Download Evidence for specific immune response against P210 BCR

Leukemia (1998) 12, 155–163
 1998 Stockton Press All rights reserved 0887-6924/98 $12.00
Evidence for specific immune response against P210 BCR-ABL in long-term remission
CML patients treated with interferon
T Oka1,2, KJ Sastry3, P Nehete3, SJ Schapiro3, JQ Guo1, M Talpaz4 and RB Arlinghaus1
Departments of 1Molecular Pathology and 4Clinical Immunology and Biological Therapy, The University of Texas, MD Anderson Cancer
Center, Houston, TX; and 3Department of Veterinary Sciences, The University of Texas at Science Park, Bastrop, TX, USA
Interferon-alpha treatment induces complete cytogenetic
remission in 25% of Philadelphia chromosome (Ph)-positive
chronic myelogenous leukemia (CML) patients. These
remissions are durable unlike remissions induced with other
therapies and yet residual leukemia is detectable in most of
these patients. Total peripheral blood mononuclear cells
(PBMCs) from CML patients in long-term remission following
interferon treatment exhibited significantly higher proliferative
responses (four- to 15-fold over background) than normals
directed against P210 BCR-ABL in extracts of transfected monkey fibroblast cells. Surprisingly, similar enhanced levels of
specific proliferative responses were observed with extracts
from cells expressing Bcr and/or Abl proteins. In contrast,
extracts from vector only or v-Mos-expressing cells had background level responses. Control monkey fibroblast cells lacking BCR-ABL expression failed to induce proliferation over
background levels. Normal individuals had no significant
responses to Bcr/Abl extracts. On the other hand, peripheral
blood mononuclear cells from allogeneic bone marrow transplant CML patients had proliferative responses to cell extracts
independent of Bcr-Abl. These data indicate that patients in
remission due to alpha-interferon treatment have significantly
higher levels of specific cellular immunoreactivity against
Bcr/Abl sequences than normal controls, which could play a
role in maintaining cytogenetic remission in Ph-positive CML
patients.
Keywords: BCR-ABL; CML; IFN-␣; cellular immunity
Introduction
The Philadelphia (Ph) chromosome was initially described as
an abnormal chromosome 22 in chronic myelogenous leukemia (CML).1,2 The Ph is the result of a reciprocal translocation
of chromosomes 9 and 22, t(9;22)(q34;q11).2 This translocation transposes the c-ABL proto-oncogene from its normal
position on chromosome 9 to a new location on chromosome
22 in the central region of the BCR gene. The hybrid BCRABL oncogene produces an abnormal 8.6 kb mRNA that
encodes the 210 kDa BCR-ABL protein (P210)3 with an activated tyrosine kinase activity.4,5 In CML, the BCR breakpoint
is generally between several small central exons termed exons
b2, b3 and b4. These exons correspond to exons 13, 14 and
15 of BCR, respectively. Depending on whether the junction
is between either exon b2 or exon b3 of BCR and exon 2 of
ABL, two different mRNA are formed with a junction structure
known as b2a2 or b3a2, respectively. The resulting P210 BCRABL protein found in CML patients and some acute lymphocytic leukemia (ALL) patients contains either 902 or 927
amino acids encoded by 5⬘ BCR sequences fused to 1096
Correspondence: RB Arlinghaus, Department of Molecular Pathology,
Box 89, University of Texas, MD Anderson Cancer Center, 1515 Holcombe Blvd, Houston, TX 77030, USA; 2Current address: Department
of Pathology, Okayama University Medical School, Shikata 2-5-1,
Okayama 700, Japan
Received 7 August 1997; accepted 24 October 1997
amino acids of 3⬘ ABL sequences. Another type of Bcr-Abl
protein detected predominantly in Ph-positive ALL contains
426 amino acids encoded by the first exon of BCR fused to
the second exon of ABL, termed e1a2, which results in the
expression of P185 BCR-ABL.6
CML is a pluripotent stem cell disease that primarily affects
myeloid lineages in chronic phase. The disease almost always
progresses to the blast crisis stage, which is the terminal stage.
Therapy of CML with interferon (IFN) is associated with frequent hematopoietic remissions and complete cytogenetic
remissions in about 25% of newly diagnosed treated patients.7
These remissions are rather durable with 92% of patients projected to remain in remission at least 8 years.8 The underlying
nature of this remarkable event is not well understood, but
two features best characterize it: (1) the durability of these
remissions, and (2) the fact that the residual disease can be
detected in most of these patients either by studies of patients’
blood and bone marrow, or by detecting BCR-ABL-positive
colonies in a clonogenic assay.7,8 Residual disease was
detected by the latter assay even in patients who had maintained complete remission for up to 3 years.9 How can we
thus explain the presence of residual clonogenic Ph-positive
cells, which apparently do not repopulate the bone marrow
but maintain a pattern of ‘dormancy’? We propose that complete cytogenetic remissions are the result of an IFN-induced
‘specific’ cellular immune response. Of interest, Chen et al10
reported that immunization of mice with synthetic peptides
comprising the b3a2 junction of P210 BCR-ABL primes peptide-specific CD4+, class-II major histocompatibility complex
(MHC) restricted T cells. T cell clones specific for the b3a2
peptide were shown to respond in the same way to partially
purified P210 BCR-ABL containing the b3a2 junction. These
results argue that BCR-ABL junction peptides and/or partially
purified P210 BCR-ABL can activate CD4+ helper T cells, and
raise the possibility that they may also activate cytotoxic T
lymphocyte (CTL) responses in vivo. Of interest, Barrett et al11
demonstrated that normal HLA DR-4 restricted human Tlymphocytes proliferate in response to decapeptides representing both b3a2 and b2a2 fusions of P210 BCR-ABL. In
the present study, we demonstrate that peripheral blood
mononuclear cells (PBMCs) from CML patients who
underwent interferon treatment and are in complete cytogenetic remission are stimulated to proliferate in response to
extracts of Bcr-Abl-expressing cells, suggesting a specific
cellular immune response against sequences of the oncoprotein P210 BCR-ABL.
