| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
,
*
Center for Cell and Gene Therapy and Departments of
Pediatrics,
Surgery, and
Molecular Virology and Microbiology, Baylor College of Medicine, Houston, TX 77030
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Restoration of T cell immunity by infusion of ex vivo expanded
EBV-specific CTL has been shown to be safe and effective in treating
and preventing EBV-related PTLD after hemopoietic stem cell or SOT
(11, 12, 13, 14, 15). EBV-specific CTL can be reactivated from
EBV-seropositive individuals by stimulation of peripheral blood T cells
with autologous irradiated EBV-transformed lymphoblastoid B cell lines
(LCL) (11, 16). Although LCL are potent APCs, they seem
incapable of inducing a primary immune response to EBV in vitro
(11, 12, 13, 14, 15, 16, 17). This has been a major problem for the high-risk
seronegative recipients of EBV-seropositive organ grafts, who are the
individuals most at risk (8). Although it has been
possible to reactivate and expand ex vivo EBV-specific CTL from SOT
recipients after their primary infection, CTL generation takes 
12
wk, leaving patients highly vulnerable during this time
(15). Ideally, CTL for high risk patients should be
generated prophylactically, preferably before transplant. Because
partially HLA-matched CTL lines from CTL banks are unlikely to survive
for long after the adoptive transfer into the allogeneic host, we have
considered alternative strategies for the ex vivo production of
autologous EBV-specific CTL from seronegative patients.
One approach is to use APCs, such as dendritic cells (DC), that are assumed to be more potent than LCL. DC have been considered to be the only APC type capable of inducing primary responses in vitro (18). An alternative approach is to selectively expand the rare circulating EBV-specific precursors, present even in EBV-seronegative patients. It has been suggested that only 30–70% of the T cells in CTL lines reactivated from seropositive individuals by stimulation with the autologous LCL are EBV specific, while the remaining cells are most likely nonspecific bystander T cells (19). Because the precursor frequency of EBV-specific T cells in seropositive individuals is high (0.1–2% of circulating T cells), the specific cells remain in the majority and their function is readily detected. In contrast, the precursor frequency of EBV-specific CTL in seronegative individuals is extremely low (<1:100,000), so that nonspecific bystander cells may dominate the cultures, and the function of the specific CTL may not be detected (20).
The aim of this study was to identify a reliable method for the generation of EBV-specific CTL from the peripheral blood of EBV-seronegative individuals awaiting SOT for future clinical application. We compared LCL with three formulations of APCs (DC pulsed with apoptotic/necrotic LCL, or with freeze-thaw lysates of LCL, or DC fused to LCL). We also analyzed the effects of selectively expanding CD25-positive T cells after activation with LCL alone, an approach previously applied in EBV-seropositive donors to enrich the population for EBV-specific CTL (19). Our results demonstrate a biologic difference between EBV-seronegative adults and EBV-seronegative children. Thus, the standard protocol for EBV-CTL generation effectively reactivated EBV-specific CTL lines from seronegative adults, but for seronegative children, an additional CD25 selection step was required. Because we were able to detect EBV-DNA in the peripheral blood of three of the four seronegative adults, we conclude that many seronegative adults may in fact have been exposed to EBV. Our method of expanding Ag-specific CTLs from naive donors may also be applied generally to the generation of tumor-specific CTL from cancer patients whose tumors have poor Ag-presenting function.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Peripheral blood samples were collected from 10 EBV-seronegative
donors (six children, 1–8 years old; four adults, 28–44 years old).
Eight adult EBV-seropositive donors were used as controls in some
experiments. Informed consent was given in all these cases by the
donors themselves or by their respective guardians. The characteristics
and serology of the seronegative individuals are summarized in Table I
.
|
PBMC, enriched using Ficoll gradient centrifugation (Lymphoprep medium 1.077, Life Technologies, Grand Island, NY), were directly processed or frozen for further analysis. Serum or plasma samples were collected and stored.
Detection of anti-VCA and anti-EBNA Abs in serum/plasma. Both IgG and IgM anti-viral capsid Ag (VCA) and anti-nuclear EBV (EBNA) Abs were determined on serum/plasma samples, using routine ELISA (Department of Pathology, Texas Children’s Hospital, Houston, TX) or immunofluorescence staining (21). Donors were considered EBV seronegative if the anti-VCA and anti-EBNA titers were <1:10 (below the level of detection).
Detection of EBV-DNA in PBMC. DNA was isolated from 5 x 106 PBMC using an anion exchange column (Qiagen, Valencia, CA). Five hundred nanograms of DNA were then tested by real-time PCR to quantitate EBV genome numbers, as previously reported (22). The sensitivity of our PCR was four genomes per microgram of DNA. The median EBV copy number detected in normal seropositive individuals was <4 (range <4–200) copies/µg DNA.
Generation of EBV-transformed B cell lines (LCL)
For LCL generation, five million fresh or frozen PBMC were infected with concentrated supernatant from the B95-8 EBV producer cell line, as previously reported (22, 23). All cell lines were cultured in complete medium consisting of RPMI 1640 (HyClone Laboratories, Logan, UT) supplemented with 10% FBS (HyClone) and 2 mmol/L L-glutamine (Life Technologies).
