| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
*
Institut National de la Santé et de la Recherche Médicale Unité 463, Nantes, France; and
Institut National de la Santé et de la Recherche Médicale Unité 395, Toulouse, France
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain
CD25+ cells and were then stimulated for 24–96 h with
previously defined peptide Ags or with autologous PBMC infected with a
canarypox viral vector encoding the total pp65 protein (ALVAC-pp65).
Subsequent immunomagnetic purification of newly CD25-expressing cells
allowed efficient recovery of T lymphocytes specific for the initial
stimuli, i.e., for the already known immunodominant epitope
corresponding to the peptides used as a model or for newly defined
epitopes corresponding to peptides encoded by the transfected pp65
protein. Importantly, we demonstrated that direct PBMC stimulation
allowed recovery not only of CD8+ memory T lymphocytes, but
also of the CD4+ memory T cells, which are known to be
crucial to ensure persistence of adoptively transferred immune memory.
Finally, our analysis of pp65-specific T cells led to the
identification of several new helper and cytotoxic epitopes. This work
thus demonstrates the feasibility of isolating memory T lymphocytes
specific for a clinically relevant protein without the need to prepare
APC, to use infectious viral strains, or to identify immunodominant
epitopes. |
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
From an immunological point of view, accumulating data in the literature tend to favor the notion that the immune response against these two viruses is essentially focused against a few proteins (5, 6, 7, 8). For example, frequent responses against the nonameric peptide GLCTLVAML (referred to as G9L) from the EBV early lytic protein BMLF1 (9) and frequent responses against the nonameric peptide NLVPMVATV (referred to as N9V) from the pp65 CMV phosphoprotein were observed among peripheral blood mononuclear cells of seropositive HLA-A*0201+ individuals (6). From a clinical point of view, two major constraints limit a broader use of T cell therapy against viruses: the delay required to obtain specific T cells and the safety of the selection procedure. In terms of delay, the selection of EBV-specific T cells requires the generation of an autologous B lymphoblastoid cell line (BLCL)3 for use as an EBV APC, followed by a coculture with autologous PBMC. Several weeks are required to obtain the BLCL and several additional weeks are required to enrich for EBV-specific T cells (10). In terms of safety, BLCL are obtained after infection of autologous PBMC with an EBV viral strain produced by the marmoset B95.8 cell line, which is cultured in the presence of FCS. Thus, this procedure uses biological material from two different xenogenic origins: simian for the cell line and bovine for the serum. A comparable level of complexity is associated with the preparation of CMV-specific T lymphocytes (3). Clearly, significant progress would be achieved by eliminating the need for APC preparations, for timing reasons, as well as the need for an infectious viral strain, for safety concerns.
In this report we describe a method that we considered initially for the following reasons: cross-linking of the Ag-specific receptor on the surface of T lymphocytes is the key signal for CD25 expression (11, 12, 13), and CD25 expression can occur even in the absence of APC (14). Consequently, we reasoned that if autologous presentation (B-T, monocyte-T, or T-T) of a particular epitope among fresh PBMC can induce CD25 expression, then even rare specific T cells could be sorted out from the total population according to this marker and can be induced to proliferate using optimal culture conditions even though their activation status may not be initially optimal. Using purification of various EBV or CMV Ag-specific T lymphocytes as a model system, we tested different methods of PBMC stimulation before CD25 selection. To test our concept, we used three types of antigenic stimuli. Initially, we tested single small peptides corresponding to known immunodominant epitopes. Subsequently, we used a pool of 10 large overlapping peptides (23 aa long with a step of 12), which included a known nonameric epitope. Finally, we also used autologous PBMC infected with a replicative defective canarypox (ALVAC) vector encoding the pp65 protein to stimulate the entire T cell repertoire specific for this protein on this particular genetic background. The data presented in this paper prove the feasibility of these approaches, which provide a valuable tool for the rapid sorting of Ag-specific T lymphocytes and the identification of new epitopes.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Blood packs were obtained from healthy HLA-A0201+ or HLA-B8+ EBV or EBV/CMV seropositive adult donors after informed consent was obtained. PBMC were separated using Ficoll density centrifugation (lymphocyte separation medium; Eurobio, Paris, France).
Peptides
The following peptides were obtained >70% pure by HPLC from Sigma-Genosys (Cambridge, UK): The HLA-A2 binding peptide AAGIGILTV (referred to as A9V) derived from the melanoma-associated MelanA/MART-1 protein (15), the HLA-A2 binding peptide NLVPMVATV (referred to as N9V) derived from the pp65 CMV phosphoprotein (6), the HLA-A2 binding peptide GLCTLVAML (refereed to as G9L) derived from the EBV early lytic protein BMLF1 (16), and the HLA-B8 binding peptide FLRGRAYGL (referred to as F9L) derived from the EBV latent protein EBNA3A (13). In addition a panel of 50 23-aa-long (23 mer) peptides (numbered 1–50) overlapping by 12 aa and spanning the entire CMV pp65 sequence was obtained from Chiron Mimotopes (Suresnes, France). Note that the immunodominant HLA-A2 nonamer N9V was included in the 23-aa-long peptide 45. Peptide stock solutions (20 mg/ml in DMSO) were diluted first to 2 mg/ml in acetic acid (1%) and second to the final concentration in RPMI 1640 culture medium (Sigma-Aldrich, St. Quentin Fallavier, France).