Materials and methods
Cell lines
COS-1 cell line (American Type Culture Collection, Rockville,
MD, USA) is a simian fibroblast-like cell expressing SV40
Immune response against P210 BCR-ABL
T Oka et al
156
large T antigen; it was cultured in Dulbecco’s modified Eagle’s
medium (DMEM) supplemented with 10% fetal calf serum
(FCS). Human Bcr-Abl-positive CML cell line K56212 and
human Bcr-Abl-negative pre-B leukemia cell line SMS-SB13
were grown in RPMI-1640 with 10% FCS. These cells were
cultured at 37°C under a humidified 5.0% CO2 atmosphere.
Patient population
PBMCs were harvested from heparanized blood of CML
patients in complete cytogenetic remission following IFN
treatment. All patients completed the necessary consent forms.
Clinical details on patients in the study are summarized in
Table 1.
Cytogenetics
A minimum of 20 metaphases in G-banding chromosome
studies were examined according to the International System
for Human Cytogenetic Nomenclature.
Transient transfection and cell extracts
COS-1 cell transient transfection was performed by the
diethylaminoethyl (DEAE)-Dextran procedure.16 About 1.0 ×
108 COS-1 cells were transfected by c-ABL, BCR or BCR-ABL
expression plasmids as well as a vector only plasmid as a control; constructs were previously described.17 Cells were harvested 60 h after transfection. Transfections longer than 48 h
generate a mixture of both the intact Bcr, Abl or Bcr-Abl proteins, and truncated protein fragments. Cells were resuspended in 25 mM potassium phosphase (pH 7.0) with 2 mM
EDTA, 100 mM NaCl, 1 mM PMSF, 50 ␮g/ml of leupeptin, 100
units/ml of aprotinin and 20 mM benzamidine. The cells were
homogenized by use of ultrasonic cell disrupter (Virsonic 300;
Virtis, NY, USA); the homogenate was centrifuged for 10 min
at 3000 g, and supernatant fluid was ultracentrifuged at
110 000 g/min for 90 min at 4°C (Beckman, Fullerton, CA,
USA; 50 Ti rotor). Supernatant proteins in cell extracts were
adjusted to a concentration that originated from 3.0 × 107
cells/ml (after transfection), aliquoted and stored at −80°C.
Cell proliferation assay
Peptide synthesis
Peptides were made using the Merrifield solid phase method14
either on a modified Vega 250 automatic peptide synthesizer
or by the ‘Bag’ method as described by Houghton.15 In either
case, removal of T-BOC blocking groups and hydrolysis of the
peptide from the resin were accomplished by hydrofluoric
acid (HF) treatment at 0°C for 1 h. After HF treatment, either
extraction was used to remove various organic compounds
and peptides were extracted from the resin with 25% acetic
acid. Amino acid sequences of peptides were D60(b3a2)GFKQSSKALQRP; D61(b2a2)-LTINKEEALQRP; D62(e1a2)AFHGDAQALQRPVA and an unrelated Mos peptide
(SLCRYLPRELSPSVDSRSC). Peptides were dissolved in PBS
at 1.0 mg/ml.
Table 1
Patients
Clinical data of patients
Age
Sex
Length of
time on
study (months)
RD
FJ
LV
SGd
RM
CS
42
37
43
31
67
34
M
M
F
M
F
M
63
46
79
36
110
87
OZ
JC
BK
DP
FA
PS
MM
39
59
57
48
31
35
37
M
M
M
F
M
F
F
57
92
84
57
—
—
—
a
Percentage of Ph1(+) cells.
106 IU/day IFN injection.
c
106 IU/2 days IFN injection.
d
CML relapsed patient.