Generation of DC
DC were generated from frozen PBMC. Briefly, PBMC were cultured in AIM-V medium (Life Technologies) supplemented with 10% human serum type AB (HABS; C-SIX Diagnostics, Germantown, WI), 1000 U/ml human rIL-4 (R&D Systems, Minneapolis, MN), and 800 U/ml human rGM-CSF (Leukine; Immunex, Seattle, WA) for 3 days, and then applied on a 48.2% Percoll gradient (v/v; Amersham Pharmacia Biotech, Piscataway, NJ) for DC enrichment (recovery of 0.5–1%). DC were then plated at 1 x 106/ml in AIM-V medium supplemented with 10% HABS and fresh IL-4/GM-CSF. On day 7, the DC were induced to mature by culture in 50% monocyte-conditioned medium (MCM) (24). Using this procedure, we were reproducibly able to obtain a population of large cells (40–66%), positive for HLA-DR and negative for lineage (CD3 for T cells, CD19 for B cells, CD16 for NK cells, CD56 for lymphokine-activated killer (LAK) cells, and CD14 for monocytes) markers (80 ± 7%). In addition, >90% of these cells expressed CD80, CD86, CD40, and HLA-ABC Ags. After maturation, the percentage of CD83+ cells ranged from 15 to 40% of the large cells.
DC loading. To induce apoptosis/necrosis, LCL were irradiated at 4000 rad and then cultured in serum-free medium for 24–72 h, after which 47 ± 14% of the cells were apoptotic and 22 ± 15% were necrotic, as measured by annexin V and propidium iodide staining, according to manufacturer’s instructions (BD PharMingen, San Jose, CA). The LCL were then incubated with DC at a 1:1 ratio for 24 h, before maturation in MCM. After a 48-h maturation, the LCL-pulsed DC were used to stimulate autologous PBMC. Freeze-thaw lysates of LCL were generated as reported by Herr et al. (25), and the lysate was added to autologous DC obtained after Percoll enrichment. After maturation for 48 h with MCM, DC were used for CTL activation.
Fusion of LCL to DC. LCL that had been transduced with a retrovirus vector expressing enhanced green fluorescent protein (eGFP) and selected for eGFP expression were used as the fusion partner, to allow for further analysis. Percoll-enriched DC were fused to eGFP-LCL using 50% polyethylene glycol (Sigma-Aldrich, St. Louis, MO), following the procedure of Gong et al. (26). Fusion hybrid DC/LCL were then plated for 48 h in AIM-V medium plus 10% HABS and fresh IL-4/GM-CSF, then used for CTL activation. Hybrid DC-LCL formation (43%) was demonstrated by acquisition of green fluorescent protein positivity by an HLA-DR-positive, lineage-negative large cell population.
Determination of cellular immunity to EBV
Cellular immunity to EBV was measured by regression assay, which measures the number of PBMC required to produce complete regression of autologous EBV-infected B cells; ELISPOT, which measures the number of cells that secrete cytokines in response to stimulation with the LCL; and by generation of EBV-specific CTL, which determines the presence and function of T cells that can be expanded in culture in response to EBV Ags.
Regression assay. The precursor frequency of EBV-specific CTL was measured by regression assay (27). Briefly, after infection with concentrated supernatant from the EBV producer cell line, B95-8 EBV, PBMC were plated in complete medium (without cyclosporin A) at doubling dilutions, from 6 x 105 to 1 x 104 cells/well of a flat-bottom 96-well plate, using six replicates per dilution. Cultures were maintained by regular feeding and examined for foci of virus-induced transformation and subsequent regression over 6 wk. The regression was expressed in terms of the minimum initial cell number required per well for a 50% frequency of regression.
ELISPOT for IFN-
-secreting cells.
The frequency of IFN--producing cells in response to LCL-specific
stimulation was assessed on PBMC in toto by ELISPOT assay, as reported
by Ambinder et al. (28). Briefly, MAHA S45 plates
(Millipore, Bedford, MA) were coated with anti-IFN- Ab 1 DIK
(Mabtech, Stockholm, Sweden) overnight and blocked with complete medium
for 1 h at 37°C. PBMC were added at doubling dilution from
1 x 105/well in presence of autologous LCL
for 24 h at 37°C and washed off, and then plates were incubated
with biotin anti-IFN- Ab 7-B6-1 (Mabtech). Appropriate controls
consisting of PBMC, LCL, and medium alone were plated and incubated
with biotin anti-IFN- Ab 7-B6-1 as well. Streptavidin-alkaline
phosphatase (Life Technologies) was added for 1 h at room
temperature, and spots were developed with nitroblue
tetrazolium/5-bromo-4-chloro-3-indolyl phosphate substrate mix
(Life Technologies). Spots were counted (Zellnet Consulting, New
York, NY) and expressed as number of spots/1 x
105 cells when dilution was linear.
Generation of EBV-specific T cell lines
Activation with LCL. As previously reported (11, 22), PBMC were activated with autologous, irradiated (4000 rad) LCL, at an E:S ratio of 40:1, in complete medium. After 10 days, viable cells were restimulated with LCL at a 4:1 E:S ratio. From days 15–17, 40 U/ml human rIL-2 (Proleukin; Chiron, Emeryville, CA) was added. Cells were then expanded by weekly restimulation with LCL (4:1 ratio) and twice weekly addition of IL-2 (40–100 U/ml) until numbers sufficient for specificity analysis were obtained. This was referred to as the standard protocol.
