Use of adenoviruses encoding CD40L or IL-2 against B cell lymphoma
Article first published online: 20 MAY 2004
Copyright © 2004 Wiley-Liss, Inc.
International Journal of Cancer
Volume 111, Issue 6, pages 910–920, 10 October 2004
How to Cite
Meziane, E. K., Bhattacharyya, T., Armstrong, A. C., Qian, C., Hawkins, R. E., Stern, P. L. and Dermime, S. (2004), Use of adenoviruses encoding CD40L or IL-2 against B cell lymphoma. Int. J. Cancer, 111: 910–920. doi: 10.1002/ijc.20332
- Issue published online: 3 AUG 2004
- Article first published online: 20 MAY 2004
- Manuscript Accepted: 25 FEB 2004
- Manuscript Received: 13 NOV 2003
Some B cell lymphomas lack important costimulatory properties that could prevent them from being used as cell based vaccines. Infection of A20 B lymphoma cells with a replication-defective adenovirus encoding murine (m) CD40L, but not mIL-2, produces an antigen presentation phenotype with upregulation of MHC Class I/II, induction of B7-1/2 molecules and production of MIL-12 and MIP-1α. Subcutaneous vaccination with irradiated Ad-mCD40L-infected- or Ad-mIL-2-infected-A20 cells generated A20-specific CD8+ T cell responses and cross reactive A20 Ig antibodies. Only vaccination with Ad-mCD40L-infected A20 cells produced a significant delay in tumor growth and long-term survival (p = 0.0039). Stronger protective immunity to A20 challenge was generated by intravenous priming with A20 cells infected with Ad-mCD40L, Ad-mIL-2 or their combination followed by a boost immunization with A20 cells activated with syngeneic fibroblasts expressing CD40L. Compared to Ad-LacZ-infected A20 priming, the combination priming was most effective followed by Ad-mCD40L and Ad-mIL-2 (p = 0.0027, p = 0.0027, p = 0.0163 respectively). Significant A20-specific CD8+ T cell-mediated cytotoxicity was only demonstrated in splenocytes from these groups of vaccinated animals. By contrast, ELISPOT assay of splenocytes from all A20 prime/boosted vaccinated groups demonstrated increases in γ-interferon release by T cells elicited by in vitro stimulation either with A20 cells or another syngeneic 2PK-3 lymphoma, indicating the presence of cross reactive immunity. Similarly anti-A20 immunoglobulin antibodies generated after vaccination were not necessarily A20 idiotype-specific. Direct therapy of pre-established tumors was achieved with the combination of Ad-mCD40L and Ad-mIL-2 given at Days 4 and 8 at the tumor site with a significant long-term survival of 85% of tumor-bearing mice (p = 0.0001). Our study strongly supports the use of Ad-CD40L and Ad-IL-2 combination therapy for the treatment of patients with B cell lymphoma. © 2004 Wiley-Liss, Inc.
Although patients with low-grade follicular B cell non-Hodgkin's lymphoma (NHL) respond well to treatment with chemotherapy and radiotherapy, they are rarely cured due to the relapses that occur after a period of months or years with loss of response to the treatment.1, 2 High dose chemo-radiotherapy can induce longer remissions3 but unfortunately carry a high early mortality due to their substantial toxicity.4 The need for new treatment modalities has driven evaluation of various immunotherapies including recombinant human interferon-α (IFN-α),5 monoclonal antibody targeting (CD20, CD19, CD22)6 or anti-tumor vaccines.7
The idiotype (Id), a unique surface immunoglobulin expressed by B cell lymphoma, is considered a suitable specific tumor associated antigen for vaccination.8 The Id is a weakly immunogenic self-antigen and immuno-modulation by fusion to more immunogenic antigens has been shown to enhance its immunogenicity. This can be achieved by fusion to the granulocyte macrophage colony stimulating factor (GM-CSF),9 the MCP-3 chemokine,10 the constant region of a foreign immunoglobulin (Ig)11, 12, 13 or to the non-toxic C fragment (FrC) of tetanus toxin.14 All these methods provide T cell help to break tolerance against the Id antigen.
The uses of these ‘adjuvant-like’ elements have been shown to improve the Id vaccine efficacy in animal models. Although the use of Id as a lymphoma-specific antigen to elicit anti-tumor immunity is an attractive target,15 the broader use of Id-based vaccines has been hampered by being patient-specific, in addition to the possibility of immune selection of negative variants, which can provide for immune escape.16
The use of whole tumor cells as anti-tumor cell-based vaccines is an alternative approach for the treatment of B cell NHL that could potentially provide several tumor-associated antigens as targets for the immune system in addition to the Id itself. B cell NHL lack the expression of co-stimulatory molecules such as B7-1 and B7-2, and presentation of such tumor antigen-derived peptides in the context of MHC molecules can result in T cell anergy.17 Immunomodulation of B cell lymphoma with costimulatory molecules such as B7-1, B7-2 and 4-1BBL18 or cytokines such as GM-CSF, interleukin-2 (IL-2) and IL-419 is one way to potentially overcome such anergy. The fact that neither B7-1 nor B7-2 alone can confer full immunogenicity to B cell lymphoma has stimulated investigations of other costimulatory molecules that may result in highly immunogenic lymphoma vaccines.
