| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Hematology-Oncology Unit and
Department of Pathology, Massachusetts General Hospital, Boston, MA 02114 and Department of Medicine, Harvard Medical School, Boston, MA 02114; and
Genzyme Molecular Oncology, Framingham, MA 01701
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
7–8 h) on the surface of HLA-A*0201+ cells.
Thus, patients with metastatic melanoma are not tolerant to gp100 Ag
based on the detection of CD8+ T cells specific for
multiple HLA-A*0201-restricted, gp100-derived
epitopes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Gene transfer strategies have shown that replication-deficient Ad is well suited to serve as a vehicle for the ex vivo transduction of human DC (17, 18, 19). Distinguishing features of Ad vectors are their capacity to generate high titers necessary for clinical studies as well as their ability to infect nondividing cells. Recent developments in vector biology are built upon an appreciation that optimal virus production is influenced by distinct early region gene products while avoiding contamination by replication competent Ad (RCA) (20, 21). The version 2 Ad2 vector incorporates a unique E4 ORF-6/pIX chimeric element which reduces the generation of RCA via homologous recombination during propagation in the 293 helper cell line. The pIX capsid protein confers thermostability to the viral particle, ensuring production of high titer viral stocks essential for clinical studies. A well-recognized limitation of Ad vectors is that pre-existing Ad immunity can impair efficient in vivo transduction after systemic administration (22). A recent phase I clinical trial in patients with metastatic melanoma indicates that doses up to 1011 infectious units (IU) of Ad2 vectors can be safely administered; in addition, some individuals can be successfully immunized after systemic administration of Ad2 vectors encoding either MART-1 or gp100 Ags (23).
MART-1 and gp100 are melanocyte lineage-restricted, nonmutated Ags
isolated through expression cloning of melanoma cDNA libraries
(24). Expression of both MART-1 and gp100 is restricted to
normal melanocytes, pigmented retinal epithelia, as well as primary
cutaneous melanoma. Immunohistochemical analysis confirms gp100
expression in 75% of distant metastases (25).
Interestingly, only tumor-infiltrating lymphocytes specific for gp100,
but not MART-1 Ag, demonstrate therapeutic efficacy when reinfused into
patients (26). Initial studies identified five
HLA-A*0201-restricted dominant epitopes (G154, G209, G280, G457, and
G476) recognized by gp100-specific CD8+ T cells
from patients with melanoma (26). Subsequent analysis
showed that several subdominant epitopes of gp100 (G177 and G570) also
exist and can elicit melanoma-reactive CTL after in vitro priming with
DC (27). Several studies suggest that most circulating
A*0201-restricted, gp100-specific T cells recognize the G280 epitope
(28, 29).
We sought to examine the epitope specificity of gp100-specific, HLA-A*0201-restricted T cells upon in vitro priming with autologous DC transduced with Ad vector encoding gp100 melanoma Ag. Optimal conditions which permit >80% transduction of DC in large scale suitable for clinical immunization protocols were established. Ad-transduced DC do not undergo any discernible maturational changes or apparent cytopathic effects (CPE). gp100 Ag-specific CTL obtained from three patients identifies the G209 and the G280 epitopes as dominant. Interestingly, none of the other higher affinity dominant (or subdominant) epitopes were recognized.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The first generation Ad2 vectors encoding the gp100 and the MART-1 human melanoma Ags have been described (30). The second generation vectors described herein, referred to as version 2 (v2) vectors, possess the same 5' CMV immediate-early promoter/enhancer and harbor a deletion in the E1 region. In addition, the entire E4 region is deleted except that a novel E4 ORF-6/pIX chimeric gene has been inserted within the E4 site and is transcribed in a right to left direction (21). The E2 and E3 regions remain intact. Viral titers of clinical grade vectors exceed 1011 IU/ml after purification by standard CsCl gradient ultracentrifugation. The particle/IU ratio of Ad2 stocks range between 3 and 12. No RCA are detectable using criteria described previously (20). Clinical grade Ad2 vector stocks are stored at -80°C and used after only one freeze-thaw cycle.
Peptides
Synthetic peptides were purchased from Biosynthesis (Lewisville, TX) or manufactured at the Massachusetts General Hospital (MGH) biopolymer core facility (Charlestown, MA). Peptides were purified to >90% by reversed-phase HPLC as confirmed by mass spectrometry. Peptides were dissolved in 10% DMSO (v/v) at 2 mg/ml, sterile filtered through a 0.20 µM membrane (Schleicher & Schuell, Keene, NH) and stored at -80°C.
T2 dissociation assay
T2 cells (174xCEM.T1 hybrid, CRL-1992) are TAP-deficient,
HLA-A*0201 lymphoblastoid cells maintained in Iscove’s media (Life
Technologies, Rockville, MD) with 10% FCS (31).