Activation using DC. PBMC were cocultured with irradiated (4000 rad) DC that had been loaded with apoptotic/necrotic LCL, or with LCL lysate, or fused with LCL, all used at an E:S ratio of 20:1. At 10-day intervals, cells were restimulated using irradiated, autologous LCL in the presence of IL-2, as per the standard protocol.
Activation using LCL and selection of CD25-positive cells. PBMC cultured for 9–11 days with irradiated, autologous LCL at an E:S ratio of 40:1 were stained with CD25-PE Ab (20 µl for 107 cells; BD Biosciences, Mountain View, CA) for 30 min at 4°C, washed, resuspended in 500 µl of sterile PBS supplemented with 1% FBS, and sorted using a FACScan flow cytometer (BD Biosciences). From seronegative children, an average of 4.4 ± 3.4% of bright CD25+ cells were selected (purity from 90 to 97%). After sorting, CD25-enriched cells were expanded using the standard protocol. In some cases, CTL culture medium contained 10% HABS instead of FBS.
Immunophenotyping
For phenotypic evaluation, cells were stained with combinations
of the following Abs: CD83 and CD56 from Immunotech (Miami, FL); CD80
and CD86 from BD PharMingen; CD3, CD4, CD8, CD14, CD16, CD19, CD25,
TCR
, TCR
, HLA-DR, and HLA-ABC from BD Biosciences. Control
samples labeled with appropriate isotype-matched Abs (BD Biosciences)
were used in each experiment. After staining, cells were evaluated on a
FACScan analyzer (BD Biosciences) equipped with the filter set for
triple fluorescence signals. In addition, the TCR-V region of the
expanded CTL was studied using the TCR-V repertoire kit (IOTest
Mark kit; Immunotech), according to manufacturer’s instructions.
Chromium release assays
The cytotoxic activity of each CTL line was evaluated in a standard 5-h 51Cr release assay, as previously reported (22, 23). Autologous LCL, HLA class I- and/or II-mismatched LCL, and autologous PHA blasts were used as the target cells. In addition, the EBV-negative K562 chronic-erythroid-leukemia or the HSB-2 T cell lymphoma cell lines were used as indicators of NK and LAK cells, respectively. Preincubation for 30 min with either the mAb W6/32 (DAKO, Carpenteria, CA) or the mAb CR3/43 (DAKO) was used to determine whether the killing was HLA class I or II restricted, respectively. If cytolytic activity against HSB-2 or K562 was high, CTL were depleted of CD56-positive cells using magnetic cell sorting, according to the manufacturer’s instructions (MACS-CD56 MicroBeads, and AS column for negative selection; Miltenyi Biotec, Bergisch Gladbach, Germany). CD56-depleted CTL were then tested for cytotoxic specificity. In some experiments, cold target inhibition assays were performed as follows: 2-fold dilutions of unlabeled competitor cells, ranging from 3.2 x 105 to 2 x 104, were incubated with a constant number (104) of 51Cr-labeled target cells and a constant number of effector cells (105 cells) in a standard 5-h 51Cr release assay.
ELISA
The IFN- human ELISA system (BD PharMingen) was used to
measure the release of cytokines in culture medium. Supernatant was
collected 24 h after specific antigenic stimulation and analyzed
according to the manufacturer’s instructions.
Statistical analysis
Student’s t test was used to determine the statistical significance of differences between samples. All data are presented as mean ± 1 SD.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
To assess the ability of LCL to induce EBV-specific CTL expansion from naive individuals, the standard protocol was first tested in our series of EBV-negative donors.
When LCL-activated cell lines from the six EBV-seronegative children
were evaluated after their fourth LCL stimulation, the majority of the
expanded cells were
CD3+CD4+ (51 ± 19%),
with <10% of the cells CD8+ or
CD56+ (mean of 8 ± 8% and 7 ± 9%,
respectively). The remaining cells were CD19-positive LCL (Table II). The total expansion over this period
ranged from 0.3- to 5-fold (mean of 2.7 ± 2.1). These T cells
lysed neither the autologous LCL (12 ± 11% at 40:1 ratio) nor
allogeneic LCL (11 ± 10%, at 40:1 E:T ratio) (Fig. 1A). However, high lysis of
the HSB-2 cell line was observed in three cases (54, 73, and 90% at
the 40:1 ratio), suggesting the presence of nonspecific LAK cells. The
lack of specific killing by T cell lines from seronegative children
correlated with negligible IFN- release in response to autologous or
allogeneic LCL, as measured by ELISA (data not shown).
|
|
B, the expanded
lymphocytes demonstrated significant killing of the autologous LCL
(68 ± 14% at a 20:1 ratio) compared with allogeneic LCL (6
± 9% at a 20:1 ratio) or the HSB-2 cell line (17 ± 17% at a
20:1 ratio). This killing was MHC class I restricted, because it was
inhibited by the anti-HLA class I mAb, W6/32 (mean inhibition,
39%; range, 17–82%). The majority of the T cells were
CD3+CD8+ (88 ± 8%),
with fewer CD3+CD4+ T cells
(11 ± 9%). A significant percentage of the
CD3+ cells had a LAK cell phenotype, coexpressing
the CD8 and CD56 Ags (30 ± 11%, Table II). Depletion of CD56
cells reduced the lysis of the HSB-2 cell line to 2%, while killing of
autologous LCL remained unmodified. In no case was lysis of autologous
PHA blasts or fibroblasts observed (data not shown), an important
specificity requirement if these CTL lines are to be used clinically.
