close

Вход

Забыли?

вход по аккаунту

?

443

код для вставкиСкачать
Int. J. Cancer: 81, 459?466 (1999)
r 1999 Wiley-Liss, Inc.
Publication of the International Union Against Cancer
Publication de l?Union Internationale Contre le Cancer
IDENTIFICATION OF A SART-1-DERIVED PEPTIDE CAPABLE OF INDUCING
HLA-A24-RESTRICTED AND TUMOR-SPECIFIC CYTOTOXIC T LYMPHOCYTES
Megumi KIKUCHI1, Masanobu NAKAO1, Yoshiko INOUE1, Kazuko MATSUNAGA1, Shigeki SHICHIJO2, Hideaki YAMANA3
and Kyogo ITOH1,2*
1Cancer Vaccine Development Division of Kurume University Research Center for Innovative Cancer Therapy, Kurume University
School of Medicine, Kurume, Japan
2Department of Immunology, Kurume University School of Medicine, Kurume, Japan
3Department of Surgery, Kurume University School of Medicine, Kurume, Japan
We have described the SART-1 gene?encoding peptides
recognized by HLA-A2601-restricted and tumor-specific cytotoxic T lymphocytes (CTLs). We now have investigated
whether SART-1 encodes peptides capable of inducing the
HLA-A24-restricted CTLs. Among the 18 different peptides
with HLA-A24-binding motifs, the SART-1690?698 peptide (EYRGFTQDF) was most strongly recognized by the HLA-A24restricted and tumor-specific CTLs established from an
esophageal cancer patient. After a third stimulation in vitro,
this peptide induced HLA-A24-restricted CTLs recognizing
the SART-1259? tumor cells in PBMCs of all HLA-A24 homozygous and the majority of HLA-A24 heterozygous cancer
patients and healthy donors tested. A similar activity, induction of CTLs from PBMCs, was observed in the Saccharomyces
cerevisiae?derived nonapeptide (EYRGFTPMF) that shares 7
amino acids with the SART-1690?698 peptide. The SART-1690?698
peptide?induced CTL activity was significantly higher in
PBMCs of HLA-A24 homozygotes than in HLA-A24 heterozygotes. The CTL precursor frequency in PBMCs after a third
stimulation in vitro with the SART-1690?698 peptide was high
(G1/200) in both cancer patients and healthy donors. The
SART-1690?698 peptide could thus be useful for specific immunotherapy of HLA-A24? cancer patients. Int. J. Cancer 81:459?
466, 1999.
r 1999 Wiley-Liss, Inc.
autologous tumor cell line KE4 in medium [45% RPMI 1640
medium, 45% AIM-V medium (GIBCO BRL, Gaithersburg, MD),
10% FCS (Equitech Bio, Ingram, TX) with 100 units/ml of IL-2
(Shionogi, Osaka, Japan) and 0.1 mM MEM non-essential amino
acid solution (GIBCO BRL)], designated as the culture medium.
These CTLs were tested for cytotoxicity against various tumor
cells in a 6-hr 51Cr-release assay at different E:T ratios (Nakao et
al., 1995), as well as for production of IFN-? in response to various
tumor cells by an ELISA (limit of sensitivity ?5 pg/ml; Otsuka,
Tokyo, Japan) as reported (Shichijo et al., 1998). VA13 cells
transfected with the 6A1?1D7 gene corresponding to nucleotide
position 1,517?2,506 of the SART-1 and the HLA-A (HLA-A2402 or
-A2601) cDNA were provided as reported (Shichijo et al., 1998)
and used as target cells after pulsing with different concentrations
of peptide. After determination of HLA-A24-restricted and tumorspecific CTL activity, the CTL line at day 33 was cryo-preserved in
a liquid nitrogen tank. The surface phenotype of the CTL line was
investigated by a direct immunofluorescence assay with FITCconjugated anti-CD3, -CD4 or -CD8 monoclonal antibody (MAb)
(Nichirei, Tokyo, Japan) as reported (Nakao et al., 1995). The CTL
line that had been cryo-preserved on day 33 of culture was thawed
in the morning of the experiments and used to provide effector cells
for identifying peptide antigens.
Many different peptide antigens recognized by HLA class
I-restricted cytotoxic T lymphocytes (CTLs) recognizing melanoma cells have been identified (Kang et al., 1995; Kawakami et
al., 1994; Traversari et al., 1992; Robbins et al., 1995; van der
Bruggen et al., 1994). Some of these are under clinical trial as
cancer vaccines, and major tumor regression has been reported in
several HLA-A1? melanoma patients who received vaccine therapy
with the MAGE-3 peptide (Marchand et al., 1995). Further,
peptide-based therapy resulted in tumor regression in HLA-A2?
melanoma patients (Nestle et al., 1998; Rosenberg et al., 1998).
Therefore, these peptides could be a new tool for specific immunotherapy of melanoma patients. However, with the exception of a
mutated CASP-8 peptide (Mandruzzato et al., 1997), no peptides
have been identified from human squamous cell carcinoma (SCC),
one of the major human malignancies. We have reported that a
nonapeptide of the SART-1 antigen might be used for specific
immunotherapy of HLA-A2601? SCC and lung adenocarcinoma
patients as a cancer vaccine (Shichijo et al., 1998). An HLA-A24
allele is present in 60% of Japanese individuals (Imanishi et al.,
1992). The SART-1 antigen possesses many peptide sequences
with an HLA-A24-binding motif. Therefore, we have investigated
whether these HLA-A24-binding peptides might be used for
specific immunotherapy of HLA-A24? SCC and adenocarcinoma
patients. We report that the SART-1690?698 peptide can induce
HLA-A24-restricted and tumor-specific CTLs.