Twenty-four hours before use, cells were split to ensure logarithm
phase growth. T2 cells were washed three times with 50 ml PBS and
resuspended in IMDM (serum free) at 5 x 105
cells/ml. After addition of 30 µM peptide plus 3 µg/ml human
ß2-microglobulin (Sigma, St. Louis, MO, or
Calbiochem, La Jolla, CA), cells were kept at room temperature for 30
min. Cells were transferred to 24-well tissue culture trays and
returned to 37°C in 5% CO2 overnight
(12 h). The next morning peptide-loaded T2 cells were
washed twice in PBS, resuspended in Iscove’s media (serum free)
containing 1 µg/ml brefeldin A (Epicentre Technologies, Madison, WI),
and returned to the incubator. At the indicated time points, cells were
removed, washed once, and stained with BB7.2 (HLA-A2-specific) mAb for
30 min on ice (32). Cells were washed in ice-cold staining
buffer (PBS/1% BSA/0.2% sodium azide) and incubated with 1 µl
secondary conjugate goat anti-mouse IgG-FITC (Caltag, South San
Francisco, CA). T2 cells loaded either with the H-2
Kb-binding peptide (SIINFEKL, OVA) or no peptide
served as the negative control. Cells were analyzed on a FACScan
(Becton Dickinson, Mountain View, CA) with Lysis II software. A minimum
of 10,000 events was analyzed for each sample. T2 cell death due to
prolonged exposure to brefeldin A limits the reliability of this assay
beyond a period of 24–30 h.
Cell collection and processing
Patients with stage IV malignant melanoma underwent leukopheresis at the MGH Blood Bank. Each patient had documented gp100-positive metastases by immunoperoxidase staining with the HMB-45 mAb and was HLA-A2 positive. The protocol was granted approval by the Dana-Farber Cancer Institute/MGH Institutional Review Board and patients provided written informed consent before study. Purified CD8+ T cells were obtained by negative selection of PBMC using a panel of mAb and magnetic beads. Briefly, PBMC were incubated with anti-CD4 (OKT4; American Type Culture Collection (ATCC), Manassas, VA), anti-HLA-DR (L243; ATCC), anti-CD11b (MY904; ATCC), anti-CD20 (1F5; ATCC), anti-CD14 (3C10; ATCC), and anti-CD56 (B159; PharMingen, San Diego, CA) at saturating concentrations for 60 min at 4°C. Cells were washed twice and then incubated with magnetic particles coated with goat anti-mouse IgG (PerSeptive Biosystems, Framingham, MA) for 2–4 h at 4°C. Cell separation was performed with a strong magnet. Cells were collected, washed twice, and placed in culture. Purified CD8+ T cells were >90% positive for expression of CD3+CD8+ and <2% positive for CD3+CD4+.
DC transduction
iDC were obtained from adherent PBMC fractions as described previously (4, 33). Briefly, adherent PBMC (obtained by a 2-h adherence step) were cultured in RPMI 1640, 1% AB+ serum, and penicillin/streptomycin with 100 ng/ml GM-CSF (Immunex, Seattle, WA) and 20 ng/ml IL-4 (Peprotech; Rocky Hill, NJ) for 6 days. Cells were harvested, washed twice in serum-free media, and resuspended in X-Vivo 15 (BioWhittaker, Walkersville, MD) at 107/ml. Cells were equilibrated to 37°C in a water bath for 20–30 min before transduction. Ad stocks were thawed on ice and added to the DC suspension at the indicated multiplicity of infection (moi). Cells were gently mixed by agitation and placed immediately back in the 37°C water bath. After a 20-min incubation, warm media (X-Vivo 15) containing GM-CSF and IL-4 were added to dilute the DC to a final concentration of 1 x 106/ml. Transduced DC were transferred to low-adherence 6-well trays (Ultralow 3471; Costar, Cambridge, MA) at 5 ml/well and maintained at 37°C in 5% CO2 for an additional 24 h. At the indicated time, DC were harvested, washed once in PBS, and used for analysis. For most studies, moi 300 was used to transduce DC. To obtain mDC, soluble trimeric human CD40 ligand (Immunex) at 1 µg/ml final concentration was added to iDC on day 6 of culture. After 48 additional h, cells were harvested and analyzed by flow cytometry.
Blockade of Ad receptor or coreceptors
Cells (DC or melanoma, MGH-LH) at 107/ml
were allowed to equilibrate in a 37°C water bath for 20 min.
Three-fold serial dilutions of purified Ad2 fiber-knob protein
(Genzyme, Cambridge, MA) (34) were prepared in serum-free
media (X-Vivo 15) in a volume of 100 µl. Azide-free mAb specific for
vß3 and
vß5 (Chemicon
International, Temecula, CA) were pooled and used at a final
concentration of 100 µg/ml. Purified mouse IgG1,
anti-keyhole
limpet hemocyanin mAb (PharMingen) was used as an isotype control. One
million cells were added and kept at 37°C for an additional 30 min.
Ad2/CMVEGFP was added at moi 300 (for DC) or moi 30 (for melanoma
cells), and the incubation was continued for an additional 30 min.
Afterward, 100-fold excess media was added, and the cells were washed
once to remove residual virus. Cells were resuspended in X-Vivo 15
(containing GM-CSF/IL-4 for DC) and cultured for an additional 24
h at 37°C in 5% CO2. Cells were analyzed by
flow cytometry as described above.
Fluorescent virus assays
Ad virus particles were labeled with carbocyanide dye (Cy3; Amersham, Arlington Heights, IL) as described previously (35). Briefly, 1012 virus particles/ml were incubated with Cy3 (1:9 v/v) for 30 min and then dialyzed to remove noncovalently bound dye. Cy3-conjugated Ad was added directly to iDC at 1000 particles/cell or to melanoma cells at 300 particles/cell. Cells were attached to poly-L-lysine-coated plates. Cy3-labeled Ad was added to cells and after a 2-min incubation period at 37°C, the cells were washed with PBS to remove residual virus. Fiber inhibition studies were performed by preincubation of cells with purified Ad2 fiber-knob protein (1 mg/ml) for 1 h before the addition of Cy3-labeled Ad. The fiber-knob protein was present during the entire incubation period. At the indicated time points, each culture was washed three times with PBS before fixation with 1% paraformaldehyde. The coverslips were counterstained with 4',6'-diamidino-2-phenylindole (DAPI; Sigma) to visualize the nuclei before mounting with Permamount (Sigma). Fluorescence was visualized using the Olympus Provis AX70 microscope (New Hyde Park, NY) equipped with a Hamamatsu digital charge-coupled device camera (Middlesex, NJ) for capturing images.
