| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Departments of
* Medical Oncology and
Immunology, Cancer Research Campaign, Paterson Institute of Cancer Research, Christie Hospital National Health Service Trust, Manchester, United Kingdom
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
release against a syngeneic
dendritic cell line transduced using a retroviral vector to express the
A20 scFv idiotype (XS52.A1.A20). Importantly, both T cell lines lysed
the A20 lymphoma cells. An immunodominant H-2Kd-restricted
CD8+ T cell peptide, DYWGQGTEL (A20[106–114]), was
identified as a naturally occurring A20 scFv epitope. A single
immunization with Ad.A20hFc but not Ad.A20 provided protection in
>40% of animals challenged with a lethal dose of the A20 tumor line
and was more effective, in this model, than a previously optimized
plasmid vaccine. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Preparing individual protein vaccines for patients is time consuming and expensive. Advances in molecular biological techniques have allowed the development of rapid methods to isolate tumor-derived V genes (5) and assemble them as a single chain variable fragment (scFv),3 with the two chains separated by a linker peptide to allow the scFv to fold properly.
Initial attempts at using DNA vaccines encoding the scFv alone induced only low levels of anti-idiotypic immunity (6, 7), necessitating the use of methods to enhance the immunogenicity of the vaccines. Immunization with DNA encoding a fusion protein of the idiotype to a foreign Ig constant region and GM-CSF was more effective, with the vaccine efficacy comparable to an equivalent protein vaccine (8). Fusion of the idiotypic DNA to fragment C of tetanus toxoid to provide Th cell epitopes also appeared to improve the vaccine, inducing strong protection against tumor challenge (9).
Although anti-idiotypic cytotoxic T cells have been generated against human B cell idiotypes (10, 11, 12), in mice the main effector mechanism appears to be the induction of anti-idiotypic Abs (13). In murine models where protective immunity is dependent on the existence of CD8+ or CD4+ T cells (14, 15, 16, 17, 18), the existence of a cytotoxic anti-idiotypic T cell has yet to be demonstrated. T cell-mediated immunity may be particularly important against tumor cell variants that no longer express cell surface idiotype, and are therefore resistant to anti-idiotypic Abs, but are able to continue to produce idiotypic peptides to present to T cells. Until it is clear exactly which effector mechanisms are needed to provide therapeutic immunity, it is prudent to investigate the best methods of inducing both cell-mediated and humoral immunity.
Recombinant adenoviral vectors allow more efficient and reliable gene transfer than naked DNA vaccines. In a number of animal models adenoviral vectors have been shown to induce both humoral and cellular immune responses to a variety of Ags, including tumor Ags (19, 20). Adenoviral vectors owe part of their efficacy to their innate ability to infect eukaryotic cells. They are also highly immunogenic, and expression of the viral genes may act as an immunological adjuvant.
It has recently been demonstrated that vaccination of mice with idiotype-encoding adenoviruses leads to an anti-idiotypic Ab response and protection from tumor challenge (21), both dependent on inclusion of a human Ig constant region. In this report we have investigated the ability of recombinant adenoviral vectors to induce both anti-idiotypic humoral and cellular immunity. In an animal model we compare the efficacy of adenoviral vectors expressing the idiotype alone, or the idiotype fused to human Fc (hFc), at inducing anti-idiotypic immunity. Both humoral responses and cytotoxic CD8+ T cells were generated and an immunogenic idiotypic T cell epitope was defined. The cytotoxic T cells were able to specifically lyse both target cells expressing the lymphoma idiotype and the B cell lymphoma line itself. Finally, we demonstrate that anti-idiotypic immunity induced by the adenoviral vector expressing Id-hFc fusion is an efficient means of inducing tumor protection and is more effective, in this model, than a previously optimized plasmid vaccine (9).
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Six- to 8-wk-old female BALB/c mice were bred and housed under specific pathogen-free conditions. All procedures were conducted in accordance with British Home Office guidelines. A20, a BALB/c B cell lymphoma line originally derived from a spontaneous reticulum cell neoplasm (22), was obtained from American Type Culture Collection (ATCC, Manassas, VA). The cell line was cultured in complete RPMI 1640 medium (10% FCS, 2 mM glutamine, 100 U/ml penicillin, 100 U/ml streptomycin, and 50 µM 2-ME). XS52, a dendritic cell (DC) line established from newborn BALB/c mouse epidermis, was grown in complete RPMI 1640 medium supplemented with recombinant murine GM-CSF at 20 U/ml (R&D Systems, Minneapolis, MN) and supernatants were collected from the syngeneic fibroblast cell line NS47 (5% v/v). Both cell lines were kindly provided by A. Takashima (University of Texas Southwestern Medical Center, Dallas, TX) (23). The 293T tumor cell line (ATCC), the HeLa cell line (ATCC), and the Cre8 cell line (kindly provided by S. Hardy, Somatix, Alameda, CA) were grown in DMEM supplemented with 10% FCS, 2 mM glutamine, 100 U/ml penicillin, and 100 U/ml streptomycin. All media were obtained from Sigma (Dorset, U.K.), and all supplements were from Life Technologies (Paisley, U.K.).
