Antibody to ras Proteins in Patients with Colon Cancer1pan-ras antibody (pan-ras Ab3) specific for both mutated and wild-type proteins (Oncogene Science). The fractions contain-ing

$Page 1: Antibody to ras Proteins in Patients with Colon Cancer1pan-ras antibody (pan-ras Ab3) specific for both mutated and wild-type proteins (Oncogene Science). The fractions contain-ing$
Vol. 1, 1071-1077. October 1995 Clinical Cancer Research 1071

Advances in Brief

Antibody to ras Proteins in Patients with Colon Cancer1

Masazumi Takahashi,2 Wei Chen,3 David R. Byrd,

Mary L. Disis, Eric S. Huseby, Huilian Qin,

Larry McCahill, Heidi Nelson, Hiroshi Shimada,

Kiyotaka Okuno, Masayuki Yasutomi,

David J. Peace, and Martin A. Cheever

Division of Oncology, Departments of Medicine [M. T., W. C.,M. L. D., E. S. H., H. Q., M. A. C.] and Surgery [D. R. B., L. M.],University of Washington, Seattle, Washington 98195; Colon and

Rectal Surgery, Mayo Clinic, Rochester, Minnesota 55905 [H. N.];

Division of Hematology/Oncology, Loyola University, Maywood,

Illinois 60153 [D. J. P.]; Department of Surgery, Yokohama City

University School of Medicine, Yokohama 236, Japan [H. S.]; andKinki University, Osaka 589, Japan [K. 0., M. Y.J

Abstract

The current study examined sera from 160 colon cancerpatients and 60 normal individuals to determine whether

antibody to mutated p21 ran protein was present. Studies

focused on the aspartic acid substitution at amino acidposition 12 (denoted D12), one of the most common muta-

tions in colon adenocarcinoma. IgA antibodies directed

against mutated p2! ras-D12 protein were detected in 51

(32%) of 160 colon cancer patients, but only in 1 (2.5%) of

40 normal individuals. The greater incidence of antibody in

cancer patients provides presumptive evidence that immu-nization to the ras proteins occurred as a result of themalignancy. Examination of sera for antibody reactivity to

wild-type p21 ras protein (denoted p2! ras-G12) as well as

p2! ran proteins bearing the D12, V12, S12, or L61 muta-tions showed that antibody detected was largely to normal

segments of the p21 ras protein. Epitope mapping, using

peptide neutralization assays with mutated or normal raspeptides as competitors, demonstrated that in 10 (67%) of 15

sera examined the antibody reactivity to p2! ras-G12 pro-

tein was neutralized by peptides near the carboxyl terminus

of p21 ras protein, but not by peptides spanning the specific

point mutation region. Antibody reactivities correlated with

peripheral blood lymphocyte count, but did not correlatewith patient age, sex, histology, stage, tumor locus, lymphnode metastasis, or serum carcinoembryonic antigen.

IntroductionThe ras oncogenes are cancer-related genes that become

activated by specific point mutations and can be identified in a

Received 5/4/95; revised 7/10/95; accepted 7/13/95.This work was supported by Grant CA54561 from the National Cancer

Institute, Department of Health and Human Services.2 Present address: Department of Surgery, Yokohama City University

School of Medicine, 3-9 Fukuura Kanazawa-ku, Yokohama 236, Japan.To whom requests for reprints should be addressed, at Division of

Oncology, University of Washington, Box 356527, Seattle, WA 98195-

6527.

wide variety of human malignancies (I, 2). Three ras genes

have been defined in the human genome, H-ras-1, K-ras-2, and

N-ras (3). These genes encode a highly conserved group of Mr

21,000 proteins, denoted as p21 ras. Activation of the ras

proto-oncogene occurs most commonly at codon 12 or codon 61

and results in corresponding single amino acid substitutions

within the p2l protein. The activating amino acid substitutions

impair the GTPase activity of the ras protein and generate

constitutively activated signaling complexes with transforming

activity (4, 5). Mutated ras proteins are implicated in malignant

transformation of many human cancers including approximately

45% of colon adenocarcinomas. The most common specific

amino acid substitutions replacing glycine at amino acid posi-

tion 12 of the wild-type p21 ras protein in colon adenocarci-

noma are aspartic acid (13%), valine (15%), cysteine (6%), or

serine (4%; Ref. 6). Only a few cases with leucine replacing

glutamine at amino acid position 61 have been reported.

Mutated p2l ras proteins are not expressed by normal

tissue and thus represent cancer-specific proteins. Mice immu-

nized to whole mutated ras proteins are able to mount a T-cell

response that is specific for the mutated segment of the molecule

(7, 8). Conversely, immunizing animals with peptides that span

the mutated segment can result in the generation of both helper/

inducer and cytolytic T-cell responses that recognize mutated

protein (9, 10). Recently, it has been demonstrated that T cells

from normal individuals can recognize peptides that span the

mutated segment of the molecule (11-13). Furthermore, mem-

ory T cells, isolated from a patient with a follicular thyroid

carcinoma, were found to specifically recognize a ras peptide

with a leucine substitution at residue 61 (14), a common muta-

tion in that tumor. Studies in our laboratory examining periph-

eral blood lymphocytes from patients with pancreatic cancer

showed that CD4� T-celI immunity to the mutated segment of

ras protein was present in some patients with pancreatic cancer.