CTL generation
Purified CD8+ T cells were cultured with autologous DC transduced with Ad2/gp100v2 at a ratio of 20:1 in RPMI 1640 (Life Technologies) plus 10% male AB+ serum, 100 U/ml penicillin, 100 µg/ml streptomycin, 10 mM HEPES, 2 mM L-glutamine, nonessential amino acids, and sodium pyruvate. A total of 5 x 105 CD8+ T cells with 2.5 x 104 transduced DC was cultured in a volume of 0.5 ml in 48-well tissue culture trays (Costar 3548) in the presence of IL-7 (Peprotech) 10 ng/ml and maintained at 37°C in 5% CO2 (27). Beginning on day 3, 50 U/ml IL-2 (Chiron Therapeutics, Emeryville, CA) was added to each well every 2–3 days. Every 7 days, responding T cells were harvested, transferred to new culture trays, and restimulated with fresh Ad2/gp100v2-transduced autologous DC. After the third restimulation, T cell lines were expanded in 24-well trays in a final volume of 1 ml medium/well. CTL were tested in cytolysis assays 5 days after each restimulation as described below.
Cytolysis assays
Cytolytic activity was measured using a standard 4-h 51Cr release assay. Briefly, T2 cells were labeled with 100 µCi Na2CrO2 (NEN, Boston, MA) at 1 x 106 cells/ml for 1 h at 37°C. Cells were washed three times, counted, and then loaded with 5 µM peptide for 30 min before addition to wells containing effector cells. Melanoma cell lines were harvested by pipetting (avoiding the use of trypsin) and labeled with 50 µCi Na2CrO2 for 1 h at 37°C. Melanoma cell lines were washed three times and used as target cells when indicated. The melanoma cell lines DM13 (gp100+, A2.1+) and DM14 (gp100-, A2-) were kindly provided by T. Darrow (Duke University Medical Center, Durham, NC) (36). Assays were performed in 96-well round-bottom microtiter trays in a volume of 0.2 ml media with 104 target cells/well. Culture trays were centrifuged for 2 min at 25 x g before incubation at 37°C in 5% CO2. After a 4-h period, each tray was centrifuged for 5 min at 200 x g, and 0.05 ml medium was transferred to a Lumaplate 96 tray (Packard, Meriden, CT) and placed in a gamma counter (TopCount; Packard). The percent specific lysis was calculated using the standard formula: % lysis = 100 x ([experimental release - spontaneous release]/[maximal release - spontaneous release]). The spontaneous release is calculated from target cells incubated without effector cells, whereas the maximal release is calculated from target cells lysed with 1% Triton X-100. All determinations were performed in duplicate, and spontaneous release was usually <20%.
Flow cytometry
DC transduced with Ad2/CMVEGFP vector were harvested, washed
once, and analyzed directly on the FL-1 channel (emission 530 nm peak
fluorescence) after 488-nm excitation on a FACScan (Becton Dickinson)
for a minimum of 10,000 events gated by forward/side scatter. Detection
of cell surface Ags was performed on DC harvested from culture, washed
twice, and incubated for 15 min in 50 µg/ml human IgG (Baxter
Diagnostics, McGraw Park, IL) to block nonspecific binding. Cells were
then washed an additional time with ice-cold PBS/1%BSA/0.2% azide. A
total of 5 x 105 cells was labeled with
optimal (pretitered) concentrations of the indicated
fluorochrome-conjugated mAb for 30 min on ice. When indicated, a
secondary goat anti-mouse IgG-FITC conjugated reagent (Caltag) was
used. Cells were washed twice and fixed with 1% paraformaldehyde
before analysis using Lysis II software. Isotype, nonbinding control
mAb were used as negative control reagents in each experiment. The
following Abs were used in these studies: anti-HLA-DR-FITC (I3;
Coulter, Palo Alto, CA), anti-CD11c-PE (Leu-M5; B. D.,
Sunnyvale, CA), anti-CD14-FITC (My4; Coulter), anti-HLA class
I-PE (W6/32; Serotec, Oxford, U.K.), anti-CD80-PE (L307.4; B.
D.), anti-CD86-FITC (FUN1; PharMingen), anti-CD83 (HB15, T.
Tedder, Duke University Medical Center), anti-CAR (RmcB; R.
Finberg, Dana-Farber Cancer Institute), anti-integrin
vß3 (1976; Chemicon),
and anti-integrin
vß5 (1961;
Chemicon).
Immunoperoxidase staining
Single-cell suspensions were incubated at 37°C at 5%
CO2 in sterile tissue culture grade chamber
slides (Falcon 4108; Becton Dickinson) and 12 h later were fixed
with 10% buffered Formalin. After washing, cells were stained with
anti-human gp100 primary mAb (HMB-45; BioGenex Laboratories, San
Ramon, CA) and a biotinylated secondary goat anti-mouse IgG.