Detection of SART-1 antigens and peptide synthesis
The expression of SART-1259 or SART-1800 antigen in various
tumor cells was detected by Western blot analysis using the
anti-SART-1259 or anti-SART-1800 antibody, respectively (Shichijo
et al., 1998). The sequence data for SART-1 are available from
EMBL/GenBank/DDBJ (accession number AB006198). There are
18 different peptide sequences capable of binding to HLA-A24 in
the deduced amino acid sequences of the SART-1800 and SART-1259
antigens (Shichijo et al., 1998; Rammensee et al., 1995). Among
these, SART-1690?698 (EYRGFTQDF) possesses the strongest HLAA24-binding motif, with tyrosine in the 2nd position and phenylalanine at the 9th position. Each amino acid (aa) of SART-1690?698 was
substituted by glycine when it was not glycine or by threonine
when it was glycine. Phenylalanine at the 9th position was also
substituted by alanine, tryptophan, leucine or isoleucine. The
control nonapeptide used in our study was SART-1785?793
(VLSGSGKSM), which was not capable of binding to HLA-A24
but was capable of binding to HLA-A2601 (Shichijo et al., 1998;
Rammensee et al., 1995). A nonapeptide (EYRGFTPMF) derived
from a membrane protein of Saccharomyces cerevisiae (SwissProt
accession number P25363) was also synthesized since it showed
Grant sponsors: Ministry of Education, Science, Sport and Culture of
Japan; Ministry of Health and Welfare, Japan; Grant numbers: 08266266,
09470271, 10153265, 10670521, 10671230, H10-genome-003.
MATERIAL AND METHODS
Generation of HLA-A24-restricted and SART-1-specific CTL line
The HLA-A24-restricted and tumor-specific CTL line was
established from PBMCs of an esophageal cancer patient (HLAA2402/A2601) by mixed lymphocyte tumor cell culture (Nakao et
al., 1995). Briefly, the patient?s PBMCs were stimulated with the
*Correspondence to: Kurume University School of Medicine, 67
Asahi-machi, Kurume, 830?0011, Japan. Fax: (81)942?31?7699.
E-mail. [email protected]
Received 7 September 1998; Revised 26 November 1998
KIKUCHI ET AL.
460
TABLE I ? CYTOTOXICITY OF THE CTL LINE AT DAY 33 AGAINST THE VARIOUS EPITHELIAL CANCER CELLS
Cell line
KE4
KE3
TE8
TE11
PC9
11?18
SKG-I
TE10
LC65A
RERF-LC-AI
LK79
LK87
QG56
KUMA-1
KMG-A
86-2
A549
TE9
RERF-LC-MS
VA13
Origin
Esophagus
Esophagus
Esophagus
Esophagus
Lung (adeno.)
Lung (adeno.)
Uterus
Esophaghus
Lung (small)
Lung (squamous)
Lung (small)
Lung (adeno.)
Lung (squamous)
Head and neck
Gallbladder
Lung (large)
Lung (adeno.)
Esophagus
Lung (adeno.)
Fibroblast
SART-1 antigen2
HLA-A1 allele
A2402/2601
A2/2402
A2402/2601
A2402/2601
A0206/2402
A0201/2402
A2402/
A2/2402
A1101/2402
A2402/
A2402/
A0207/1101
A2601/
A2603/3302
A2601/
A1101/
A2603/3001
A3302/
A1101/
n.e.
SART-1259
SART-1800
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
?
% specific lysis by
the CTL line3
(E/T ratio)
20
10
5
27
32
23
31
39
27
25
38
6
3
3
2
2
9
9
2
1
5
6
2
22
20
20
21
33
22
18
30
3
3
1
4
2
6
5
2
2
3
5
7
18
13
16
12
22
17
10
22
3
1
4
3
2
3
2
0
2
2
1
1
1The genotypes of HLA-A alleles have been previously reported (Rammensee et al., 1995).?2Expression
of SART-1259 antigen in the cytosol fraction or SART-1800 antigen in the nuclear fraction was determined by
Western blot analysis with polyclonal anti-SART-1259 or anti-SART-1800 antibody, respectively. Some of
the results are shown in Figure 1.?3The cytotoxicity of the CTL line at day 33 against the various epithelial
cancer cells was determined by a 6-hr 51Cr-release assay at 3 different effector-to-target cell ratios. The
mean of the triplicate assays is shown. n.e., not expressed.
high homology with SART-1690?698. These synthesized peptides
were kindly provided by Dr. M. Kanaoka (Sumitomo, Osaka,
Japan), and the purity was 85% to 95% in most peptides.
Induction of CTLs by peptides
PBMCs (2 ? 106) from esophageal or lung cancer patients or
healthy volunteers were incubated with 10 然 of a peptide in one
well of a 24-well plate containing 2 ml of culture medium. PBMCs
were also stimulated with 2 or 50 然 of the peptide in certain
experiments. HLA class I alleles of the PBMCs were serotyped by
the conventional method of Shichijo et al. (1998). At days 7 and 14
of culture, cells were collected, washed and stimulated with
irradiated (50 gray) autologous PBMCs as APCs that had been
pre-incubated with the same peptide at the same dose for 2 hr
followed by washing with PBS. The ratio of responder to stimulator
cells was 10:1. Effector cells were harvested at day 21 of culture,
and most were immediately tested for the ability to produce IFN-?
in response to various target cells by an ELISA and for cytotoxicity
by a 6-hr 51Cr-release assay in triplicate (Shichijo et al., 1998).
Deviations of the triplicate determinations were usually ?20 % of
mean values. Some effector cells were provided for CTL precursor
frequency analysis. In brief, these PBMCs were plated at 1, 2, 5, 10,
20, 40 and 80 cells per well of 96-well micro-culture plates and
cultured with cloning medium (25% RPMI 1640 medium, 55%
AIM-V medium, 20% FCS, 100 units/ml IL-2, 10 痢/ml PHA and
0.1 mM MEM non-essential amino acid solution) in the presence of
irradiated (50 gray) allogenic PBMCs (2 ? 105 cells/well) donated
by 3 healthy volunteers as feeder cells. In certain experiments,
these PBMCs were plated at 100, 200 and 500 cells per well. The
culture medium without feeder cells was used for maintenance of
the culture. Cells from each well were tested between 8 and 14 days
of culture for IFN-? production in response to various target cells.