Antibodies
The rat anti-human IL-2R -chain (CD25) Ab 33.B.3.1 was
kindly provided by Dr. J. Carcagne (Pasteur-Mérieux, Lyon,
France). Anti-CD3, -CD4, -CD8, -CD19, and anti-HLA DQ mAbs are from
Immunotech (Marseille, France) and the anti-IFN-
-FITC is from BD
PharMingen (San Diego, CA).
CD25+ depletion of fresh PBMC
The small fraction of "contaminating" CD25+ cells already present among unstimulated PBMC was depleted as follows: fresh PBMC (200 x 106/ml) were incubated for 20 min at 4°C in 2 ml of PBS containing 5% human serum (HS), 2 mM EDTA, and 20 µg/ml 33.B.3.1. Cells were then washed twice with 30 ml of ice-cold PBS/HS/EDTA (centrifugations were performed at 4°C at 300 x g) and the pellet was resuspended in PBS/HS/EDTA (80 µl for 10 x 106 cells). Goat anti-rat IgG MicroBeads (Miltenyi Biotec, Bergisch Gladbach, Germany) were added (20 µl for 10 x 106 cells), and the cell suspension was mixed gently and incubated for 15 min at 4°C. The cells were then washed in 25 ml of PBS/HS/EDTA and were resuspended in the same buffer (500 µl for 100 x 106 cells). Depletion of CD25+ cells was performed using the VarioMACS (Miltenyi Biotec) with an AS column according to the supplier’s instructions.
Stimulation of CD25-depleted PBMC with peptides
The CD25-depleted PBMC fractions were loaded for 2 h at 1 x 107/ml with 10 µM of G9L (Do1, Do2, and Do3), 1.25 µM of F9L (Do4), 10 µM of N9V (Do6, Do8, and Do12), or 5 µM of a peptide mixture (40–49; Do8 and Do12) in RPMI 1640 alone in a 15-ml polypropylene tube (Sarstedt, Newton, NC). Then, cells were washed twice and cultured in 80 cm2 tissue culture flasks (Nunc, Copenhagen, Denmark) at 2.5 x 106/ml in RPMI 1640, 8% HS, 1% L-glutamine, and 50 µg/ml gentamicin without cytokine for 24–96 h.
Stimulation of CD25-depleted PBMC with ALVAC (canarypox virus)
For pp65 stimulation, one-fourth of the CD25- PBMC fractions was infected at 1 x 107 cells/ml with recombinant ALVAC-pp65 expressing the pp65 protein for 60 min at 37°C in RPMI 1640 alone in a 15-ml polypropylene tube (Sarstedt) at a multiplicity of infection (MOI) of 5:1. After infection, cells were washed once and cocultured with uninfected CD25- PBMC in 80 cm2 tissue culture flasks (Nunc) at 2.5 x 106/ml in RPMI 1640, 8% HS, 1% L-glutamine, and 50 µg/ml gentamicin without cytokine for 24–96 h.
Selection of Ag-specific T cells
After 24–96 h of stimulation with peptides or recombinant ALVAC-pp65 infected autologous PBMC, cells were incubated with the 33.B.3.1 anti-CD25 mAb in the same conditions as for the CD25 depletion, i.e., 20 min at 4°C at a concentration of 2 x 108cells/ml in the presence of 20 µg/ml 33.B.3.1. After incubation, cells were washed twice and incubated with goat anti-rat MicroBeads (80 µl of cold PBS/HS/EDTA and 20 µl of goat anti-rat MicroBeads (Miltenyi Biotec) for 1 x 107 cells) for 15 min at 4°C. After washing, cells were resuspended (500 µl for 108 cells) in cold (4°C) PBS/HS/EDTA. The CD25+ selection was performed using the VarioMACS on a MS+/RS column (Miltenyi Biotec) according to the supplier’s instructions. The CD25+ fraction was then stimulated with pooled allogeneic feeder cells (5 x 106 irradiated (35 Gys) PBMC and 5 x 105 irradiated (35 Gy) BLCL) in the presence of 1 µg/ml leukoagglutinin-A (Sigma-Aldrich) and 150 U/ml rIL-2 (Proleukin, Adesleukine; Chiron, Amsterdam, Pays-Bas). Before specificity assays, cells lines were cultured in rIL-2 alone (150 U/ml) for at least 3–6 wk.