Peroxidase detection using diaminobenzidine (Vectastain Elite; Vector
Laboratories, Burlingame, CA) was performed according to instructions
supplied by the manufacturer. MART-1-specific mAb (A103) was purchased
from BioGenex Laboratories. The anti-HLA-DR mAb (L243) culture
supernatant was used as a positive control.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The viral constructs used in this study are shown schematically in
Fig. 1
. iDC were obtained by culture of
adherent PBMC in media containing 1% human AB+
serum plus GM-CSF and IL-4. After 6 days of culture, phenotypic
characterization by flow cytometry yields a uniform population of DC
that express high levels of HLA class II Ags, CD11c, CD86, and low
levels of CD80 and CD83 (see below). Expression of CD14 is variable as
determined with the My4 mAb but is always 10–50-fold lower than levels
present on fresh PBMC. Expression of the CD3, CD19, and CD56 lineage
markers are always negative. Using phenotypic and morphologic criteria
established by other investigators (37), the DC obtained
after 6 days of culture under these conditions are judged to be
immature. Final yields of iDC obtained at day 6 are 5–10% of the
starting number of PBMC.
|
A, iDC are susceptible to
transduction with Ad vector in a dose-dependent manner. At 48 h
postinfection, it appears that moi from 100 to 500 is optimal for the
detection of EGFP as monitored by flow cytometry. For example, at moi
500, 93% of DC were positive for the expression of enhanced green
fluorescent protein (EGFP) whereas uninfected cells were <1%
positive. The mean fluorescence intensity (MFI) of EGFP expression was
6577 among DC transduced at moi 500; in contrast, control uninfected
iDC exhibited a MFI of 25. We have consistently seen efficient EGFP
expression (>80% positive cells) with iDC from >20 normal donors. We
have noted that moi >500 typically results in a decrease in MFI with
some CPE evident. For example, moi 2500 results in 82% positive cells
at 48 h with an MFI 3694.
|
B,
detection of EGFP was optimal between 24 and 48 h postinfection.
In the representative experiment, EGFP was undetectable at 2 h,
but was clearly evident by 14 h postinfection. At 24 h, 95%
of DC were expressing EGFP with an MFI of 1807; at the 48-h time point,
98% of DC were judged positive with an MFI of 1319. By 168 h
postinfection, only 77% of cells (MFI 342) were positive (data not
shown). At each time point measured, uninfected DC population cultured
under identical conditions served as a negative control. The MFI of
each control DC population was always <10. In Fig. 2C,
representative fluorescence histograms show the time course of EGFP
levels in human DC. Similar results have been obtained in our
laboratories using Ad2/CMVß-gal vectors to infect human DC by
detection with the fluorescein
di-ß-D-galactopyranoside) substrate using flow
cytometry (data not shown). Ad transduction of human DC is independent of the coxsackie Ad receptor
Since human DC are susceptible to infection with Ad only at
relatively high moi (Fig. 2), we sought to determine whether the
coxsackie Ad receptor (CAR) pathway is intact (38). Using
the RmcB mAb, which has specificity for the human CAR, it appears that
human DC (n = 7 donors) are deficient in the expression
of CAR as monitored by flow cytometry (Fig. 3A). As a positive control,
the melanoma cell line (MGH-LH) expresses significant levels of CAR and
is highly susceptible to Ad virus infection. For example, at moi 3,
95% of the melanoma cells are transduced with Ad2/CMVEGFP, whereas
human DC require moi >300 to achieve this level of transgene
expression (data not shown). Both human DC and the melanoma line
express detectable amounts of the Ad coreceptors
vß3 and
vß5 as determined by
flow cytometry, suggesting that integrins are not limiting Ad
transduction of DC.
|
B). In contrast, CAR did not block the
transduction of human DC at moi 300. In fact, we have reproducibly seen
modest enhancement in Ad transduction of DC with the addition of Ad2
fiber-knob protein. One possible explanation for this finding would be
the presence of a small contaminant in the fiber-knob preparation that
activates the endocytic compartment of DC to facilitate uptake by Ad
vector. To assess the contribution of integrin-mediated
internalization, Ab-mediated blockade of
vß3 and
vß5 was performed
using a pool of mAb specific for both integrins. Ad transduction of
both cell types was partially blocked by the anti-integrin mAbs
(Fig. 3B). The MFI of transduced DC was reduced 57% in the
presence of anti-integrin mAb (control mAb reduced by 6%), whereas
the MFI of transduced MGH-LH was reduced 48% by anti-integrin mAb
(control mAb reduced by 16%). This observation suggests a role for
vß3 and
vß5 integrins in Ad
vector gene transfer in human monocyte-derived iDC. Localization of fluorescent-labeled Ad virions in transduced human DC
To confirm the above observations, Cy3-labeled Ad particles were
used to infect CAR-negative iDC and the CAR-positive human melanoma
line. Fluorescent photomicrographs of representative fields illustrate
the binding, uptake, and subcellular localization of Ad after
incubation at 37°C for either 2, 15, or 90 min. After the indicated
incubation period, the cells were gently washed to remove residual
unbound Ad, counterstained with DAPI, and immediately processed for
microscopic inspection. Fig. 4 shows the
kinetics of distribution of viral vector in each cell type after
incubation with Cy3-labeled Ad. A 2-min incubation (denoted as 0 min)
is sufficient for Ad binding and internalization by the melanoma cell
line. In contrast, little fluorescence is evident either on or within
the cytoplasm of DC after the 2-min incubation. When Cy3-labeled Ad was
allowed to remain on DC for longer periods (15 or 90 min), uptake and
perinuclear localization of Ad vector was clearly visible.