The well was considered positive if it contained effector cells
producing a much higher level (?100 pg/ml) of IFN-? in response
to KE4 tumor cells (HLA-A24? and SART-1259? and SART-1800?)
compared to the IFN-? level without target cells or in response to
either QG56
(HLA-A24?, SART-1259? and SART-1800?) or VA13
?
(HLA-A24 and SART-1259? and SART-1800?). The data were
FIGURE 1 ? Recognition of SART-1259 antigen by CTLs. The CTL
line at day 33 was tested for its ability to produce IFN-? by recognition
of VA13 cells transfected with the 6A1?1D7 gene encoding the
SART-1259 antigen and HLA-A2402 or -A2601 cDNA. One-hundred
nanograms of 6A1?1D7 cDNA and of HLA-A2601 or HLA-A2402
cDNA were co-transfected into VA13 cells (1 ? 104), incubated for 2
days, followed by testing of the ability to stimulate IFN-? production
by CTLs (2 ? 104) in triplicate assays. The supernatant was collected to
measure IFN-? activity by an ELISA as reported (Shichijo et al., 1998).
Mean values are shown.
analyzed by the minimum ?2 method with 95% confidence
intervals. The CTL precursor frequency was calculated by determining the cell concentration at which 37% of the wells were negative,
SART-1 PEPTIDE INDUCES CTL
461
using a graph of the percentage of negative wells on the ordinate
and the responder cell concentration on the abscissa (Taswell,
1981). Some effector cells harvested at day 21 were also provided
for enrichment of CD8? T cells by Dynabeads coupled with
anti-CD8 MAb (Dynal, Oslo, Norway), followed by testing for
cytotoxicity to KE4 and the other tumor cells in a 6-hr 51Cr-release
assay. The frequency of CD8? T cells among the unseparated
effector cells or the purified fraction was 10% to 20% or 90%,
respectively.
Expression of HLA-A24 molecule
PBMCs were stained with anti-HLA class I (W6/32) MAb
recognizing a monomorphic region of the HLA class I molecule or
anti-HLA-A24 (A11.1M) MAb reacting with the HLA-A24 molecule (ATCC, Rockville, MD), respectively. Expression of the
HLA class I or -A24 molecule was evaluated using a FACScan
(Becton Dickinson, Mountain View, CA) as reported (Nakao et al.,
1995).
RESULTS
Establishment of HLA-A24-restricted and tumor-specific CTL line
The CTLs from PBMCs of an esophageal cancer patient (KE4,
HLA-A2402/2601) began to exhibit HLA-A24?-restricted and
tumor-specific cytotoxicity after being stimulated 5 times with KE4
autologous tumor cells. Representative results are shown in Table I.
The CTLs at day 33 lysed the 7 HLA-A2402? tumor cells,
including 4 esophageal SCCs, 2 lung adenocarcinomas and 1
cervical SCC. All of these tumor cells expressed both SART-1259
and SART-1800 antigens. In contrast, the CTLs failed to lyse any of
the 3 HLA-A2402? epithelial cancer cells expressing SART-1800
but not SART-1259 (Table I) or the HLA-A24? PHA blastoid cells
expressing SART-1800 but not SART-1259 (data not shown). The
CTLs also did not lyse either VA13 fibroblast cells or any of the 8
HLA-A24? epithelial cancer cells, including HLA-A2601? or
A2603? tumor cells, regardless of SART-1259 expression (Table I).
The surface phenotype of this CTL line was CD3?4? 8? (data not
shown). The CTL line produced a significant level of IFN-?
by recognition of VA13 cells transfected with both HLA-A2402
cDNA and the 6A1?1D7 gene encoding the SART-1259 antigen but
not of VA13 cells transfected with HLA-A2601 and the 6A1?1D7 as
a control (Fig. 1). These results suggest that this CTL line
recognized mainly the SART-1259? tumor cells in an HLA-A24restricted manner. This CTL line was used in the following
experiments.
Recognition of peptide by CTLs
We tested the reactivity of the CTL line to the 18 different
peptides with HLA-A24-binding motifs that were originated from
SART-1800 and SART-1259 antigens (Fig. 2). The CTL line produced
significant levels (?400 pg/ml) of IFN-? by recognition of
HLA-A2402-transfected VA13 cells that had been pulsed with the
SART-1690?698, SART-1746?755 or SART-1758?766. These 3 peptides
are expected to be shared by both the SART-1259 and SART-1800
antigens. The highest level of IFN-? production was consistently
observed in the CTLs by recognition of the SART-1690?698 (EYRGFTQDF) in the different experiments. CTL activity was dependent
on the dose of the SART-1690?698 peptide used for pulsing, and the
IFN-? production observed ranged from a low of 10 nM to a
maximum of 50 然 of the peptide (Fig. 3).
Determination of aa required for CTL-mediated recognition
The substituted nonapeptides along with the parental
SART-1690?698 were tested for the ability to stimulate IFN-?
production by this CTL line. Representative results are shown in
Figure 4. The ability of the SART-1690?698 peptide decreased when
tyrosine at position 2 (2Y-G), glutamine at position 7 (7Q-G) and
FIGURE 2 ? Recognition of peptides by CTLs. The CTL line at day
33 was tested for its ability to produce IFN-? by recognition of
HLA-A2402-transfected VA13 cells that had been pulsed with one of
the 18 different peptides or of HLA-A2601-transfected VA13 cells with
some of these peptides as a negative control. The peptide (10 然) was
loaded for 2 hr to VA13 cells (1 ? 104) transfected with HLA-A2402 or
-A2601, followed by testing of the ability to stimulate IFN-? production by CTLs. For IFN-? production, CTLs (2 ? 104) were added and
incubated for 18 hr in triplicate assays and the culture supernatant was
collected for measurement of IFN-? by ELISA. The background level
of IFN-? production by CTLs in response to HLA-A2601-transfected
cells plus peptide was 100 to 200 pg/ml. Results in response to
HLA-A2402-transfected cells are shown.
