relative MOI ratio (Figure 2). Likewise, infection of lymphoma cells with Ad- Two days after the initial infection with adenovirus, the CD154 also resulted in high expression levels of murine infected lymphoma stimulator cells were treated with 100 CD154 provided MOI ratios were у100 (Figure 2). However, ng/ml mitomycin C for 1 h at 37°C to block proliferation. infection of lymphoma cells with Ad-hCD154 did not result These cells then were washed three times to remove the mito- in significant expression of human CD154, even at MOI ratios mycin C. 105 mitomycin C-treated cells were then placed in of 1000 (Figure 2). Infection with Ad-CD154 at MOI ratios ranging from 100 to 1000 resulted in transgene expression by 20 ± 15% to 85 ± 18% of the CD19+ gated B cells (n = 7) (Figure 3a). Moreover, inspection of the cell subsets infected with Ad-CD154 revealed that murine CD154 was expressed predominantly on CD19-positive cells (Figure 3b). In contrast, relatively few CD19-negative cells, comprised mostly of CD3+ T cells, expressed CD154 following infection with Ad-CD154, even at high MOI ratios (Figure 3b). Time course analysis also showed that CD154 transgene expression reached maximum levels approximately 2 days after infection and lasted for at least 7 days after infection (Figure 4).

Ad-CD154 induced B cell phenotypic changes

Figure 1 HPLC chromatographs of Ad-CD154 and Ad-hCD154 We examined whether infection with Ad-CD154 or Ad- purified adenovirus. Histograms demonstrate the optical density at hCD154 resulted in phenotypic changes in infected lym- 260 nm over time of the material eluted from high performance liquid phoma B cells. In contrast to non-infected, Ad-LacZ infected, chromatography of anion-exchange-purified Ad-CD154 (upper histogram) or Ad-hCD154 (lower histogram). The peak in each spec- or Ad-hCD154 infected lymphoma B cells, only Ad-CD154 trograph corresponds to that of the virus vector as indicated by the infected lymphoma B cells were induced to express immune arrow. accessory molecules, such as CD80, CD86, CD54, CD70, and

Leukemia T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1453 reaction (MLR). Normal donor T cells were purified to Ͼ93% purity by depleting CD19, CD14, CD56, CD16, and HLA-DR- positive cells from isolated blood mononuclear cells of normal donors. Upon co-culture with mitomycin-C-treated Ad- CD154-infected lymphoma cells, the allogeneic T cells were induced to undergo proliferation at rates significantly greater than that observed with non-infected lymphoma cells or lym- phoma cells infected with Ad-LacZ or Ad-hCD154 (Figure 7a). Moreover, Ad-CD154-modified lymphoma cells induced these cells to produce significantly greater amounts of IFN-␥ than did lymphoma cells infected with any other adenovirus vector (Figure 7b).

Autologous T cell responses

Similarly, Ad-CD154-modified lymphoma B cells were tested for their ability to stimulate autologous T cells. Each MLR was performed in serum-free culture medium to minimize the potential for presenting foreign serum antigens. Autologous CD3+ blood T cells incorporated significantly greater amounts of 3H-thymidine upon co-culture with autologous Ad-CD154- infected lymphoma cells than they did when co-cultured with non-infected cells or lymphoma cells infected with Ad-LacZ (Figure 8a). Furthermore, this response was significantly inhibited by an anti-HLA mAb, W6/32, consistent with the response being restricted by MHC class I antigens. We also examined for phenotypic changes in lymph node T cells from these patients with follicle center lymphoma fol- lowing ex vivo adenovirus infection. We found that nodal T Figure 2 Infection of HeLa and lymphoma B cells with Ad- cells were induced to express MHC class II antigen (HLA-DR) hCD154 or Ad-CD154. (a) 5 × 105 HeLa cells (left graph) or 2.5 × and (CD71) following infection of the lym- 6 10 lymphoma B cells (right graph) were infected at increasing MOI phoma B cells with Ad-CD154 (Figure 8b). Two days after the ratios with either Ad-CD154 (squares) or Ad-hCD154 (circles) for 2 ± + days. HeLa cells and CD19+ lymphoma cells were then stained for initial infection, an average of 36% ( 10% s.d.) of the CD3 T cells in Ad-CD154 infected patient samples (n = 3) were surface CD154 with murine or human CD154 species-specific mAb. + (b) 5 × 105 HeLa cells (upper histograms) or 2.5 × 106 lymph node HLA-DRbright. The proportion of CD3 cells that expressed mononuclear cells (lower histograms) were infected with either Ad- these activation antigens following infection with Ad-CD154 CD154 or Ad-hCD154 at an MOI ratio of 10 for HeLa cells and 1000 + + was significantly greater than that of the CD3 cells of lymph for lymphoma cells. Two days after infection, HeLa cells and CD19 node cells infected with control Ad-LacZ virus (m = 16% ± lymphoma cells were stained for CD154 and then analyzed by flow = ± cytometry. 5% s.d.) or non-infected control cells (m 13% 3% s.d.) (P Ͻ 0.05, Bonferroni test).