b
The protocol, based on our previous method18 is shown in
Figure 1. About 30 ml of whole heparanized blood was collected by venous puncture from either Ph-positive CML
patients undergoing IFN-␣ treatment and in complete cytogenetic remission or healthy volunteers. Peripheral blood
mononuclear cells were separated by Ficoll–Hypaque
(Histopaque 1077; Sigma, St Louis, MO, USA) centrifugation
and were tested for proliferative responses. Three kinds of
BCR-ABL junction peptides and unrelated mos synthetic peptide were tested. In addition, several kinds of cell extracts
including K562, SMS, and COS-1 cells transfected with BCR,
c-ABL or BCR-ABL genes.17 Cells were suspended at 2 × 106
cells/ml of RPMI-1640 supplemented with 10% FCS. A 100 ␮l
aliquot of cell suspension was dispensed into each well of a
U-bottom 96-well microtiter plate and incubated with differ-
Treatment
rIFN␣
rIFN␣
rIFN␣
rIFN␣
rIFN␣
rIFN␣
rIF␥
rIFN␣
rIFN␣
rIFN␣
rIFN␣
BMT
BMT
BMT
Initial
dose
9.0 MIU q db
10.0 MIU q d
8.0 MIU q d
8.9 MIU q d
6.0 MIU q d
10.0 MIU q d
0.5 mg q d
9.4 MIU q d
10.0 MIU q d
4.4 MIU q d
8.5 MIU q d
—
—
—
Last
dose
9.2 MIU
4.0 MIU q
8.0 MIU q
8.9 MIU
6.0 MIU q
5.0 MIU q
—
9.4 MIU q
4.0 MIU q
4.4 MIU q
6.0 MIU q
—
—
—
Duration of
response
(months)
d
d
odc
od
d
d
d
d
Recent cytogenetics
(×103/mm3)
WBC
Lymph
Pha
21
19
48
17
96
43
3.4
10.3
2.6
2.8
3.9
6.2
0.9
1.6
0.9
1.1
0.1
2.0
0.0
0.0
0.0
60.0
5.5
0.0
28
44
57
48
—
—
—
2.4
6.8
3.0
2.5
4.7
8.5
6.4
0.9
0.3
1.5
1.0
1.6
3.6
3.0
10.0
0.0
5.0
0.0
0.0
0.0
0.0
Immune response against P210 BCR-ABL
T Oka et al
Western blotting analysis
Western blotting analysis was performed according to the
technique of Towbin et al.20 Cell extracts were lysed in boiling
sodium dodecyl sulfate (SDS) sample buffer for 5 min, separated by 6.5% SDS/PAGE, transferred electrophoretically to
polyvinylidene difluoride membrane (Immobilon; Millipore,
Bedford, MA, USA) and then treated with 1:20 000 dilution of
anti-ABL (8E9) monoclonal antibody.21 Immunoreactive bands
were visualized with peroxidase-labeled goat anti-mouse Ig
following reaction with the substrate of the enhanced chemiluminescence (ECL) Western blotting system (Amersham,
Arlington Heights, IL, USA) and subjected to autoradiography.
Results
Extracts from Bcr-Abl expressing cells stimulate
proliferation of PBMC from CML remission patients
Figure 1
Proliferation assay protocol for PBMCs. The procedure
is described in Materials and methods. RBCs, red blood cells; PHA,
phytohemagglutinin; IL2, interleukin 2.
ent concentrations of synthetic peptides, cell extracts or
medium alone in a total volume of 200 ␮l. Peptide or extracts
were added at 20 ␮l volume. For junction peptides, a 1:10
dilution of peptide stock would yield a concentration in the
assay of 10 ␮g/ml. All treatments were done in triplicate wells.
The cultures were incubated for 72 h at 37°C in a humidified
5% CO2 atmosphere, and during the last 16–18 h, 1 ␮Ci of
3
H-thymidine (6–7 Ci/m mol; ICN Biochemicals, Costa Mesa,
CA, USA) was added. Cells were harvested onto filter strips
for estimating 3H-incorporation and counted using a liquid
scintillation counter.
T cell purification
In some experiments, purified T cells were used in the proliferation assay. T cells were purified by the L-leucine methyl ester
(Leu-OMe) method as described by Thiele et al19 combined
with the IgG conjugate immunobeads method. Briefly after
separation by the Ficoll–Hypaque method, PBMCs were resuspended in RPMI 1640, containing 10% FCS and cultured in
plastic petri dishes for 1 h at 37°C. Adherent cells bound to
the culture matrix were pooled as the macrophage–monocyte
fraction. Non-adherent cells were treated with 5 mM Leu-OMe
for 50 min in room temperature to lyse residual macrophages
and monocytes and washed twice with Dulbecco’s BSS followed by treatment with 10 mg/ml of DNase-I for 30 min at
37°C. After washing twice with PBS, B cells were absorbed
with IgG conjugate immunobeads. The eluted fraction was
used as the purified T cell fraction.
Total PBMCs from CML patients in complete cytogenetic
remission as a result of IFN-␣ treatment (Table 1) were tested
for proliferative responses against each of three kinds of BCRABL junction peptides: D60 (b3a2); D61 (b2a2); D62 (e1a2)
and an unrelated mos peptide. The protocol is described in
Figure 1. To our surprise except in a few patients, none of the
BCR-ABL junction peptides at any of the three concentrations
tested consistently stimulated proliferation at levels higher
than PBMCs incubated in medium alone or with a control mos
peptide (Figure 2a). However, cell extracts from COS-1 cells
transfected with pSG-BCR-ABL plasmid (COS P210) strongly
stimulated the proliferation of PBMCs from these CML patients
(Figure 2a and b). Cell extracts from K562 cells consistently
stimulated proliferation at some dilution conditions but at
lower levels than BCR-ABL COS-1 cell extracts (Table 2). Significant proliferation was not observed in several controls
including cell extracts of Bcr-Abl negative SMS cells, nontransfected COS-1 cells and COS-1 cells transfected with the
vector plasmid pSG5 lacking a gene insert (Figure 2a, Table
2). In addition, PBMCs from healthy volunteers showed no
significant response against any Bcr-Abl junction peptide or
cell extracts (Figure 3).
