expression of muc1 and muc2 mucins in gastric carcinomas ... · vol. 4, 2605-2614, november 1998...

Vol. 4, 2605-2614, November 1998 Clinical Cancer Research 2605

3 The abbreviations used are: MAb, monoclonal antibody: ABC. avidin-biotinylated horseradish peroxidase complex.

Expression of MUC1 and MUC2 Mucins in Gastric Carcinomas: Its

Relationship with the Prognosis of the Patients’

Tamiharu Utsunomiya, Suguru Yonezawa,2

Hideie Sakamoto, Hiroshi Kitamura,

Shuichi Hokita, Takashi Aiko, Sadao Tanaka,

Tatsuro Irimura, Young S. Kim, and Eiichi SatoDepartments of Pathology [T. U.. S. Y., H. S., H. K., E. 5.1 andSurgery [S. H., T. A.], Kagoshima University School of Medicine.Kagoshima 890-8520, Japan; Department of Pathology, Kagoshima-shi Medical Association Hospital, Kagoshima 890-0064, Japan[S. T.]; Department of Cancer Biology and Molecular Immunology.Faculty of Pharmaceutical Sciences, The University of Tokyo, Tokyo1 13-0033, Japan [T. I.); and Gastrointestinal Research Laboratory,Veterans Affairs Medical Center and University of California. SanFrancisco, California 94121 [Y. S. K.]

ABSTRACT

Our previous immunohistochemical studies for the cx-

pression of MUC1 mucin antigen (which was detected by

monoclonal antibody DF3) and MUC2 mucin antigen (which

was detected by polyclonal antibody anti-MRP) in pancre-

atic and intrahepatic bile duct tumors demonstrated that

invasive carcinoma with poor outcome showed a pattern of

MUC1+ and MUC2- expression, whereas many of the

noninvasive tumors with favorable outcome showed a pat-

tern of MUC1- and MUC2+ expression. To clarify the

relationship between the expression of these mucin antigens

and the biological properties of gastric cancers, the expres-

sion of MUC1 and MUC2 mucin antigens was examined

immunohistochemically in 136 patients with gastric cancer

invading the submucosa or the deeper layer, and the sur-

vival of the antigen-positive and antigen.negative patient

groups was compared using the Kaplan-Meier method. For

MUC1 mucin expression, different glycoforms of MUC1

were examined using four monoclonal antibodies (NCL-

MUC-1-CORE, DF3, MY.1E12, and HMFG-1). The patients

with MUC1+ mucin antigen staining in the carcinoma

Received 4/16/98; revised 8/10/98: accepted 8/17/98.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisement in accordance with 18 U.S.C. Section 1734 solely toindicate this fact.

I Supported in part by Grants-in-Aid 08457064, 09470053, and09877038 and the Monbusho International Scientific Research Program,

Special Cancer Research from the Ministry of Education, Science.Sports and Culture, Japan [to S. Y.]; by the Kodama Memorial Fund forMedical Research, Japan [to S. Y.]; by an Otsuka Pharmaceutical Co.,Ltd. grant, Gastric Mucin Research Forum. Japan [to S. Y.]; and byUSPHS Grant CA24321 from the National Cancer Institute, the Depart-ment of Veterans Affairs Medical Research Service, and the Theodore

Betz Foundation [to Y. S. K.].

2 To whom requests for reprints should be addressed. at Second Depart-

ment of Pathology, Faculty of Medicine. Kagoshima University, 8-35-I

Sakuragaoka, Kagoshima 890-8520, Japan. Phone: 81-99-275-5270:Fax: 81-99-265-7235: E-mail: [email protected].

showed significantly worse survival than those with MUC1 -

mucin antigen staining. In contrast, the patients with

MUC2+ mucin antigen staining in the carcinoma showed

significantly better survival than those with MUC2- mucin

antigen staining. In conclusion, MUC1 antigen expression

was associated with a poor outcome in patients with gastric

cancer, irrespective of its glycosylation status, and MUC1 is

thus considered to be a useful prognostic factor for poor

outcome in patients. In contrast, MUC2 antigen expression

is a prognostic factor associated with a favorable outcome in

patients. In addition, combined evaluation of the MUC1 and

MUC2 mucin staining is clinically useful to predict outcome

in patients with gastric cancer.