Identification of the idiotype from the A20 cell line and generation of A20 single chain (A20 scFv) constructs
H and L chain variable sequences from the A20 cell line were obtained by RT-PCR amplification from total A20-derived RNA using primers based on published H chain genes for VH (24), and with V gene family-specific primers for VK. Following cloning and sequencing, the idiotypic sequences were identified and assembled to encode a scFv with a flexible linker sequence, in a two-step procedure, as previously described (5). Appropriate cloning sites were added to the 5' end of the A20 scFv to allow subcloning of the A20 scFv into eukaryotic and prokaryotic expression vectors. For expression from adenoviral, plasmid eukaryotic, and retroviral expression vectors, the leader sequence from the human oncostatin-M gene was used (25).
The assembled A20 scFv was cloned into the plasmid pVAC2.FrC as a SfiI-NotI fragment, allowing expression of A20 scFv as a fusion protein with the fragment C (FrC) of tetanus toxin (pVAC2.A20.FrC) (9).
The A20 scFv was also cloned into the adenoviral shuttle vector, pAdlox (kindly provided by Dr. S. Hardy) as a ClaI-NotI fragment for production of recombinant adenoviruses. Two A20-scFv-expressing vectors were generated. One contained only the A20 scFv (pAd.A20). The second vector (pAd.A20.hFc) contained the A20 scFv linked to the Fc fragment of human IgG1, which itself was generated as a NotI-XbaI fragment by RT-PCR from total human PBMC-derived RNA. These vectors allow the expression of the A20 scFv alone (pAd.A20) or as a fusion protein with hFc (pAd.A20.hFc). A third adenoviral vector, containing the scFv B1.8 (25), specific for the haptens 4-hydroxy-3-nitrophenyl acetyl/4-hydrox-3-5-iodo-phenylacetyl, also as a fusion with hFc (pAd.B1.8hFc), was generated for use as a control scFv.
For retroviral expression, the A20 scFv and the B1.8 scFv were also cloned as ClaI-NotI fragments into the vector rKat (26), into which an internal ribosome entry site (IRES) and the cDNA of enhanced green fluorescent protein (GFP) had previously been cloned (rKat.A20.IRES.GFP and rKat.B1.8.IRES.GFP) (33).
For bacterial expression, the A20 scFv was cloned into a pUC119-based vector (5) with the pel b leader sequence removed by restriction enzyme digestion (NcoI-HindIII) to allow intracellular localization of the A20 scFv. The expressed A20 scFv protein contained a c-myc and a hexahistidine tag to allow for detection of the scFv by Western blotting and purification using a metal affinity column.
Generation of recombinant adenovirus expressing the A20 scFv: Ad.A20 and Ad.A20hFc
Replication defective E1-E3-deleted 
5 adenovirus was kindly
provided by Dr. S. Hardy. Recombinant viruses were constructed as
previously described (27) by cotransfection of CRE8 cells
with 5 viral DNA and SfiI-digested pAdlox.A20 or
pAdlox.A20.hFc. Recombinant viruses were passaged twice in CRE8 cells
to reduce contamination of residual 5 adenovirus. CRE8 cell lysates
were prepared by freeze-thawing the cells three times, and recombinant
adenoviral particles were purified on CsCl gradients and titered in
cytopathic effect assays.
Confirmation of A20 scFv expression from the plasmid pVAC2.A20FrC and the adenovirus Ad.A20hFc
To confirm the expression of the A20 scFv from pVAC2.A20.FrC, 293T cells were transiently transfected with pVAC2.A20.FrC using a standard CaPO4 transfection protocol. After 48 h medium from the transfected cells was collected, and the cells were harvested and lysed in lysis buffer (PBS/0.5% Nonidet P-40/0.2 mM PMSF (Sigma)). The media and cell lysates were then analyzed for the presence of the scFv fusion protein A20.FrC by Western blotting, with the A20.FrC fusion protein successfully identified in the medium from pVAC2.A20FrC-transfected cells but not in the untransfected cells.
To confirm successful expression from the adenoviral vector Ad.A20hFc, HeLa cells were infected with AdA20hFc (multiplicity of infection, 10). After 3 days the supernatants were collected from the infected cells and protein expression was confirmed by Western blot analysis for the expression of hFc.
Western blot analysis
Protein samples were fractionated by reducing SDS-PAGE (15%) and electroblotted onto nitrocellulose. The blot was blocked for at least 1 h at room temperature with blocking buffer (5% milk powder in PBST (PBS/0.1% Tween 20)), and probed for 1 h with the detecting Ab (HRP-conjugated anti-hFc (Sigma)) or polyclonal antitetanus sera from immune individual or 9E10 anti-myc monoclonal, diluted 1/20 in blocking buffer. The blot was washed three times in PBST and developed using an ECL detection system (Amersham Pharmacia Biotech, Little Chalfont, Buckinghamshire, U.K.) or incubated with an HRP-conjugated anti-hFc or HRP-conjugated goat anti-mouse (Sigma), as appropriate, before developing.
Production of A20 Ig and A20 scFv protein
Idiotypic IgG2a, secreted from A20 cells, was concentrated from tissue culture supernatant by precipitation with 50% saturated ammonium sulfate and purified on a HiTrap Protein G HP column (Amersham Pharmacia Biotech), according to the manufacturer’s instructions. Purified A20 Ig was analyzed by Western blotting, as described.