Studies focused on the aspartic acid substitution (denoted D12)

as the most common mutation. Patients were found in whom T

cells responded to both ras D12 peptides and p21 ras-D12

protein (15). Existent specific T-celI reactivity to mutated ras

peptides and proteins imply that the patient’s own tumor acts as

a source of immunizing protein and that mutated ras proteins

can elicit immune responses in humans.

Often, helper/inducer T cells and antibodies can respond to

the same protein. The current study examined whether antibod-

ies specific to ras proteins exist in sera of patients with colon

cancer. Sera from 160 patients with colon cancer and 60 normal

donors were examined to evaluate antibody reactivity to p21 ras

proteins using ELISA. Studies focused on the aspartic acid

substitution (D12). IgA antibodies directed against p21 ras-Dl2

protein could be detected in colon cancer patients to a substan-

tially greater extent than in normal individuals. However, the

majority of antibody detected was to the normal but not the

mutated portion of p21 ras proteins. The greater incidence of

antibody in cancer patients provides evidence that immunization

to the ras proteins occurred as a result of the malignancy.

Antibody correlated to peripheral blood lymphocyte counts.

Research. on March 18, 2021. © 1995 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

1072 Antibody to ras Proteins in Colon Adenocarcinoma

The abbreviations used are: HRP, horseradish peroxidase; CEA, car-

cinoembryonic antigen.

Whether or not the presence of antibody correlates with out-

come has not yet been evaluated.

Materials and Methods

Sera. Sera were collected from 155 patients with newly

diagnosed colon cancer, 5 patients with recurrent colon cancer,

and 60 healthy normal individuals. One hundred sixteen patients

from Yokohama City University Hospital (Yokohama, Japan)

and Kinki University Hospital (Osaka, Japan), and 44 patients

from the University of Washington Medical Center (Seattle,

WA), Loyola University Medical Center (Chicago, IL), and

Mayo Clinic (Rochester, MN) were studied. Blood (10 ml) from

each colon cancer patient or normal individual was obtained

with informed consent. Sera were stored at -20#{176}C prior to

testing.

p2! ras Proteins. Purified wild-type p21 K-ras-G12 pro-

tein and mutated K-ras-D12 protein were purchased from On-

cogene Science (Manhasset, NY). These were purified without

SDS and BSA. Endotoxin was removed using the endotoxin

removal device (END-X B15; Cape Cod, Woods Hole, MA).

Mutated H-ras-V12 protein, H-ras-S12 protein, and H-ras-L61

protein were expressed in Escherichia coli and purified in our

laboratory. In brief, plasmids capable of expressing p21 ras

protein in a eukaryotic host were constructed by cloning a

synthetic C-Ha-ras gene into the pGH-L9 plasmid (16, 17). The

synthetic gene was altered by cassette mutagenesis to encode

various activated forms of p21 ras including valine 12 (V-12),

serine 12 (5-12), and leucine 61 (L-61). E. coli (strain HB 101)

was transfected with plasmid containing coding sequences for

one of the three activated p21 ras proteins (18). The transfected

bacteria were expanded, and the detergent lysates of transfected

bacteria were purified by size on a Sephadex 075 (SG75)

column. The fractions containing M� 21,000 ras protein were

concentrated by Amicon filtration (PM1O). The ras protein was

further purified by HPLC (Waters 625 LC System; Millipore,

Milford, MA) on a DEAE ion exchange column (Bio-Rad

Laboratories, Richmond, CA). Fractions containing ras protein

were analyzed using SDS-PAGE (Pharmacia PhastSystem;

Pharmacia LKB Biotechnology, Uppsala, Sweden) and Western

blotting. The primary antibody was a monoclonal mouse IgG

pan-ras antibody (pan-ras Ab3) specific for both mutated and

wild-type proteins (Oncogene Science). The fractions contain-

ing p2 1 ras were pooled, dialyzed, sterile filtered, and dried, and

endotoxin was removed as previously described. Protein con-

centration was determined by spectrophotometer (Bio-Rad Pro-

tein Assay; Bio-Rad Laboratories). Recombinant myoglobin (Mr

19,000; Sigma) was used as negative controls.

Synthetic ran Peptides. Five peptides derived from the

amino acid sequence of the wild-type p21 K-ras-G12 protein

were constructed and used in the current studies. All peptides

are 18 amino acids in length. Two ras peptides, YKLVVVGAG-

GVGKSALTI (p4-21) and GETCLLDILDTAGQEEYS (p48-

61), were constructed to span the specific point mutation region

at codon 12 or 61. The other three peptides, DTKQAQDLARSY-

GIPFIE (p125-142), KTRQRVEDAFYTLVREIR (plzV7-l64),

and EIRQYRLKKISKEEKTPG (p162-179) with an a-helix

constructing protein molecule surface, are from the carboxyl

terminus amino acid sequence of the wild-type p21 K-ras-G12

protein. Peptides were synthesized by Dr. Patrick S. H. Chou

(Biopolymer Facility, Department of Immunology, University

of Washington) using 9-fluorenylmethoxycarbonyl chemistry in

an automated peptide synthesizer (Model 433A; Applied Bio-

systems, Inc., Foster City, CA) and purified by HPLC.