Preincubation of cells with Ad2 fiber-knob protein was performed to
assess CAR-mediated transduction of Cy3-labeled Ad particles. As
expected, transduction of the human melanoma cell line was
substantially decreased with prior incubation of Ad2 fiber-knob protein
at all time points studied. In contrast, fiber-knob protein had no
discernible effect on Cy3-Ad uptake by and localization within DC after
a 15- or 90-min incubation period. Thus, these data suggest that the
dwell time of Ad vector particles is critical for the efficient
transduction of DC and perhaps other CAR-negative cells.
|
To assess the potential use of Ad-transduced DC for cancer vaccine
therapy, we monitored the expression of two major lineage-restricted
melanocyte Ags in DC after ex vivo transduction with Ad2/gp100v2 and
Ad2/MART-1v2. DC were transduced with each vector at moi 300 as
described above and were cultured in GM-CSF and IL-4 containing media
for an additional 24–48 h. To assess the feasibility of large-scale
transduction required for clinical studies, we utilized 10–50 x
106 DC. In large-scale experiments performed
under the conditions detailed above, transduction efficiency of >80%
was routinely measured as assessed by immunoperoxidase detection of
single-cell suspensions using mAb directed to gp100 or MART-1. In a
representative experiment shown in Fig. 5, DC were transduced with
replication-deficient recombinant Ad at moi 300. Forty-eight hours
later, single-cell suspensions of DC were processed for
immunoperoxidase staining. More than 95% of Ad2/MART-1v2-transduced DC
were judged positive for MART-1 expression by detection with A103 mAb.
Ad2/gp100v2-transduced DC scored 92% positive when stained with the
HMB-45 mAb. As a positive control, virtually all DC were positive for
MHC class II (HLA-DR) after staining with the L243 mAb; in contrast, DC
incubated with a nonbinding isotype control mAb (or no primary Ab) were
nonreactive. As a control, DC transduced with Ad2/CMVEV were stained
positive using the L243 mAb; in contrast, staining with either
anti-gp100 or anti-MART-1 mAb was judged to be nonreactive.
|
To assess the possible effect of replication-defective Ad on DC
maturation, we studied the expression of various cell surface markers
on Ad-transduced and nontransduced DC cultured in parallel. DC obtained
after 6 days of culture were harvested and split into three groups. DC
were either transduced with Ad2/gp100v2, exposed to soluble trimeric
human CD40 ligand (nontransduced), or left untreated. The CD40
ligand-treated DC served as a positive control to assess the mDC
phenotype. All three groups were recultured in parallel under identical
conditions in X-Vivo 15 serum-free media containing GM-CSF and IL-4. As
shown in Fig. 6, day 6 DC have the
phenotype characteristic of iDC: HLA class I+,
HLA-DR+, CD11c+,
CD14lo, CD80lo,
CD83lo, CD86+. Exposure to
CD40 ligand promotes terminal maturation of DC and the up-regulation of
various markers including HLA class I Ags: HLA-DR, CD80, CD83, and
CD86. Ad transduction does not appear to promote terminal maturation as
assessed by the expression levels of the cell surface markers studied.
In Fig. 6, the MFI of HLA-DR expression on uninfected day 8 DC (MFI,
1042) is not significantly different when compared with
Ad2/gp100v2-transduced DC (MFI, 1059); in contrast, CD40 ligand-treated
DC express greater amounts of HLA-DR cell surface Ag (MFI, 1336)
consistent with terminal maturation. We have consistently noted
(n = 8 experiments) that Ad-transduced DC retain the
phenotypic markers (and relative levels) present on uninfected DC. We
conclude that Ad infection with replication-deficient virus does not
adversely affect HLA class I Ag expression nor the levels of
costimulatory molecules or adhesion molecules such as CD54 and CD58
(data not shown). Moreover, Ad-transduced DC remain viable for periods
up to 7 days in culture as monitored by exclusion of propidium iodide
and detection of esterase activity after loading with the cell
permeable calcein AM ester (data not shown).
|
To evaluate the immunogenicity of Ad-transduced DC, purified
CD8+ T lymphocytes from patients with melanoma
were restimulated in vitro with autologous DC transduced with
Ad2/gp100v2. The three patients studied in detail were
HLA-A*0201-positive and had stage IV disease with documented
gp100-positive metastasis. Responding T cells of each patient were
restimulated weekly using autologous DC expressing gp100 and expanded
in the presence of IL-2. To assess for reactivity to gp100 Ag, effector
T cells were tested in a standard 51Cr release
assay. Specific killing of the gp100-positive melanoma cell line DM13
(HLA-A2.1, -31, -B13, -18) was noted using effector cells from all
three patients. In contrast, the gp100-negative melanoma cell line DM14
(HLA-A11, -28, -B5, -8, -Cw2, -4) was not lysed (Fig. 7A). Effector T cells obtained
under these conditions do not exhibit significant lysis of K562 cells
and are Ag specific as assessed by cold target inhibition assays (Fig. 7B). Specifically, cold DM13, but neither DM14 nor K562, can
compete for lysis of 51Cr-labeled DM13 melanoma
cells. Addition of an HLA class I-specific mAb (W6/32) completely
blocks recognition of melanoma cell line DM13 by CTL from all three
patients (data not shown). To assess the epitope specificity of
effector cells from each donor, T2 cells pulsed individually with a
panel of peptides from gp100 were used as target cells (Fig. 7C). Interestingly, only the G209 and G280 peptides were
recognized from among the entire panel of seven dominant and
subdominant epitopes. CTL from patients 057-02 and 057-03 lysed T2
targets pulsed with either the G209 or the G280 peptide. Patient 057-01
CTL killed only T2 targets pulsed with the G280 epitope. Furthermore,
the remaining HLA-A*0201-restricted epitopes (G154, G177, G457, G476,
G570) were not recognized by CTL from any patient studied.