FIGURE 3 ? Dose dependence of the SART-1690?698 peptide. Different
doses (0.001 to 100 然) of the SART-1690?698 peptide were pulsed to
HLA-A2402- or HLA-A2601-transfected VA13 cells and tested for the
ability to stimulate IFN-? production by CTLs at day 33. Background
IFN-? production (50 to 200 pg/ml) in response to HLA-A2601transfected VA13 cells with the peptide was subtracted.
aspartic acid at position 8 (8D-G) were substituted with glycine,
respectively. Similarly, the ability decreased when phenylalanine at
position 9 was substituted with alanine (9F-A) or leucine (9F-L)
but not glycine (9F-G) or isoleucine (9F-I).
462
KIKUCHI ET AL.
Induction of CTLs from HLA-A24 donors
The 3 peptides recognized by the CTL line were tested for the
ability to induce CTLs recognizing HLA-A24? tumor cells from
PBMCs of HLA-A24 cancer patients and healthy donors. PBMCs
that had been stimulated in vitro 3 times with the SART-1690?698
peptide produced higher levels of IFN-? by recognition of the
FIGURE 4 ? Determination of aa required for CTL-mediated recognition. Each amino acid of the SART-1690?698 peptide was substituted with
glycine or, when it was glycine, threonine. Phenylalanine at position 9
was also substituted with alanine, isoleucine or leucine. These substituents along with the parental nonapeptide (10 然) were loaded for 2 hr
to VA13 cells (1 ? 104) transfected with HLA-A2402 or HLA-A2601,
followed by testing of the ability to stimulate IFN-? production by
CTLs.
KE4 tumor (HLA-A24?, SART-1259? and SART-1800?) but not
QG56 (HLA-A24?, SART-1259? and SART-1800?) tumor cells or
VA13 (HLA-A24?, SART-1259? and SART-1800?) fibroblast cells in
all 6 HLA-A24 homozygotes (n ? 4 in healthy donors and n ? 2 in
cancer patients) tested (Fig. 5). The increased IFN-? production
was dependent on the increased number of effector cells (Fig. 6).
These CTLs recognized the peptide used for stimulation: the
PBMCs stimulated with the SART-1690?698 peptide produced a higher
level of IFN-? by recognition of this peptide on HLA-A2402-
FIGURE 6 ? Dose dependence of peptide-induced CTL activity by
the PBMCs shown in Figure 5 was studied at different E:T ratios in
triplicate assays. Results of PBMCs of healthy donor (HD) 1 and HD 7
are shown. Mean values of the triplicate assays are shown. Similar
results were obtained with other donors.
FIGURE 5 ? Induction of CTLs by the SART-1690?698 peptide. PBMCs (2 ? 106) from healthy donors (HD 1 to HD 8) and esophageal (Pt 1 to Pt
3) or lung cancer (Pt 4) patients were incubated with the SART-1690?698 peptide (10 然) in a well of a 24-well plate containing 2 ml culture
medium. At days 7 and 14 of culture, cells were collected, washed and stimulated with irradiated autologous PBMCs as APCs, which had been
pre-incubated with the same nonapeptide (10 然) for 2 hr. Cells were harvested at day 21 of culture and tested for their ability to produce IFN-? in
response to KE4 tumor cells (HLA-A2402?, SART-1259?, SART-1800?), QG56 (HLA-A24?, SART-1259?, SART-1800?), and VA13 (HLA-A24?,
SART-1259?, SART-1800?) at an E:T ratio of 4:1 in triplicate assays. Mean values of triplicate assays are shown, and background IFN-? production
by the effector cells alone was subtracted. Deviations of the triplicate determinations were usually ?20% of mean values.
SART-1 PEPTIDE INDUCES CTL
FIGURE 7 ? Recognition of the SART-1690?698 peptide by CTLs.
PBMCs were stimulated by the SART-1690?698 peptide 3 times in vitro.
PBMCs were tested for their ability to produce IFN-? by recognition of
HLA-A2402-transfected VA13 cells that had been pulsed with the same
peptide or HLA-A2601-transfected VA13 cells with the peptide as a
control at an E:T ratio of 4:1 in triplicate assays. Representative results
of healthy donor (HD) 1 in Figure 5 are shown. The mean value is
shown, and background IFN-? production by the effector cells alone
(100 pg/ml) was subtracted.
transfected VA13 cells but not on HLA-A2601-transfected VA13
cells (Fig. 7).
In HLA-A24 heterozygotes, PBMCs from 5 of the 6 healthy
donors or 4 of the 6 cancer patients also produced higher levels of
IFN-? by recognition of the KE4 tumor cells as compared with
those of either the QG56 or VA13 cells after having been stimulated
3 times with the SART-1690?698 peptide (Fig. 5). However, the levels
of IFN-? produced by the PBMCs of heterozygotes were significantly lower than those of homozygotes in all cases tested. PBMCs
from none of the 6 HLA-A24?/A24? cases (n ? 3 in healthy
donors, n ? 3 in cancer patients) produced a higher level of IFN-?
by recognition of the KE4 tumor cells than by recognition of either
the QG56 or VA13 cells (data not shown).
The SART-1746?755 and SART-1758?766 peptides were also tested
for induction of CTLs in HLA-A24? donors. However, none of
them consistently induced the HLA-A24-restricted CTLs under the
conditions employed (data not shown).
463
FIGURE 8 ? Cytotoxicity of peptide-induced CD8? T cells. CD 8? T
cells were enriched from PBMCs after the 3rd stimulation with
SART-1690?698 and tested for cytotoxicity against various target cells
in a 6-hr 51Cr-release assay at 3 different E:T ratios. Target cells were
KE4 tumor cells, QG56 lung SCC and EBV-transformed B cells of
healthy donor (HD) 1 (HLA-A24/A24, SART-1359? and SART-1800?),
EBV-transformed B cells of HD 5 (HLA-A24/A2, SART-1359? and
SART-1800?) and EBV-transformed B cells of the other HD (HLA-A1/
A26, SART-1259? and SART-1800?). Representative results from PBMCs
of HD 1 in Figure 5 are shown.
of the tumor cells expressing the 3 molecules together after
having been stimulated 3 times with the SART-1690?698 peptide
(Table II). They also produced low but significantly higher
levels of IFN-? by recognition of the 2 tumor cells expressing
HLA-A24 and SART-1800 but not SART-1259. In contrast, they
failed to recognize any of the HLA-A24? tumor cells. The CTLs of
patient 3 (an HLA-A24 heterozygote) produced low but significantly higher levels of IFN-? via recognition of the tumor
cells expressing the 3 molecules together but not by recognition of
the tumor cells expressing HLA-A24 and SART-1800 but not
SART-1259.