For bacterial expression of A20 scFv, the pUC119.A20 plasmid was transformed into Escherichia coli strain BL21, grown from single colonies to an OD600 of 0.9, and induced with 1 mmol isopropyl-D-thiogalactopyranoside (BDH Laboratory Supplies, Poole, Dorset, U.K.) for 18 h at 30°C. Bacterial pellets were sonicated for three cycles of 30 s in the presence of 8 M urea (BDH Laboratory Supplies), followed by ultracentrifugation at 100,000 x g. The protein was refolded by resuspending in alkaline dialysis buffer (50 mM KH2PO4, 50 mM NaCl, 1 mM EDTA (pH 10.7)) for 0.5 h, followed by titration to pH 8. The protein was dialyzed overnight in PBS, concentrated in a Vivapore10/20 column (Sartorius, Epsom, U.K.), and analyzed by Western blotting.
Retroviral transduction of XS52 cell line
Recombinant retrovirus was produced by transient CaPO4 transfection of 293T cells with rKat.A20.IRES.GFP or rKat.GFP as a negative control (Cell GeneSys, Foster City, CA), and the plasmid pKat, which encodes the retroviral packaging genes. Retrovirus-containing supernatants were harvested after 48 and 72 h, filtered through 0.45-µm pores, and used immediately.
For retroviral transduction 1 x 105 XS52 cells/well were plated in a six-well plate and allowed to adhere. A total of 1 ml XS52 medium was then replaced with 1 ml of viral supernatant containing 2 µg/ml polybrene (Sigma), and the cells were incubated overnight at 37°C/5%CO2. The retroviral supernatant was removed, the cells were incubated with fresh medium for 8 h, and the retroviral transduction process was then repeated.
Cloning of XS52 A20 scFv cell line
XS52 cells retrovirally transduced with rKat.A20.IRES.GFP were plated at decreasing concentrations from 5 cells/well, with a feeder layer of 104 irradiated NS47 cells/well (7000 rad), to a total volume of 200 µl XS52 medium in U-bottom 96-well tissue culture plates. The XS52 clones were initially screened for GFP expression using a fluorescent microscope and then analyzed by flow cytometry. One clone (XS52.A1.A20), which showed high expression of GFP as well as DC markers, was selected for use in T cell assays. Confirmation of A20 scFv transcription by XS52.A1.A20 was achieved by PCR amplification of total cellular cDNA using oligonucleotide primers specific for the A20VH gene.
XS52 cells were also retrovirally transduced with rKat.B1.8.IRES.GFP (XS52.B1.8).
Flow cytometric analysis of cell surface markers
XS52.rkatA20IRESGFP clones were stained with a panel of R-PE-conjugated mAbs against surface markers characteristic of DC including H-2Kd, I-Ad, CD54 (BD PharMingen, Heidelberg, Germany), CD40, CD80, and CD86 (Serotec, Oxford, U.K.). Isotype-matched mAbs were used as controls. Immunostained cells were analyzed on a FACScan flow cytometer (BD Biosciences, Oxford, U.K.) using PCLysys software (BD Biosciences).
Adenoviral transduction of XS52 line
XS52 cells were plated at 2 x 106 cells in 2 ml XS52 medium. The cells were allowed to adhere and then infected with Ad.A20, Ad.A20hFc, or Ad.GFP at a multiplicity of infection of 250. After 18 h medium was replaced and the cells were used for T cell stimulation following a 72-h incubation. Successful adenoviral transduction was confirmed by the expression of the reporter gene GFP.
Immunization of mice with adenoviral vectors encoding A20 scFv and generation of T cell lines against A20 scFv
Preliminary experiments using Ad.B1.8hFc established the optimal route and dose of adenoviral vaccine. To determine whether the administration of Ad.A20 constructs could induce Ag-specific immune responses, BALB/c mice were immunized with 109 PFU per mouse of Ad.A20 or Ad.A20hFc in 100 µl PBS, with 50 µl given s.c. and 50 µl injected i.m.
To evaluate specific immune responses to the A20 scFv, blood was collected at day 21 postinjection by cardiac puncture under terminal anesthesia and the spleens were obtained. Splenocytes were pooled according to the vaccine used and cocultured (2 x 106 cells/ml) with irradiated (3000 rad) adenovirus-infected XS52 cells (2 x 105 cells/ml) in a six-well tissue culture plate. T cells were restimulated weekly, initially with adenovirus-infected XS52 cells. When the T cell lines were found to specifically proliferate to XS52.A1.A20, an A20 scFv retrovirally transduced XS52 cell line, they were maintained by restimulation with these cells at weekly intervals, at an effector:stimulator ratio of 10:1, in the presence of recombinant human IL-2 (20 U/ml) (PeproTech, London, U.K.).