Generation of EBV-BLCL
BLCL were generated for each donor by culturing 2 x 106 PBMC in 100 µl of RPMI and 10% FCS with 500 µl of EBV-containing supernatant from the virus-producing B95.8 marmoset cell line. Cultures were performed in the presence of 1 µg/ml cyclosporin A. After 24 h, 2 ml of RPMI 1640 containing 10% FCS, 1% L-glutamine, and 50 µg/ml gentamicin was added to each well.
Target cells
Autologous or allogeneic BLCL and PHA blasts were either
infected with a recombinant vaccinia virus or loaded with a peptide.
Chromium labeling was performed before the loading and after the
infection. For infection we used recombinant vaccinia viruses encoding
for the CMV phosphoprotein pp65 (WR-pp65) or the CMV immediate-early
protein IE1 (WR-IE1). The vectors, vCP260 (ALVAC-CMV-pp65); vP1214
(WR-pp65); vCP244 (ALVAC-IE1-Exon4); and vP893 (WR-IE1), were kindly
provided by Dr. J. Tartaglia (Virogenetics Corporation, Troy, NY).
Target cells (1 x 106) were infected in
RPMI 1640 alone at a MOI of 10 for 60' at 37°C in polypropylene tubes
(Sarstedt). After infection, cells were diluted to 8 x
105 cells/ml in RPMI 1640 and 10% FCS for
overnight incubation. For loading, target cells were incubated for 30
min at 37°C in the presence of 10 µM of G9L, F9L, A9V, or N9V, or
10 µM of peptides 40–49, and were washed twice in RPMI-FCS. For
human CMV (HCMV)-infected target cells, HCMV AD169 was propagated in
MRC-5 human fibroblasts. Virus was collected when cytopathic effects
were >90%. Supernatants were clarified by centrifugation at 1500
x g for 10 min at 4°C and were stored at -80°C until
use. Virus titer was determined by PFU titration in human foreskin
fibroblasts (American Type Culture Collection, Manassas, VA) according
to standard procedures. U373 myasthenia gravis (MG) astrocytoma cells
were a gift from S. Michelson (Institut Pasteur, Paris, France). Both
MRC5 and U373 MG cells were phenotyped HLA-A2 by the Laboratoire
Central d’Immunologie-Rangueil (Prof. Ohayon, Toulouse, France).
Before cytotoxic assays, U373 MG and MRC-5 cells were incubated
overnight with IFN- (1000 U per 106 cells) and
were infected with HCMV (MOI of 3) or mock infected for 4
h.
Cytotoxic assay
Infected or loaded target cells were labeled with 100 µCi
Na251CrO4
(1 h for BLCL and PHA blasts and for the last 2 h of HCMV
infection for U373 MG and MRC5) at 37°C, washed three times, and
plated (3000–5000 cells/well) with effector cells at the
indicated E:T ratio in a 96-well round-bottom culture plate. After
4 h of incubation at 37°C, 25 µl of supernatant from each well
was removed, mixed with 100 µl of scintillation fluid (Optiphase
"Supermix"; Wallac, Courtaboeuf, France), and counted in a 
scintillation counter (Wallac). Each test was performed in triplicate.
Results are expressed as percentage of lysis, according to the
following formula: (experimental release - spontaneous
release)/(maximal release - spontaneous release) x 100,
where experimental release represents mean counts per minute released
from the target cells in the presence of effector cells; spontaneous
release represents that from target incubated without effectors; and
maximum release represents that from target incubated with 1% Triton
X-100.
Proliferation assays
Resting T cell taken >3 wk after the last stimulation were
cocultured in 96-microwell flat-bottom culture plates at a 1:1
responder:stimulator ratio (25,000 T/25,000 B) for 72 h with the
irradiated (35 Gy) autologous or allogeneic target BLCL described
above. Six hours before harvesting, 1 µCi of
[3H]thymidine was added to each well, and
[3H]thymidine uptake was then measured in a
liquid scintillation counter (Wallac). Results are expressed as the
mean of triplicate cultures.
Preparation of the CD25-depleted responder population
(referred to as the CD25- fraction), peptide
stimulation, and CD25+ cell recovery (referred to as the
CD25+ fraction) (Fig. 1
)
Because the frequency of T cells specific for a single MHC and peptide complex is expected to be low, the few PBMC already expressing CD25 had first to be eliminated. The recovery after CD25 depletion was in mean 83.5% ± 11.4%. In preliminary experiments we tested the effect of different peptide stimulation conditions on total CD25+ recovery. The CD25- fraction was either pelleted or adjusted to 107 cells/ml for 2 h before incubation with peptide (2.5 or 10 µM), incubation was performed either 18 or 72 h at 2.5 x 106/ml, and cells were then either incubated in RPMI 1640 supplemented with 5% HS or in X-vivo 15 serum-free culture medium. No dramatic differences were observed between these different conditions on total CD25+ recovery. As a negative control, CD25- fractions were processed without stimulation. Through six experiments performed with the peptide G9L, the CD25+ frequencies varied from 1/600 to 1/286 (1/540 in mean), whereas for the six negative controls, CD25+ recovery varied from 1/800 to 1/300 (1/466 in mean). That the number of CD25+ cells recovered after incubation with or without peptide were very similar was somewhat unexpected. Whether this was due to contamination of the initial CD25- fraction by CD25+ cells that escaped the depletion procedure, to nonspecific induction of the CD25 Ag during the incubation period at high cellular concentration, to delayed specific induction of the CD25 Ag among lymphocytes recently stimulated in vivo, or to nonspecific binding of 33.B.3.1 on some CD25-depleted cells before the positive selection is not known. Experiments are currently being performed to directly address this question.