|
Most dominant epitopes exhibit high affinity for MHC class I
molecules (39). Previous studies using a competition assay
with a radiolabeled high-affinity HLA-A*0201-restricted peptide have
established a rank order of binding affinity for the gp100 epitopes
(26, 27). The G154 and G476 peptides are classified as
high-affinity (IC50, <50 nM) epitopes, whereas
the remaining five peptides were of intermediate affinity
(IC50, 50–500 nM) for HLA-A*0201. We chose to
investigate this issue using a live cell assay with the TAP-deficient
HLA-A*0201+ T2 lymphoblastoid cell line to
measure the dissociation rate of
A*0201/ß2-microglobulin/peptide complexes. T2
cells were loaded with individual peptides in the presence of exogenous
human ß2-microglobulin (and in the absence of
FCS). After overnight culture at 37°C, the loaded T2 cells were
washed vigorously and recultured at 37°C in serum-free media
containing brefeldin A to block transport of newly synthesized class I
molecules to the cell surface. At each time point, T2 cells were
removed, washed, and stained with the BB7.2 mAb to quantitate folded
forms of HLA-A*0201 cell surface molecules. The 
MFI obtained at each
time point (MFI = MFI with peptide - MFI with no peptide)
is depicted in Fig. 8 for each gp100
epitope as well as a reference hepatitis B virus (HBV) core 18-27 (C18)
epitope previously shown to have high binding affinity for HLA-A*0201
(40). Linear regression analysis was used to obtain the
measured half-life (t1/2) at
37°C of each peptide when complexed with HLA-A*0201. The rank order
of peptides from gp100 as determined by this analysis corresponds
closely to that seen by others using different methods (Table I): G154
(t1/2 = 24.7 h) > G177
(13.9 h) > G570 (11.2 h) > G209 (8.25 h) > G280 (7.0
h), > G457 (6.35 h). We were unable to accurately assess the G476
peptide because of poor solubility; however, it has been characterized
previously as the second most avid binder to HLA-A*0201. The reference
high-affinity HBV core C18 peptide has a half-life of 22.8
h.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Several previous reports (17, 18, 19) show that human DC are
susceptible to Ad transduction ex vivo. Results from our study provide
additional information about optimization of Ad transduction of human
DC. First, using Ad2 viral stocks with a particle:IU ratio 3–12:1
there appears to be no absolute requirement for liposomes to enhance
transduction (data not shown). Second, it appears that Ad-transduced DC
exhibit no unusual phenotypic changes or evidence to suggest
differentiation to the mature state. This is of potential interest
since iDC have unique chemokine receptor and cytokine/chemokine
profiles which contribute to their homing properties (42).
In addition, iDC are very efficient at capturing and processing Ags; in
contrast, mDC are less efficient at Ag capture so timing would be
critical for proper loading of Ag ex vivo before readministration
(43, 44). Third, we provide formal documentation of dose
response and time course of transgene expression in human DC transduced
with Ad vector. Human DC transduced with the second-generation Ad2
vectors described herein at moi 100–500 are viable and show no
evidence of CPE. However, careful titration of Ad stocks is essential
since moi >500 cause CPE of DC and result in suboptimal Ag expression.
One possible explanation for the CPE is the toxicity of the penton
(pIII) protein evident at higher particle inoculations. Our finding
that human DC are devoid of CAR provides one explanation for the higher
moi requirement for iDC transduction. The blocking experiment using
anti-integrin mAbs supports the notion that
vß3 and
vß5 act as coreceptors
(45) for Ad infection of human DC. We emphasize that,
unlike vaccinia virus (46), replication-deficient Ad
vectors are not cytopathic for human DC, ensuring Ag presentation
beyond 24 h after transduction.
We were surprised to find that iDC are susceptible to transduction by
subgroup C (Ad2) Ad vectors, despite their CAR-negative phenotype.
Several studies have clearly demonstrated the requirement for CAR
expression in Ad vector transduction of multiple cell types (35, 47, 48). For example, the basolateral localization of CAR by
respiratory epithelium probably limits transduction when Ad vector is
administered in aerosolized or aqueous form in the lumen
(49). In a second study, a panel of 14 human melanoma cell
lines were surveyed for susceptibility to Ad-mediated transduction;
expression of CAR, but neither
vß3 or
vß5 integrins,
correlated with transgene expression among the cell lines tested
(50). Elegant studies by Leopold et al. (35)
using Cy3-labeled Ad clearly show the temporal-spatial distribution of
Ad vector on CAR-positive epithelial A549 cells. Ad internalization in
A549 cells occurs with a t1/2 2.5 min
and, by 60 min postinfection, >80% virus is localized to the nucleus.