The T cell lines were cloned by plating T cells at 1, 5, or 10 cells/well in a 96-well U-bottom plate in the presence of 2 x 104 irradiated XS52.A1.A20 (3000 rad) and 2 x 104 irradiated syngeneic splenocytes (3000 rad) as feeder cells, in a final volume of 200 µl. Recombinant human IL-2 was added at a concentration of 100 U/ml. Following a second stimulation with XS52.A1.A20, successfully expanded clones were screened for their ability to lyse XS52 or XS52.A1.A20 cells by transferring 50 µl of resuspended medium containing the T cells to plates containing 51Cr-labeled cells. Clones able to specifically lyse XS52.A1.A20 cells were further expanded by restimulation with XS52.A1.A20 cells and retested in titrated 51Cr release assays for their ability to lyse XS52 cells loaded with the A20-derived peptides.
Peptides
Potential H-2Kd binding peptides, derived
from the A20 scFv idiotype (Fig. 1
A) were chosen according to
an epitope prediction program developed by Parker et al.
(28). This program did not yield any peptides with
significant binding affinities for H-2Dd
molecules. The five A20 scFv-derived peptides with the highest
predicted binding affinities for H-2Kd molecules
(Fig. 1B) were synthesized and purified by HPLC (Alta
Bioscience, University of Birmingham, Birmingham, U.K.).
|
Cytolytic activity of the T cell lines was tested by 51Cr release assay. Target cells consisted of XS52 cells, XS52 cells infected with recombinant adenovirus, A20 cells, XS52.A1.A20 cells, XS52.B1.8 cells, and, in later assays, XS52 pulsed with A20 peptides. Target cells were labeled with 51Cr (100 µCi) for 1.5 h and then plated in V-bottom 96-well plates at 1 x 103 cells/well. In the case of XS52 peptide-pulsed cells, peptide was added to the cells with the 51Cr at a final concentration of 20 µg/ml. Purified anti-CD8 and anti-CD4 Abs (BD PharMingen) were added in some instances to the T cells 30 min before the addition of the target cells at a concentration of 20 µg/ml. Purified isotype controls were used at the same concentration. Effector cells were added at various E:T ratios to a final volume of 150 µl, with each ratio represented in triplicate. Maximum 51Cr release was determined from supernatants of cells that were lysed with 100 µl PBS/2% Tween 20. Spontaneous release was obtained by incubating target cells in medium alone. The plates were incubated at 37°C/5% CO2 for 4 h. A total of 90 µl of supernatant was collected from each well and counted in a Topcount scintillation counter (Packard Instrument, Berkshire, U.K.). The percentage of lysis was calculated as follows: % lysis = [(specific release - spontaneous release)/maximum release - spontaneous release)] x 100.
Proliferation assay
Stimulator cells consisting of irradiated XS52, XS52.A1.A20, XS52.B1.8 (3000 rad), and A20 cells (7000 rad) were plated out in triplicate in U-bottom 96-well plates at a concentration of 5 x 104 cells/well. T cells (5 x 105 cells/well) were added to a total volume of 200 µl. Purified anti-CD8 and anti-CD4 Abs (BD PharMingen) were added in some instances to the T cells 30 min before the addition of the target cells at a concentration of 20 µg/ml. Purified isotype controls were used at the same concentration. Plates were incubated at 37°C/5% CO2 for 4 days. During the last 18 h of incubation cultures were pulsed with 1.0 µCi/well [3H]thymidine. Cells were then harvested using a Packard Cell Harvester onto unifilter plates. Cell-associated radioactivity was measured by scintillation counting (Topcount scintillation counter).
ELISPOT
T cells were also assessed in an ELISPOT assay using an IFN-
ELISPOT kit (IDS, Tyne and Wear, U.K.). T cells were plated out in
duplicate in the wells of a 96-well hydrophobic, high protein-binding
Immobilon-P membrane plate (Millipore, Gloucester, U.K.) coated
overnight with IFN- capture Ab, diluted in carbonate-bicarbonate
buffer (pH 9.6). Irradiated XS52.A1.A20, XS52, and A20 cells were used
as stimulators. After 48 h at 37°C/5%
CO2, the plates were washed vigorously with
PBS/0.05% Tween 20 to remove the cells. IFN- production by T cells
was detected by a 3-h incubation with biotinylated detection Ab diluted
in PBS/1% FCS, followed by a 1.5-h incubation with alkaline
phosphatase-conjugated streptavidin. Spots were visualized by the
addition of 5-bromo-4-chloro-3-indolyl phosphate/nitroblue tetrazolium
substrate buffer, and the number of spots was quantified using
computer-assisted video image analysis (Image-Pro Plus; Media
Cybernetics, Silver Spring, MD).
ELISA
Vaccine-induced anti-A20 scFv and anti-A20 Ig Ab responses were assessed by ELISA. The A20 proteins, diluted in borate buffer (100 mM boric acid, 150 mM NaCl, pH 8.5) were coated to the wells overnight at 4°C. Purified monoclonal mouse IgG2a (BD PharMingen) was used to check the specificity of the immune response. Dilutions of sera were made in PBS/2% milk protein with bound mouse Ig detected by HRP-conjugated goat anti-mouse Fc (Sigma). Isotype-specific detectors were used for the quantification of anti-idiotypic Abs of the IgG1, IgG2a, and IgG2b subclasses (AMS Biotechnology, Abingdon, U.K.). The plates were developed using BM Blue POD substrate (Roche, Mannheim, Germany).