Cytotoxicity of CD8? T cells
Our results suggest that the SART-1690?698 peptide consistently
induced HLA-A24-restricted CTLs recognizing SART-1 259? tumor
cells after being stimulated 3 times in vitro only, in PBMCs from
both HLA-A24? cancer patients and healthy donors. CD 8? T cells
were then enriched from the PBMCs after the 3rd stimulation with
SART-1690?698 and tested for cytotoxicity against various target
cells in a 6-hr 51Cr-release assay. Representative results are shown
in Figure 8. CD8? T cells showed significant levels of lysis to KE4
tumor cells only but not to any of the other target cells, including
QG56 tumor cells or the autologous Epstein-Barr virus?transformed B-cell line.
Induction of CTLs by Saccharomyces-derived peptide
The mechanism of epitope mimics was suggested in the case of
the MART-1/melan-A27?35 peptide that could induce HLA-A2restricted CTLs in PBMCs in vitro (Loftus et al., 1996). We have
tested the ability of the S. cerevisiae?derived peptide (EYRGFTPMF) to induce HLA-A24-restricted CTLs in PBMCs. This
peptide shared 7 aa with SART-1690?698, including the expected
agretopic aa at positions 2 and 9. Representative results are shown
in Figure 9 and Table II. This Saccharomyces-derived peptide
induced the HLA-A24-restricted CTLs that recognized mainly the
SART-1259? cancer cells. The pattern of the reactivity of CTLs
induced by the Saccharomyces-derived peptide was similar to that
induced by SART-1690?698.
Recognition of SART-1800? tumor cells
We then determined whether the peptide-induced CTLs
showed no reaction against tumor cells expressing HLA-A24 and
SART-1800 but not SART-1259, using an IFN-? release assay (Table
II). The PBMCs of healthy donor 1 and patient 2 (HLA-A24
homozygotes) produced the highest levels of IFN-? by recognition
HLA-A24 expression and CTL activity
The above results indicate that the SART-1690?698 peptide?
induced CTL activity in PBMCs from HLA-A24 homozygotes was
about 2 to 4 times higher than that from HLA-A24 heterozygotes.
To understand the underlying mechanism, expression levels of the
KIKUCHI ET AL.
464
TABLE II ? RECOGNITION OF TUMOR CELLS BY PEPTIDE-INDUCED CTLs
IFN-? production by CTLs in response to tumor cells (pg/ml)1
Effector
(PBMC)
HD1
Pt 2
Pt 3
HLA
A24/24
HLA-A24?
SART-1259?SART-1800?
Peptide
SART-1690?698
EYRGFTPMF
A24/24 SART-1690?698
EYRGFTPMF
A24/2 SART-1690?698
HLA-A24?
SART-1259?SART-1800?
HLA-A24?
SART-1259?SART-1800?
HLA-A24?
SART-1259?SART-1800?
KE4
KE3
TE11
TE10
LC65A
QG56
LK87
86-2
VA13
397 (293)2
294
224 (196)2
244
169 (69)2
344 (240)
304
100 (72)
265
138 (38)
350 (246)
371
164 (136)
216
231 (131)
145 (41)
229
69 (41)
182
55 (0)
202 (98)
252
90 (62)
101
38 (0)
97
97
0
17
23
102
80
8
10
8
104
13
28
88
100
82
5
12
90
98
1PBMCs
from an HLA-A24 homozygous healthy donor (HD 1, Fig. 5), aesophageal cancer patient (Pt 2) and an HLA-A24 heterozygous
esophageal cancer patient (Pt 3) were stimulated with the SART-1690?698 peptide or Saccharomyces cerevisiae?derived peptide (EYRGFTPMF) (10
然). After the third stimulation, CTL activity was measured. Details of the methods are shown in the legend of Figure 5.?2IFN-? production by
recognition of KE4 tumor cells after subtraction of IFN-? production by recognition of 86-2 tumor cells is shown in parentheses.
FIGURE 9 ? CTL activity induced by the Saccharomyces-derived peptide. PBMCs (2 ? 106) were stimulated with the S. cerevisiae?derived
peptide (EYRGFTPMF) 3 times and tested at day 21 for their ability to produce IFN-? in response to KE4 tumor cells, QG56 and VA13 cells at
different E:T ratios in triplicate assays. Representative results of healthy donor (HD) 2 (HLA-A24/24), HD 6 (HLA -A24/33) and HD 9
(HLA-A26/31) are shown. Mean value is shown.
HLA-A24 molecule on PBMCs of HLA-A24 homozygotes were
compared with those of HLA-A24 heterozygotes. Expression
levels of the HLA-A24 molecule in homozygotes were significantly higher than those of heterozygotes, as determined by
comparison of the mean channels of HLA-A24 expression after
staining of PBMCs with anti-HLA-A24 MAb (136, 184, 160 and
137 mean channels in the 4 homozygotes vs. 97, 77 and 111 in the 3
heterozygotes). In contrast, similar levels of HLA class I expression were observed in these 7 donors when PBMCs were stained
with anti-HLA class I MAb (W6/32) recognizing a monomorphic
portion of the HLA class I (data not shown).