The interaction between CD40 and CD40 ligand (CD40L) and the provision of IL-2 are considered the main mechanisms by which CD4+ T cells provide help for priming naive CD8+ T cells and expanding and sustaining CD8+ cytotoxic T lymphocyte (CTL) responses, respectively.20 CD40L, which is expressed by activated CD4+ T cells, has been shown to enhance the ability of B cells to act as antigen presenting cells (APCs).21 Interaction of CD40L with the CD40 molecule expressed by B cell lymphoma can lead to expression of a number of costimulatory and adhesion molecules such as B7-1, B7-2, ICAM-1 and LFA-3 that are important for T cell activation.22 Expressing CD40L or IL-2 in the lymphoma cells may lead to effective generation of CD4+ T cell help. Recent studies have shown that immuno-modulation of the A20 B cell lymphoma with CD40L can result in upregulation of costimulatory molecules, T cell activation and protection against tumor challenge.23 Other studies have focused on transfer of genes for different cytokines and more importantly IL-2.24 It has been shown, for example, that injection of tumor cells genetically engineered to secrete IL-2 can result in the stimulation of CTL responses against parental tumor.25 Secretion of low levels of IL-2 by tumor cells has also been shown to enhance their immunogenicity and induce a long-term protective immune response against tumor challenge.26
Although encouraging results were obtained by immunization using adenovirus-modified tumor cells or other methods such as DNA or viral vaccinations, the levels of specific immunity induced by these vectors has generally been insufficient to achieve a complete protection against tumor challenge. Improvement has been obtained using prime-boost vaccination strategies that minimize anti-vector responses leading to enhanced efficacy of anti-cancer vaccines against many tumor antigens27 including CEA28 and B cell leukaemia.29
Our study attempts to modify the A20 B cell lymphoma using a replication-defective adenoviral vector expressing either mouse CD40L (Ad-mCD40L) or IL-2 (Ad-mIL-2). We also investigate the immunogenicity and tumorigenicity of modified lymphoma cells. The APC activation status of the A20 cells was assessed by cell surface expression of MHC and B7 molecules and production of murine IL-12 and MIP-1α. Vaccination protocols using adenoviral-modified tumor cells (Ad-mCD40L or Ad-mIL-2 or their combination) with or without a boost immunization using A20 cells activated by fibroblast cells expressing mCD40L, were analyzed for humoral, cell mediated, tumor protective and therapeutic immunity.
MATERIAL AND METHODS
Female BALB/c (H-2d) mice, 6–8 weeks of age were purchased from Harlan (United Kingdom). All procedures were carried out according to approved protocols and in accordance with recommendations for proper care and use of laboratory animal approved by the British Home Office.
A20, a BALB/c B cell lymphoma line originally derived from a spontaneous reticulum cell neoplasm30 was obtained from the American Type Culture Collection (ATCC, Manassas, VA). The cells grow in complete RPMI-1640 (Sigma, Dorset, UK) medium (CM) (10% FCS), penicillin (50 U/ml), streptomycin (50 U/ml), L-glutamine (2 mM), and 2-mercaptoethanol (50 mM) at 37°C in a 5% CO2 atmosphere. The 2PK-3 cell line, a BALB/c B cell lymphoma (ATCC), was grown in DMEM media (Sigma, St. Louis, MO) supplemented with 10% FCS, glutamine and antibiotics as above. The NK sensitive YAC-1 lymphoma cell line (ECACC) was grown in complete RPMI media (CM). The 293 T cell line (ATCC) is a 293 human embryonic kidney cell line (ATCC) transformed with human adenovirus type 5 (Ad5) DNA. The Cre8 cell line that produces high levels of Cre recombinase was kindly provided by S. Hardy (Somatix, Alameda, CA).31 The 293, 293T and Cre8 cell lines were grown in DMEM media supplemented with 10% FCS. XS52, a dendritic cell line (DC) established from a new-born BALB/c mouse epidermis was grown in complete RPMI-1640 medium supplemented with recombinant mGM-CSF (20 U/ml) (R&D Systems, Abingdon, UK) and supernatants collected from a syngeneic fibroblast cell line NS47 (5% v/v) that grows in CM. Cell lines XS52 and NS47 were provided by A. Takashima (University of Texas South-Western Medical Centre, Dallas, TX). XS52-A20-A1 cell line, an XS52-derived dendritic cell line retrovirally transduced to express the A20 scFv, was generated in our laboratory.11 This cell line was grown in the same medium as XS52.