Table 2 summarizes the proliferative responses of PBMCs
from patients and normal controls to the various cell extracts.
The stimulation index (SI) is defined as the ratio of average
counts per minutes (c.p.m.) of each experimental condition
over an average c.p.m. of PBMCs in medium alone. PBMCs
from the CML patients showed a three to 11 times higher proliferative response to COS P210 cell extracts compared to
COS-1 extract lacking P210. PBMCs from six normal healthy
individuals showed no significant response to extracts of COS
P210 and that of control COS-1 cells. The SI of PBMCs treated
with phytohemagglutinin (PHA) ranged from 17.21 to 206.15
in samples from CML patients and those of healthy volunteers
with no significant differences in responses between the two
groups.
Of interest, patient SG who was relapsing had a very low
response to COS P210 (Table 2), suggesting a possible negative correlation of cellular immunity towards Bcr-Abl with leukemia in this patient. Whether these results reflect a role for
cellular immunity in the control of the leukemia in the IFNtreated remission patients requires further study.
Several experiments were performed to determine whether
T cells were responsible for the observed stimulated proliferation. The results indicated that purified T cells alone did not
157
Immune response against P210 BCR-ABL
T Oka et al
158
Figure 2
Specific proliferative responses to cell extracts expressing
P210 BCR-ABL in PBMCs from Ph-positive CML patients with IFN-␣induced complete cytogenetic remission. (a) PBMCs from CML patient
LV with IFN-␣ treatment induced complete cytogenetic remission (see
Table 1 for details). PBMCs from patient LV in this experiment
responded to PHA (45 623 c.p.m. ± 1378). PBMCs were cultured in
the medium alone or with IL-2 (10 ␮g/ml), or in the presence of various peptides or cell extracts. The following abbreviations are shown in
the abscissa: BCR3-ABL2 12mer peptide (D60), or BCR2-ABL2 12mer
peptide (D61), BCR1-ABL2 12mer peptide (D62), mos peptide, or cell
extracts from K562 cells, SMS cells, COS-1 cells transfected with pSGBCR-ABL (COSp210), COS-1 cells transfected with vector plasmid
pSG5 (COSpSG5), COS-1 cells transfected with v-mos plasmid
(COSv-mos), or COS-1 cells not-transfected (COS). The bar graphs
represent results from several dilution series. The cultures were
assayed after 72 h incubation using 3H-thymidine (6–7 Ci/mmol)
incorporation during the last 16 to 18 h. Bars indicate the means standard deviation (s.d.) of triplicate values of the c.p.m. (b) PBMCs from
IFN remission patient DP lack response to vector-only expressing
COS-1 cells. In this experiment PBMCs from this patient responded
to PHA (3596 c.p.m. ± 1711).
Immune response against P210 BCR-ABL
T Oka et al
Table 2
Patients
RD
FJ
LV
SGd
RM
CS
OZ
JC
BK
Nor
Nor
Nor
Nor
Nor
Nor
1
2
3
4
5
6
159
Proliferative responses of PBMCs from interferon-treated CML patients with BCR-ABL-related proteins
Ratioc
Stimulation index (SI)
COSa+ba COSpSGb COSp210
IL2
K562
SMS
COS
COSabl
COSbcr
4.07
4.84
6.75
2.26
ND
2.52
ND
2.30
ND
1.49
1.30
3.97
1.92
4.46
2.92
1.07
1.46
0.69
0.82
0.84
0.62
0.30
ND
1.55
ND
0.80
ND
0.98
0.68
0.99
0.47
2.05
1.51
1.11
1.16
1.59
ND
ND
ND
ND
ND
5.58
7.10
2.78
4.63
ND
ND
ND
ND
ND
5.09
7.68
4.40
3.75
ND
ND
ND
ND
1.81
5.31
7.63
3.27
5.48
ND
ND
ND
ND
ND
ND
ND
ND
ND
6.37
3.66
11.50
1.52
17.22
4.24
9.70
2.94
5.68
6.50
5.38
11.62
3.23
8.40
2.81
8.74
2.53
3.75
4.39
2.67
1.87
2.42
3.64
3.85
0.57
0.56
0.95
0.94
0.86
1.05
0.82
0.79
1.01
0.61
0.86
1.23
1.16
1.16
1.64
0.92
1.32
1.39
ND
1.17
1.08
0.69
1.03
1.58
ND
1.20
1.07
0.57
1.08
1.64
1.43
1.14
0.89
0.80
0.92
1.49
ND
0.88
0.97
0.61
1.20
1.56
1.57
0.53
0.66
0.66
0.82
0.80
1.35
0.47
0.40
0.72
0.62
0.58
SI, average (c.p.m. experimental)/average (c.p.m. cell only).
SI (PHA) ranged from 17.21 to 206.15.
a
COS-1 cells transfected with BCR and c-ABL.
b
COS-1 cells transfected with vector only (pSG5).
c
Ratio of SI of COSp210 over COS.
d
CML relapsed patient.
ND, not done.