|
Cells (5 x 106) were incubated with anti-CD4 or anti-CD8 mAb diluted at 1/40 (BioAtlantic, Nantes, France) for 30 min at 4°C, washed twice, and incubated at a 4:1 bead:cell ratio with Dynabeads M-450 sheep anti-mouse IgG (Dynal Biotech, Oslo, Norway) according to manufacturer’s instructions. The positive fraction was then stimulated with pooled allogeneic feeder cells in the presence of leukoagglutinin-A and rIL-2 as described above. Before specificity assays, the cell lines were cultured in rIL-2 alone (150 BRMP U/ml) for at least 3–6 wk.
Flow cytometry
Binding of anti-CD4 anti-CD8 mAb (Immunotech) was
revealed by FITC-conjugated rat anti-mouse IgG antiserum (green
fluorescence; rabbit anti-mouse-FITC; BioAtlantic). Five
thousand labeled cells were analyzed on a FACScan flow cytometer (BD
Biosciences, Mountain View, CA) using LYSIS II software. All mAbs were
used at previously tested saturating concentrations. For determination
of IFN--producing pp65-specific T lymphocytes, intracellular
cytokine assessment using flow cytometry was performed as previously
described (17), with minor modifications. Briefly, resting
T cells (taken 20 days after the last stimulation) were cocultured with
the indicated target cells at a concentration of 2.5 x
106/ml (E:T ratio of 1:1) in the presence of 10
µg/ml brefeldin A (Sigma-Aldrich) for 5 h at 37°C in a
humidified 5% CO2 incubator. After incubation,
cells were washed with cold PBS, resuspended in PBS containing 1 mM
EDTA (Amresco, Solon, OH), incubated for 10 min at 37°C, washed with
cold PBS, fixed in PBS containing 4% (w/v) paraformaldehyde (EMS, Ft.
Washington, PA) for 10 min at room temperature. Finally, cells were
washed in PBS and 1% BSA (w/v; Sigma-Aldrich). For staining,
permeabilization was performed with PBS, 0.1% BSA, and 0.1% saponin
(w/v; Sigma-Aldrich). After staining with Abs PE-CD3, PE-CD19,
(Immunotech), or FITC-conjugated mouse anti-human IFN- (BD
PharMingen) for 30 min at 4°C, cells were washed twice in PBS
containing 0.1% BSA and 0.1% saponin, once in PBS, and were then
resuspended in PBS containing 1% paraformaldehyde, and stored in the
dark at 4°C before analysis.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
)
CD25-depleted PBMC from HLA-A*0201+ or
HLA-B8+ donors were stimulated with the peptides
G9L, F9L, or N9V (see Fig. 1
and Materials and Methods).
After amplification using a procedure that preserves the initial
diversity of the population amplified (18), the
CD25+ fractions were tested for cytotoxic
activity against specific and irrelevant target cells. Results are
shown in Fig. 2, a–c, for G9L, F9L, and N9V
stimulation, respectively. In Fig. 2a, the
CD25+ fraction showed specific recognition of the
HLA-A*0201/G9L antigenic complex only, because neither the
HLA-A*0201+ target BLCL loaded with peptide A9V
or N9V nor the HLA-A*0201- target BLCL loaded
with the G9L peptide were recognized. Low background recognition of
HLA-A*0201+ BLCL loaded with an irrelevant
peptide was expected because only a small minority of BLCL cells
express proteins of the lytic cycle from which the G9L peptide is
derived. In Fig. 2b, the CD25+
fraction killed the HLA-B8+ but not the
HLA-B8- target BLCL loaded with F9L, and did not
kill the HLA-B8+ target BLCL loaded with A9V or
G9L. Consequently, this T cell population contained CTL specific for
the HLA-B8/F9L antigenic complex. Note that all BLCL used in this study
were obtained by transformation with the EBV strain derived from the
marmoset B95.8 cell line, which encodes an equivalent EBNA3A epitope
(F9I instead of F9L) that is not recognized when endogenously presented
(19). Thus, HLA-B8+ BLCL are not
recognized unless they are loaded with the wild-type peptide. Finally,
in Fig. 2c, the CD25+ fraction killed
the HLA-A2+ but not the
HLA-A2- target BLCL loaded with N9V, and did not
kill the HLA-A2+ target BLCL loaded with A9V or
G9L (Fig. 2c). Consequently, this population contained CTL
specific for the HLA-A*0201/N9V antigenic complex.
|
)
The above data proved the efficiency of our procedure and
validated the concept that PBMC stimulation with a single peptide
followed by a CD25 selection step allows the recovery of T cells with
known specificity. Next, we reasoned that if an epitope could still be
recognized by memory T lymphocytes when presented among many others,
then such a procedure could be used for the random search of new
specificities. To test this hypothesis, selection of
HLA-A*0201/N9V-specific T lymphocytes was used as a model.