INTRODUCTIONThe synthesis and secretion of mucin are common features

of glandular epithelial tissues, and it has been reported that

many cancer-associated antigens are mucin antigens. Mucins are

high molecular weight glycoproteins with oligosaccharides at-

tached to the apomucin protein backbone by O-glycosidic link-

ages.

Our previous immunohistochemical studies for MUC 1 mu-

cm expression detected by MAb3 DF3 and MUC2 mucin ex-

pression detected by polyclonal antibody anti-MRP in pancre-

atic and intrahepatic bile duct tumors demonstrated that invasive

carcinoma with a poor outcome showed a pattern of MUC 1 +

and MUC2- expression, whereas many of the noninvasive

tumors with a favorable outcome showed a pattern of MUC I -

and MUC2+ expression. Therefore, we have been interested in

studying the relationship between the expression of these mu-

cins and the biological properties of various human cancers.

Our recent study indicated that gastric carcinoma patients

with DF3+ expression in their carcinoma cells showed signif-

icantly decreased survival compared to patients who were neg-

ative for DF3 expression (1). MAb DF3, which was used in the

study described above, identifies MUC 1 apomucin� however,

MAb DF3 binding to the MUC1 core protein may be enhanced

by the presence of carbohydrates (2). There are differences in

the glycosylation of MUC1 in the different human tumors and

normal cell types. MUC1 expressed by breast carcinomas is

poorly glycosylated in the MUC! mucin, whereas normal breast

tissue shows little or no expression of the MUC 1 mucin core

peptide (3-5). This phenomenon is explained in part by the

finding that MUC1 core peptide epitopes are masked by carbo-

hydrate side chains produced by normal breast epithelial cells

(4, 6), whereas the carbohydrate side chains of MUCI produced

by breast adenocarcinomas are shorter or less densely distrib-

uted than those produced by normal cells (7). On the other hand,

Research. on May 12, 2020. © 1998 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

http://clincancerres.aacrjournals.org/

2606 MUC1 and MUC2 Mucins in Gastric Carcinoma

Table I Antibodies used to detect antigens�’

Antigen Reagent used to detect the antigen

MUC1 mucinCore peptide of MUCI mucinCore peptide of MUC1 mucin

Sialylated MUC1 mucinFully glycosylated MUCI mucin

MUC2 mucinCore peptide of MUC2 mucin

(1 NCL-MUC-l-CORE, mouse IgG, hybridoma culture supernatant(Novocastra Laboratories Lt.): DF3, mouse IgG. ascites, Centocor

CAI5-3 (Toray-Fuji Bionics, Tokyo, Japan); MY.lEl2, mouse IgG,ascites (developed in the laboratory of Dr. T. Inmura), HMFG- 1 , mouseIgG. hybridoma culture supernatant [developed in the laboratory of Dr.J. Taylor-Papadimitriou (Imperial Cancer Research Fund Laboratories,London, United Kingdom)]: anti-MRP, purified rabbit IgG (developedin the laboratory of Dr. Y. S. Kim).

I, Sialyl oligosaccharides on the core peptide may be required for

DF3 antigenicity.

colorectal carcinomas show a high level of expression of fully

glycosylated MUC 1 mucin in the advanced stages or in the

metastatic lesions (8). Thus, it is of interest to investigate the

relationship between the glycosylation of MUC1 mucin ex-

pressed in gastric cancer tissues and the outcome in patients

with gastic carcinoma.

In the present study, we examine the expression of different

glycoforms of MUC 1 mucins by using four MAbs (NCL-MUC-

1-CORE, DF3, MY.lEl2, and HMFG-l) in gastric cancer us-

sues and show that MUC 1 expression is associated with a poor

outcome in patients with gastric cancer, irrespective of their

glycosylation status. We also examine the expression of MUC2

mucin in the same gastric cancer tissues and show that MUC2

mucin expression is a favorable prognostic factor in patients

with gastric carcinoma.