In agreement with this report, we find that transduction of the
CAR-positive melanoma line is sensitive to inhibition by Ad fiber-knob
protein. In addition, we show that CAR-negative DC can be efficiently
transduced under conditions that promote virus cell contact under
conditions of high cell density in serum-free media for 20 min at
37°C. It appears that the
vß3 and
vß5 integrins do play
a role in viral uptake by DC based on inhibition studies with mAb;
however, despite integrin blockade, we still observed significant
Ad transduction. A distinguishing feature of iDC is the
constitutive high rate of macropinocytosis (43, 51). We
postulate that macropinocytosis of Ad vector allows internalization and
delivery to an endosomal compartment before cytoplasmic transport. If
macropinocytosis is the primary mechanism underlying CAR-independent
transduction, then DC would appear to have a selective advantage for
infection over most other cell types that lack the high-affinity
receptor.
Tolerance to self nonmutated Ags is a critical issue when designing optimal vaccine formulations for certain malignancies such as melanoma (52). Virtually all candidate melanoma-specific tumor Ags under clinical study are self nonmutated proteins with restricted patterns of tissue expression and include Mage-1/3, tyrosinase, TRP-1/2 as well as MART-1 and gp100 (24). Using tumor-infiltrating lymphocytes as an enriched source of tumor-specific CD8+ T cells expanded in IL-2, Kawakami et al. (26) identified five dominant HLA-A*0201-restricted peptides (G154, G209, G280, G457, and G476) within gp100. Using peptide-pulsed DC, Tsai et al. (27) identified the G177 and G570 epitopes as subdominant based on primary in vitro stimulation of normal donor CD8+ T cells that were shown to specifically kill HLA-A*0201+ melanoma lines. Thus, for the gp100 Ag, there are seven known epitopes restricted to HLA-A*0201 that are potential candidates for inclusion into a subunit vaccine formulation. We chose to use Ad-transduced DC to expand gp100-reactive CD8+ T cells in vitro from patients with metastatic melanoma known to harbor gp100+ metastases. We were surprised to discover that only two epitopes, G209 and G280, are dominant. Interestingly, the G209 and G280 epitopes are among the lower affinity peptides for HLA-A*0201. Since we were unable to detect CD8+ T cells specific for the remaining higher affinity (G154, G177, G476, and G570) peptides, this suggests either deletion or perhaps anergy as the mechanism for tolerance. Furthermore, G457 has the weakest binding to HLA-A*0201 as measured in the competition assay and the T2 dissociation assay; we found no measurable G457-specific response in any of the patients studied. We conclude that, from among the panel of seven epitopes identified for HLA-A*0201, G209 and G280 represent good candidates for further study. Moreover, clinical studies in patients with metastatic melanoma provide clear evidence that both G209 and G280 are immunogenic in vivo and can elicit melanoma-reactive CTL (53).
Based on the results presented in this report, a phase I/II clinical trial utilizing Ad-transduced DC encoding gp100 and MART-1 is underway to assess the safety, immunogenicity, and efficacy in patients with metastatic melanoma. Calculation of precursor frequencies of peptide-specific CTL should allow one to further assess the hierarchy of dominant and subdominant epitopes encoded by gp100. Several recent reports confirm the relative potency of DC as an adjuvant capable of breaking tolerance to self Ags (12, 54, 55) and provide a rationale for clinical trials with DC which utilize self nonmutated Ags such as gp100 and MART-1.
We detail the use of replication-deficient recombinant Ad to transduce human DC ex vivo. We provide evidence that gene transfer to human DC is efficient at moi 100–500 of Ad2 vector, and that transduced DC do not exhibit any phenotypic or maturational changes. Ad infection of human DC occurs independently of CAR, but integrins expressed by DC appear to participate in viral uptake. Tracking studies with fluorescent-labeled Ad document the rapid internalization and distribution of virus within DC. Using autologous DC transduced with Ad2/gp100v2 to stimulate and expand gp100-reactive T cells from patients with melanoma, we show that the G209 and G280 (HLA*0201-restricted) epitopes are dominant. We conclude that melanoma patients who harbor gp100+ metastases are not tolerant to gp100 based on the detection of melanoma-reactive CTL.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Gerald P. Linette or Dr. Frank G. Haluska, Massachusetts General Hospital, Hematology-Oncology Unit, 55 Fruit Street, Boston, MA 02114.
3 Abbreviations used in this paper: iDC; immature dendritic cell; moi, multiplicity of infection; Ad, adenovirus; EGFP, enhanced green fluorescent protein; CAR, coxsackie Ad receptor; MFI, mean fluorescence intensity; mDC, mature dendritic cell; RCA, replication competent Ad; IU, infectious unit; DAPI, 4',6'-diamidino-2-phenylindole; CPE, cytopathic effect; HBV, hepatitis B virus.