CD25- PBMC from two CMV seropositive individuals
(donors 8 and 12) were stimulated under the conditions described
above, but with a mixture of 10 23 mer (5 µM) peptides spanning
one-fifth of the pp65 protein instead of the single N9V peptide (the 23
mer 45 encompasses the nonamer N9V). After amplification, the
CD25+ fractions were tested against
HLA-A2+ and HLA-A2- BLCL
loaded individually with each of the 23 mer peptides present within the
pool used for stimulation, or with N9V peptide (see Fig. 3). No
significant response was observed against peptides loaded or unloaded
HLA-A2- BLCL (Fig. 3). No response against any
target was detected in the CD25+ fraction of PBMC
cultured in the absence of the peptide pool (not shown). In contrast,
the CD25+ fraction selected after stimulation
with the mixture of 23 mer peptides showed two readily detectable
cytotoxic responses: one against HLA-A2+ BLCL
loaded with the 23 mer 45, which contain the N9V sequence, and the
other against the nonamer N9V itself. Consequently, these data
demonstrated that the nonameric epitope N9V, even when provided within
one of 10 overlapping 23 mer, could still be recognized by
HLA-A*0201/N9V-specific T lymphocytes. Thus, our procedure seems
particularly attractive for the screening of a limited set of peptides
covering a limited protein area.
|
and 5)
Finally, keeping in mind the objective to apply such a T
cell selection strategy in a clinical setting, we tested a canarypox
viral vector to induce expression of the entire pp65-protein among
fresh CD25- PBMC. Naturally attenuated canarypox
(ALVAC) constructs have been shown to be an efficient tool in the
induction of protective immunity in vivo (20, 21, 22) and also
in the ex vivo activation of cytotoxic T lymphocytes (23).
These vectors, which retain the pancytotropism of most pox viruses, are
unable to productively replicate in nonavian species, and thus
eliminate the safety concerns that exist for vaccinia vectors.
CD25- PBMC from donors 8, 12, 15, 20, 22, and 26
were stimulated with ALVAC-pp65 according to the protocol described
Materials and Methods. For donors 8 and 12, this assay was
performed in parallel with the other procedures described above, i.e.,
a stimulation with the pool of peptides (40–49) or the peptide N9V
alone. After amplification, the CD25+ fractions
were tested against an HLA-A2+ BLCL either loaded
with a peptide (N9V or G9L), infected with a canarypox vector
(ALVAC-pp65 or ALVAC-IE1), or infected with a recombinant vaccinia
virus (WR-pp65 or WR-IE1). Results obtained with donor 8 are reported
in Fig. 4 as the percentage of specific lysis observed at an E:T ratio
of 10:1. For the three cases, CD25-selected PBMC recognized the
HLA-A2+ BLCL when loaded with the N9V but not
with the G9L control peptide. They also recognized the same target BLCL
when it was infected with the recombinant vaccinia virus WR-pp65 but
not with the recombinant virus expressing the immediate early protein
IE1 (similar results were obtained with donor 12). Determination of
pp65 expression levels by intracellular staining after BLCL infection
with ALVAC-pp65 or WR-pp65 indicated that canarypox vectors infect BLCL
significantly less efficiently than vaccinia vectors (data not shown).
Accordingly, although HLA-A2+ BLCL infected with
the ALVAC-pp65 showed some recognition compared with ALVAC-IE1
infected BLCL, the level of cytotoxicity observed was well below that
obtained against recombinant vaccinia virus-infected BLCL. Remarkably,
the three CD25+ fractions had a comparable level
of cytotoxic activity against the target cells either loaded with a
single peptide or infected with a vector encoding the entire protein.