Serum Antibody Detection Using ELISA. The purified

ras proteins were diluted in carbonate buffer (pH 9.6), and 50 pi

of a 0.02-p.g protein solution were added to Immulon II poly-

styrene ELISA plate wells (Dynatech Laboratories Inc., Chan-

tilly, VA) and incubated for 18 h at 4#{176}C.Human sera were

diluted to 1:25 in 0.2% casein buffer (1.58 g/liter Tris-HC1, 9

g/liter NaCl, 3.72 g/liter EDTA, 0.2 g/liter Thimerisol, and

0.05% NP4O) and rocked for 18 h at 4#{176}C.The monoclonal

mouse IgG anti-ras antibody (pan-ras Ab3; Oncogene Science)

specific for both mutated and wild-type ras proteins was diluted

1:500 in 0.2% casein buffer as a positive control and incubated

in the same manner as the human sera. Excess ras protein was

washed six times from each well. Two hundred p�l of 0.2%

casein buffer without NP4O were added to each well and incu-

bated for 3 h at room temperature for blocking. After washing,

100 pA of the diluted human sera, diluted pan-ras antibody, or

buffer alone were added to the appropriate wells and incubated

overnight at 4#{176}C.The next day the plates were washed six times.

As the secondary antibody, antihuman IgA HRP3 (Accurate

Chemicals, Westbury, NY) diluted 1:500 in 0.2% casein buffer

with 0.05% NP4O and 1% mouse sera was added. Antimouse

immunoglobulin HRP (Amersham, Arlington Heights, IL) di-

luted in 0.2% casein buffer with 0.05% NP4O was used for

positive control wells. The plates were incubated for 2 h at room

temperature. After the incubation the plates were washed six

times and developed with equal parts tetramethylbenzidine per-

oxidase substrate and peroxidase Solution B (Kirkegaard and

Perry Laboratory, Inc., Gaithersburg, MD). After 40 mm, the

development reaction was stopped by the addition of I N HCI.

Absorbency was read at a wavelength of 450 nm. All evalua-

tions were performed in quadruplicate. The positive antibody

control was a monoclonal mouse IgG pan-ras antibody specific

for both mutated and wild-type proteins.

Peptide Neutralization ELISA. Sera from 15 patients

with detectable antibody to p21 ras-G12 in the previous ELISA

study were selected by the level of absorbency, between 0.4 and

1.0, for this study. Patient sera as primary antibody were pre-

incubated for 1 h with or without ras peptides (100 �i.g/ml) or

p21 ras (1 ji�ml) as competitors. Other procedures for the

ELISA were as follows as detailed above. The percentage of

inhibition of the absorbency by the competitor was calculated.

Antibody Detection Using Western Blotting. Immuno-blotting with chemoluminescence was used to verify antibody

reactivity. Depending on the stock concentration of the recom-

binant protein, between 80 and 100 ng purified protein were

analyzed in each lane. Briefly, protein was transferred to nitro-

cellulose (Hybond-C; Amersham) and then blocked in 5% BSA-

0.05% NP4O in PBS for 1 h. The blot was incubated with human

sera diluted 1:200-1:1000 in 1% BSA-0.1% NP4O in PBS



51/160

(32%)

A

2.0

5)

aCS

�1.oCI�

CS

0.2

0

1/40

(2.5%)

mean + 3SD:::::::i:::: of normal

individuals

normal

individuals

...u_fU-tfl’lI’IlTflLFLu...

colon cancer

patients

Clinical Cancer Research 1073

(washing buffer) for 18 h at 4#{176}Cand then washed six times. The

secondary antibody antihuman immunoglobulin HRP (Amer-

sham) was added to the blot at a concentration of 1:10,000 and

incubated at room temperature for 1 h, followed by a final

washing. Control blots were incubated with pan-ras Ab3 (On-

cogene Science) at 1:4,000 dilution and then 1:10,000 dilution

of antimouse immunoglobulin HRP (Amersham). The blots

were developed in equal parts of enhanced chemiluminescence

detection reagents (Amersham) for 50 to 120 s, after which they

were exposed to Hyperfilm-ECL (Amersham), and the film was

developed.

Statistical Analysis. The correlation between antibody

and p21 ras proteins and clinical parameters were analyzed.

Student’s t test for paired and unpaired data was used to com-

pare the statistical significance of the differences between the

percentage of patients with and without antibody in relation to

age, sex, tumor locus (colon and rectum), histology (well dif-

ferentiated and poorly differentiated), stage, peripheral blood

lymphocyte count, lymph node metastasis, and serum CEA.

Results

Detection of Serum Antibody to p21 ras-D12 Protein in

Colon Cancer Patients and Normal Individuals. Normal

wild-type ras protein contains glycine at amino acid position 12

(denoted G12). Initial studies focused on detection of antibody

reactivity to the aspartic acid substitution (D12), as one of the

most common mutations in colon cancer. Sera from 160 colon

cancer patients and 40 normal individuals were tested for IgA

antibody reactivity to p21 K-ras-D12 proteins using an ELISA.

IgA antibodies directed against p2! ras-D12 protein were de-

tected in 51 (32%) of 160 colon cancer patients, but only in 1

(2.5%) of 40 normal individuals using a cutoff of 3 SDs above

the mean of normal individuals as the definition of a positive

response (Fig. 1). The greater incidence of IgA antibody to p21

K-ras-D12 protein in sera of colon cancer patients provides

evidence that immunization to the ras proteins occurred as a

result of the malignancy.