Tumor challenge
Two groups of BALB/c mice were immunized once on day 0 with the recombinant adenoviruses Ad.A20 and Ad.A20hFc (109 PFU in 100 µl PBS split s.c. and i.m.). One group of mice was immunized with 50 µg pVAC2A20 i.m. on days 0, 21, and 42 (based on previous experiments with the plasmid (9)). Two control groups were used, one vaccinated with Ad.B1.8hFc (109 PFU in PBS) and one with PBS alone. On day 42 1 x 106 A20 tumor cells were injected s.c. The mice were monitored for up to 130 days.
Statistical analysis
Quantitative results are expressed as mean ± SD. The log rank test was used to analyze the statistical differences between the vaccine groups.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Idiotypic sequences identified from the A20 lymphoma cell line
were subcloned as a scFv into various expression cassettes (Fig. 1
A). To generate recombinant idiotype-encoding
adenoviral vectors for these studies the A20 scFv was subcloned into
the adenoviral expression vector, pAdlox. A second plasmid,
pAdloxA20hFc, was also constructed. This directs the expression of the
A20 scFv as a fusion to the Fc fragment of IgG1. The xenogeneic Fc
fragment was included as the result of studies elsewhere demonstrating
the importance of foreign proteins in the induction of effective
anti-idiotypic immunity (9, 13).
Generation and cloning of an XS52 cell line stably expressing the A20 scFv
To generate an A20 scFv-expressing target for use in T cell assays, XS52 cells were transduced with the retrovirus rKat.A20.IRES.GFP. The presence of the reporter gene GFP allowed confirmation by fluorescent microscopy of successful retroviral transduction and a simple means by which to clone idiotype-expressing XS52 cells. Several clones were established and further analyzed by flow cytometry for the expression of GFP (98%) and DC surface markers (class I (98%), class II (98%), ICAM-I (98%), B7.1 (98%), B7.2 (98%), CD40 (98%)). A representative clone, designated XS52.A1.A20, expressing high levels of DC markers was chosen for future use. Confirmation of the presence of A20 scFv DNA was achieved by PCR analysis using A20 scFv VH primers of total cellular cDNA isolated from XS52.A1.A20 cells.
Optimization of dosing schedule of scFv-expressing adenoviral vectors
To establish the optimal dosing schedule of scFv-expressing recombinant adenovirus, mice were vaccinated with Ad.B1.8hFc. This virus encodes a scFv specific for the hapten 4-hydroxy-3-nitrophenyl acetyl/4-hydrox-3–5-iodo-phenylacetyl, which is expressed as a fusion with hFc. Results of experiments with this vector demonstrated that the optimal dose for anti-B1.8 Ab response was 109 PFU, split i.m./s.c., with a peak Ab response at day 49. Intravenous administration of the virus was less effective and repeat immunization over a short period of time was not beneficial. Nevertheless, after 23 wk of rest, boosting with the same adenoviral construct was effective (data not shown).
Vaccination with idiotype-encoding Ad vectors induces idiotype-specific T cell immunity
Attempts to induce idiotype-specific T cell responses in vitro
using splenocytes from naive mice were unsuccessful. To determine
whether vaccination of mice with a single dose of an idiotype-encoding
adenoviral vector is able to induce an idiotype-specific CTL response
in vivo, BALB/c mice were immunized with 109 PFU
of Ad.A20 or Ad.A20hFc. Twenty-one days later pooled splenocytes from
the immunized mice were cultured in vitro with XS52 cells infected with
the recombinant adenoviral vector AdA20. In vitro T cell expansion
using the A20 cell line was unsuccessful. Cultures were assayed for
cytotoxic activity against a panel of target cells in a standard
51Cr release assay (Fig. 2). T cells from both Ad.A20- and
Ad.A20hFc-vaccinated mice showed specific lytic activity against the
idiotype-expressing cell line XS52.A1.A20, with negligible lysis of the
untransduced XS52 line. The T cell lines were also able to
significantly lyse the A20 lymphoma cell line used to derive the A20
scFv. Both adenoviral vectors induced similar levels of cytotoxic
activity against XS52.A1.A20 and A20 cells. High, nonspecific cytotoxic
activity was seen against adenoviral-transduced XS52 cells. There was
some cytotoxic activity of the T cells against the control scFv,
B1.8.
|
). No proliferative
response was recorded against the A20 tumor line (data not shown).