Figure 3
PBMCs from a healthy volunteer did not proliferate in response to COS P210 extracts. PBMCs from this normal donor responded
to PHA (38 400 c.p.m. ± 459). D60, D61, D62 represent BCR-ABL peptides as in Figure 2a. D155-P is a phosphotyrosine BCR peptide
(VSPSPTTpYRMFRDK); D153 the same BCR peptide made without phosphorylation of the tyrosine residue pSG5; pSG5 is a commercial vector
lacking a coding insert; COS v-Mos are transiently transfected COS-1 cells expressing p37v-mos. P210 is the same as COS P210.
Immune response against P210 BCR-ABL
T Oka et al
160
have increased proliferation in response to COS P210 compared to COS-1 alone but purified T cells mixed with an identical number of irradiated PBMCs from the same patient gave a
positive response to COS P210 over and above COS-1 extracts
(results not shown).
Statistical analysis (Student’s t-test; df, 13) indicated that significantly higher specific proliferative responses to COS P210
were observed for CML patients than for normal subjects (t =
3.2, P ⭐ 0.01). Moreover, proliferative responses generated
from K562 cell extracts were significantly higher for CML
patients than for normal subjects (t = 2.4, P ⭐ 0.05). In
addition, proliferative responses generated from cell extracts
of K562 and COS P210 were significantly higher than those
from COS cell extracts for patients (t = 2.4, P ⭐ 0.05 and t =
3.7, P ⭐ 0.01, respectively) but not for normal subjects.
Extracts of cells expressing Bcr and Abl protein
sequences stimulate PBMC proliferation from CML
remission patients
Because we observed proliferative responses against P210 but
not junction BCR-ABL peptides in our assay, we tested
whether extracts of cells expressing either Bcr or Abl protein
sequences would be recognized by PBMCs from IFN
remission patients. Surprisingly, extracts from COS-1 cells
transfected with pSGc-ABL, pSGc-BCR or both, stimulated the
proliferation of CML patient PBMCs over that of background,
but did not stimulate that of healthy volunteers (Table 2).
These results indicate that PBMCs from CML remission
patients behave as if they have been immunologically primed
to recognize normal Bcr and Abl sequences.
Student’s t-test (df = 13) indicated that significant differences
were observed between patients and normal subjects in their
proliferative responses against extracts from COS cells transfected with Abl (t = 4.8, P ⭐ 0.01), Bcr (t = 5.3, P ⭐ 0.001),
and a combination of Abl and Bcr (t = 4.2, P ⭐ 0.01).
PBMCs from allogeneic transplant CML patients lack a
specific response to Bcr-Abl
In an effort to determine whether CML patients who had
received a bone marrow transplant also possessed the ability
to proliferate in response to Bcr-Abl extracts, we assayed
PBMCs from three bone marrow transplant patients in
remission. PBMCs from allogeneic bone marrow transplanted
CML patients showed uniformly high levels of proliferation in
response to extracts alone or extracts expressing Bcr, Abl and
Bcr-Abl (Table 3). Similar activation was observed with P210
BCR-ABL junction peptides and the mos peptide (data not
shown). In these experiments, there was no significant difference between average ratios of SI value obtained with extract
of COS-1 cells transfected with P210 BCR-ABL plasmid and
that of non-transfected COS-1 cells (Table 3). Samples from
CML transplant patients and healthy volunteers had high
responses to PHA, indicating that T cells within the PBMC
preparation were responsive to antigen stimulation.
Immunoblotting analysis of Bcr/Abl expressing cells
To establish that COS-1 cells transfected with the pSG5 BCRABL plasmid expressed P210 BCR-ABL, we performed Western blotting analysis using an anti-Abl (8E9) antibody.21 The
results showed an intense band of P210 BCR-ABL in transfected but not control COS cell extracts. COS P210 cells contained many degradated forms of the Bcr-Abl protein besides
intact P210 BCR-ABL and P145 c-ABL. In contrast, K562 cells
had relatively intact P210 BCR-ABL and P145 c-ABL proteins
and much lower amounts of degraded BCR-ABL products
(Figure 4). Similar levels of degraded Bcr/Abl proteins were
observed in COS cells transfected with either BCR or ABL or
both cDNAs (not shown).
Discussion
CML is a pluripotent stem cell disease that primarily affects the
myeloid lineage in chronic phase. IFN-␣ induces a complete
cytogenetic remission in 25% of treated patients.7 These
remissions are durable unlike remissions induced by other
therapies and yet residual leukemia is detectable in most of
these patients.8 Understanding the mechanisms that govern
this remission may help to reveal mechanisms of remission in
leukemia in a broader way and help design improved therapies for minimal residual leukemia. However, the nature of
this remarkable event is not well understood. Our cell proliferation data are indicative of a statistically significant specific
immune response directed against proteins expressed in COS
P210 cells; this response was detected in PBMCs from CML
patients in complete cytogenetic remission due to interferon
treatment. Significant proliferation was also detected with
extracts from a CML cell line (K562) expressing P210 BCRABL. However, stimulation of proliferation was much more
efficient with COS P210 extracts than with K562 cell extracts.