ELISA results were verified by Western blotting analysis

in a selected subset of patients. Sera from five colon cancer

patients who had the antibody reactivity to the p21 K-ras-Dl2

proteins (absorbency from 0.66 to 1.62) in ELISA were corn-

pared to sera from five colon cancer patients and five normal

individuals who had no detectable antibody reactivity (absor-

bency less than 0.03). The results of immunoblot showed that

the serum antibody reactivity to p21 ras-Dl2 in the cancer

patients was as marked as that of a pan-ras mAb, whereas colon

cancer patients with no antibody in ELISA and the normal

individual showed no detectable evidence of antibodies to p21

ras-D12 (data not shown).

For Most Colon Cancer Patients, Serum AntibodiesResponding to Mutated p2! ras-D12 Protein Also Respond

to Normal p2! ras-G12 Protein. The above results show that

serum antibody to mutated p21 ras-D12 protein can be detected

in sera of colon cancer patients. Whether or not the detected

serum antibodies were against the mutated D12 segment or

nonmutated segments of the p2! ras-D12 protein was examined.

Sera from the same colon cancer patient population were exam-

ined for antibody reactivity to both mutated p21 K-ras-D12

Fig. I Serum IgA antibody to p21 ras-D12 protein can be detected in

colon cancer patients with much greater incidence than in normal

individuals. Sera from 160 colon cancer patients and 40 normal mdi-viduals were tested for IgA antibody responses to p21 K-ras-D12

proteins using an ELISA, as described in ‘ ‘ Materials and Methods.”

The definition of a positive antibody response was determined using a

cutoff of 3 SDs above the mean of normal individuals.

proteins and to normal wild-type p21 K-ras-G!2 proteins using

an ELISA. Results (Table 1) show that 41 (26%) of 160 patients

exhibited detectable antibody reactivity to both the p21 K-ras-

D12 and p21 K-ras-G12 proteins. Sera from 10 (6%) of 160

colon cancer patients responded to mutated p21 K-ras-D12

protein, but not to the normal p21 K-ras-G12 protein. Antibody

response to the normal wild-type p21 K-ras-G12 protein but not

to the mutated p21 K-ras-D12 protein was detected in only 7

(4%) of 160 colon cancer patients, and responses were low by

absorbancies (Table 1).

Antibody from a Small Percentage of Patients Re-sponds Only to Mutated p2! ras Proteins. The above data

show that antibody in the majority of colon cancer patients

recognizes normal p21 K-ras-Gl2 protein. In only a small

number of colon cancer patients was the serum antibody reac-

tivity against only the mutated p2! ras-D!2 protein (Table 2).

To determine specific antibody reactivity to other mutated p2!

ras proteins, we retested sera from 142 of the 160 colon cancer

patients (adequate sera was not available from 18 patients) as

well as from 60 normal individuals. ELISAs were performed

with a larger panel of proteins, including four additional ras

proteins (p21 H-ras-Gl2, ras-V12, ras-S12, and ras-L61 pro-

teins) and myoglobin as negative controls. Sera from 9 (6%) of

142 patients who had antibody reactivity to only mutated p21

K-ras-D12 responded specifically to p21 ras-D12 protein, but

not to normal p2! ras-G12 protein or other mutated p21 ras-

V12, 512, and L61 proteins as well as myoglobin. A represen-

tative patient is shown in Fig. 14. Thus, the epitopes recognized

by the p21 ras-D12 protein-reactive antibody are most likely

directed against the mutated D12 segment of the p21 ras-D12

proteins. In sera from 4! colon cancer patients who had antibody

reactivity to both mutated p21 K-ras-D12 and normal p21 K-ras-

D12 proteins, the results showed that sera from these patients




Table / For most patients, antibody that responds to mutated p21

K-ras-D12 protein also responds to normal p21 K-ras-G12 protein”

Antibodies to p21

K-ras protein% of

D12 mutations Normal positive patients No.

+ + 25.6 41/160+ - 6.3 10/160

- + 4.4 7/160

- - 63.8 102/160

‘a Sera from 160 colon cancer patients were examined for antibodyresponses to both mutated p2! K-ras-D12 proteins and to normal wild-

type p21 K-ras-G!2 protein using an ELISA, as described in “Materials

and Methods. ‘ ‘ The definition of a positive antibody response wasdetermined using a cutoff of 3 SD above the mean of normal individuals.

responded to all of the mutated or normal ras proteins. A

representative patient is shown in Fig. 2B.

Antibody Responses to p2! ras Proteins in Colon Can-cer Patients Can Be Blocked by Peptides from the Carboxyl

Terminus of p21 ras Protein. The above data show that theantibody reactivity to p21 ras proteins in the majority of re-

sponding patients was directed against normal common seg-

ments rather than mutated segments. Sera from 15 patients who

had the antibody reactivity to both the wild-type and mutated

p2! ras proteins were further tested using the peptide neutral-

ization assay. Five 18-rner ras peptides (p4-21, p48-65, p125-

142, p147-164, and p162-179) were used. The results (Fig. 3)

demonstrated that in 10 (67%) of 15 antibodies, reactivity to p21

ras protein was inhibited more than 25% by either peptide

p147-164 or p162-179 near the carboxyl terminus of the p21

ras protein. None of the reactivity in these patients was inhibited

more than 25% by peptide p4-21 or p48-65, which span the

hot point mutation region at codons 12 or 61. Thus, epitopes of

antibodies to the normal portion of p21 ras protein in some

colon cancer patients were localized near the carboxyl terminus

of the p2l ras protein.