Received for publication September 2, 1999. Accepted for publication January 6, 2000.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
. J. Exp. Med. 179:1109.v-ß5 and CD36, and cross-present antigens to cytotoxic T lymphocytes. J. Exp. Med. 188:1359.This article has been cited by other articles:
|
|
 |
|
  M. A. Purbhoo, Y. Li, D. H. Sutton, J. E. Brewer, E. Gostick, G. Bossi, B. Laugel, R. Moysey, E. Baston, N. Liddy, et al. The HLA A*0201-restricted hTERT540-548 peptide is not detected on tumor cells by a CTL clone or a high-affinity T-cell receptor Mol. Cancer Ther., July 1, 2007; 6(7): 2081 - 2091. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. Chapiro, S. Claverol, F. Piette, W. Ma, V. Stroobant, B. Guillaume, J.-E. Gairin, S. Morel, O. Burlet-Schiltz, B. Monsarrat, et al. Destructive Cleavage of Antigenic Peptides Either by the Immunoproteasome or by the Standard Proteasome Results in Differential Antigen Presentation J. Immunol., January 15, 2006; 176(2): 1053 - 1061. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. P. Linette, D. Zhang, F. S. Hodi, E. P. Jonasch, S. Longerich, C. P. Stowell, I. J. Webb, H. Daley, R. J. Soiffer, A. M. Cheung, et al. Immunization Using Autologous Dendritic Cells Pulsed with the Melanoma-Associated Antigen gp100-Derived G280-9V Peptide Elicits CD8+ Immunity Clin. Cancer Res., November 1, 2005; 11(21): 7692 - 7699. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Yang, K.-Y. Tsang, and J. Schlom Induction of Higher-Avidity Human CTLs by Vector-Mediated Enhanced Costimulation of Antigen-Presenting Cells Clin. Cancer Res., August 1, 2005; 11(15): 5603 - 5615. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Yang and F. G. Haluska Treatment of Melanoma with 5-Fluorouracil or Dacarbazine In Vitro Sensitizes Cells to Antigen-Specific CTL Lysis through Perforin/Granzyme- and Fas-Mediated Pathways J. Immunol., April 1, 2004; 172(7): 4599 - 4608. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Mukouyama, N. K. Janzen, J. M. Hernandez, J. S. Lam, R. Caliliw, A. Y. Wang, R. A. Figlin, A. S. Belldegrun, and G. Zeng Generation of Kidney Cancer-Specific Antitumor Immune Responses Using Peripheral Blood Monocytes Transduced With a Recombinant Adenovirus Encoding Carbonic Anhydrase 9 Clin. Cancer Res., February 15, 2004; 10(4): 1421 - 1429. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Tuettenberg, H. Jonuleit, T. Tuting, J. Bruck, V. Biermann, S. Kochanek, J. Knop, and A. H. Enk Early Adenoviral Gene Expression Mediates Immunosuppression by Transduced Dendritic Cell (DC): Implications for Immunotherapy Using Genetically Modified DC J. Immunol., February 1, 2004; 172(3): 1524 - 1530. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. R. Parkhurst, C. DePan, J. P. Riley, S. A. Rosenberg, and S. Shu Hybrids of Dendritic Cells and Tumor Cells Generated by Electrofusion Simultaneously Present Immunodominant Epitopes from Multiple Human Tumor-Associated Antigens in the Context of MHC Class I and Class II Molecules J. Immunol., May 15, 2003; 170(10): 5317 - 5325. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Yang, G. P. Linette, S. Longerich, and F. G. Haluska Antimelanoma Activity of CTL Generated from Peripheral Blood Mononuclear Cells After Stimulation with Autologous Dendritic Cells Pulsed with Melanoma gp100 Peptide G209-2M Is Correlated to TCR Avidity J. Immunol., July 1, 2002; 169(1): 531 - 539. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 H. Tsao, P. Millman, G. P. Linette, F. S. Hodi, A. J. Sober, M. A. Goldberg, and F. G. Haluska Hypopigmentation Associated With an Adenovirus-Mediated gp100/MART-1-Transduced Dendritic Cell Vaccine for Metastatic Melanoma Arch Dermatol, June 1, 2002; 138(6): 799 - 802. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. M. Kirkwood, J. G. Ibrahim, J. A. Sosman, V. K. Sondak, S. S. Agarwala, M. S. Ernstoff, and U. Rao High-Dose Interferon Alfa-2b Significantly Prolongs Relapse-Free and Overall Survival Compared With the GM2-KLH/QS-21 Vaccine in Patients With Resected Stage IIB-III Melanoma: Results of Intergroup Trial E1694/S9512/C509801 J. Clin. Oncol., May 1, 2001; 19(9): 2370 - 2380. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
D. Rea, M. J. E. Havenga, M. van den Assem, R. P. M. Sutmuller, A. Lemckert, R. C. Hoeben, A. Bout, C. J. M. Melief, and R. Offringa Highly Efficient Transduction of Human Monocyte-Derived Dendritic Cells with Subgroup B Fiber-Modified Adenovirus Vectors Enhances Transgene-Encoded Antigen Presentation to Cytotoxic T Cells J. Immunol., April 15, 2001; 166(8): 5236 - 5244. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Vincent, B.-G. Harvey, S. M. Hogan, C. J. Bailey, R. G. Crystal, and P. L. Leopold Rapid Assessment of Adenovirus Serum Neutralizing Antibody Titer Based on Quantitative, Morphometric Evaluation of Capsid Binding and Intracellular Trafficking: Population Analysis of Adenovirus Capsid Association with Cells Is Predictive of Adenovirus Infectivity J. Virol., February 1, 2001; 75(3): 1516 - 1521. [Abstract] [Full Text]
|
|
|
|
|
|
S. Stager, D. F. Smith, and P. M. Kaye Immunization with a Recombinant Stage-Regulated Surface Protein from Leishmania donovani Induces Protection Against Visceral Leishmaniasis J. Immunol., December 15, 2000; 165(12): 7064 - 7071. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 W. C. Russell Update on adenovirus and its vectors J. Gen. Virol., November 1, 2000; 81(11): 2573 - 2604. [Full Text]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