Ad-mCD40L and Ad-LacZ replication-defective recombinant adenoviruses with E1, E3 deletions were propagated in 293 cells and purified on double CsCl gradients as described previously.32 Ad-mIL-2 was generated through Cre-lox recombination as described previously.31 Briefly, Cre8 cells were cotransfected with Ψ5-helper virus and a shuttle vector pAdlox-mIL-2. Recombinant adenoviruses were passaged twice in Cre8 cells to reduce contamination of residual Ψ5 virus. After a good cytopathic effect, cell lysates were generated by freeze-thawing the cells 3 times and supernatants were purified by CsCl gradients. The purified virus was titered using cytopathic effect assay and the mIL-2 production was assessed using an IL-2 enzyme linked immuno-sorbent assay (ELISA) kit (DIACLONE, Besançon, France).
Adenoviral infection of A20 cells
To establish the optimal dosing of adenovirus, A20 cells were infected with Ad-mCD40L, Ad-mIL-2 or Ad-LacZ adenoviruses at different multiplicity of infection (MOI) using Lipofectamine reagent (Invitrogen Life Technologies, Paisley, UK) in combination with the adenovirus to improve the infection efficiency. The Lipofectamine was used in all infections with the different adenoviruses (including Ad-GFP and Ad-LacZ) to exclude non-specific effects of liposomes. The virus was mixed with Lipofectamine in serum-free media and incubated at room temperature for 15 min before adding A20 cells. Complete media was added to the cells after 24 hr. Expression of mCD40L was assessed by flow cytometry and found to be dose-dependent. We have shown that infection of A20 cells with Ad-mCD40L at 1,000 MOI + Lipofectamine resulted in at least 74% mCD40L expression compared to 22% expression using the Ad-mCD40L alone. Supernatants from A20 cells infected with Ad-mIL-2 at 1,000 MOI + Lipofectamine were analyzed by ELISA for mIL-2 production (DIACLONE) and the cells were able to produce at least 275 pg/ml mIL-2. A control adenovirus encoding LacZ (E. coli β-galactosidase) was also used and LacZ expression was assessed using an X-Gal staining assay kit (GTS, San Diego, CA) according to the manufacturer's instructions. Ad-LacZ infecting 98% of A20 cells did not produce any detectable levels of IL-2. Supernatants from infected cells were also analyzed for mIL-12 and MIP-1α production using mIL-12 (P 40) and MIP-1α ELISA kits (R&D, Quantikine M).
A20 cells were stained with a panel of anti-mouse R-PE conjugated mAbs against cell surface molecules including MHC Class I, MHC Class II, CD40L (CD154) (BD Pharmingen, Heidelberg, Germany), B7-1 (CD80) and B7-2 (CD86) (Serotec, Oxford, UK). Matching isotype controls were used. Immunostained cells were analyzed on a FACscan machine (BD Biosciences, Oxford, UK) and PCLysis software.
Mixed lymphocyte reaction
Irradiated APCs (A20, A20-Ad-GFP, A20-Ad-mCD40L [12,000 Rad]) were incubated with purified CD4+ T cells (Mouse CD4 Dynabead and DETACHaBEAD kits, Dynal Biotech, Oslo, Norway) obtained from C57 BL/6 mice. The diluted APCs were plated out in triplicate in a 96 U-bottomed plate at 0.5 × 105 cells/well and the T cells were added at different APC:T cells ratios. T cells + PHA was used as a control for the purity of T cells. The plates were labeled with 3H-thymidine after 3 days incubation at 37°C and left incubating for an additional 18 hr. The cells were harvested (Filtermate Cell Harvester, Packard Instruments, Berkshire, UK) and thymidine uptake was determined.
Retroviral transduction of NS47 cells with mCD40L gene
NS47 cell line was retrovirally transduced with rKat-mCD40L. The retrovirus was produced by a transient CaPO4 transfection of 293T cells with rKat-mCD40L and pKat plasmid DNA that contains the retroviral packaging genes. Supernatant containing retrovirus was harvested after 48 and 72 hr, filtered through 0.2 μm filters. One milliliter of viral media was mixed with 8 μg/ml Polybrene (Sigma) and added immediately to the NS47 cells plated at 1 × 105 cells/well in a 6-well plate the night before. The cells were incubated with the retroviral media for 8 hr then in fresh media overnight. This process was repeated twice.
Cloning of NS47-mCD40L cell line
The generated NS47-mCD40L cell line was analyzed by flow cytometry for mCD40L expression and 10% of the highly positive cells were sorted 3 times before cell cloning. The sorted cells were plated at very low concentration (0.5 cells/well) in flat-bottomed 96-well plates in 200 μl CM media. The growing clones were kept in culture and analyzed for mCD40L expression by flow cytometry. One of the highly positive clones (NS47-mCD40L-CL5) was selected for further studies.