CD95 (Figure 5). In addition, Ad-CD154 infected lymphoma cells demonstrated reduced expression of CD27, as previously Discussion noted for CD40 stimulated normal and leukemia B cells.8 We also investigated whether infection with Ad-CD154 This is the first report of successful transfer to human affected lymphoma cell proliferation or survival in vitro. Equal primary follicle center lymphoma B cells. We prepared high- numbers of Ad-CD154 or non-infected lymphoma B cells titer adenovirus vectors of high purity for infection of mono- were cultured in AIM-V medium and assessed for viability and nuclear cells isolated from the lymph nodes of patients with cell number over time. There were no significant differences follicular B cell lymphoma. Nearly all CD19-positive lym- detected between the numbers of Ad-CD154-infected vs non- phoma B cells within lymph node cell preparations could be infected lymphoma cells over time (Figure 6a). Moreover, the made to express a transgene within 24 h after infection. This viability of adenovirus-infected cells was not significantly dif- required that we infect the cells at MOI ratios that were ferent from that of non-infected cells (Figure 6b). Furthermore, approximately 10-fold higher than that required for effective we found that ratios of B cells to T cells remained constant gene transfer into HeLa cells. Nevertheless, we found that the over time after infection, indicating that infection with Ad- CD19-positive lymphoma cells were among the most sensitive CD154 did not reduce the relative numbers of T cells within cells to adenovirus infection within such lymph node prep- the mononuclear lymph node cell population (data not arations. On the other hand, the T cells within these prep- shown). arations were relatively resistant to infection (Figure 3). The relative sensitivity of lymphoma B cells to infection with aden- ovirus vectors may obviate having to isolate lymphoma B cells Allogeneic mixed lymphocyte reactions from other cell types prior to infection to achieve selective gene transfer into neoplastic B cells. We examined whether Ad-CD154-infected lymphoma cells Despite having similar titer and purity, the adenovirus vec- could stimulate allogeneic T cells in a mixed lymphocyte tor encoding human CD154 failed to effect expression of

Leukemia T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1454

Figure 3 Ad-CD154 infection of lymphoma cells. 2.5 × 106 lymph node mononuclear cells were infected with Ad-CD154 at the indicated MOI ratio. Two days after infection, cells were stained for surface CD154 and analyzed by flow cytometry. (a) CD154 mAb staining (open histograms) compared to isotype control mAb staining (shaded histograms) for CD19-positive cells are plotted. The percentage of cells expressing CD154 in addition to the MFIR of CD154 (in parentheses) is indicated for each histogram. Data are from a single representative patient sample. (b) The percentage of CD19-positive B cells and CD3-positive T cells expressing CD154 are also plotted. Error bars represent standard deviations (n = 7).