MATERIALS AND METHODSTissue Samples. Surgical specimens of 1 36 gastric can-

cers (from 86 males and 50 females) invading the submucosal

layer or the deeper layer were obtained from the files of the

Department of Pathology, Kagoshima University School of

Medicine and the Department of Pathology, Kagoshima-shi

Medical Association Hospital. The mean age of the patients was

62.5 years (range, 32-91 years). The pathological staging was

achieved using the unified Japanese Classification of Gastric

Carcinoma (9). Histologically, there were 68 cases of intestinal-

type cancer (T,, 17 cases; T,, 46 cases; and T3, 5 cases) and 68

cases of diffuse-type cancer (T,, 5 cases; T,, 48 cases; and T3,

15 cases). Lymph node metastasis was found in 87 patients (46

patients with intestinal-type cancer and 41 patients with diffuse-

type cancer). Clinical outcome data were available for 136

gastric cancer patients. The patients who died of other diseases

were excluded from the analysis in the current study. All spec-

imens were fixed in formalin, embedded in paraffin, and cut into

4-iim-thick sections for immunohistochemistry in addition tothe usual H&E staining.

Antibodies. The antibodies used to detect the mucin an-

tigens are listed in Table 1 . For MUC 1 expression, different

glycoforms of MUC1 were examined by four MAbs (NCL-

NCL-MUC- 1-COREDF3”MY. lEl2HMFG-l

Anti-MRP

MUC-l-CORE, DF3, MY.lEl2, and HMFG-l). MAb NCL-

MUC-l-CORE recognizes the MUC1 mucin core peptide [(TR-

PAPG) Data sheet; Novocastra Laboratories, Ltd., New Castle,

United Kingdom]. MAb DF3 identifies the MUC1 core peptide;

however, the MAb DF3 binding to protein may be enhanced by

the presence of sialyl oligosaccharides (2). A new MAb,

MY.lEl2, is specific for sialylated-MUC1 mucin (10). The

binding of MAb HMFG-l to core protein epitopes is influenced

by the carbohydrate chains (4), and HMFG-l detects fully

glycosylated MUC1 mucin, and its binding is particularly af-

fected by sialic acid (4, 1 1, 12).

For simple representation, MUC1-CORE, DF3, MY.lEl2,

and HMFG-l antigens were used for the MUCI mucin antigens

detected by MAbs NCL-MUC-l-CORE, DF3, MY. lEl2, and

HMFG-l in this study, respectively. MRP antigen was used for

the MUC2 mucin antigen detected by polyclonal antibody anti-

MRP.

Biotinylated affinity-purified horse antimouse IgG, goat

antirabbit IgG, and ABC were purchased from Vector Labora-

tories (Burlingame, CA), as was the Vectastain ABC Kit.

Staining Procedure. Immunohistochemical stainings

were done on formalin-fixed, paraffin-embedded tissue sections

by an immunoperoxidase method using the ABC (13) as de-

scribed previously (1, 14, 15). Briefly, each section was depar-

affinized with xylene. Endogenous peroxidase was blocked by

incubating the sections in 0.3% hydrogen peroxide in absolute

methanol at room temperature for 30 mm. After hydration in

decreasing concentrations of ethanol in water, the sections were

washed in 0.01 mol/liter PBS (pH 7.4), and then 2% horse or

goat serum in PBS was applied for 30 mm at room temperature

to prevent nonspecific staining. In the staining using each anti-

body, the sections were incubated with dilutions of the primary

antibodies (NCL-MUC-l-CORE, 1:100; DF3, 1:200; MY.lEl2,

1:200; HMFG-l, 1:100; and anti-MRP, 1:600) in PBS with 1%

BSA for 16 h at 4#{176}C.The sections were washed three times with

PBS, incubated with the biotinylated secondary antibodies, and

washed three times with PBS. All of the sections then received

ABC for 30 mm. After washing with PBS three times, the

sections were finally reacted with diaminobenzidine substrate

for 10-30 mm for visualization, rinsed with tap water, counter-

stained with hematoxylin, and mounted. Reaction products were

not present when nonimmune serum or PBS was used instead of

the primary antibodies.