Activation of A20 cells using NS47-mCD40L-CL5
To activate the A20 cells, irradiated NS47-mCD40L-CL5 cells (5,000 Rad) were plated out and allowed to adhere. A20 cells were added at a ratio of (1:1) and incubated for 48 hr. The cells were analyzed for B7-1/2 expression and the supernatants were analyzed for mIL-12 and MIP-1α production.
Production of A20 protein (IgG2a)
The IgG2a immunoglobulin secreted by the A20 cells was concentrated from A20 cell culture supernatants by precipitation using 314 g/L ammonium sulphate (NH4SO4). The protein was affinity purified using a HiTrap Protein G HP column (Amersham Pharmacia Biotech, Little Chalfont, Buckinghamshire, UK) and was analyzed by Western blotting and SDS-PAGE.
Vaccination and tumor challenge
Groups of 10 mice/group received subcutaneous (s.c.) injections of 2 × 106 irradiated (8,000 Rad) A20 cells or A20 cells modified with adenoviruses. At Day 21, 3 mice from each group were sacrificed; spleens and blood were collected for in vitro assays. At Day 35 the remaining animals were challenged s.c. with 106 A20 cells in the opposite flank. Tumor size was measured 3 times per week (tumor volume = (width)2 × (length)/2) and animals were killed when tumors reach 1.24 cm.3
Prime boost strategy
Groups of mice (n = 16) were primed (i.v.) with 106 irradiated A20 cells, A20 modified with adenoviruses (Ad-mCD40L and/or Ad-mIL-2) or A20 cells activated with NS47-CD40L. Seven days later 8 mice/group were boosted (i.v.) with 106 irradiated NS47-CD40L-activated A20 cells. At Day 15 post-priming, 3 mice from the prime-boosted groups and the primed groups were sacrificed for in vitro studies, and at Day 23 all the remaining mice were challenged (s.c.) with 106 A20 cells. Tumor growth and survival were followed.
Groups of mice (n = 7) were injected (s.c.) with 0.5 × 106 A20 cells. Four days later mice were injected at the site of the tumor with PBS, 109 p.f.u. Ad-LacZ, Ad-mCD40L or Ad-mIL-2 followed by another intratumoral (i.m.) injection of the same viruses at Day 8 using only 108 p.f.u. of virus. Mice were followed for survival.
Survival in protective and active treatments was analyzed with standard Kaplan-Meier plots33 using SPSS program. Student's t-test was used to analyze the differences between test groups. A value of p < 0.05 was considered significant.
Splenocytes were stimulated in vitro for 5 days with irradiated A20 cells that have been activated with NS47-mCD40L-CL5 and analyzed for lytic activity against a panel of target cell lines using a standard 51Cr-release assay. When blocking the CD4 or CD8, 20 μg/ml of the αCD4 or αCD8 antibodies (BD Pharmingen) were incubated with the T cells 30 min before adding target cells.
γ-IFN enzyme-linked immuno spot assay (ELISPOT)
T cells were assessed for γ-IFN using an ELISPOT kit (IDS, Tyne and Wear, UK), according to the manufacturer's instructions, using 96-well hydrophobic, high protein binding Immobilon-P membrane plate (Millipore, Gloucester, UK) ELIspot plates.
Detection of anti-A20 immune response in sera by ELISA
To assess the humoral immune responses in vaccinated mice, ELISA assays were carried out to detect the presence of anti-A20 (IgG2a) in the pooled sera from 3-mice/group. The plates were coated with A20 protein (whole Ig), hIgG1 (Sigma) or monoclonal mIgG2a (Serotec, Oxford, UK). Immune responses were detected using HRP-conjugated goat anti-mouse Isotype IgG1 and IgG2b detecting antibodies (AMS Biotechnology, Abington, UK) and the plates were developed using BM blue POD substrate (Roche, Mannheim, Germany).
Adenovirus-induced immune activation of A20 lymphoma cells
To assess the activation of the A20 cells after infection with Ad-mCD40L, supernatants from infected cells were analyzed by ELISA for the presence of mIL-12 and MIP-1α, which is important in producing inflammation, attracting effector cells and activating APCs such as dendritic cells (DCs). High concentrations of both mIL-12 and MIP-1α were detected in supernatants from cells infected with Ad-mCD40L (>1,500 pg/ml for mIL-12 [P 40]; >240 pg/ml for MIP-1α) with no detectable levels in supernatants obtained from non-infected cells or cells infected with Ad-GFP (Fig. 1) or Ad-mIL-2 (data not shown). After 48 hr, the infected cells were analyzed by flow cytometry for the expression of cell surface molecules important for antigen presentation. A representative experiment of more than 10 experiments is shown in Figure 2 where signaling through CD40L resulted in significant upregulation of MHC Class I and Class II molecules and complete induction of the costimulatory molecules B7-1 and B7-2. Infection of A20 cells with Ad-GFP (99% positive) or Ad-mIL-2 resulted neither in upregulation of MHC molecules nor in induction of costimulatory molecules (data not shown).