Analysis of the Correlation between Serum Antibody

and the Patient Characteristics and Clinical Parameters.The existence of antibody to p21 ras protein was analyzed in

relation to a variety of clinical parameters available retrospec-

tively. Antibody responses did not correlate with patient age,

sex, locus, histology, stage, lymph node metastasis, or serum

CEA (P > 0.05). Detection of antibody reactivity to p21 ras

proteins correlated with the peripheral blood lymphocyte count.

Antibody was detected in 2 (16%) of 13 patients with a lym-

phocyte count of <1000, and in 4 (55%) of 7 patients with a

lymphocyte count >3000 (P < 0.05). However, data were not

available for all patients. Prospective studies with more patients

are needed to validate the conclusions concerning clinical cor-

relations. Data concerning the presence or absence of specific

ras mutations in tumors and clinical outcome were not available

for analysis, but should be studied prospectively.

Discussion

ras oncogenes are activated by point mutations that result

in the expression of p21 ras proteins with single-substituted

amino acids, altered tertiary structure, and transforming activity

Table 2 Antibody from a small percentage of patienonly mutated p2! ras protein”

ts responds to

Mutated % ofp21 ras proteins positive patients No.

p2! K-ras-D12 6.3 9/142p21 H-ras-V12 4.2 6/142p21 H-ras-S12 4.9 7/142

p21 H-ras-L61 3.5 5/142a Sera from 142 patients with colon cancer and 60 normal individ-

uals were tested to determine specific antibody to the mutated segment

of ras proteins using an ELISA, as described in ‘ ‘ Materials and Meth-ods. ‘ ‘ The definition of a positive antibody response was determined

using a cutoff of 3 SDs above the mean of normal individuals.

(1, 2, 4, 5). ras proteins have been studied as a potential

tumor-specific antigen because ras mutations are common in

human malignancy, and the common mutations are limited in

number and location in the molecule. Importantly, the common

mutations present in human malignancy are also commonly

observed in murine malignancies, so that results in murine

models can be extrapolated to humans. Immunity to ras can be

elicited in animal models by immunization to ras peptides and

protein. Our previous animal studies have shown that under

well-defined circumstances, murine CD8� CTL.s and CD4�

helper T cells specific for the mutated segment of p2! ras

protein can be generated. Immunizing animals with peptides that

span the mutated segment can result in the generation of helper!

inducer and cytolytic T-cell responses that recognize intact

mutated protein (9, 10). CTLs specifically immune to mutated

ras peptides can lyse cells transformed by the ras protein with

the same mutation (10). Thus, ras protein represents a tumor-

specific antigen.

The observation that p21 ras protein can be immunogenic

in mice stimulated the current studies to determine whether

immunity to p21 ras protein is present in patients with colon

cancer. The current study showed that 32% of a panel of colon

cancer patients had IgA antibodies against ras proteins. The

greater incidence of antibody in cancer patients provides strong

evidence that immunization to the protein occurred as a result of

the malignancy. Immunoglobulin class switching from 1gM to

IgA or IgG generally requires T cell help, which is usually

cognate. Cognate help means that B cells and T cells recognize

different epitopes, but that the epitopes are on the same mole-

cule. The antibody response to p21 ras detected in colon cancer

patients was an IgA response, therefore, T cell help was most

probably involved. The major antibody response to p21 ras is to

epitopes located in the nonmutated COOH terminus segment

rather than to the mutated segments that are unique to the

mutated ras protein. Only a small number of colon cancer

patients had a detectable antibody response to only the mutated

ras-D12 proteins as compared to a panel of mutated or wild-type

ras-G12 proteins. One issue is whether the T cell help consid-

ered necessary for immunoglobulin class switch is to mutated or

nonmutated segments. Helper T-cell responses specific for the

mutated portion of ras proteins have been elicited in animal

models. T-cell responses to nonmutated segments have not been

studied in animal models. Human T-cell studies by our group

and others have shown that human CD4� T cells are capable of



0.6

0.4

0.2

0

C

Patient A Patient B

0

Cc/�


antigens antigens

Fig. 2 Antibody from a small percentage ofcolon cancer patients responds specifically to mutated p21 ras-Dl2 proteins. Sera from 142 colon cancer

patients and 60 normal individuals were examined for specific IgA antibody responses to mutated p21 K-ras-D12 in an ELISA, as described in

‘ ‘Materials and Methods.’ ‘ A panel of normal or mutated p21 ras proteins and myoglobin was used as control proteins. Sera from 9 (6%) of 142

patients responded to only mutated p21 K-ras-D12 protein. A, representative patient. Sera from 4! of 142 colon cancer patients who had antibody

responses to mutated p21 K-ras-D!2 responded to all of the mutated or normal ras proteins. B, representative patient. The definition of a positive

antibody response was determined using a cutoff of 3 SDs above the mean of normal individuals.

recognizing ras proteins in a mutation-specific manner (12-15).

T cells specifically recognizing mutated ras peptides have been

isolated from patients with follicular thyroid carcinoma and

colorectal cancer (14, 19-22). Thus, although the majority of

antibody detected was to nonmutated portions of ras protein, the

T cell help may be to the mutated segment in some patients.

To elicit antibody responses and CD4� T cell help for

antibody responses, ras protein must access the immune system.