human CD154 on lymphoma B cells. Since both human and murine CD154 were driven by the heterologous CMV promoter/enhancer, had similar Kozak sequences, and lacked 3Ј non-translated sequences that could mitigate transcript stability, the differential expression of these transgenes is unlikely due to differences in transcriptional regulation or translation initiation. In addition, infection of HeLa cells with either Ad-hCD154 or Ad-CD154 resulted in similar, high-level expression of the CD154 on the plasma membrane at the same MOI ratio. Thus, the failure of hCD154 cannot be due to differences between adenovirus preparations or in the sensitivity of our detection methods. Still, lymphoma B cells failed to express the human CD154 even at MOI ratios that were sufficient to allow for high-level expression of the murine homologue. These results suggest that a post-transcriptional block precludes expression of human CD154 on the surface of the lymphoma B cells. The inability to effect expression of the human CD154 on lymphoma B cells contrasts with other reports indicating that normal or neoplastic B cells can express cell-surface 27–30 Figure 4 Time-course of CD154 transgene expression. 5 × 106 CD154. Conceivably, the lymphoma B cells examined in lymphoma cells were infected with Ad-CD154 at an MOI ratio of 300. this study represented cells at different states of B cell acti- Aliquots were sampled after 1 to 7 days infection and CD19-positive vation or differentiation, in which they were precluded from B cells were examined for CD154 transgene expression. Both the per- expressing this surface molecule. Alternatively, there may be centage of cells expressing CD154 (line graph) and the MFIR (bar heterogeneity among neoplastic B cells in their ability to graph) of CD154 are shown. The data are from one representative patient sample. express cell-surface CD154. Consistent with this, we find that neoplastic plasma cells from patients with can be made to express human CD154 following infection with Ad-hCD154 infection (data not shown).

Leukemia T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1455

Figure 5 Lymphoma B cell phenotype following adenovirus infection. 2.5 × 106 lymphoma cells were infected either Ad-CD154 (upper histograms) or Ad-hCD154 (lower histograms) at an MOI ratio of 300. Following 2 days infection, CD19-positive cells were analyzed for expression of a panel of surface antigens (indicated above histograms). The surface antigen expression for Ad-CD154 or Ad-hCD154 infected cells (hatched histograms) and Ad-LacZ infected cells (open histograms) are plotted. In addition, histograms of cells stained with fluorochrome- conjugated isotype-control antibody (shaded histograms) are shown. The phenotypic data from one representative patient sample are depicted.

Figure 6 Cell count and viability time course following Ad-CD154 infection. 5 × 106 lymphoma cells were either non-infected (squares) or infected with adenovirus at an MOI of 300 (circles). Aliquots were sampled from 1 to 7 days and both cell concentration (a) and viability (b) were determined. Error bars represent standard deviations (n = 4).

Human CD154 may exist as a soluble molecule31,32 that can bind cells that bear its cognate receptor, namely CD40. Consistent with this, we noted that CD40-bearing leukemia B cells could be stained with mAbs specific for human CD154 after a brief incubation at 4°C with non-purified lysates of Ad- hCD154-infected 293 cells that contained soluble CD154 (data not shown). Conceivably, the surface expression of human CD154 on B cells noted by others similarly may have been secondary to the cytophilic-binding of soluble CD154 to CD40. However, because soluble human CD154 protein is not present in the highly purified preparations of Ad- hCD154 used in these studies, we were not able to detect any CD154 protein other than that made by infected cells. Nevertheless, we found that follicle center lymphoma B cells could be transduced with Ad-CD154, allowing for high- level expression of murine CD154. Because murine CD154 is effective in ligating human CD40,33 such cells were also induced to express important immune-costimulatory mol- 7 ecules, such as CD80 and CD54, that are upregulated follow- Figure 7 Allogeneic mixed lymphocyte reaction. 10 lymphoma ing ligation of CD40.7 cells were either non-infected (mock) or infected with Ad-CD154, Ad- hCD154, or Ad-LacZ at an MOI ratio of 300 for 2 days followed by Expression of such accessory molecules was associated with mitomycin C treatment. 105 lymphoma cells were then plated with an enhanced capacity of Ad-CD154-infected cells to stimulate 105 purified normal donor T cells in 96-well U-bottom plates as indi- allogeneic or autologous blood T cells in mixed lymphocyte cated. (a) Following 5days culture, cells were pulsed with 3H-thymid- reactions. In contrast, follicle center lymphoma B cells trans- ine and proliferation determined 12–16 h later. (b) Culture supernatant ␥ duced with adenovirus encoding the irrelevant protein, ␤- aliquots were taken after 2 days culture and tested for IFN- secretion galactosidase, or Ad-hCD154 did not undergo such pheno- by ELISA. The data are from one representative patient. Error bars rep- resent standard deviations of triplicate measurements. typic changes and could not stimulate even normal allogeneic