Ad-mCD40L- or Ad-GFP-infected A20 cells were evaluated for their ability to act as stimulator cells in an allogeneic mixed lymphocyte reaction (MLR). Adenovirally-transduced A20 cells were irradiated and co-cultured with purified CD4+ T cells from allogeneic C57 BL/6 mice. As shown in Figure 3, a high proliferative activity was obtained after incubating T cells with Ad-mCD40L-activated A20 cells but not Ad-GFP-infected cells. To assure that the purified CD4+ T cells used in our assay are free of residual syngeneic antigen presenting cells, we cultured these T cells with the addition of PHA. Negligible proliferation was recorded. Data are shown as mean of triplicates ± SE. It is evident that Ad-mCD40L-treated A20 are potent APCs.
Vaccination with adenovirus-infected A20 cells
To investigate the use of A20 modified cells as a vaccine, mice were immunized s.c. with 2 × 106 irradiated A20-Ad-mCD40L, A20-Ad-mIL-2, A20-Ad-LacZ or non-infected A20 cells. Sera and splenocytes were obtained 21 days after vaccination. As shown in Figure 4, ELISA activity vs. the specific A20 IgG2a showed higher titers of IgG1 (>1:400) and IgG2b (1:360) in mice vaccinated with either A20-Ad-mCD40L or A20-Ad-mIL-2 than A20-Ad-LacZ (1:200) or non-infected A20 mice (1:200). The sera also react with an irrelevant murine IgG2a monoclonal antibody but not mIgG1, ovalbumin or human IgG1 (data not shown). This suggests that at least a part of the reactivity is not anti-idiotype specific.
To study CTL responses, spleen cells from vaccinated mice were restimulated in vitro with A20 lymphoma cells but no specific cytotoxicity could be generated. We used A20 cells co-cultured previously with a syngeneic fibroblast cell line expressing CD40L, which induced upregulation of MHC molecules, induction of B7-1 (37%) and B7-2 (51%) costimulatory molecules and secretion of IL-12 and MIP-1α cytokines (at least 50 pg/ml). Spleen cells from both A20-Ad-mCD40L and A20-Ad-mIL-2 groups showed a specific lytic activity against the A20 cells, with negligible lysis against the syngeneic 2PK-3 lymphoma cell line (Fig. 5). Spleen cells from the control groups (A20-Ad-LacZ or non-infected A20) failed to lyse any of the target cells. The specific lytic activity vs. A20 cells was inhibited by anti-CD8 (but not anti CD4), blocking antibodies and was clearly independent of NK activity for the negligible lysis of the NK-sensitive YAC-1 lymphoma cells.
Immunized mice were challenged s.c. with lethal doses of viable A20 cells on Day 35 post-vaccination as in previous studies.11 Treatment with the A20-Ad-mCD40L delayed significantly the development of tumors compared to the A20-Ad-mIL-2, A20-Ad-LacZ and non-infected A20 treated groups (Fig. 6). Significant long-term survival advantage of more than 45% for up to 140 days was only seen in mice vaccinated with the A20-Ad-mCD40L when compared to mice vaccinated with non-modified A20 cells (p = 0.0039) and A20-Ad-LacZ (p = 0.0085). All other survivals were not significant (A20-Ad-mIL-2vs. A20: p = 0.1929, A20-Ad-mIL-2vs. A20-Ad-LacZ: p = 0.2350, A20-Ad-LacZvs. A20: p = 0.7421, A20-Ad-mCD40Lvs. A20-Ad-mIL-2: p = 0.3563).
Investigations of immune responses using prime-boost strategy
As demonstrated in the previous experiment, s.c. vaccination of mice with Ad-mCD40L- or Ad-mIL-2-infected A20 cells generated anti-A20 immunity although only the former induced statistically significant protection against tumor challenge. Intravenous vaccination was used in subsequent experiments because this may allow the modified A20 tumor cells better interaction with T cells and other APCs for antigen presentation and subsequent efficient T cell activation and expansion by local IL-2 production. We sought to optimize the efficacy of anti-tumor protection, specific antibody and T cell immunity by investigating the use of A20 lymphoma cells treated with Ad-mCD40L, Ad-mIL-2 and their combination as a priming vaccination as well as a boost immunization with NS47-mCD40L-activated A20 cells. All treatments were administered by i.v. route of delivery and with an earlier tumor challenge. At Day 15, 3 mice from each primed and prime-boosted group were sacrificed for immunological assays. At Day 22, 5 mice from each group were challenged with 106 of viable A20 cells and followed for survival.