Normally, p2l ras protein is affixed to the inner aspect of the

cell membrane and is thus not considered to be a secreted

protein. However, p21 ras can be readily detected in vivo under

a variety of circumstances (23), including presence in the ex-

tracellular tumor environment (24). Thus, p21 ras protein at the

site of tumor deposition is likely to be available in the external

cell environment to be processed and presented by autologous

APC expressing class II MHC antigens, including B cells with

immunoglobulin receptors specific for ras protein.

Although human antibody reactivity to ras proteins in

cancer patients has not been previously reported, the formation

of antibodies directed against p53, HER-2!neu, and other onco-

genie proteins have been described (24-33). Whether the de-

tection of antibody to p21 ras could serve as a surrogate marker

for tumors expressing ras mutations is still being evaluated. The

ras mutations of individual patient tumors has not been deter-

mined. Thus, studies correlating specific ras mutations with

antibody formation have not been performed. Similar to ras, p53

is activated by point mutation. Serum antibody to p53 has been

reported to be correlated with the presence of mutations in

particular protein domains; however, the antibody detected ap-

pears to react with the wild-type p53 protein (31-33). In contrast

to ras, HER-2/neu is amplified and the protein is overexpressed.

Antibody to normal HER-2/neu has been detected (24, 28). In

this circumstance CD4� helper T-cell responses to normal

HER-2/neu protein sequences have been detected (24). Thus, it

is not necessary to postulate that ras-specific T-cell responses

need to be to the mutated segment. Theoretically, rapidly ne-

crosing tumor could release normally sequestered internal pro-

tein for recognition by the immune system. This would allow the

development of an immune reaction directed against a nonmu-

tated protein. Studies examining helper T-cell responses to both

normal and mutated ras protein might reveal the extent to which

the mutated segment is recognized as a cancer-specific antigen

in the tumor-bearing host.

An initial purpose for the current study was to determine

whether the detection of human antibodies to mutant p21 ras

proteins might be used as a surrogate marker for the presence of

tumors expressing mutated ras proteins. The majority of anti-

body was to nonmutated p21 ras protein. However, the existence



100�

75-

50�

25�

$

$

I

I

0

n=15

I

:. I

I

I#{149} T UII

.25�

.‘� ,co

I I I I I

�q�# ‘�#�4� q

I p2lras peptides I

of antibody specific for mutated ras has not been ruled out in a

minority of patients. Although the majority of antibodies were to

the nonmutated segment of ras, antibody responses to the non-

mutated segment may provide a method to identify individuals

who have been exposed to mutated p21 ras protein and might

yet be shown to provide an early indication of incipient or occult

malignancies undetectable by conventional methods.

The correlation of antibody formation to prognosis is not

known and should be evaluated. Investigations to determine

whether antibodies to other oncogenic proteins correlate to

prognosis or other important disease parameters are still in their

initial phases. Initial studies correlating antibody to p53 to

prognosis of lung cancer have demonstrated that the presence of

antibody to p53 did not correlate with survival (34). Other

reports have suggested antibody formation to p53 is a poor

prognostic indicator in breast cancer (35). However, it is not

clear whether antibody to p53 or the presence of mutant p.53 is

the primary cause of a poor prognosis. The examination of

antibody to ras might provide important information in similar

studies. Mutant K-ras in colon cancer does not predict the

patient’s response to chemotherapy or survival (36). Thus, an-

tibody to ras might be more easily studied as an independent

variable. In the current studies, the evaluation of the clinical


competitor

correlation of antibody existence showed that antibody reactiv-

ity did not correlate with patient age, sex, histology, stage, or

serum CEA, but did correlate with lymphocyte number in pe-

ripheral blood drawn before operation. Lymphocyte number and

antibody reactivity were evaluated at only one point in time, and

data were not available on all patients. Studies with larger

numbers of patients are needed to validate the conclusions

$ concerning clinical correlations. Whether changes occur serially

and whether serial changes in antibody levels can be used as

indicators of changes in disease status also should be evaluated

in prospective studies.

Acknowledgments

We thank Kevin Whitham for assistance in the preparation of the

#{149} manuscript.

Fig. 3 Antibody responses to p21 ras proteins in colon cancer patients

can be blocked by peptides from the carboxyl terminus of p2! ras

proteins. Sera from 15 colon cancer patients who had antibody re-

sponses to both the wild-type and mutated p2! ras proteins were tested

using the peptide neutralization assay. Five 18-mer ras peptides with an

a-helix constructing protein molecule surface were chosen as compet-

itors. Sera from patients were preincubated either with or without ras

peptides (100 p.gIml) for I h and tested in an ELISA, as described in

“Materials and Methods.” The percentage of inhibition of the absor-bance by the competitor was calculated.

References

1. Bos, J. ras oncogenes in human cancer: a review. Cancer Res., 49:

4682-4689, 1989.

2. Barbacid, M. ras genes. Annu. Rev. Biochem., 56: 779-827, 1987.

3. Stavnezer, E., Fogh, I., and Wigler, M. Three human transforming

genes are related to the viral ras oncogenes. Proc. Natl. Acad. Sci. USA,

80: 2!12-2!!6, 1983.

4. Seeburg, P. H., Colby, W. W., Capon, D. J., Goeddel, D. V., and

Levinson, A. D. Biological properties of human c-Ha-ras I genes

mutated at codon 12. Nature (Lond.), 312: 71-75, 1984.