Finally, the CD25+ fractions from ALVAC-pp65
stimulated PBMC of donors 8, 20, 22, and 26 were tested against the
human HLA-A2 U373 MG or MRC-5 cell lines infected or not with HCMV
AD169. Results presented in Fig. 5 demonstrate that all CD25-selected
fractions showed specific cytotoxic activity against these actual
HCMV-infected target cells. Although a systematic analysis at the
clonal level will be necessary to document the different specificities
present among T cell population selected with the pool of peptide and
the ALVAC vector, taken together, these results stress the dominance of
the response directed at the HLA-A2 N9V CMV epitope. In total, specific
T lymphocytes were isolated after stimulation with N9V, the peptide
pool, or ALVAC-pp65 from 3 of 5, 2 of 3, and 6 of 11 of the donors
tested, respectively.
|
|
)
Memory/effector CD4+ (Th1-type) and
CD8+ T lymphocytes have the capability to secrete
effector cytokines like IFN- following short-term antigenic
restimulation. To obtain a minimal estimation of the enrichment of
pp65-specific T lymphocytes among the CD25+
fraction, fresh PBMC and resting ALVAC-pp65-stimulated PBMC (tested 20
days after the last stimulation) were tested for their capacity to
secrete IFN- following stimulation against the autologous BLCL alone
or transfected with pp65 or IE1 (an example of such a determination for
donor 8 is shown in Fig. 6a). Results obtained for the five
donors tested are shown in Fig. 5b: minimal estimation of
pp65-specific T lymphocytes ranged from 7.9 to 76% (Fig. 6b). Among pp65-unselected PBMC of the same donors, the
frequencies of pp65-specific cells ranged from 0.2 to 2.2% (data not
shown). To evaluate the speed of amplification of the populations of
interest, the mean doubling time of 28 CD25+
fractions was evaluated and found to be 33.3 h. Fig. 6c
shows the theoretical growth of a CD25+ fraction
and the particular example of donor 8 ALVAC-pp65-selected cells. In
this case, the pp65-specific T lymphocytes were amplified 583-fold
within 13 days.
|
)
To determine whether CD25+ fractions
selected after ALVAC-pp65 stimulation contained pp65-specific T
lymphocytes other than the HLA-A*0201/N9V-specific
CD8+ cells described above, they were further
separated into CD4+ and
CD8+ fractions. After immunomagnetic separation
and amplification, each fraction was tested for purity according to
their phenotype and for specificity using proliferation and cytotoxic
assays. The results are shown in Fig. 7. The CD4+
fraction, whose purity is indicated in the top panel, was tested
against the autologous BLCL either loaded with the N9V or an irrelevant
peptide (G9L) or transfected with a vaccinia virus encoding pp65
(WR-pp65) or IE1 (WR-IE1). As shown, the CD4+
fraction contained T lymphocytes specific for the pp65-transfected
autologous BLCL only. This result proved that canarypox virus-infected
HLA class II PBMC-bearing cells, likely B lymphocytes or monocytes,
were responsible for this CD4+ T lymphocyte
activation. As expected, no reactivity was detected against the HLA
class I-restricted epitope N9V. The CD8+ fraction
was cytotoxic for the autologous BLCL transfected with the WR-pp65
vaccinia virus and also with the autologous BLCL loaded with the
peptide N9V, but not against the autologous BLCL transfected with the
WR-IE1 vaccinia virus or loaded with the G9L control peptide. These
data demonstrated that pp65-specific T lymphocytes were present in both
the CD4+ and CD8+ fraction
and the data also confirm the immunodominance of the N9V epitope among
pp65-specific CD8 T cells. From the 11 donors tested, specific T cells
could be recovered after stimulation with ALVAC-pp65 in six cases and
both CD4+- and
CD8+-specific T cells were obtained in five of
these six responding donors.
|
)
To precisely document the presence of CD4-specific T lymphocytes
among pp65-selected PBMC, CD4 T cell clones were derived by limiting
dilution from the bulk cultures of four different donors (8, 12, 15, 20). After cloning, each clone was tested against the autologous
BLCL loaded with one of the 50 peptides covering the entire pp65
sequence. Fig. 8 shows an example of peptide identification.
Twenty-seven of the 28 CD4+ T cell clones tested
showed the pattern of reactivity presented, strongly suggesting that
this particular CD4+ population contained only
one or a few specific distinct T cell clones. Proliferation was
observed against the autologous BLCL only when it was loaded with
peptide 4 and not with any other peptide from the panel (Fig. 8a). Further testing with five overlapping shorter peptides
covering the sequence of peptide 4 allowed identification of the
minimal peptide L12Q, whose reactivity is shown in Fig. 8b.
Finally, this recognition was abrogated in the presence of an
HLA-DQ-specific mAb (not shown) and was observed against
HLA-DQ0602+ target BLCL only (Fig. 8c), thus demonstrating that HLA-DQ0602 was the restricting
element. Using this approach, CD4+ T cell clones
specific for peptides 4, 14, 34, and 45 were identified from donors 8,
12, 15, and 20, and CD8+ T cell clones specific
for peptides 25, 30, and 33 were identified from donors 8 and 20. Table I summarizes the panel of pp65 epitopes
identified during this study. In the case of donors 8, 12, and 15, the
testing of shorter peptides derived from the 23 mer initially used for
screening allowed identification of the minimal epitopes underlined in
Table I.
|
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
following a short-term
antigenic restimulation with synthetic peptides. To purify
IFN--secreting cells, the authors developed a so-called affinity
matrix technology, which first consists in creating an affinity matrix
for IFN- on the cell surface using Ab-Ab conjugates directed against
CD45 and IFN- (anti-IFN-CD45). Then, specific T lymphocytes
are allowed to secrete IFN- for a short period of time, which then
relocate on the Ab-Ab conjugates. Next, IFN- is stained with a
PE-conjugated IFN--specific Ab, and finally, magnetic activated cell
sorting using anti-PE Ab microbeads can enrich PE-labeled cells.