Following the demonstration of a specific proliferative response to
XS52.A1.A20 line these cells were subsequently used to restimulate and
maintain the T cell lines.
|
To determine the restriction element involved in the specific
anti-idiotypic activity demonstrated by the T cell lines generated
from the mice vaccinated with the A20 scFv adenoviral constructs, both
cytotoxic and proliferative assays were conducted in the presence of
monoclonal blocking Abs against CD4 and CD8 molecules. These Abs are
known to block the interaction between CD8+/class
I MHC and CD4+/class II MHC, leading to
inhibition of proliferative and lytic activity of CD4/CD8-restricted T
cells. Fig. 4A shows that the
specific proliferative activity was completely inhibited by
anti-CD8 Ab, with the proliferative activity unaffected by
anti-CD4 Ab or isotype-matched control Abs. The specific lytic
activity of these T cell lines was also inhibited by the addition of
anti-CD8 Ab (Fig. 4, B and C). Furthermore,
the lytic activity of these T cell lines against the A20 tumor cell
lines was also completely inhibited by the CD8 Ab, demonstrating the
CD8+/class I involvement in this recognition
(Fig. 4C). Phenotypic analysis revealed that
CD8+ T cells accounted for 85% of both T cell
populations, with the remainder of the cells being
CD4+ (data not shown).
|
Five potential H-2Kd-binding peptides (Fig. 1B) were identified from the idiotypic sequence using the
program developed by Parker et al. (28) and individually
analyzed for their recognition by the T cell lines in a
51Cr release assay. As shown in Fig. 5, only the A20[106–114] scFv-derived
peptide, DYWGQGTEL, was recognized by both Ad.A20 (Fig. 5A)
and Ad.A20hFc (Fig. 5B) T cell lines. To test whether T cell
populations against the other four epitopes were present at very low
frequencies in these T cell lines we used an ELISPOT assay to evaluate
each peptide for its ability to stimulate the production of IFN-. As
depicted in Fig. 6, T cell lines derived
from Ad.A20-vaccinated mice secreted IFN- in response to stimulation
with XS52.A1.A20, XS52 cells pulsed with A20[106–114], and A20
cells. No IFN- production was recorded when the T cells were
stimulated with XS52 cells pulsed with the other four peptides. T cell
lines derived from mice vaccinated with Ad.A20.hFc showed a similar
pattern of IFN- production against XS52.A1.A20, A20[106–114], and
A20 cells (data not shown).
|
|
T cells lines from mice vaccinated with Ad.A20 and Ad.A20hFc were
cloned using XS52.A1.A20 cells as stimulators and irradiated
splenocytes as a feeder layer, in the presence of 100 U/ml recombinant
human IL-2. The cells were restimulated on two further occasions and
then screened in a 51Cr release assay, with 14 of
96 clones tested able to lyse XS52.A1.A20 cells and not XS52 cells.
These specific clones were expanded with irradiated XS52.A1.A20 cells
and tested in a second 51Cr release assay for
their ability to specifically lyse XS52 loaded with the A20-derived
peptides. Six successfully expanded T cell clones were able to lyse
XS52 cells loaded with A20[106–114], XS52.A1.A20, and the A20 tumor
cell line, but not XS52 cells loaded with the other four peptides. The
specific lytic activity of one representative T cell clone, T.B1,
derived from mice vaccinated with Ad.A20, is shown in Fig. 7A. This T cell clone was
further analyzed in an ELISPOT assay. Again, specific IFN- release
was demonstrated in response to stimulation by XS52.A1.A20 cells (Fig. 7B) and to stimulation by A20[106–114], but not the other
peptides.
|
Serum was obtained from mice 21 days after vaccination with
the adenoviral vectors and the idiotype-encoding plasmid, pVAC2.A20FrC.
Anti-idiotypic Ab responses were assessed by ELISA using both A20
Ig purified from tissue culture supernatant and A20 scFv purified from
E. coli. Although the idiotype-encoding adenoviral vectors
were equivalent at inducing T cell responses in the vaccinated mice,
there was a clear difference in the ability of the vectors to induce an
anti-idiotypic Ig Ab response, with mice vaccinated with Ad.A20hFc
but not Ad.A20 or pVAC2.A20FrC developing Abs to the purified A20 Ig.
Ad.A20 did induce an Ab response to the recombinant A20 scFv, but at a
much lower titer than that induced by Ad.A20hFc. The isotype profile of
this Ab response was also examined (Fig. 8). Vaccination of mice with Ad.A20hFc
induced anti-A20 Ig IgG1, -2a, and -2b responses, which is
consistent with a Th1- and Th2-type T cell response. In contrast, the
anti-A20 scFv response was predominantly IgG2b (data not
shown).