A single immunization with A20 cells treated with Ad-mIL-2, Ad-mCD40L or their combination elicited very little activity vs. A20 immunoglobulin in ELISA (Fig. 7). After the boost with NS47-CD40L-activated A20 cells after priming with A20, A20-Ad-mCD40L, A20-Ad-mIL-2 or A20-Ad-mCD40L + Ad-mIL-2, however, there was significantly increased anti-A20 Ig activity. Sera from mice pre-immunized with the A20 cells modified with (Ad-mCD40L + Ad-mIL-2) and boosted with the activated A20 cells showed the greatest increase in both anti-A20 immunoglobulin IgG1 (Fig. 7a) and IgG2b (Fig. 7b) immune responses compared to other treatments. Antibody titers for the prime-boost were 1/400 compared to 1/100 for the prime alone (data not shown). Mice primed with NS47-CD40L-activated A20 cells, however, did not show any significant increase in generated immune responses after boosting. The pattern and magnitude of the A20 Ig responses are also seen vs. an irrelevant IgG2a ELISA (data not shown).
Consistent with the generation of non-A20-specific cross-reactive immunity are the results from γ-interferon ELISPOT assays (Fig. 8). Fresh splenocytes obtained from primed and prime-boosted mice 15 days after the first vaccination were incubated with A20 cells, the syngeneic 2PK-3 B cell lymphoma or YAC-1 T cell lymphoma for 24 hr. γ-IFN-producing T cells were clearly detectable in splenocytes obtained from mice that received any A20 priming plus a boost with the NS47-CD40L-activated A20 cells. A smaller priming alone effect was evident in mice treated with NS47-CD40L-activated A20 cells, Ad-mIL-2 or combined-adenovirus treated-A20 cells. This does not seem to be a specific A20 response because 2PK-3 B lymphoma cells were equally able to stimulate this response whereas YAC-1 cells could not.
By contrast it was possible to measure the highest specific CTL activity against A20 targets after in vitro restimulation with NS47-CD40L-activated A20 cells of splenocytes from animals that received A20-Ad-mCD40L priming and boosting with activated A20 cells or the adenovirus-combination A20 priming with or without the boost (Fig. 9). These groups of effectors also showed increased lysis of the XS52-A20-A1 target cells expressing the A20 idiotype11 compared to the XS52 DC line implicating at least a proportion of this CTL activity was specific for the A20 idiotype (data not shown). In no group of effectors (except from the A20-Ad-mIL-2/boosted group for unclear reasons) was there any significant effect of immunization on cytotoxicity vs. the NK-sensitive YAC-1 targets or the irrelevant lymphoma 2PK-3. The A20-specific cytotoxicity was shown to be blocked by CD8 but not CD4-specific antibodies (data not shown).
To investigate the efficacy of this prime-boost strategy on tumor protection, 5 mice from each immunized group were challenged s.c. with lethal doses (106) of viable A20 cells on Day 22 post-vaccination. As shown in Figure 10, all mice that were primed and boosted with NS47-CD40L-CL-5-activated A20 cells died at early stages of the experiment (p = 0.1258). Mice primed with A20-Ad-mIL-2 and boosted with activated A20 cells exhibited 20% significant long-term survival compared to mice primed with non-modified A20 cells (p = 0.0163) and mice primed with A20-Ad-mCD40L showed 40% significant long-term survival (p = 0.0027). Priming with A20 cells infected with both Ad-mCD40L and Ad-mIL-2 adenoviruses followed by a boost with the activated A20 cells resulted in 80% significant long-term survival compared to mice primed with non-modified A20 cells (p = 0.0027) and to mice primed with Ad-mIL-2 (p = 0.0316). This group, however, did not show a statistically significant difference compared to the group primed with Ad-CD40L (p = 0.1578), which showed 40% survival
Active therapy by intratumoral injection of adenoviruses
After confirming the effect of Ad-mCD40L and Ad-mIL-2 modification of A20 cells on inducing protective immunity and reducing the tumorigenicity of the viable cells, we evaluated the therapeutic effect of these recombinant adenoviruses on pre-established tumors. The A20 tumors were established in groups of mice followed by injections of Ad-mCD40L, Ad-mIL-2 or the combination of both adenoviruses into the tumor site at Days 4 and 8 after tumor challenge. None of the mice treated with Ad-mCD40L+Ad-mIL-2 developed tumors and only one mouse died due to unknown reasons. All mice from the control groups treated with either PBS or Ad-LacZ developed tumors, however, and died at early stages of the experiment. Significant long-term survival of mice treated with either Ad-mCD40L (70%; p = 0.0008) or Ad-mIL-2(25%; p = 0.0001) is shown in Figure 11. The combination therapy of both adenoviruses to treat pre-established tumors has resulted in 85% significant long-term survival compared to control groups (p = 0.0001).