The efficiency of this procedure has been shown for Flu 58–66
peptide-specific IFN--producing T cells and for recombinant tetanus
toxoid Th2-type IL4-secreting CD4+ T
lymphocytes.
The tetramer technology is limited to T cells with known specificities.
The affinity matrix technology is limited to T cells that secrete a
particular cytokine. This former limitation can become a significant
concern if one wants to recover T cells from all components of a
particular memory T cell repertoire. For example, it has been recently
demonstrated that immunological memory is displayed by distinct T cell
subsets: CD45RA-CCR7+
cells corresponding to lymph node-homing cells lacking inflammatory and
cytotoxic function (defined by the authors as central memory T cells)
and CD45RA-CCR7- cells
corresponding to tissue-homing cells having various effector functions
and, in particular, the ability to secrete IFN-, IL-4, and IL-5. The
authors defined these cells as effector memory T cells. Because
different memory subsets display different cytokine profiles, this
could render their global purification even more complicated using the
affinity matrix technology.
Clonal amplification is the basis of the T cell response, and CD25 expression is a prerequisite for T cell proliferation. Curiously, although CD25 selection had been considered >10 years ago for the enrichment of specific T cells before cloning (22), to our knowledge, no systematic approach for direct amplification of Ag-specific T cells using this principle has been developed so far. The end point of the present work was to find a clinically suitable strategy able to select in any genetic background the memory T cell repertoire specific for a viral protein.
We demonstrated that virus-specific memory T cells can be purified through CD25 selection after direct stimulation of PBMC with a peptide, a mixture of peptide, and finally a viral vector encoding an entire protein. In particular, direct purification of pp65-specific CD8+ and CD4+ T cells after a single PBMC stimulation with a canarypox viral vector encoding the entire pp65 protein strongly suggests that this method is probably bound to become the most straightforward approach to purifying specific memory T lymphocytes against a protein of interest, irrespective of their genetic background. After ALVAC-pp65 stimulation, pp65-specific T lymphocytes could be isolated from six of 11 of the CMV seropositive donors tested. Five of these six donors presented both a CD8+ and a CD4+ positive response. Several reasons can account for the fact that pp65-specific T lymphocytes were not isolated from all the donors tested. Although early studies have suggested that the human CTL response to CMV is dominated by CTL against pp65 (28), the major immediate early protein (IE-1) has also been recognized as an important CTL target; thus, for some donors, the frequency of pp65-specific T cells may have been too low to be isolated by our technique. In addition, Kern et al. (29) demonstrated that in some individuals, CD8+ T cells recognized IE-1 but not pp65. Nevertheless, we do not favor the latter explanation because in Kern’s experience, all donors nonresponsive to pp65 were HLA-A2 negatives, contrary to all the donors tested in the present study.
Although we demonstrated the possibility of probing the T cell repertoire for the presence of yet unknown specificities, other strategies may be consider to this end. Intracellular cytokine staining and ELISPOT assays have been used successfully for enumeration and characterization of Ag-specific CD4+ and CD8+ T cells (1, 2), and indeed have proved to be quite useful for T cell epitope mapping (3). If obtaining the corresponding Ag-specific T cell populations in culture is not necessary, these latter strategies are obviously less cumbersome. In contrast, in as far as live Ag-specific T cell purification is required, these methods cannot be used. To isolate as many Ag-specific T cells as possible, two other methods are currently practical: the use of tetrameric complexes or the affinity matrix technology. As stated previously, tetramer technology is limited to T cells with known specificity, and the affinity matrix technology is limited to T cells that secrete a particular cytokine. Each method presents some advantages and some drawbacks, depending on the application. For example, tetramer are best suited to enumerating T cells with already known specificity and T cells that are independent of their functional status (30, 31). In contrast, the affinity matrix technology enables selective isolation of Ag-specific T cells with particular cytokine-mediated effector function (25).