Leukemia T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1456 exogenous , such as interleukin-4.18,19 As such, it appears that CD40 signaling does not independently stimulate the growth of follicle center lymphoma B cells. The capacity of Ad-CD154-infected lymphoma cells to stimulate autologous T cells in MLR indicates that such cells may be useful for inducing autologous immune recognition of lymphoma cells. Interestingly, high proportions of nodal T cells were induced to express activation antigens, such as HLA-DR or CD71, following infection of the lymphoma cells with Ad-CD154 (Figure 8). In contrast, there were no detect- able increases in expression of such activation antigens on the T cells of lymphoma cells infected with Ad-LacZ. This suggests that many of the T cells infiltrating the lymphoma tissue may have the capacity to react against lymphoma-associated anti- gens and/or adenovirus antigens provided that the lymphoma cells first are modified into proficient antigen presenting cells. As such, infection of lymphoma cells ex vivo with Ad- CD154 could be used in a strategy similar to that used recently in patients with CLL.14 Infusion of Ad-CD154 infected autologous CLL cells resulted in acute production of Th-1-type cytokines, such as interleukin-12 and interferon-gamma, increases in the absolute numbers of blood T cells, and reductions in numbers of circulating leukemia cells and lymph node sizes. Moreover, infusion of autologous Ad-CD154- infected CLL cells increased the numbers of T cells capable of recognizing both infected and non-infected leukemia cells in vitro. Conceivably, this strategy may also be used for patients with follicle center lymphoma, allowing for induction of a host anti-lymphoma immune response against putative lymphoma-associated antigens. Further studies are required to evaluate the safety and efficacy of this approach in follicle center B cell lymphoma, potentially allowing for a novel form Figure 8 Autologous T cell responses. (a) 107 lymphoma cells of active immune of patients with this disease. were either non-infected (mock) or infected with Ad-CD154 or Ad- LacZ at an MOI ratio of 300 for 2 days and then treated with mitomy- cin C. 105 lymphoma cells were plated with 105 purified autologous T cells in 96-well U-bottom plates, as indicated. After 5days in cul- Acknowledgements ture, 3H-thymidine was added to each of the wells and the amount 3 of H-thymidine incorporated into DNA was determined 12–16 h These studies funded in part by PO1-CA81534, PO1- later. Error bars represent standard deviations of triplicate measure- ments from one representative patient sample. (b) 2.5 × 106 lymphoma CA34233, and PO1-CA72103 from the NIH cells were infected with either Ad-CD154 (hatched histograms) or Ad- LacZ (open histograms) at a MOI ratio of 300 for 2 days. CD3-express- ing T cells were then analyzed for CD71 and HLA-DR expression References (indicated above histograms) by FACS. Isotype-control mAb stained cells are also plotted (shaded histograms). The data are from one rep- resentative patient sample. 1 Banchereau J, Bazan F, Blanchard D, Briere F, Galizzi JP, van Kooten C, Liu YJ, Rousset F, Saeland S. The CD40 antigen and its ligand. Annu Rev Immunol 1994; 12: 881–922. 2 Laman JD, Claassen E, Noelle RJ. Functions of CD40 and its T cells in the MLR. The failure of Ad-hCD154-infected lym- ligand, gp39 (CD40L). Crit Rev Immunol 1996; 16: 59–108. phoma cells to upregulate expression of co-stimulatory mol- 3 Cantwell M, Hua T, Pappas J, Kipps TJ. Acquired CD40-ligand ecules or to stimulate T cell responses is consistent with their deficiency in chronic lymphocytic leukemia. Nat Med 1997; 3: 984–989. inability to express hCD154 following infection with Ad- 4 Castle BE, Kishimoto K, Stearns C, Brown ML, Kehry MR. Regu- hCD154. lation of expression of the ligand for CD40 on T helper lympho- Despite the fact that Ad-CD154 infected lymphoma B cells cytes. J Immunol 1993; 151: 1777–1788. had phenotypic changes similar to those of CD40-activated 5Hermann P, Blanchard D, de Saint-Vis B, Fossiez F, Gaillard C, B cells, we did not observe any changes in lymphoma cell Vanbervliet B, Briere F, Banchereau J, Galizzi JP. Expression of a proliferation or survival following infection with Ad-CD154. 32-kDa ligand for the CD40 antigen on activated human T lym- phocytes. Eur J Immunol 1993; 23: 961–964. This result contrasts with those of others that found either 6 Roy M, Waldschmidt T, Aruffo A, Ledbetter JA, Noelle RJ. The 34 35 stimulation or inhibition of lymphoma cell growth follow- regulation of the expression of gp39, the CD40 ligand, on normal ing CD40 ligation. Such differences may be secondary to and cloned CD4+ T cells. J Immunol 1993; 151: 2497–2510. methods used to induce CD40 signaling. On the other hand, 7 Ranheim EA, Kipps TJ. Activated T cells induce expression of it could reflect heterogeneity among the different types of B /BB1 on normal or leukemic B cells through a CD40-dependent cell in their response to CD40 signaling. Consist- signal. J Exp Med 1993; 177: 925–935. 8 Ranheim EA, Kipps TJ. Tumor factor-alpha facilitates ent with this, prior studies by others on follicle center lym- induction of CD80 (B7–1) and CD54 on human B cells by acti- phoma also noted no change in growth or survival of the neo- vated T cells: complex regulation by IL-4, IL-10, and CD40L. Cell plastic B cells following CD40 ligation in the absence of Immunol 1995; 161: 226–235.