Other investigations to study the effect of adenoviral modification on viable tumor cells exhibited that the Ad-mCD40L- or Ad-mIL-2-infection of A20 cells significantly reduced their tumorigenicity with the best survival of cells obtained with the combination of the 2 adenoviruses in the treatment (data not shown).
B cell lymphoma express a specific Ig Id that can act as a tumor-specific target but the tumor cells themselves lack the appropriate signals to productively activate specific T cells. Indeed, presentation of tumor antigen epitopes in the absence of co-stimulation may result in T cell anergy due to lack of IL-2 production by stimulated T cells.17 The challenge is to modify the tumor cells or idiotypic antigen to enhance the antigen presentation capacity in an efficient way. We have recently investigated the ability of recombinant adenovirus encoding a lymphoma (A20) Id gene fused to a human IgG1 Fc helper determinant that can induce an anti-idiotypic immune response capable of protecting syngeneic mice from tumor challenge.11 The use of idiotype vaccines is limited, however, by the fact that Id is individual lymphoma-specific. Modulation of B cells lymphoma and their use as cell-based vaccines has been investigated by previous studies and the use of lymphoma cells genetically engineered to produce GM-CSF19 has been successful but failed in the generation of any humoral immunity. The aim of our study was to test whether modification of lymphoma cells by either expression of IL-2 or CD40L using replication-defective adenoviral vectors can produce improved vaccine efficacy. The advantage of this approach is the potential for immunization with multiple target antigens that derive from the many genetic changes that accumulate during the natural history of a B cell lymphoma.
The adenoviral infection of A20 cells required Lipofectamine probably due to low or no expression of alpha V integrins34 or Coxsackie-adenovirus receptor (CAR)35 on these cells. This resulted in high frequency transduction without deleterious effects on cell growth. The A20 cells express CD40 and infection with Ad-mCD40L mimicked the CD40-CD40L interaction on APCs whereby there is an upregulation of MHC and costimulatory molecules B7.1/B7.2 and production of mIL-12 and MIP-1α.32, 36 The A20 cells, which expressed B7.1/B7.2 are thus available to interact with CD28 on T cells and deliver the obligate second signal for specific T cell activation and expansion. IL-12 can promote the development of Th1 CD4+ T cells, which facilitate maturation of CD8+ CTLs whereas MIP-1α promotes CD8+ T cell migration and their interaction with DCs.37, 38 Ad-mCD40L-activated A20 B cells are potent APCs as judged from the strong specific proliferative activity in an MLR. Infecting the cells with Ad-mIL-2, which resulted in mIL-2 production as a means to compensate for that normally produced during B7/CD28 interaction, had no effect on the surface phenotype of either MHC or costimulatory molecules. Local production of mIL-2 by the A20 cells in close proximity with the encountered T cells may provide for an efficient A20-specific T cell activation and expansion.
Subcutaneous vaccination of mice with irradiated adenovirally-modified A20 cells has generated anti-A20 Ig IgG1 and IgG2b antibodies and A20-specific CD8+ T cells from A20-Ad-mCD40L and A20-Ad-mIL-2 vaccinated mice but not A20-Ad-LacZ or non-modified A20 cells vaccinated mice. Equal IgG1 antibody titers were recorded for each of these 2 groups, however, the IgG2b titers were higher in the A20-Ad-mCD40L vaccinated group. In addition the IgG1 titers were about 4-fold higher than that of the IgG2b. This antibody specificity cannot be exclusively attributed to the variable region of the A20 IgG2a, because it was also able to react against a normal mouse IgG2a. This may explain the recent failure of Levy group23 in detecting anti-A20 idiotypic antibodies in animals vaccinated with the A20-Ad-mCD40L cells using a recombinant single chain Fv A20 protein that only expresses the heavy and light chain variable regions. Tumor protection after tumor challenge has yielded a delay in tumor growth in mice vaccinated with A20-Ad-mCD40L with significantly improved survival compared to the A20 and A20-Ad-LacZ control groups. Vaccination with the A20-Ad-mIL-2 had no significant effect on growth or survival suggesting that the Ad-mCD40L-modified lymphoma cells are superior to the A20-Ad-mIL-2, however, the available assays for immunogenicity do not show any gross differences in CTL responses. There were, however, subtle differences in the humoral immune recognition of the A20 IgG2a for mice vaccinated with mIL-2 or mCD40L modified A20 cells. The contributions of cell mediated or humoral immunity in protecting vs. lymphoma challenge have been documented.11, 12