In general terms, the CD25 strategy is not limited by a structural
criterion (that is, the MHC-peptide complex of the tetramer) nor by a
specific functional status (that is, the ability to secrete a
particular cytokine). Moreover, anti-CD25 mAb are widely available
compared with tetrameric complexes or the Ab-Ab conjugates required by
the affinity matrix technology. Beyond their particular advantages for
specific application, it is, to date, difficult to compare the
intrinsic performance of each protocol in term of efficiency for
purification. Indeed, although data concerning Ag-specific T cell
detection are numerous, data concerning Ag-specific T cells
purification using these protocols are sparse. In any case, efficiency
of cell purification is greatly influenced by the percentage of
specific cells in the sample (32). Because in the context
of a clinical application we don’t have the choice of the initial
sample, purity is likely to remain a poorly predictable parameter. To
our understanding, the only way to ensure 100% purity of the selected
population would be to clone directly T cells after the purification
step and to test individual clones for specificity before pooling them
to obtain a polyclonal, yet pure, Ag-specific T cell population. In
addition, in the context of allotransplantation, one can consider the
possibility of using the same protocol, but for negative selection,
deleting alloreactive T cells from the sample before positive selection
of viral Ag-specific T cells. Aside from purity, the total number of T
cells recovered is also an important parameter to be considered.
Indeed, after selection, Ag-specific T cells have to be amplified to
reach sufficient number for further testing and/or for reinjection. For
many years, nonspecific stimulation procedures have been used to
amplify T lymphocytes (33, 34). When optimal, such
procedures allow amplification of all T cells present in the culture
and consequently do not affect their initial diversity
(18). These methods rely on the use of large excess of
autologous (when available) or allogeneic feeder cells made of PBMC,
BLCL, a polyclonal T cell activator such as PHA or an anti-CD3, and
IL-2. The growth rate of T lymphocytes cultured under these conditions
corresponds to a doubling time of between 24 and 35 h. According
to previously published results, therapeutic doses of Ag-specific T
cells probably comprise between 108 (4 x
107/m2 for EBV in the
experience of C. Rooney et al. (2) and several billions in
the case of CMV in the experience of S. Ridell et al.
(3)). In the example presented in Fig. 6c,
1.3 x 108 pp65-selected T cells were
obtained within 13 days with a purity of 64% as detected by IFN-
production, starting with 108 PBMC. At present,
the exact composition a virus-specific T cell repertoire should have to
protect or cure a patient is not really known. That is, how many clones
against how many Ags; what we have called the "sufficient T cell
repertoire" (8). Nonetheless, whatever the number of
Ag-specific T cell targeted, it is likely that the time required for
their in vitro amplification will not be reduced to less than 2–3 wk.
In any case, in the context of clinical application, both the success
of the Ag-specific T cell isolation (which will be affected by the
frequency and the total number of Ag-specific T cells in the sample)
and the biological security of the preparation (associated with the
origin of the biological reagent used) will be directly dependant on
the initial volume of blood available. The larger the sample, the
higher the number of T cell precursors as well as the number of
autologous cells are available for use as feeder cells. Finally, the
new clinical possibilities offered by the strategy presented can be
rapidly tested because its relies on methods and reagents, namely
immunomagnetic sorting, canarypox vector, and anti-CD25 mAbs, that
have already been validated for clinical applications.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Addres correspondence and reprint requests to Dr. H. Vié, Institut National de la Santé et de la Recherche Médicale Unité 463, Institut de Biologie, 9 Quai Moncousu, 44035 Nantes cedex, France. E-mail address: hvie{at}nantes.inserm.fr
3 Abbreviations used in this paper: BLCL, B lymphoblastoid cell line; HS, human serum; MOI, multiplicity of infection; HCMV, human CMV; MG, myasthenia gravis.
Received for publication October 13, 2000. Accepted for publication August 9, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain (CD25)-positive selection. J. Immunol. 164:4924.
of Epstein-Barr virus. J. Virol. 69:4872.[Abstract]
expression among V- subsets: Evidence for tight regulation of the combinatorial diversity. Ann. N. Y. Acad. Sci. 756:84.[Medline]
interferon. J. Virol. 74:10249.This article has been cited by other articles:
|
|
 |
|
  G. Li Pira, L. Bottone, F. Ivaldi, R. Pelizzoli, F. Del Galdo, L. Lozzi, L. Bracci, A. Loregian, G. Palu, R. De Palma, et al. Identification of new Th peptides from the cytomegalovirus protein pp65 to design a peptide library for generation of CD4 T cell lines for cellular immunoreconstitution Int. Immunol., May 1, 2004; 16(5): 635 - 642. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. A. Papanicolaou, J.-B. Latouche, C. Tan, J. Dupont, J. Stiles, E. G. Pamer, and M. Sadelain Rapid expansion of cytomegalovirus-specific cytotoxic T lymphocytes by artificial antigen-presenting cells expressing a single HLA allele Blood, October 1, 2003; 102(7): 2498 - 2505. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Allart, J. Lule, B. Serres, T. Jones, J.-L. Davignon, F. Malecaze, and C. Davrinche Impaired Killing of HCMV-Infected Retinal Pigment Epithelial Cells by Anti-pp65 CD8+ Cytotoxic T Cells Invest. Ophthalmol. Vis. Sci., February 1, 2003; 44(2): 665 - 671. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||