|
Adenovirus encoding A20hFc protects against challenge with a lethal dose of A20 cells
To determine the capacity of adenoviruses expressing the A20
idiotype to induce protective antitumor immunity, mice were
preimmunized with 109 PFU of adenovirus and then
challenged with a lethal dose of tumor cells. Another group of mice was
vaccinated with the plasmid pVAC2.A20.FrC at a dose previously shown,
in another lymphoma model (9), to protect against tumor
challenge. Mice vaccinated with PBS or Ad.B1.8.hFc, a recombinant
adenovirus expressing a control scFv, were not protected from tumor
challenge, nor were mice preimmunized with the plasmid pVAC2.A20FrC
(Fig. 9). Mice preimmunized with AdA20
had a significant delay in tumor development compared with the mice
vaccinated with AdB1.8hFc (p = 0.002). However,
>40% of the mice pretreated with AdA20hFc were effectively protected
against A20 challenge (p = 0.0003), remaining
tumor free for >250 days. Furthermore, mice pretreated with Ad.A20hFc
had a significant survival benefit to those pretreated with Ad.A20
(p = 0.022).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In this study we have shown that vaccination of mice with a single dose of recombinant adenovirus encoding the A20-derived idiotype alone, or as a fusion to hFc, is capable of inducing potent idiotype-specific CD8+ T cell responses. Only the adenovirus encoding the A20 fused to hFc is able to induce anti-idiotypic Abs able to recognize A20 Ig. Importantly, we were able to show strong idiotype-specific cytotoxic T cell activity; these T cells were also capable of lysing the A20 tumor cells themselves. The majority of murine models examining the use of idiotype vaccines have demonstrated the importance of induction of anti-idiotypic Abs in protecting mice against tumor challenge (13, 29). While T cells have been shown to play a role in protection in murine lymphoma models (14, 15, 16, 17, 18), no animal model to date has demonstrated the induction of idiotype-specific cytotoxic T cells following idiotypic vaccination. One possibility is that some tumor idiotypes do not contain CTL-binding motifs. However, idiotype-specific CTLs have been clearly demonstrated against human idiotypes (10, 11, 12), and in some murine models the tumor protection cannot be fully attributed to an anti-idiotypic Ab response (15, 17, 18), suggesting that T cell-mediated immunity may play a role in tumor eradication. It is possible that the in vitro systems used to expand and identify idiotype-specific T cells in such studies may need to be optimized to demonstrate such an immune response. In our study, identification of idiotype-specific T cells relied on the use of a syngeneic DC line, XS52 retrovirally transduced to express the A20 idiotype. Interestingly, expansion of T cells from vaccinated mice in vitro using the A20 lymphoma cell line itself was not successful, possibly because lymphoma cells are poor APCs in comparison to DC.
T cells from mice vaccinated with adenoviral vectors were able to lyse, with equal efficiency, XS52 cells adenovirally transduced to express the idiotype, a control idiotype, or GFP, reflecting the induction of antiadenoviral-specific CTL response in vaccinated mice (data not shown). There was also some lysis of XS52 cells, retrovirally transduced to express a control scFv, B1.8, although the lytic activity of T cells against the A20 scFv was clearly stronger. This cross-reacting activity is not surprising, as the B1.8 scFv and the A20 scFv have some sequence homology. Indeed, the B1.8 scFv peptide predicted by the Parker computer program to be the strongest H-2Kd binder is identical to A20[106–114] apart from the last two amino acids (DYWGQGTTV). This cross-reactivity does not result in cross-protection; animals vaccinated with Ad.B1.8hFc were not protected from tumor challenge with A20 cells.
An important finding of our study was the identification of an immunodominant class I binding epitope in the idiotypic region of the A20 Ig. Five potential MHC class I (H-2Kd)-binding peptides were identified using a peptide-binding prediction database. Subsequent analysis revealed that both T cell lines from the vaccinated mice were able to recognize XS52 cells pulsed with the A20[106–114] peptide, in both a cytotoxic and an ELISPOT assay. However, the T cell did not recognize XS52 cells pulsed with any of the other four peptides. Analysis of the peptide specificity of the T cell lines by the more sensitive ELISPOT assay took place after some weeks of in vitro culture and restimulation, which may have led to the expansion of the T cell population specific for the immunodominant peptide. Therefore, it remains to be demonstrated whether the A20[106–114] peptide is the only idiotypic peptide induced by vaccination, or if further peptide epitopes will be identified, by the use of assays such as ELISPOT, in T cell populations isolated from vaccinated mice before in vitro expansion.
Both Ad.A20 and AdA20hFc adenoviral vectors appear to be equally capable of inducing anti-idiotypic T cell responses. However, assessment of the T cell lines by ELISPOT after a long period of in vitro culture suggests that the frequency of idiotype-specific T cells is greater in the T cell line generated from mice vaccinated with Ad.A20. Whether this reflects the in vivo frequency of anti-idiotypic T cells or is the result of long-term culture needs further investigation.
We have been able to generate T cell clones specific for the A20 idiotype which are also able to lyse the A20 tumor cells and a syngeneic DC line pulsed with the A20 immunodominant peptide, A20[106–114]. These T cell clones should enable further investigation of the use of idiotype-specific T cells. It would be of interest to determine whether idiotype-specific T cells, capable of lysing the parent lymphoma line, are also capable of treating tumor-bearing mice.
This study investigated the use of adenoviral vectors expressing the
idiotype either alone or as a fusion to hFc. While both
idiotype-encoding adenoviral vectors induced similar idiotype-specific
T cell responses, the vector expressing the idiotype-hFc fusion is
clearly superior at inducing an anti-idiotypic Ab response (Fig. 8). Before analysis the anti-idiotypic T cells were expanded in
vitro with a DC line. The quantitative assessment of T cell responses
before in vitro expansion, using sensitive assays such as ELISPOT, may
elucidate any qualitative differences between the two vectors in their
ability to induce anti-idiotypic T cell responses. Furthermore,
although mice vaccinated with either Ad.A20 or Ad.A20hFc had a
significant delay in tumor development compared with mice vaccinated
with Ad.B1.8hFc (p = 0.002 and
p = 0.0003, respectively), mice vaccinated with
Ad.A20hFc had a clear survival advantage with >40% remaining tumor
free for >120 days. It appears, then, that in accordance with other
lymphoma models the induction of an anti-idiotypic Ab response is
important for tumor protection in the A20 model. Interestingly, mice
vaccinated with Ad.B1.8hFc developed cross-reacting Abs to the A20 Ig,
with equivalent levels of anti-A20 Ig Abs of the IgG1 and -2a
isotype, but half the titer of the IgG2b isotype, to that induced by
Ad.A20hFc. This raises the possibility that it is Abs of the IgG2b
isotype that are critical for tumor protection here. Future work will
determine whether the anti-idiotypic humoral response is the only
arm of the immune system contributing to tumor protection in mice
vaccinated with Ad.A20hFc. Nevertheless, for some B cell tumors, such
as myeloma, which do not express surface Ig, it seems likely that the
induction of an idiotype-specific cytotoxic immune response will be
useful.