Evaluation of the Results by Scoring. The results of the

antibody stainings were evaluated by the percentage of posi-

tively stained neoplastic cells. The immunostaining result was

considered positive if at least 5% of the cells were stained.

When less than 5% of neoplastic cells were stained, the immu-

nostaining result was considered negative. To compare the ex-

pression of antigens in a more comprehensive manner, the

degree of positive expression was graded as follows: (a) +,

5-50% of the neoplastic cells were stained; and (b) + + , >50%

of the neoplastic cells were stained.

Statistical Analysis. For an evaluation of the expression

rates of mucin antigens, statistical analysis using Fisher’s exact

test was performed. A probability of P < 0.01 or P < 0.05 was

considered statistically significant. The survival of the patients



Clinical Cancer Research 2607

Tabl e 2 Expression of the muc in antigens in no nneoplastic gastric mucosa

MUC1-CORE (%) DF3 (%) MY.lEl2 (%) HMFG-l (%) MRP (%)

Foveolar epithelium (n = 64) 0 22 1 1 39 0Fundic gland (n = 42) 38 100 98 81 0

Pyloric gland (,z = 32) 0 0 25 38 0

Intestinal metaplasia (n = 35) 0 0 0 0 100

was compared between the group with positive mucin antigen

expression and the group with negative mucin antigen expres-

sion according the Kaplan-Meier method. The significance of

the difference in survival between the two groups was tested

using the log-rank test.

The correlation of mucin expression with nodal status or

histology (intestinal-type cancer versus diffuse-type cancer) was

analyzed statistically. The correlation of the degree of staining

with the depth of invasion or with the prognosis of patients was

also analyzed statistically.

Combined Evaluation of MUC1 and MUC2 Expression.

Combined evaluation of MUC1 and MUC2 expression was

examined in every MUC1 mucin (MUC1-CORE, DF3,

MY. 1E12, and HMFG-l) and MUC2 mucin (MRP). The sur-

vival of the patients was compared between the group with the

MUC1 + and MUC2- expression pattern and the group with

the MUC1 - and MUC2+ expression pattern.

RESULTS

Expression of the Mucin Antigens in Nonneoplastic

Gastric Mucosa

Table 2 shows the expression rate of each mucin antigen in

the nonneoplastic epithelium of foveolar epithelium (64 cases),

the fundic gland (42 cases), the pyloric gland (32 cases), and

intestinal metaplasia (35 cases). MUC1-CORE antigen was ex-

pressed in the fundic glands in 16 of 42 (38%) specimens

examined, but it was not expressed in the other areas in all of the

cases. DF3 antigen was expressed in the fundic glands in all of

the specimens examined, but it was not expressed in the other

areas in many cases. However, DF3 antigen expression was

observed in the foveolar epithelium in 14 of 64 (22%) cases.

MY.lEl2 antigen was expressed in the fundic glands in 41

(98%) specimens, in foveolar epithelium in 7 (1 1%) specimens,

and in the pyloric glands in 8 (25%) specimens. HMFG-l

antigen was expressed in the fundic glands in 34 (81%) speci-

mens, in foveolar epithelium in 25 (39%) specimens, and in the

pyloric glands in 12 (38%) specimens. MRP antigen was not

expressed in the normal gastric mucosa, except for a faint

positive staining in the fundic glands in all of the specimens

examined.

In the intestinalized gastric mucosa, which was seen in 35

of 64 (55%) specimens, MUC1-CORE, DF3, MY.1E12, and

HMFG- 1 antigens were not expressed, even in the cases show-

ing positive MUC 1 mucin antigen expression in the foveolar

epithelium. On the other hand, MRP antigen was expressed in

the supranuclear region of the goblet cells in all of the speci-

mens.