We then addressed the question of whether a higher dose of the
SART-1690?698 peptide induced higher CTL activity. PBMCs were
stimulated 3 times with 2, 10 or 50 然 of the peptide, followed by the
CTL assay at day 21. The peptide-induced CTLs at 2, 10 and 50 然
produced 80, 288 and 248 pg/ml IFN-? in response to the KE4 tumor
cells in the case of an HLA-A24 homozygote and 42, 70 and 10 pg/ml
in the case of a heterozygote at an E:T ratio of 5:1, respectively. Similar
results were obtained for the other HLA-A24? donors.
CTL precursor frequency
We analyzed the CTL precursor frequency in PBMCs of
HLA-A24 homozygotes and a heterozygote that had been stimulated 3 times by the SART-1690?698 peptide. This assay was carried
out to determine whether there were fewer precursor CTLs
recognizing the SART-1690?698 in the heterozygote and to identify
any differences between cancer patients and healthy donors. The
CTL precursor frequency was 1/77 in an HLA-A24 homozygous
cancer patient, 1/174 in an HLA-A24 homozygous healthy donor
and 1/192 in an HLA-A24 heterozygous healthy donor (Fig. 10). In
contrast, it was less than 1/30,000 when unstimulated PBMCs were
provided for CTL precursor frequency analysis.
DISCUSSION
Our results demonstrate that the SART-1690?698 (EYRGFTQDF)
peptide could induce HLA-A24-restricted CTLs which recognized
the SART-1259? tumor cells in PBMCs of all HLA-A24 homozygotes and the majority of HLA-A24 heterozygotes tested in vitro in
both cancer patients and healthy donors. Four different tumorrejection antigens (tyrosinase, ?-catenin, gp100 and p15) have
been reported, in melanomas, to possess antigenic peptides recognized by HLA-A24-restricted CTLs (Kang et al., 1995; Robbins et
al., 1995). Some other tumor-rejection antigens identified from
melanomas (MAGE-1, MAGE-3, PRAME and MART-1) might
encode HLA-A24-binding antigenic peptides recognized by CTLs
(Celis et al., 1994; Stevens et al., 1995; Ikeda et al., 1997; Tanaka
et al., 1997). The SART-1259 antigen is expressed in the majority of
esophageal and head-and-neck SCCs, half of lung adenocarcinomas (Shichijo et al., 1998) and one-third of uterine cancers (data
not shown). The HLA-A24 allele is present in approximately 60%
of Japanese [the vast majority (95%) of them being genotypically
HLA-A2402] (Tokunaga et al., 1997), 20% of Caucasians and 12%
of Africans (Imanishi et al., 1992). Therefore, the SART-1690?698
peptide (EYRGFTQDF) might be useful for specific immunotherapy of relatively large numbers of HLA-A24? cancer patients
with SCCs and adenocarcinomas as a cancer vaccine and as a
SART-1 PEPTIDE INDUCES CTL
465
HLA-A24?
FIGURE 10 ? CTL precursor frequency analysis. The CTL precursor
frequency of PBMCs from HLA-A24 homozygotes [healthy donor
(HD) 1 and patient (Pt) 1] or a heterozygote (HD 6) who had been
stimulated 3 times by the SART-1690?698 peptide was analyzed, and that
of unstimulated PBMCs in HD 1 was also measured as a negative
control. Details are described in ??Material and Methods??.
peptide antigen in vitro to induce CTLs for adoptive cellular
therapy.
The CTL line used for determining the peptides was established
from PBMCs of an esophageal cancer patient by mixed lymphocyte tumor cell culture. This CTL line lysed HLA-A24? epithelial
cancer cells expressing SART-1259 cytosol and SART-1800 nuclear
proteins together
(Shichijo et al., 1998). In contrast, none of the
?
HLA-A24 epithelial cancer cells expressing SART-1800 but not
SART-1259 were susceptible to lysis by this CTL line. These results
support the hypothesis that the SART-1259 antigen, but not SART1800, is a major source of tumor-rejection antigen, primarily
because of its location and pattern of distribution (Shichijo et al.,
1998).
SART-1690?698 exhibits the highest stimulation of IFN-? production by the CTL line among the 18 different peptides with
HLA-A24-binding motifs tested. None of the modified peptides
tested had a markedly higher stimulation of IFN-? production by
the CTL line. Instead, substitution of this peptide at position 2, 7, 8
or 9 decreased the stimulation of IFN-? production. These results
along with those from others (Ibe et al., 1996; Tokunaga et al.,
1997) suggest that amino acids at positions 2 and 9 are critical for
binding of the SART-1690?698 peptide to the HLA-A24 molecule,
while those at positions 7 and 8 are important for T cell?mediated
recognition of the peptide.
The activity of SART-1690?698 in dose-response assays is rather
weak, reaching maximum above 10 痢/ml, suggesting the possible
existence of a superagonistic peptide capable of either inducing the
CTLs at a much lower concentration or increasing CTL responses
to tumor cells (Vergelli et al., 1997; Valitutti et al., 1996). Although
we failed here to find such a peptide through residue substitution,
our results do not deny the presence of better peptides for CTL
induction, which should be important clinically.
The SART-1690?698 peptide induced HLA-A24-restricted CTLs
reacting with SART-1259? tumor cells in vitro after only 3
PBMCs from cancer patients and
stimulations in
healthy donors. CTL precursor frequencies after the 3rd stimulation
were high in a cancer patient and in healthy donors, though slightly
higher in the patient than in the healthy donors. The MART-1/MelanA37?35 peptide induces in vitro the HLA-A2-restricted CTLs recognizing melanoma cells and melanocytes in PBMCs from both
melanoma patients and healthy donors with a much earlier
induction in melanoma patients (Rivoltini et al., 1995). We also
reported that MART-1/Melan-A37?35 induced HLA-A2-restricted
CTLs in PBMCs of auto-immune uveitis patients with VogtKoyanagi-Harada disease (Sugita et al., 1996). The MART-1/
Melan-A antigen is a tissue-specific antigen expressed in normal
melanocytes (Kawakami et al., 1994). In contrast, the SART-1259
antigen is not tissue-specific (Shichijo et al., 1998). It is expressed
in the cytosol of SCCs and adenocarcinomas but not in normal
cells, except testis and fetal liver (Shichijo et al., 1998). The
SART-1800 antigen, the other source for the SART-1690?698 peptide, is
also not expressed in any unstimulated normal cells or tissues
except testis and fetal liver (Shichijo et al., 1998). The SART-1800
antigen, however, is expressed in proliferating cells, including PHA
blasts or EBV-transformed B-cell lines (Shichijo et al., 1998). In
addition, the CTLs induced by SART-1690?698 recognized, to some
extent, when the reactivity was measured by an IFN-? release
assay, the HLA-A24? tumor cells expressing SART-1800 but not
SART-1259. Therefore, CTL precursors reacting to this peptide and
originated from the SART-1800 antigen might be present in circulating blood.