The specific cytolytic activity correlated with high IFN- release in
response to stimulation with the autologous LCL (2140 ± 540
pg/ml x 106 cells). Although the CTL from
seronegative adults demonstrated a phenotype and cytotoxic activity
comparable with CTL generated from our series of normal healthy
EBV-seropositive donors, they grew with slower kinetics. The total fold
expansion after the fourth stimulation was 2.8 ± 0.5 (range
2.3–3.3) vs 5.5 ± 4.2 (range 0.6–12) in CTL generated from
healthy donors. EBV-seronegative adults may be EBV carriers
To investigate why LCL were able to activate and expand EBV-specific CTL in the group of EBV-seronegative (and presumed uninfected) adults, we tested our donors for the presence of EBV-DNA and EBV-specific T cell memory.
As shown in Table I, Abs to VCA and EBNA were undetectable in plasma or
serum samples from all EBV-seronegative donors. However, EBV-DNA was
sporadically detected in the peripheral blood of three EBV-seronegative
adults (10–60 genome copies/µg DNA). The mean level of EBV-DNA that
we have previously found in normal seropositive individuals was 10
genomes/µg DNA (median <4, range <4–200 copies/µg DNA). Negative
controls run for each PCR assay excluded the possibility of
contamination or cross-contamination. EBV-DNA was never detected in the
fourth adult or in any of the seronegative children.
To further quantitate the EBV-specific immune response, we used the
regression assay, developed by Rickinson et al. (27). In
both seropositive and seronegative donors, PBMC exposed to EBV in vitro
developed foci of proliferating EBV-transformed B cells within the
first 2 wk of culture. Regression of the culture was seen in all
EBV-seropositive donors and in three of the four EBV-seronegative
adults: the foci of EBV transformation degenerated in the following 3
wk and failed to give rise to LCL (Table I). In contrast, none of the
EBV-seronegative children showed regression, even in the presence of a
high number of PBMC. This suggested that at least three of the four
EBV-seronegative adults tested had recent or persistent infection with
EBV, without developing serological evidence of this encounter.
Finally, we investigated the frequency of EBV-specific CTL by ELISPOT
assay (Table I). In three seronegative adults, the number of T cells
that secreted IFN- in response to LCL stimulation was 651 ±
110/105 PBMC. Similar frequencies were detected
in EBV-seropositive healthy donors (723 ± 180
spots/105 PBMC). The fourth adult had a frequency
of EBV-specific precursor (20 spots/105 cells)
within the range observed in the EBV-seronegative children (101 ±
100 spots/105 cells). These low numbers indicate
a lack of EBV-specific memory T cells in the peripheral blood of these
individuals. For the fourth seronegative adult, the presence of
regression and the ability to develop an EBV-specific CTL using the
standard protocol, despite negativity for EBV-DNA and low frequency of
IFN- spots, may be explained by the presence of CTL precursors that
cross-react with EBV Ags (molecular mimicry).
EBV-Ag-loaded DC induce EBV-specific CTL from some, but not all, EBV-seronegative donors
To determine whether more potent APCs would be able to activate EBV-specific CTL from the EBV-negative children, unresponsive to standard LCL expansion, we used DC loaded with EBV Ags as stimulators. Strategies for the loading of these DC with EBV Ags were to coculture them with apoptotic/necrotic LCL (29), or with lysates of LCL (25), or to fuse them with LCL (26).
EBV-specific CTL could be generated from only two of the four
EBV-seronegative children tested, using DC loaded with
apoptotic/necrotic LCL. The CTL lines from these two children were
predominantly CD4+ T cells (72 ± 12%)
(Table II) and killed the autologous LCL more then the allogeneic LCL
and the HSB-2 cell line (78 ± 19% vs 11 ± 11% vs 33
± 15% at 20:1 ratio, respectively) (Fig. 2A). These CTL lines also
secreted IFN- in response to the autologous LCL (528 vs 159
pg/ml x 106 cells in presence of autologous
LCL vs allogeneic LCL, respectively). No lysis of autologous PHA blasts
was observed (data not shown). In the other two individuals, there was
no difference in the killing of autologous or allogeneic LCL and HSB-2,
suggesting that only NK/LAK cell-mediated activity had been expanded
(data not shown). Failure to generate EBV-specific T cells was also
observed when PBMC from EBV-seronegative children were activated with
DC loaded with autologous LCL lysates (data not shown). When hybrid
DC-LCL were used as activators, EBV-specific CTL were generated in just
one of the EBV-seronegative children tested, and in this child the
killing of the autologous LCL (94% at 20:1 ratio) was blocked by Abs
to both MHC class II and MHC class I (killing reduced to 44 and 75%,
respectively).
|
B, the expanded T cell lines
exhibited specific cytotoxic activity. EBV-specific CTL were generated reproducibly from EBV-seronegative children by selection of CD25-positive T cells
While EBV-Ag-loaded DC were able to expand EBV-specific CTL in
some EBV-seronegative children, the procedure was not reproducible and
therefore not suitable for use in a clinical trial. Therefore, we
determined whether EBV-specific CTL activity could be revealed by
selectively expanding T cells that expressed activation markers in
response to stimulation with autologous LCL. The CD25 Ag is the
-chain of the IL-2R, which is up-regulated on the surface of T
lymphocytes after they encounter cognate Ag. In the past, this Ag has
been used to deplete alloreactive T cells from stem cell allografts.