Western blot analysis detected larger amounts of P210 BCRABL degradation products in COS P210 cell extracts, com-
Table 3
Proliferative response of PBMCs from allogeneic bone marrow transplanted CML patient’s PBMC to cell extracts expressing BCRABL-related proteins
Patients
PS
MM
FA
Ratioc
Stimulation index (SI)
K562
SMS
COS
COSabl
COSbcr
COSa+ba
COSpSGb
COSp210
1.14
2.02
2.45
—
—
6.58
4.75
2.96
7.00
3.86
6.17
6.26
4.23
4.03
6.82
4.39
4.66
6.62
—
—
2.80
3.19
4.80
7.84
SI, average (c.p.m. experimental)/average (c.p.m. cell only).
SI (PHA) ranged from 50.52 to 111.54.
a
COS-1 cells transfected with BCR and c-ABL.
b
COS-1 cells transfected with vector only (pSG5).
c
Ratio of SI of COSp210 over COS.
Normal healthy controls are shown in Table 2.
0.67
2.38
1.12
Immune response against P210 BCR-ABL
T Oka et al
Figure 4
Western blot analysis of P210 BCR-ABL expressed in
K562 cells, COS-1 cells transfected with pSG-BCR-ABL and control
COS-1 cells. The expression of P210 BCR-ABL was detected by antiAbl 8E9 monoclonal antibody. An equal amount of protein was
applied to each gel lane.
pared to that in K562 cells (Figure 4). This may suggest that
higher expression of Bcr-Abl degradation products present in
the transfected COS-1 cell extracts compared to K562 cell
extracts might contribute to the higher efficiency of presentation of Bcr-Abl peptides by antigen-presenting cells contained within the PBMC preparation. The specificity of these
phenomena observed in PBMCs of IFN-treated CML patients
strongly suggests that IFN treatment stimulates cell-mediated
immunity against BCR-ABL expressing leukemic cells within
CML patients.
Because of the complexity of these assays and the reagents
used, we wanted to test the reproducibility of these findings
by another group of individuals. Dr Talpaz’s immunology
group were able to confirm these findings but used 5–7-day
incubation periods with PBMCs instead of 3 days as in this
study (X Qui, C Newell, K Cheng and M Talpaz, unpublished results).
Surprisingly, extracts from COS-1 cells expressing normal
Bcr and Abl proteins or both also induced significant proliferation of PBMCs from CML long-term remission patients. In
contrast, vector only transfected cells and mos-transfected
lacked stimulatory activity. These findings suggest that Bcr and
Abl protein sequences are recognized by memory T cells as
a result of IFN treatment.
We cannot eliminate the possibility that the cell-mediated
immune responses are generated against proteins induced in
COS-1 cells by P210 BCR-ABL. This possibility is in our view
unlikely because Bcr and Abl overexpression would have to
induce the same or related antigens. Also, these experiments
were not done with stably transfected cells but with COS-1
cells transiently transfected for only 3 days in which less than
10% of the cells are expressing either Bcr-Abl or Bcr/Abl
proteins.
Before beginning these studies, we expected that BCR-ABL
junction peptides would stimulate proliferation of PBMCs
from IFN responders. However, such junction peptides consistently failed to activate PBMCs from IFN responders. These
findings suggest that the response to COS P210 extracts is due
to sequences within Bcr or Abl proteins, but not to junction
sequences. In this regard we observed that extracts of COS-1
cells expressing Bcr, Abl or both Bcr and Abl proteins induced
significant proliferation of PBMCs from IFN responders. The
response was similar in magnitude to that observed with COS
P210 extracts. These results were puzzling to us at first. However, these findings could be explained by previous observations made by others in which normal proteins such as
HER2/neu expressed in high amounts appeared to be sufficient to activate T cells.22 A mechanism for recognition of
normal antigens has been proposed by Ioannides et al23 and
later extended by Fisk et al.22 Their results showed that a 9mer peptide from HER-2 (369–377) was efficient in sensitizing
human CEM cells for lysis by CTL lines that were ovarian
tumor-specific. Disis et al24 demonstrated induction of CTLs
from normal lymphocytes with peptides derived from the
HER-2/neu proto-oncogene. These results suggest that over
expression of a normal antigen could in some situations
impart immunogenicity to the protein and lead to a leukemic
cell-specific CTL response. Therefore in the case of CML, IFN
treatment may enhance the cell-mediated immune response
system in a way that allows efficient induction and expression
of helper and cytotoxic T lymphocytes directed against BcrAbl-expressing cells. Possibly high expression of the Bcr-Abl
protein antigen leads to high amounts of presentation by class
I and II molecules in the presence of IFN, thereby overcoming
the natural lack of T cell activation by normal gene products.
That this interpretation may have some merit is supported by
our observation that extracts of COS-1 cells overexpressing
Bcr or Abl proteins provide high levels of protein fragments
(Figure 4) that could be presented by antigen-presenting cells
within the PBMC preparation, leading to T cell activation.
Thus, either Bcr protein fragments or Abl protein fragments
would trigger memory T cells that were initially primed by
leukemic cells expressing Bcr-Abl proteins, but COS-1 cells
not expressing high amounts of Bcr or Abl proteins would not
yield extracts that stimulated PBMC proliferation.
Of interest in this regard, SMS cells are a human leukemia
cell line expressing moderate levels of Bcr and Abl proteins
yet extracts from these cells did not stimulate proliferation of
PBMCs from IFN-treated patients (Table 2). This result
reinforces the interpretation that it is overexpression combined
with protein degradation of Bcr and Abl sequences that lead
to activation of patient PBMCs.