Other protein-derived peptides have a similar sequence of the
MART-1/Melan-A37?35 peptide; these appeared to induce CTL
activity in PBMCs from melanoma patients and healthy donors
(Loftus et al., 1996). Database analysis results showed that one
candidate nonapeptide (EYRGFTPMF) was encoded by a membrane protein of S. cerevisiae, a well-characterized non-pathogenic
yeast (Benit et al., 1992). This Saccharomyces-derived nonapeptide shares 7 aa with the SART-1259-derived peptide, including
positions 2 and 9 necessary for binding to the HLA-A24 molecule.
Only 2 aa at positions 7 and 8 are different from each other.
Humans exhibit strong cell-mediated immunity against yeast
(Romani, 1997). S. cerevisiae is present in many different fermented foods and beverages; therefore, the humans may well be
exposed to this membrane protein on a daily basis. This Saccharomyces-derived peptide induced HLA-A24-restricted CTLs to react
to SART-1259? tumor cells in HLA-A24? PBMCs. Therefore, the
mechanism of epitope mimics might also be involved in the
induction of CTLs by the SART-1690?698 peptide.
Our results have shown that peptide-induced CTL activity in
PBMCs from HLA-A24 homozygotes was significantly higher
than that from HLA-A24 heterozygotes. The expression level of
HLA-A24 antigen on PBMCs, but not the dose of peptide,
appeared to be involved. In addition, the use of IFN-?-pre-treated
PBMCs as APCs results in increased CTL activity (data not
shown). This suggests that HLA-A24 homozygous cancer patients
are more suitable than HLA-A24 heterozygous cancer patients for
specific immunotherapy with the SART-1690?698 peptide. Furthermore, this could be important for determining an adjuvant to be
injected together with the peptide. Adjuvants that induce IFN-?,
IFN-? or the other cytokines that increase HLA class I molecules or
these cytokines themselves might be preferable from this point of
view.
ACKNOWLEDGEMENTS
We thank Dr. D. Yang for providing the CTL line used and Dr. K.
Orita, Executive Director of Hayashibara Biochemical Lab., for
providing natural human INF-?.
466
KIKUCHI ET AL.
REFERENCES
BENIT, P., CHANET, R., FABRE, F., FAYE, G., FUKUHARA, H. and SOR, F.,
Sequence of sup61-RAD18 region on chromosome III of Saccharomyces
cerevisiae. Yeast, 8, 147?153 (1992).
CELIS, E., FIKES, J., WENTWORTH, P., SIDNEY, J., SOUTHWOOD, S., MAEWAL,
A., GUERCIO, M.F.D., SETTE, A. and LIVINGTON, B., Identification of
potential CTL epitopes of tumor-associated antigen MAGE-1 for five
common HLA-A alleles. Mol. Immunol., 31, 1423?1430 (1994).
IBE, M., MOORE, Y.I., MIWA, K., KANEKO, Y., YOKOTA, S. and TAKIGUCHI,
M., Role of strong anchor residues in the effective binding of 10-mer and
11-mer peptides to HLA-A2402 molecules. Immunogenetics, 44, 233?241
(1996).
IKEDA, H., LETHE, B., LEHMANN, F., BAREN, N.V., BAURAIN, J.F., DE SMET,
C., CHAMBOST, H., VITALE, M., MORETTA, A., BOON, T. and COULIE, P.G.,
Characterization of an antigen that is recognized on a melanoma showing
partial HLA loss by CTL expressing an NK inhibitory receptor. Immunity, 6,
199?208 (1997).
IMANISHI, T., AKAZA, T., KIMURA, A., TOKUNAGA, K. and GOJOBORI, T.,
Allele and haplotype frequencies for HLA and complement loci in various
ethnic groups. In: K. Tsuji, M. Aizawa and T. Sasazuki (eds.), HLA 1991,
Vol. 1, pp. 1065?1220, Oxford Scientific Publications, Oxford (1992).
KANG, X., KAWAKAMI, Y., EL-GAMIL, M., WANG, R., SAKAGUCHI, K.,
YANNELI, E., APPELLA, E., ROSENBERG, S.A. and ROBBINS, P.F., Identification of a tyrosinase epitope recognized by HLA-A24-restricted, tumorinfiltrating lymphocytes. J. Immunol., 155, 1343?1348 (1995).
KAWAKAMI, Y., ELIYAKU, S., SAKAGUCHI, K., ROBBINS, P.F., RIVOLTINI, L.,
YANNELLI, J.R., APPELLA, E. and ROSENBERG, S.A., Identification of the
immunodominant peptides of the MART-1 human melanoma antigen
recognized by the majority of HLA-A2-restricted tumor infiltrating lymphocytes. J. exp. Med., 180, 347?352 (1994).
LOFTUS, D.J., CASTELLI, C., CLAY, T.M., SQUARCINA, P., MARINCOLA, F.M.,
NISHIMURA, M.I., PARMIANI, G., APPELLA, E. and RIVOLTINI, L., Identification of epitope mimics recognized by CTL reactive to the melanoma/
melanocyte-derived peptide MART-1(27?35). J. exp. Med., 184, 647?657
(1996).
MANDRUZZATO, S., BRASSEUR, F., ANDRY, G., BOON, T. and VAN DER
BRUGGEN, P.A., CASP-8 mutation recognized by cytolytic lymphocytes on
a human head and neck carcinoma. J. exp. Med., 186, 785?793 (1997).