Several groups have demonstrated that the depletion of donor
CD25-positive cells after coculture with recipient cells is able to
eliminate alloreactive response and reduce the incidence of
graft-versus-host disease (30, 31). Recently, Ibisch et
al. (19) demonstrated that EBV-specific CTL can be
purified from EBV-seropositive donors by positive selection for CD25
expression 6 days after stimulation with autologous LCL. Therefore, we
selected CD25-positive cells 9–11 days after stimulation by cell
sorting, in the anticipation that this would select the EBV-specific
CTL precursors.
At the time of selection, the percentage of CD25+
T cells in the cultures was 4.4 ± 3.4%, compared with 37 ±
20% in seropositive or adult-seronegative donors. After the CD25
selection, the CTL lines expanded rapidly even when very few cells were
recovered (the total expansion from sorting on day 9–11 up to the
seventh stimulation was 240-fold, range from 20 to 769). Fig. 3 shows the gate used for the CD25
sorting and the expansion of a representative CTL line generated from
one of the six EBV-seronegative children. Using this procedure,
EBV-specific CTL were successfully generated from all six
EBV-seronegative children. In addition, this procedure was successfully
repeated in two of these children, confirming its reproducibility.
Interestingly, the majority of the CTL were
CD3+CD4+ (71 ± 27;
Table II), although in three cases 19–25% of the cells expressed
CD3+CD8+. In all cases, the
CTL demonstrated specific killing of the autologous LCL compared with
allogeneic LCL (46 ± 20% vs 2 ± 3% at a 20:1 ratio; Fig. 4A), which was reduced in the
presence of MHC class II Abs (mean 29 ± 13%). In the three lines
containing CD8+ T cells, the killing was also
blocked by anti-class I Abs (data not shown). LAK activity (mean
21 ± 20% at 20:1 ratio) was seen in three CTL lines that had
>10% of CD56+ T cells. Depletion of these
CD56+ cells eliminated LAK killing of HSB-2
(7 ± 5%), without significant modification of autologous LCL
lysis (42 ± 18%) (Fig. 4B). Moreover, lysis of the
autologous LCL was not inhibited by the presence of increasing
concentration of cold HSB-2 cells or of cold allogeneic LCL (Fig. 5C). As suggested by the
cytotoxicity assay, supernatant collected 24 h after stimulation
with autologous LCL showed high IFN- release (2471 ± 1190
pg/ml x 106 cells) compared with allogeneic
LCL (872 ± 12 pg/ml x 106 cells). The
TCR-combinatorial diversity of these CD25-selected CTL was analyzed by
mAbs recognizing TCR-V-specific regions. As observed for CTL
generated from our series of normal donors, CTL generated using the
CD25 selection procedure presented considerable heterogeneity, with no
specific over- or underrepresentation of any V family (Table III).
|
|
|
|
C and Table II). When
the CD25-selected CTL line was evaluated for TCR-V subsets, the
pattern of the T cell repertoire was superimposable with CTL generated
with the standard protocol (Table III, donor 6). CD25-sorted CTL are EBV specific, even when FBS is present in the culture medium
It has been suggested that stimulation of CTL in the presence of
FBS can result in FBS-specific killing (32, 33). When PBMC
from EBV-seropositive donors were reactivated with LCL in presence of
either 10% FBS or 10% HABS, there was little difference in the
increase in expression of CD25, which peaked at 9–11 days (Table IV). However, the possibility of
activation of FBS-specific CTL is greater when the initial frequency of
the responding EBV-specific T cells is low. To determine whether
EBV-specific CTL generated in FBS medium contained FBS-specific CTL, we
tested the CD25-sorted CTL for the ability to kill autologous LCL grown
in HABS. For comparison, CTL generated by the standard protocol were
tested in parallel. All CTL were able to lyse both FBS-grown and
HABS-grown LCL, although the latter were lysed less well (Fig. 5, D, E, and F). However, as shown in
Fig. 5, cold target competition experiments clearly showed that
HABS-grown autologous competitor LCL inhibited the lysis of autologous
FBS-grown target LCL. This was true for CTL generated from the
EBV-seronegative adult, when either the standard protocol or the CD25
selection approach was used (Fig. 5, A and B,
respectively). A similar high level of inhibition was seen when CTL
from EBV-seronegative children were tested (Fig. 5C).
|
shows CTL expanded from
two EBV-seronegative children in the presence of FBS (A and
C) or HABS (B and D). Interestingly,
as observed for normal EBV-seropositive donors (data not published),
CTL generated in HABS usually have lower cytotoxic activity against the
HSB-2 (Fig. 6B) compared with CTL generated in
FBS-containing medium (Fig. 6A). However, in one of the CTL
lines expanded in HABS medium, a higher number of
CD16+ cells was observed (16% resulting in HSB-2
killing of 30% at a 20:1 ratio, Fig. 6D) as compared
with the CTL expanded from the same individual in presence of FBS (5%
resulting in HSB-2 lysis of 9% at 20:1 ratio, Fig. 6C).