5. Der, C. J., Finkel, T., and Cooper, G. M. Biological and biochemical

properties of human rasH genes mutated at codon 61. Cell, 44: 167-176,

1986.

6. Capella, G., Cronauer-Mitra, S., Pienado, M. A., and Perucho, M.

Frequency and spectrum of mutations at codons 12 and 13 of the

c-K-ras gene in human tumors. Environ. Health Perspect., 93: 125-31,

1991.

7. Peace, D. J., Smith, J. W., Disis, M. L., Chen, W., and Cheever,M. A. Induction of T cells specific for the mutated segment of oncogenicP21 ras protein by immunization in vivo with the oncogenic protein. J.

Immunother., 14: 110-114, 1993.

8. Cheever, M. A., Chen, W., Disis, M. L., Takahashi, M., and Peace,

D. J. T-cell immunity to oncogenic proteins including mutated ras and

chimeric bcr-abl. In: J. C. Bystryn, S. Ferrone, and P. Livingston (eds.),Specific Immunotherapy of Cancer with Vaccines, vol. 690, pp. 10!-

112. New York: New York Academy of Sciences, 1993.

9. Peace, D. J., Chen, W., Nelson, H., and Cheever, M. A. T cell

recognition of transforming proteins encoded by mutated ras proto-

oncogenes. J. Immunol., /46: 2059-2065, 1991.

10. Peace, D. J., Smith, J. W., Chen, W., You, S-G., Cosand, W. L.,

Blake, J., and Cheever, M. A. Lysis of ras oncogene-transformed cells

by specific cytotoxic T lymphocytes elicited by primary in vitro immu-

nization with mutated ras peptide. J. Exp. Med., 179: 473-479, 1994.

11. Skipper, J., and Stauss, H. J. Identification of two cytotoxic T

lymphocyte-recognized epitopes in the ras protein. J. Exp. Med., 177:

1493-1498, 1993.

12. Gedde-Dahl, T., Eriksen, J. A., Thorsby, E., and Gaudernack, G.1-cell responses against products of oncogenes: generation and charac-

terization of human 1-cell clones specific for p21 ras-derived synthetic

peptides. Hum. Immunol., 33: 266-274, 1992.

l3. Jung, S., and Schluesener, H. J. Human I lymphocytes recognize a

peptide of single point-mutated, oncogenic ras proteins. J. Exp. Med.,

173: 273-276, 1991.

14. Gedde-Dahl, 1., Spurkland, A., Eriksen, J. A., Thorsby, E., Gaud-

ernack, G., and Gedde-Dahi, 1. Memory 1 cells of a patient with

follicular thyroid carcinoma recognize peptides derived from mutated

p21 ras(Gln -� Leu6l). Int. Immunol., 4: 1331-1337, 1992.

15. Qin, H. L., Chen, W., Takahashi, M., Disis, M. L., Byrd, D. R.,

Fenton, R., Bertram, K., McCahill, L., Peace, D. J., and Cheever, M. A.




CD4� T cell immunity to mutated ras protein in pancreatic and coloncancer patients. Cancer Res., 55: 2984-2987, 1995.

16. Miura, K., Kamiya, H., Tominaga, M., Inoue, Y., Ikehara, M.,

Noguchi, S., Nishimura, S., and Ohtsuka, E. Overproduction of cellular

and activated Ha-ras proteins by mutating a synthetic gene. Chem.

Pharm. Bull. (Tokyo), 35: 4878-4882, 1987.

17. Ikehara, M., Ohtsuka, E., Tokunaga, 1., Taniyama, Y., Iwai, S.,

Kitano, K., Miyamoto, S., Ohgi, 1., Sakaragawa, Y., Fujiyama, K.,

Ikari, 1., Kobayashi, M., Miyake, 1., Shibahara, S., Ono, A., Ueda, 1.,

Tanaka, 1., Baba, H., Miki, 1., Sakurai, A., Oishi, 1., Chisaka, 0., andMatsubara, K. Synthesis of a gene for human growth hormone and its

expression in Escherichia coli. Proc. Natl. Acad. Sci. USA, 81: 5956-

5960, 1984.

18. Miura, K., Inoue, Y., Nakamori, H., Iwai, S., Ohtsuka, E., Ikehara,

M., Noguchi, S., and Nishimura, S. Synthesis and expression of asynthetic gene for the activated human c-Ha-ras protein. Jpn. Cancer

Res., 77: 45-5!, 1986.

19. Fossum, B., Gedde-Dahl, 1., III, Breivik, J., Eriksen, J. A., Spurk-

land, A., Thorsby, E., and Gaudernack, G. p21-ras-peptide-specific

T-cell responses in a patient with colorectal cancer. CD4� and CD8� Icells recognize a peptide corresponding to a common mutation (l3Gly

-4 Asp). Int. J. Cancer, 56: 40-45, 1994.

20. Fossum, B., Breivik, J., Meling, G. I., Gedde-Dahl, 1., III, Hansen,

I., Knutsen, I., Rognum, 1. 0., Thorsby, E., and Gaudernack, G. AK-ras l3Gly -� Asp mutation is recognized by HLA-DQ7 restricted I

cells in a patient with colorectal cancer. Modifying effect of DQ7 on

established cancers harbouring this mutation? Int. J. Cancer, 58: 506-

511, 1994.

21. Gedde-DahI, I., III, Olsen, B. H., Ihorsby, E., and Gaudernack, G.

Oncogene-derived peptides: a new class of tumor rejection antigens?