Interestingly, this study failed to demonstrate a protective benefit resulting from vaccination of mice with the plasmid pVAC2.A20.FrC, thus failing to confirm the results of King et al. (9), who demonstrated that mice vaccinated with an equivalent plasmid in an identical dosing schedule were protected from subsequent tumor challenge. Similar plasmids have been effective in other lymphoma models (8, 9, 13). The reason for this discrepancy is unclear. The plasmid used in this study was successfully expressed in eukaryotic cells, although there may be insufficient in vivo expression from the plasmid to induce effective immunity in this tumor model. However, the superiority of the adenoviral vector encoding the A20 scFv-hFc fusion over the plasmid vaccine in this model is clear.
Although it has been possible to identify class I epitopes against human B cell lymphomas, most T cell responses against murine lymphoma idiotypes have been limited to proliferative responses. The identification of lymphoma-derived T cell epitopes raises the possibility of using peptide vaccines, which have been relatively successful in certain tumors, such as melanoma (30, 31). A recent study of idiotypic sequences derived from human B cell lymphomas found >70% of predicted HLA-A2 binding peptides were derived from framework region motifs (32). Some of these peptides were able to induce tumor-specific CTL responses. If the majority of lymphoma-derived epitopes are from Ig framework regions this may obviate the need for patient-specific therapies. The A20[106–114] peptide identified in the present study overlaps the complementary-determining region-3 and the framework region, with the first H-2Kd anchor residue being derived from the complementary-determining region-3. Thus it appears, in this model at least, amino acid residues from the hypervariable region itself are important in T cell-mediated immunity to the idiotype.
One potential concern with the use of adenoviral vectors in clinical trials is that preexisting immunity to the vector, as the result of natural exposure, will preclude successful adenoviral vaccination. Timmerman et al. (21) were able to demonstrate that prevaccination of mice with an irrelevant adenovirus vector did not impair tumor protection induced by subsequent vaccination with an idiotype-expressing adenovirus. Other groups working with model tumor Ags and adenoviral (19, 20) report similar findings.
In conclusion, our results demonstrate that recombinant idiotype-encoding adenoviral vectors are more efficient at inducing anti-idiotypic immunity than DNA vaccines, and they also support the inclusion of genes encoding xenogeneic proteins in the development of such vaccines for clinical trials. Furthermore, we have demonstrated CTL induction and identified the peptide epitope recognized by these T cells. The role of CTL in tumor protection remains to be elucidated.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Robert E. Hawkins, Department of Medical Oncology, Cancer Research Campaign, Paterson Institute for Cancer Research, Christie Hospital National Health Service Trust, Manchester M20 4BX, U.K. E-mail address: RHawkins{at}picr.man.ac.uk
3 Abbreviations used in this paper: scFv, single chain variable fragment; hFc, human Fc; IRES, internal ribosome entry site; GFP, green fluorescent protein; DC, dendritic cell.
Received for publication May 29, 2001. Accepted for publication February 7, 2002.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  D. J. Betting, K. Kafi, A. Abdollahi-Fard, S. A. Hurvitz, and J. M. Timmerman Sulfhydryl-Based Tumor Antigen-Carrier Protein Conjugates Stimulate Superior Antitumor Immunity against B Cell Lymphomas J. Immunol., September 15, 2008; 181(6): 4131 - 4140. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. B. Fredriksen and B. Bogen Chemokine-idiotype fusion DNA vaccines are potentiated by bivalency and xenogeneic sequences Blood, September 15, 2007; 110(6): 1797 - 1805. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. W. Schjetne, A. B. Fredriksen, and B. Bogen Delivery of Antigen to CD40 Induces Protective Immune Responses against Tumors J. Immunol., April 1, 2007; 178(7): 4169 - 4176. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Schiavo, D. Baatar, P. Olkhanud, F. E. Indig, N. Restifo, D. Taub, and A. Biragyn Chemokine receptor targeting efficiently directs antigens to MHC class I pathways and elicits antigen-specific CD8+ T-cell responses Blood, June 15, 2006; 107(12): 4597 - 4605. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. U. Lundin, P. O. Hofgaard, H. Omholt, L. A. Munthe, A. Corthay, and B. Bogen Therapeutic effect of idiotype-specific CD4+ T cells against B-cell lymphoma in the absence of anti-idiotypic antibodies Blood, July 15, 2003; 102(2): 605 - 612. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