However, this LAK activity is not a concern since our previous
experience with hemopoietic stem cell transplant recipients showed that
infusion of CTL demonstrating in vitro HSB-2 killing was not associated
with in vivo toxic effects or signs of graft-versus-host disease
(11).
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The relative ease of generating CTL from seronegative adults compared
with children may be explained by their increased exposure to multiple
environmental Ags. As a result, the CTL repertoire of adults becomes so
wide that CTL precursors that cross-react with EBV Ags are elicited
(molecular mimicry). This might be true for one of the seronegative
adults, from whom we were able to grow EBV-specific CTL lines, but who
tested EBV negative by PCR and by ELISPOT (Table I). In three other
adults, the more likely explanation is that they have in fact been
exposed to EBV, but have failed to mount a serological response to the
virus, because EBV-DNA could be detected in peripheral blood.
Furthermore, using the regression assay, which is a quantitative
measure EBV-specific immunity, T cells from three of the four
seronegative adults produced regression of virus-transformed B cells,
with estimated CTL precursor frequencies within the range detected in
normal seropositive individuals. EBV has previously been detected in a
seronegative adult (34). This individual carried an EBV
transformation deletion mutant that was detected in saliva for more
than 7 years.
Conversely, in the EBV-seronegative children, neither EBV-DNA nor LCL regression was observed at any cell concentration, indicating that they were truly naive. The consequence of this difference was that multiple stimulations with the autologous LCL (standard protocol) reactivated EBV-specific CTL from the four seronegative adults but from none of the six children. The effector cells from the seronegative adults are true CTL, showing clear evidence of MHC restriction and target cell specificity. Several investigators have previously reported that LCL can enhance NK or LAK activity in EBV-seronegative donors, and apparent EBV-specific cytotoxic activity obtained from seronegative donors has often not been clearly differentiated from NK- or LAK-mediated activity (17, 33, 35). We observed a similar phenomenon in three of our seronegative children from whom cytolytic T cell lines were elicited using the standard protocol. The cytotoxic activity was unrestricted by HLA Ags, was increased in the presence of blocking Abs to HLA, and was completely removed by depletion of CD56+ cells (data not shown). The clinical value of cells with NK/LAK properties or containing predominantly NK/LAK activity remains questionable.
Our comparison of different approaches for reactivation of EBV-specific CTL from EBV-seronegative individuals revealed several unanticipated results. Initially, we assumed that reactivation of EBV-specific CTL from genuinely naive pediatric donors would require Ag presentation by DC, since they are reported to be the APCs most capable of activating primary immune responses in vitro (18). Therefore, we used a variety of techniques to load DC with EBV Ags for presentation. But although our loaded DC had the phenotypic characteristics of mature APC (large lineage-negative, DR+ cells expressing CD80, CD86, and CD83), and readily reactivated EBV-specific CTL from adults, the induction of primary responses to EBV was not reproducibly obtained in the seronegative children. Instead we found that LCL alone were sufficient for the pediatric samples, provided there was early selection of CD25+ cells. While our results do not prove that DC are unnecessary for the activation of a primary immune response in vitro (since DC precursors were not eliminated from the cultures), they indicate that enrichment or specialized activation and maturation of these cells is not required. It is also likely that activation of PBMC with loaded DC, followed by a CD25+ selection step, would reproducibly produce EBV-specific CTL from seronegative children. However, use of LCL represents a simpler strategy for generating EBV-specific CTL from small children and babies, from whom the large amounts of blood required for DC generation may be difficult to obtain.
There are means other than the use of CD25 selection by which
EBV-specific precursor cells could be selected and expanded. Cloning
approaches have successfully been applied for similar purposes, but
they are laborious and have a significant failure rate. Moreover,
polyclonal responses are less likely to permit the outgrowth of epitope
loss variants than monoclonal T cells (36). The CD25
(IL-2R) activation marker is particularly useful for the application we
propose, because it is expressed only on T cells that have encountered
cognate Ag. In fact, it was recently shown that LCL-activated T cells
from EBV-seropositive donors that were positively selected for CD25
have increased overall EBV specificity with only slight decrease of the
overall diversity of the T cell line (19). From all
EBV-seronegative donors tested in our study, the selective expansion of
CD25+ cells 9–11 days after LCL stimulations
allowed the successful and reproducible generation of EBV-specific T
lymphocytes from fresh or frozen PBMC. For optimum protection, these
CD25-sorted cells should recognize a wide spectrum of EBV-associated
Ags. The antigenic repertoire may be more restricted when this
technique is applied to EBV-seronegative individuals, because fewer
precursor clones may be available for selection and expansion. However,
analysis of TCR-V usage showed that these CTL were highly
heterogeneous. The most remarkable difference when this approach was
applied in EBV seronegatives was the prevalent expansion of
CD4+ cells. This difference is more likely to be
due to true naivety, rather than to an age difference, since the CTL
lines we generated from EBV-seropositive children contained normal
CD8+ T cell numbers (22). Limited
experience of the antitumor and antiviral activity of
CD4+ EBV-specific T cell lines exists. Therefore,
only in vivo administration will provide definitive data on the
efficacy of the CD4+ T cells expanded from the
EBV-seronegative individuals using the CD25 selection. Recently, Metes
et al. (35) reported the successful generation of
EBV-specific CTL from two EBV-seronegative individuals by adding rIL-12
during the first round of stimulation. Although promising, this
approach requires testing in EBV-seronegative children. We also show
that DC, which indeed produce IL-12, can induce the generation of CTL
in EBV-seronegative individuals, but in our hands this technique was
not reproducibly effective in children. If CD4+
cells generated with the CD25 procedure do not have in vivo effects,
IL-12 can be added to increase the frequency of
CD8+ cells.