Further investigations to improve the efficacy of the generated vaccines and enhance the induced humoral and cellular immune responses were carried out using a prime-boost strategy27, 28, 29 combining the adenovirus-modified cell-based vaccines with the NS47-CD40L-activated A20 tumor cells, which we have shown to be activated by the induction of B7 molecules and production of IL-12 and MIP-1α. This heterologous prime-boost regimen has further enhanced humoral and cellular immune responses. IgG1 and IgG2b anti-A20 immunoglobulin immune responses were efficiently boosted by the NS47-CD40L-activated A20 cells in mice primed with either A20-Ad-mCD40L, A20-Ad-mIL-2 or A20-Ad-mCD40L + Ad-mIL-2, with the highest increase recorded when mice pre-immunized with A20-Ad-mCD40L + Ad-mIL-2. Mice primed with NS47-CD40L-activated A20 cells did not show any increase in antibody production. This may be attributed to the homologous boosting. It has already been reported that heterologous prime-boost regimens with 2 vaccines generate stronger CD4+ T cell responses than homologous boosting.39 All generated cellular immune responses were efficiently enhanced after boosting with the NS47-CD40L-activated A20 cells including mice primed with non-modified A20 cells as confirmed by the ELISPOT assays. Interestingly, the generated T cells did not only react against the A20 tumor cells but also against the syngeneic 2PK-3 lymphoma cell line as demonstrated by γ-IFN production. It is important to note that T cell simulation by 2PK-3 cell line were more efficient where pre-immunization of mice involved modification of tumor cells with Ad-mCD40L either alone or in combination with Ad-mIL-2. It is possible that vaccination with A20 cells modified with Ad-mCD40L generates immune responses against shared lymphoma antigens restricted by the same MHC molecules. Although 2PK-3 cell line was able to stimulate T cells to produce high levels of γ-IFN as shown by the ELISPOT assay, cytotoxic T cells failed to lyse these cells. It seems that γ-IFN release by specific recognition of antigens shared by the 2 lymphoma cell lines is probably mediated by CD4+ T helper cells. Inhibition studies using CD4 blocking Ab should clarify this issue although it is possible that the ELISPOT assay is just more sensitive than a standard 51Cr release assay. By contrast, enhanced A20-specific CD8+ cytotoxic T cell responses were detected in groups of mice pre-immunized with A20-Ad-mCD40L+Ad-mIL-2 followed by activated A20 boosting. The priming with A20 cells infected with the combination of both Ad-mCD40L and Ad-mIL-2 viruses followed by a boost with activated A20 cells has the most significant effect in tumor protection.
The influence of CD40L and IL-2 combination is consistently better in both active therapy and protective models with CD40L usually better than IL-2. Such activation of CD40+ tumor cells through CD40L not only enhances the APC function of the transfected cells resulting in a direct priming of tumor-specific T cells, but also activates other host APCs expressing CD40 molecule, such as DCs and macrophages.32 Through the CD40-CD40L interaction there is enhanced costimulation and cytokine production. This cross priming activity potentiates anti-tumor immunity by the recruitment and activation of effector CD8+ T cells. IL-2, on the other hand, can act by several mechanisms, including activation of NK cells, induction of lymphokine-activated killer (LAK) cells,40 or support of Ag-dependent activation or expansion of CD8 T cells.41 Thus, provision of IL-2 should potentially provide therapeutic benefit as it compensates for that needed to maintain and expand CD8 T cells and support prolonged anti-tumor CTL responses that do not depend upon CD4 T cell help.42
Studies clearly implicate lymphoma-specific CTL in tumor immunity although the role of other types of effector γ-IFN T cells or anti-Ig antibodies are not fully understood.23 The combination of Ad-mCD40L and Ad-mIL-2 for lymphoma modification is superior to either virus in protection and active therapy studies. The use of such vaccines in humans has already begun and initial promising results suggest that the combination of Ad-hCD40L and Ad-hIL-2 to transduce B cell non-Hodgkin's lymphoma cells can result in enhanced T cell activation and induction of cytotoxic immune response in vivo directed against unmodified tumor cells.43 The only concern for this vaccination strategy is the high dose adenovirus that is needed to infect these B lymphoma cells. Much higher doses of adenovirus (as high as 15,000 m.o.i.) were used in previous studies and did not have adverse effects on human lymphoma cells.43 In addition, other delivery systems such as Herpes simplex virus (HSV)44 and adeno-associated virus (AAV)45 vectors should also be investigated. Our studies show a significant improved tumor protection by boosting with NS47-CD40L-activated A20 cells. This opens further opportunity for optimal vaccination protocols avoiding stimulation of anti-vector immunity.
- 8Role of anti-idiotype vaccines in the treatment of human follicular lymphoma. Expert Rev Anticancer Ther 2001; 1: 89–96..