In contrast to the proliferative response exhibited by the
PBMCs from IFN responders, PBMCs from allogeneic bone
marrow transplantation patients were reactive to every antigen
tested including zenotypic stimulation, suggesting that the
immune reactivity in bone marrow transplant patients is an
allo-antigen-specific mixed lymphocyte reaction, which is distinct from that mediated by Bcr/Abl sequences.
161
Immune response against P210 BCR-ABL
T Oka et al
162
Bcr-Abl junction peptides have been shown to elicit a CD4
response in a healthy, normal individual.25 In these studies a
17 amino acid b3 a2 junction peptide was used to generate
a Bcr-Abl-specific T cell line from a healthy donor that proliferated in response to the peptide. Proliferative responses
have also been observed in normal donor cells (three of seven)
tested against a 25 amino acid b3 a2 peptide.26 In this report,
four of four HLA-A3 donor cell populations also produced
cytotoxic T lymphocytes (CTLs) in vitro in response to a junction peptide mixture. These CTLs lysed an allogeneic HLAA3-matched leukemia cell line pulsed with the same peptide
mixture. The best inducer was a KQSSKALQR (b3 a2
peptide)/ATGFKQSSK (b3 peptide) mixture. We cannot
explain why the long-term remission patients tested in our
study did not consistently respond to Bcr-Abl junction peptides. The amount of peptide used in our study could be a
factor as the concentration of junction peptide used in one
study26 (50 ␮g/ml) was five times higher than the concentration tested in our study (Figure 2a). Also, longer exposures
to PBMCs (we used 3 days) from IFN remission patients might
be a factor. Nevertheless, our results suggest that IFN may
enhance presentation of normal antigens in general as suggested by the relative high frequency of autoimmune symptoms seen in some of these patients (Talpaz, unpublished
results). Thus, these patients may generate an antileukemic
response due to enhanced recognition of Bcr and Abl gene
products.
Our data raise questions about whether or not CML patients
with active disease are able to mount an immune response
against BCR-ABL expressing leukemic cells. One possible
immune deficit is the lack of appropriate helper T cell function
necessary to elicit an effective immune response. Barrett et
al11 showed that normal individuals with DR4 class II haplotype had T cells that proliferated in response to the two forms
of Bcr-Abl junctions: b2a2 and b3a2. However, individuals
with DR 1,2,3 and 5 did not. Possibly, IFN treatment may
activate a T cell immune response in those patients that have
the DR4 haplotype. Of interest, only about 25% of CML
patients respond positively to IFN treatment. It will be of interest to determine whether IFN-sensitive patients share the same
HLA pattern. Another possibility is that BCR-ABL expression
down-regulates class I and/or class II MHC gene expression,
but in some patients, IFN treatment is able to overcome this
down-regulation. IFN might function in another way by overcoming T cell anergy against Bcr-Abl antigen presentation. In
this case, IFN would activate cell-mediated immune responses
against any cell that overexpresses Bcr-Abl proteins. Other
possibilities include lack of appropriate HLA expression in
CML patients and IFN treatment would activate the pertinent
HLA allele, or BCR-ABL expression down-regulates cytotoxic
T cell function and IFN stimulates CTL function. The latter
possibility seems unlikely, as patients in chronic phase are
typically free of immune suppression symptoms.
This research has important implications with regard to
possible immunotherapy of CML. If immunologic peptides
derived from Bcr or Abl proteins specific to CML could be
identified, those peptides could form the basis of a future
immunotherapeutic vaccine. Fragmented P210 BCR-ABL
could also be used for this purpose. In addition, if CTL effector
clones can be isolated that are specific for P210 BCR-ABL
expression in HLA-matched cell lines and/or Bcr/Abl peptidetreated cells from the same patient, these clones would be
useful for in vitro purging of marrow cells from CML patients
and may also be useful for treating allogeneic bone marrow
transplant patients. Further investigations are underway to
expand this study and to examine basic immunologic
parameters.
Acknowledgements
This work was supported by NIH grants CA65611 and
CA16672 and Hubert L Stringer Chair. We would like to gratefully acknowledge the technical assistance of Drs Dai Lu and
Jiaxin Liu as well as the assistance of Tammy Trlicek for manuscript preparation. We also thank Dr Brad McIntyre for his
help in preparing purified T cells and monocytes.
References
1 Nowell PC, Hungerford DA. A minute chromosome in human
chronic granulocytic leukemia. Science 1960; 132: 1497–1499.
2 Rowley JD. A new consistent chromosomal abnormality in chronic
myelogenous leukemia identified by quinacrine fluorescence and
Giemsa staining. Nature 1973; 243: 290–293.
3 Shtivelman E, Lifschitz B, Gale RP, Canaani E. Fused transcript of
abl and bcr genes in chronic myelogenous leukemia. Nature 1985;
315: 550–554.
4 Konopka JB, Watanabe SM, Witte ON. An alteration of the human
c-abl protein in K562 leukemia cells unmasks associated tyrosine
kinase activity. Cell 1984; 37: 1035–1042.
5 Kloetzer W, Kurzrock R, Smith L, Talpaz M, Spiller M, Gutterman
J, Arlinghaus R. The human cellular abl gene product in the
chronic myelogenous leukemia cell line K562 has an associated
tyrosine protein kinase activity. Virology 1985; 110: 230–238.