MARCHAND, M. and 20 OTHERS, Tumor regression responses in melanoma
patients treated with a peptide encoded by gene MAGE-3. Int. J. Cancer, 63,
883?885 (1995).
NAKAO, M., YAMANA, H., IMAI, Y., TOH, Y., TOH, U., KIMURA, A., YANOMA,
S., KAKEGAWA, T. and ITOH, K., HLA-A2601-restricted CTLs recognized a
peptide antigen expressed on squamous cell carcinoma. Cancer Res., 55,
4248?4252 (1995).
NESTLE, F.O., ALIJAGIC, S., GILLIET, M., SUN, Y., GRABBE, S., DUMMER, R.,
BURG, G. and SCHADENDORF, D., Vaccination of melanoma patients with
peptide- or tumor lysate-pulsed dendritic cells. Nature (Med.), 4, 328?332
(1998).
RAMMENSEE, H.G., FRIEDE, T. and STEVANOVIC, S., MHC ligands and
peptide motifs: first listing. Immunogenetics, 41, 178?228 (1995).
RIVOLTINI, L., KAWAKAMI, Y., SAKAGUCHI, K., SOUTHWOOD, S., SETTE, A.,
ROBBINS, P.F., MARINCOLA, F.M., SALGALLER, M.L., YANNELI, J.R. and
APPELLA, E., Induction of tumor-reactive CTL from peripheral blood and
tumor-infiltrating lymphocytes of melanoma patients by in vitro stimulation
with an immunodominant peptide of the human melanoma antigen MART-1.
J. Immunol., 154, 2257?2265 (1995).
ROBBINS, P.F., EL-GAMIL, M., LI, Y.F., TOPALIAN, S.L., RIVOLTINI, L.,
SAKAGUCHI, K., APPELLA, E. and ROSENBERG, S.A., Cloning of a new gene
encoding an antigen recognized by melanoma-specific HLA-A24-restricted
tumor-infiltrating lymphocytes. J. Immunol., 154, 5944?5950 (1995).
ROMANI, L., The T cell response against fungal infections. Curr. Opin.
Immunol., 9, 484?490 (1997).ROSENBERG, S.A. and 15 OTHERS, Immunologic and therapeutic evaluation of synthetic peptide vaccine for the
treatment of patients with metastatic melanoma. Nature (Med.), 4, 321?328
(1998).
SHICHIJO, S., NAKAO, M., IMAI, Y., TAKASU, H., KAWAMOTO, M., NIIYA, F.,
YANG, D., TOH, Y., YAMANA, H. and ITOH, K., A gene encoding antigenic
peptides of human squamous cell carcinoma recognized by cytotoxic T
lymphocytes. J. exp. Med., 187, 277?288 (1998).
STEVENS, E.J., JACKNIN, L., ROBBINS, P.F., KAWAKAMI, Y., EL-GAMIL, M.,
ROSENBERG, S.A. and YANNELLI, J.R., Generation of tumor-specific CTLs
from melanoma patients by using peripheral blood stimulated with allogenic melanoma tumor cell lines. Fine specificity and MART-1 melanoma
antigen recognition. J. Immunol., 154, 762?771 (1995).
SUGITA, S., SAGAWA, K., MOCHIZUKI, M., SHICHIJO, S. and ITOH, K.,
Melanocyte lysis by cytotoxic T lymphocytes recognizing the MART-1
melanoma antigen in HLA-A2 patients with Vogt-Koyanagi-Harada disease. Int. Immunol., 8, 799?803 (1996).
TANAKA, F., FUJIE, T., TAHARA, K., MORI, M., TAKESAKO, K., SETTE, A.,
CELIS, E. and AKIYOSHI, T., Induction of antitumor cytotoxic T lymphocytes
with a MAGE-3-encoded synthetic peptide presented by human leukocytes
antigen-A24. Cancer Res., 57, 4465?4468 (1997).
TASWELL, C., Limiting dilution assays for the determination of immunocompetent cell frequencies. I. Data analysis. J. Immunol., 126, 1614?1619
(1981).
TOKUNAGA, K. and 13 OTHERS, Sequence-based association analysis of HLA
class I and II alleles in Japanese supports conservation of common
haplotypes. Immunogenetics, 46, 199?205 (1997).
TRAVERSARI, C., VAN DER BRUGGEN, P., LUESCHER, I.F., LURQUIN, C.,
CHOMEZ, P., VAN PEL, A., DE PLAEN, E., AMAR-COSTESEC, A. and BOON, T.,
A nonapeptide encoded by human gene MAGE-1 is recognized on HLA-A1
by cytolytic T lymphocytes directed against tumor antigen MZ2-E. J. exp.
Med., 176, 1453?1475 (1992).
VALITUTTI, S., MULLER, S., DESSING, M. and LANZAVECCHIA, A., Different
responses are elicited in cytotoxic T lymphocytes by different levels of T
cell receptor occupancy. J. exp. Med., 183, 1917?1921 (1996).
VAN DER BRUGGEN, P., BASTIN, J., GAJEWSKI, T., COULIE, P.G., BOEL, P., DE
SMET, C., TRAVERSARI, C., TOWNSEND, A. and BOON, T., A peptide encoded
by human gene MAGE-3 and presented by HLA-A2 induces cytolytic T
lymphocytes that recognize tumor cells expressing MAGE-3. Europ. J.
Immunol., 24, 3038?3043 (1994).
VERGELLI, A., HEMMER, B., KALBUS, M., VOGT, A.B., LING, N., CONLON, P.,
COLIGAN, J.E., MCFARLAND, H. and MARTIN, M., Modification of peptide
ligands enhancing T cell responsiveness imply large numbers of stimulatory
ligands for autoreactive T cells. J. Immunol., 158, 3736?3752 (1996).
Документ
Категория
Без категории
Просмотров
2
Размер файла
126 Кб
Теги
443
1/--страниц
Пожаловаться на содержимое документа