The characteristics of the T lymphocytes expanded from EBV-seronegative individuals by stimulation with autologous LCL have previously been evaluated, and it was found that when FBS was used in the culture medium many of the expanded lymphocytes recognize FBS-associated Ags (33). The risk of expanding FBS-specific CTL may not be relevant when the frequency of the precursors to be expanded is high, as for EBV-seropositive individuals. In the hemopoietic stem cell transplant setting, we have extensively demonstrated the antiviral and antitumor effects of EBV-specific CTL, even though they were expanded in FBS-containing medium (11, 12). In contrast, when EBV-specific CTL precursor frequencies are low, competing expansion of FBS-specific CTL is more likely. This problem was ruled out, since autologous LCL grown in HABS medium were lysed at similar levels to FBS-grown LCL and acted as competitive target.
In conclusion, the definition of the immune status of both donor and recipient at the time of transplant is crucial for specifically identifying patients at high risk for developing PTLD. The correct identification of EBV-negative individuals will also avoid the necessity of generating EBV-specific CTL using alternative and more expensive approaches. For truly EBV-negative individuals, the selection of cells expressing CD25 after stimulation with LCL represents a reproducible, simple, and effective protocol for the generation of EBV-specific CTL from naive individuals. This technique may be applied to immunotherapy of other disease and malignancies in which the Ag-specific precursor frequency is low, either because a tumor fails to present Ags to the immune system or because a patient is immunologically naive.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 Address correspondence and reprint requests to Dr. Cliona M. Rooney, Center for Cell and Gene Therapy, Baylor College of Medicine, 6621 Fannin Street, MC 3-3320, Houston, TX 77030. E-mail address: crooney{at}bcm.tmc.edu
3 Abbreviations used in this paper: PTLD, posttransplant lymphoproliferative disorder; DC, dendritic cell; EBNA, anti-nuclear EBV; eGFP, enhanced green fluorescent protein; HABS, human serum type AB; LAK, lymphokine-activated killer; LCL, lymphoblastoid B cell line; MCM, monocyte-conditioned medium; SOT, solid organ transplant; VCA, viral capsid Ag.
Received for publication June 5, 2001. Accepted for publication November 2, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
-chain (CD25)-positive selection. J. Immunol. 164:4924.This article has been cited by other articles:
|
|
 |
|
  H. M. Long, J. Zuo, A. M. Leese, N. H. Gudgeon, H. Jia, G. S. Taylor, and A. B. Rickinson CD4+ T-cell clones recognizing human lymphoma-associated antigens: generation by in vitro stimulation with autologous Epstein-Barr virus-transformed B cells Blood, July 23, 2009; 114(4): 807 - 815. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Quintarelli, G. Dotti, B. De Angelis, V. Hoyos, M. Mims, L. Luciano, H. E. Heslop, C. M. Rooney, F. Pane, and B. Savoldo Cytotoxic T lymphocytes directed to the preferentially expressed antigen of melanoma (PRAME) target chronic myeloid leukemia Blood, September 1, 2008; 112(5): 1876 - 1885. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Liu, H.-E. Claesson, Y. Mahshid, G. Klein, and E. Klein Leukotriene B4 activates T cells that inhibit B-cell proliferation in EBV-infected cord blood-derived mononuclear cell cultures Blood, March 1, 2008; 111(5): 2693 - 2703. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Tanijiri, T. Shimizu, K. Uehira, T. Yokoi, H. Amuro, H. Sugimoto, Y. Torii, K. Tajima, T. Ito, R. Amakawa, et al. Hodgkin's Reed-Sternberg cell line (KM-H2) promotes a bidirectional differentiation of CD4+CD25+Foxp3+ T cells and CD4+ cytotoxic T lymphocytes from CD4+ naive T cells J. Leukoc. Biol., September 1, 2007; 82(3): 576 - 584. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