6 Hermans AS, Heisterkamp N, von Lindern M, vanBaal S, Meijer
D, van der Plas D, Wiedemann LM, Groffen J, Bootsma D, Grosveld G. Unique fusion of bcr and c-abl genes in Philadelphia chromosome positive acute lymphoblastic leukemia. Cell 1987; 51:
33–40.
7 Kantarjian HM, Deisseroth AB, Kurzrock R, Estrov Z, Talpaz M.
Chronic myelogenous leukemia; a concise update. Blood 1993;
82: 691–703.
8 Dhingra K, Kurzrock R, Baine R, Eastman S, Kantarjian H, Gutterman JU, Talpaz M. Minimal residual disease in interferontreated chronic myelogenous leukemia; results and pitfalls of
analysis based on polymerase chain reaction. Leukemia 1992; 6:
754–757.
9 Talpaz M, Estrov Z, Kantarjian H, Ku S, Foteh A, Kurzrock R. Persistence of dormant leukemic progenitors during interferoninduced remission in chronic myelogenous leukemia. J Clin Invest
1994; 94: 1383–1389.
10 Chen W, Peace DJ, Rovira D. T-cell immunity to the joining region
of P210 BCR-ABL protein. Proc Natl Acad Sci USA 1992; 89:
1468–1472.
11 Barrett AJ, Jiang YZ, Kars A, Gordon AA, Datta A. HLA DR-4
restricted T-lymphocytes recognize decapeptides representing the
novel fusion region of the BCR-ABL protein in CML. J Cell
Biochem 1992; 16D (Suppl.): 26 (Abstr.).
12 Lozzio CB, Lozzio BB. Human chronic myelogenous leukemia
cell line with positive Philadelphia chromosome. Blood 1975; 45:
321–334.
13 Smith RG, Dev VG, Shannon WA Jr. Characterization of a novel
pre-B leukemia cell line. J Immunol 1981; 126: 596–602.
14 Merrifield RB. Solid phase peptide synthesis. J Am Chem Soc
1963; 85: 2149–2154.
15 Houghten RA. General method for the rapid solid-phase synthesis
of large numbers of peptides; specificity of antigen–antibody interaction at the level of individual amino acids. Proc Natl Acad Sci
USA 1985; 76: 4350–4354.
16 Sambrook J, Fritsch EF, Maniatis T. Molecular Cloning: A Laboratory Manual. Cold Spring Harbor Laboratory Press: Cold Spring
Harbor, 1989, p 16.42.
17 Lu D, Liu J, Campbell M, Guo JQ, Heisterkamp N, Groffen J,
Canaani E, Arlinghaus RB. Tyrosine phosphorylation of p160BCR
by P210 BCR-ABL. Blood 1993; 82: 1257–1263.
18 Nehete PN, Johnson PC, Shapiro SJ, Arlinghaus RB, Sastry KJ.
Immune response against P210 BCR-ABL
T Oka et al
19
20
21
22
Cross reactive T-cell proliferative responses to V3 peptides corresponding to different geological HIV-1 isolates in HIV-seropositive
individuals. J Clin Immunol 1996; 16: 115–124.
Thiele DL, Kurosaka M, Lipsky PE. Phenotype of the accessory
cell necessary for mitogen-stimulated T and B cell responses in
human peripheral blood; delineation by its sensitivity to the lysosomotropic agent, L-leucine methyl ester. J Immunol 1993; 131:
2282–2290.
Towbin H, Staehelin T, Gordon J. Electrophoretic transfer of proteins from polyacrylamide gels to nitrocellulose sheets: procedure
and some applications. Proc Natl Acad Sci USA 1979; 76:
4350–4354.
Guo JQ, Lian JY, Xian YM, Lee M-S, Deisseroth AB, Stass SA,
Champlin RE, Talpaz M, Wang JYJ, Arlinghaus RB. BCR-ABL protein expression in peripheral blood cells of chronic myelogenous
leukemia patients undergoing therapy. Blood 1994; 83: 3629–
3637.
Fisk B, Blevins TL, Wharton JT, Ioannides CG. Identification of an
23
24
25
26
immunodominant peptide of HER-2/neu proto-oncogene recognized by ovarian tumor-specific cytotoxic T lymphocyte lines. J
Exp Med 1995; 181: 2109–2117.
Ioannides CG, Ioannides MG, O’Brian CA. T cell recognition of
oncogene products: a new strategy for immunotherapy. Mol Carcinogen 1992; 6: 77–82.
Disis ML, Smith JW, Murphy AE, Chen W, Cheever MA. In vitro
generation of human cytolytic T cells specific for peptides derived
from the HER-2/neu proto-oncogene protein. Cancer Res 1994;
54: 1071–1076.
ten Bosch GJA, Joosten AM, Kessler JH, Melief CJM, Leeksma OC.
Recognition of peptides corresponding to the joining region of
p210BCR-ABL protein by human T cells. Leukemia 1995; 9:
1344–1348.
Bocchia M, Korontsvit T, Xu Q, Mackinnon S, Yang SY, Sette A,
Scheinberg DA. Specific human cellular immunity to bcr-abl
oncogene-derived peptides. Blood 1996; 87: 3587–3592.
163