Transplant. Proc., 25: 1385-1386, 1993.

22. Gedde-Dahl, I., III, Spurkland, A., Fossum, B., Wittinghofer, A.,Thorsby, E., and Gaudernack, G. I cell epitopes encompassing the

mutational hot spot position 61 of p21 ras. Promiscuity in ras peptide

binding to HLA. Eur. J. Immunol., 24: 410-414, 1994.

23. Niman, H., Thompson, A., Yu, A., Markman, M., Willems, J.,

Herwig, H., Habib, N., Wood, C., Houghten, R., and Lerner, R. Anti-peptide antibodies detect oncogene-related proteins in urine. Proc. Natl.

Acad. Sci. USA, 82: 7924-7928, 1985.

24. Disis, M. L., Calenoff, E., McLaughlin, G., Murphy, A. E., Chen,

W., Groner, B., Jeschke, M., Lydon, N., McGlynn, E., Livingston, R. B.,Moe, R., and Cheever, M. A. Existent I cell and antibody immunity to

HER-2/neu protein in patients with breast cancer. Cancer Res., 54:

16-20, 1994.

25. Hassapoglidou, S., and Diamandis, E. P. Antibodies to the p53

tumor suppressor gene product quantified in cancer patient serum with

a time-resolved immunofluorometric technique. Clin. Biochem., 25:

445-449, 1992.

26. Labrecque, S., Naor, N., Thomson, D., and Matlashewski, G. Anal-ysis of the anti-p53 antibody response in cancer patients. Cancer Res.,

53: 3468-3471, 1993.

27. Lubin, R., Schlichtholz, B., Bengoufa, D., Zalcman, G., Tr#{233}daniel,

J., Hirsch, A., de-Fromentel, C. C., Preudhomme, C., Fenaux, P.,Fournier, G., Mangin, P., Laurent-Puig, P., Pelletier, G., Schlumberger,

M., Desgrandchamps, F., Le Due, A., Peyrat, J. P., Janin, N., Bressac,B., and Soussi, I. Analysis of p53 antibodies in patients with variouscancers define B-cell epitopes of human p53: distribution on primary

structure and exposure on protein surface. Cancer Res., 53: 5872-5876,

1993.

28. Pupa, S. M., M’enard, S., Andreola, S., and Colnaghi, M. I. Anti-

body response against the c-erbB-2 oncoprotein in breast carcinoma

patients. Cancer Res., 53: 5864-5866, 1993.

29. Sorokine, I. Presence of circulating anti-c-mvb oncogene product

antibodies in human sera. mt. J. Cancer, 47: 665-669, 1991.

30. Ben-Mahrez, K., Sorokine, I., Thierry, D., Kawasumi, T., Ishii, S.,

Salmon, R., and Kohiyama, M. Circulating antibodies against c-rnvc

oncogene product in sera ofcolorectal cancer patients. Int. J. Cancer, 46:

35-38, 1990.

31. Davidoff, A., Iglehart, J., and Marks, J. Immune response to p53 is

dependent upon p53/HSP7O complexes in breast cancer. Proc. NatI.

Acad. Sci. USA, 89: 3439-3442, 1992.

32. Schlichtholz, B., Legros, Y., Gillet, D., Gaillard, C., Marty, M.,

Lane, D., Calvo, F., and Soussi, 1. The immune response to p53 in

breast cancer patients is directed against immunodominant epitopes

unrelated to the mutational hot spot. Cancer Res., 52: 6380-6384, 1992.

33. Labrecque, S., Naor, N., Thomson, D., and Matlashewski, G. Anal-

ysis of the anti-p53 antibody response in cancer patients. Cancer Res.,

53: 3468-3471, 1993.

34. Winter, S. F., Minna, J. D., Johnson, B. E., Takahashi, 1., Gazdar,

A. F., and Carbone, D. P. Development of antibodies against p53 in lungcancer patients appears to be dependent on the type of p53 mutation.

Cancer Res., 52: 4168-4174, 1992.

35. Peyrat, J. P., Bonneterre, J., Lubin, R., Vanlemmens, L., Fournier,

J., and Soussi, 1. Prognostic significance of circulating p53 antibodiesin patients undergoing surgery for locoregional breast cancer. Lancet,

345: 621-622, 1995.

36. Markowitz, S., Hines, J. D., Lutterbaugh, J., Myeroff, L., Mackay,

W., Gordon, N., Rustum, Y., Luna, E., and Kleinerman, J. Mutant K-rasoncogenes in colon cancers do not predict patient’s chemotherapy

response or survival. Clin. Cancer Res., I: 441-445, 1995.



1995;1:1071-1077. Clin Cancer Res M Takahashi, W Chen, D R Byrd, et al. Antibody to ras proteins in patients with colon cancer.

Updated version

http://clincancerres.aacrjournals.org/content/1/10/1071

Access the most recent version of this article at:

E-mail alerts related to this article or journal.Sign up to receive free email-alerts

Subscriptions

Reprints and

[email protected] at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

Permissions

Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/1/10/1071To request permission to re-use all or part of this article, use this link



http://clincancerres.aacrjournals.org/cgi/alerts

mailto:[email protected]



Documents

Antibody to ras Proteins in Patients with Colon Cancer1pan-ras antibody (pan-ras Ab3) specific for both mutated and wild-type proteins (Oncogene Science). The fractions contain-ing