Leukemia T cell activation following infection of lymphoma B cells with Ad-CD154 MJ Cantwell et al 1457 9 Grewal IS, Flavell RA. CD40 and CD154 in cell-mediated immu- 23 Blanche F, Cameron B, Barbot A, Ferrero L, Guillemin T, Guyot nity. Ann Rev Immunol 1998; 16: 111–135. S, Somarriba S, Bisch D. An improved anion-exchange HPLC 10 Mackey MF, Barth RJ Jr, Noelle RJ. The role of CD40/CD154 inter- method for the detection and purification of adenoviral particles. actions in the priming, differentiation, and effector function of Gene Therapy 2000; 7: 1055–1062. helper and cytotoxic T cells. J Leukoc Biol 1998; 63: 418–428. 24 Huyghe BG, Liu X, Sutjipto S, Sugarman BJ, Horn MT, Shepard 11 Aruffo A, Farrington M, Hollenbaugh D, Li X, Milatovich A, Non- HM, Scandella CJ, Shabram P. Purification of a type 5recombi- oyama S, Bajorath J, Grosmaire LS, Stenkamp R, Neubauer M. The nant adenovirus encoding human by column chromato- CD40 ligand, gp39, is defective in activated T cells from patients graphy. Hum Gene Ther 1995; 6: 1403–1416. with X-linked hyper-IgM syndrome. Cell 1993; 72: 291–300. 25Shabram PW, Giroux DD, Goudreau AM, Gregory RJ, Horn MT, 12 Korthauer U, Graf D, Mages HW, Briere F, Padayachee M, Mal- Huyghe BG, Liu X, Nunnally MH, Sugarman BJ, Sutjipto S. Ana- colm S, Ugazio AG, Notarangelo LD, Levinsky RJ, Kroczek RA. lytical anion-exchange HPLC of recombinant type-5adenoviral Defective expression of T-cell CD40 ligand causes X-linked particles. Hum Gene Ther 1997; 8: 453–465. immunodeficiency with hyper-IgM. Nature 1993; 361: 539–541. 26 Becker TC, Noel RJ, Coats WS, Go´mez-Foix AM, Alam T, Gerard 13 Kato K, Cantwell MJ, Sharma S, Kipps TJ. Gene transfer of CD40- RD, Newgard CB. Use of recombinant adenovirus for metabolic ligand induces autologous immune recognition of chronic lym- engineering of mammalian cells. MethCell Biol 1994; 43: 161– phocytic leukemia B cells. J Clin Invest 1998; 101: 1133–1141. 189. 14 Wierda WG, Cantwell MJ, Woods SJ, Rassenti LZ, Prussak CE, 27 Blossom S, Chu EB, Weigle WO, Gilbert KM. CD40 ligand Kipps TJ. CD40-ligand (CD154) gene therapy for chronic lympho- expressed on B cells in the BXSB mouse model of systemic lupus cytic leukemia. Blood 2000; 96: 2917–2924. erythematosus. J Immunol 1997; 159: 4580–4586. 15Dilloo D, Brown M, Roskrow M, Zhong W, Holladay M, Holden 28 Clodi K, Asgary Z, Zhao S, Kliche KO, Cabanillas F, Andreeff M, W, Brenner M. CD40 ligand induces an antileukemia immune Younes A. Coexpression of CD40 and CD40 ligand in B-cell lym- response in vivo. Blood 1997; 90: 1927–1933. phoma cells. Br J Haematol 1998; 103: 270–275. 16 Kikuchi T, Crystal RG. Anti-tumor immunity induced by in vivo 29 Desai-Mehta A, Lu L, Ramsey-Goldman R, Datta SK. Hyperexpres- adenovirus vector-mediated expression of CD40 ligand in tumor sion of CD40 ligand by B and T cells in human lupus and its role in pathogenic autoantibody production. J Clin Invest 1996; 97: cells. Hum Gene Ther 1999; 10: 1375–1387. 