W. H. Lim, S. Kireta, G. R. Russ, and P. T. H. Coates Human plasmacytoid dendritic cells regulate immune responses to Epstein-Barr virus (EBV) infection and delay EBV-related mortality in humanized NOD-SCID mice Blood, February 1, 2007; 109(3): 1043 - 1050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. Beck, M. S. Topp, U. Koehl, E. Roilides, M. Simitsopoulou, M. Hanisch, J. Sarfati, J. P. Latge, T. Klingebiel, H. Einsele, et al. Generation of highly purified and functionally active human TH1 cells against Aspergillus fumigatus Blood, March 15, 2006; 107(6): 2562 - 2569. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. H. Gudgeon, G. S. Taylor, H. M. Long, T. A. Haigh, and A. B. Rickinson Regression of Epstein-Barr Virus-Induced B-Cell Transformation In Vitro Involves Virus-Specific CD8+ T Cells as the Principal Effectors and a Novel CD4+ T-Cell Reactivity J. Virol., May 1, 2005; 79(9): 5477 - 5488. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Liu, J. L. Arbiser, A. Holmgren, G. Klein, and E. Klein PSK and Trx80 inhibit B-cell growth in EBV-infected cord blood mononuclear cells through T cells activated by the monocyte products IL-15 and IL-12 Blood, February 15, 2005; 105(4): 1606 - 1613. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A.-L. Ponsonby, I. van der Mei, T. Dwyer, L. Blizzard, B. Taylor, A. Kemp, R. Simmons, and T. Kilpatrick Exposure to Infant Siblings During Early Life and Risk of Multiple Sclerosis JAMA, January 26, 2005; 293(4): 463 - 469. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. E. Davis, M. A. Sherritt, M. Bharadwaj, L. E. Morrison, S. L. Elliott, L. M. Kear, J. Maddicks-Law, T. Kotsimbos, D. Gill, M. Malouf, et al. Determining virological, serological and immunological parameters of EBV infection in the development of PTLD Int. Immunol., July 1, 2004; 16(7): 983 - 989. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
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]
|
|
|
|
 |
|
 I. Kang, T. Quan, H. Nolasco, S.-H. Park, M. S. Hong, J. Crouch, E. G. Pamer, J. G. Howe, and J. Craft Defective Control of Latent Epstein-Barr Virus Infection in Systemic Lupus Erythematosus J. Immunol., January 15, 2004; 172(2): 1287 - 1294. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. Bickham, K. Goodman, C. Paludan, S. Nikiforow, M. L. Tsang, R. M. Steinman, and C. Munz Dendritic Cells Initiate Immune Control of Epstein-Barr Virus Transformation of B Lymphocytes In Vitro J. Exp. Med., December 1, 2003; 198(11): 1653 - 1663. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. Nikiforow, K. Bottomly, G. Miller, and C. Munz Cytolytic CD4+-T-Cell Clones Reactive to EBNA1 Inhibit Epstein-Barr Virus-Induced B-Cell Proliferation J. Virol., November 15, 2003; 77(22): 12088 - 12104. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
P. J. Amrolia, G. Muccioli-Casadei, E. Yvon, H. Huls, U. Sili, E. D. Wieder, C. Bollard, J. Michalek, V. Ghetie, H. E. Heslop, et al. Selective depletion of donor alloreactive T cells without loss of antiviral or antileukemic responses Blood, September 15, 2003; 102(6): 2292 - 2299. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. M. Posavad, A. Wald, N. Hosken, M. L. Huang, D. M. Koelle, R. L. Ashley, and L. Corey T Cell Immunity to Herpes Simplex Viruses in Seronegative Subjects: Silent Infection or Acquired Immunity? J. Immunol., April 15, 2003; 170(8): 4380 - 4388. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C.M. Straathof, C. M. Bollard, C. M. Rooney, and H. E. Heslop Immunotherapy for Epstein-Barr Virus-Associated Cancers in Children Oncologist, February 1, 2003; 8(1): 83 - 98. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
B. Savoldo, M. H. Huls, Z. Liu, T. Okamura, H.-D. Volk, P. Reinke, R. Sabat, N. Babel, J. F. Jones, J. Webster-Cyriaque, et al. Autologous Epstein-Barr virus (EBV)-specific cytotoxic T cells for the treatment of persistent active EBV infection Blood, December 1, 2002; 100(12): 4059 - 4066. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. Omiya, C. Buteau, H. Kobayashi, C. V. Paya, and E. Celis Inhibition of EBV-Induced Lymphoproliferation by CD4+ T Cells Specific for an MHC Class II Promiscuous Epitope J. Immunol., August 15, 2002; 169(4): 2172 - 2179. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
), autologous LCL (
), and HLA class I- and
II-mismatched LCL (x) were labeled with 51Cr and
incubated with CTL for 5 h at the E:T ratios indicated.
A, The mean activity of T cell lines expanded from six
EBV-seronegative children. B, The mean activity of T
cell lines expanded from four EBV-seronegative adults. HSB-2 provides a
measure of lymphokine-activated cell-mediated killing.
). The competitor:target cell ratio is indicated on the
x-axis. D–F, The standard
51Cr release assay against FBS-grown autologous LCL (filled
bars) and HABS-grown autologous LCL (gray bars). A and
D, CTL generated using the standard protocol from one
representative EBV-seronegative adult. B and
E, CTL from the same individual generated using the CD25
selection protocol. C and F, CTL
generated by the CD25 selection approach from one representative
EBV-seronegative child. This assay shows that CD25-selected CTL exhibit
a strong EBV-specific component beyond the FBS-specific response.