2063–2073. 17 Nakajima A, Kodama T, Morimoto S, Azuma M, Takeda K, Osh- 30 Schattner EJ, Mascarenhas J, Reyfman I, Koshy M, Woo C, Fried- ima H, Yoshino S, Yagita H, Okumura K. Antitumor effect of CD40 man SM, Crow MK. Chronic lymphocytic leukemia B cells can ligand: elicitation of local and systemic antitumor responses by express CD40 ligand and demonstrate T-cell type costimulatory IL-12 and B7. J Immunol 1998; 161: 1901–1907. capacity. Blood 1998; 91: 2689–2697. 18 Schultze JL, Cardoso AA, Freeman GJ, Seamon MJ, Daley J, Pinkus 31 Kato K, Santana-Sahagu´n E, Rassenti LZ, Weisman MH, Tamura GS, Gribben JG, Nadler LM. Follicular lymphomas can be induced N, Kobayashi S, Hashimoto H, Kipps TJ. The soluble CD40 ligand to present alloantigen efficiently: a conceptual model to improve sCD154 in systemic lupus erythematosus. J Clin Invest 1999; 104: their tumor immunogenicity (published erratum appears in Proc 947–955. Natl Acad Sci USA 1995; 92: 10818). Proc Natl Acad Sci USA 32 Younes A, Snell V, Consoli U, Clodi K, Zhao S, Palmer JL, Thomas 1995; 92: 8200–8204. EK, Armitage RJ, Andreeff M. Elevated levels of biologically active 19 Schultze JL, Seamon MJ, Michalak S, Gribben JG, Nadler LM. soluble CD40 ligand in the serum of patients with chronic lym- Autologous tumor infiltrating T cells cytotoxic for follicular lym- phocytic leukaemia. Br J Haematol 1998; 100: 135–141. phoma cells can be expanded in vitro. Blood 1997; 89: 3806– 33 Lane P, Traunecker A, Hubele S, Inui S, Lanzavecchia A, Gray D. 3816. Activated human T cells express a ligand for the human B cell- 20 Hsu FJ, Benike C, Fagnoni F, Liles TM, Czerwinski D, Taidi B, associated antigen CD40 which participates in T cell-dependent Engleman EG, Levy R. Vaccination of patients with B-cell lym- activation of B . Eur J Immunol 1992; 22: 2573– phoma using autologous antigen-pulsed dendritic cells. Nat Med 2578. 1996; 2: 52–58. 34 Andersen NS, Larsen JK, Christiansen J, Pedersen LB, Christo- 21 Renschler MF, Wada HG, Fok KS, Levy R. B-lymphoma cells are phersen NS, Geisler CH, Jurlander J. Soluble CD40 ligand induces activated by peptide ligands of the antigen binding receptor or by selective proliferation of lymphoma cells in primary mantle cell anti-idiotypic antibody to induce extracellular acidification. Can- lymphoma cell cultures. Blood 2000; 96: 2219–2225. cer Res 1995; 55: 5642–5647. 35Funakoshi S, Longo DL, Beckwith M, Conley DK, Tsarfaty G, Tsar- 22 Cantwell MJ, Sharma S, Friedmann T, Kipps TJ. Adenovirus vector faty I, Armitage RJ, Fanslow WC, Spriggs MK, Murphy WJ. Inhi- infection of chronic lymphocytic leukemia B cells. Blood 1996; bition of human B-cell lymphoma growth by CD40 stimulation. 88: 4676–4683. Blood 1994; 83: 2787–2794.

Leukemia