Expression of the Mucin Antigens in Carcinoma Tissue

Fig. 1 shows representative immunohistochemical stain-

ings in MUC1 + (A) and MUC1 - (B) carcinomas and in

MUC2+ (C) and MUC2- (D) carcinomas.

Table 3 shows the expression rates of each mucin antigen

in carcinoma tissues by the depth of tumor invasion , , T2, and

T3) and the degree of staining. The total expression rate of the

DF3 antigen was not correlated with the depth of tumor inva-

sion, but the + + expression rate of the DF3 antigen was

significantly higher in T2 (35%) and T3 (15%) than it was in T,

(0%). The total expression rate of the MY. 1E12 antigen was

significantly higher in T2 (78%) than it was in T, (55%). In

particular, the + + expression rate of the MY.lEl2 antigen was

significantly higher in T2 (52%) than it was in T, (23%). The

expression rate of the MRP antigen was significantly higher in

T, (64%) than it was in T2 (30%) or T3 (30%). Both the + and

+ + expression rates of the MRP antigen were significantly

higher in T, (41 and 23%) than they were in T2 (19 and 1 1%).

The expression of the other MUC1-CORE and HMFG-l anti-

gens was not correlated with the depth of tumor invasion.

Intratumoral heterogeneity of antigen expression was ob-

served from nest to nest in several cases of intestinal-type

cancer, whereas it was not seen in the diffuse-type cancer.

However, there was no apparent difference between the degree

of expression in the whole area and that in the invasive edge.

Expression of the MUC1-CORE Antigen. In the 136

patients with gastric carcinomas, the MUC1-CORE antigen was

expressed in 18 (13%) patients, but it was not expressed in 1 18

(87%) patients. The patients with positive MUC1-CORE ex-

pression in the carcinoma showed significantly poorer survival

than those with no MUC1-CORE expression (Fig. 2A; P <

0.01), although the expression rates of MUC1-CORE showed no

significant relationship to the depth of tumor invasion (Table 3).

Expression of the DF3 Antigen. DF3 antigen was ex-

pressed in 75 (55%) patients, but it was not expressed in 61

(45%) patients. The patients with positive DF3 expression in the

carcinoma showed significantly poorer survival than those with

no DF3 expression (Fig. 2B; P < 0.01). As mentioned above,

the + + expression rate of the DF3 antigen was significantly

higher in the T2 and T3 cases than it was in the T, cases (Table 3).

Expression of the MY.1E12 Antigen. MY. 1E12 antigen

was expressed in 100 (74%) patients, but it was not expressed in

36 (26%) patients. The patients with positive MY. 1E12 expres-

sion in the carcinoma showed significantly poorer survival than

those with no MY.1E12 expression (Fig. 2C; P < 0.05). As

mentioned above, the expression rate of the MY. lEl2 antigen in

the T2 cases was significantly higher than that in the T, cases

(Table 3).



.;�

�‘

;, �

0�

--/.

I..

-

: �

, ,.�‘ -

‘A

U)

��: �

.� 4�1

‘). �1.-..-

r 1

M � 44..�l �

‘t-�� F

‘ �

Fig. I Representative immuno-histochemical stainings in MUC1(MY.lEl2)-positive (A) and -neg-ative (B) carcinomas and in MUC2(MRP)-positive (C) and -negative(D) carcinomas.

B

D

,._�)

C


.iJ

/9��

�

‘�,.

‘�1��.,

I

�

.,-

�

Expression of the HMFG-1 Antigen. HMFG-l antigen

was expressed in 105 patients (77%), but it was not expressed in

3 1 patients (23%). The patients with positive HMFG-l expres-

sion in the carcinoma showed significantly poorer survival than

those with no HMFG-l expression (Fig. 2D; P < 0.05), al-

though the expression rates of the HMFG-l antigen showed no

significant relationship to the depth of tumor invasion (Table 3).

Expression of the MRP Antigen. MRP antigen was

expressed in 48 (35%) patients, but it was not expressed in 88

(65%) patients. The patients with positive MRP expression in

the carcinoma showed significantly better survival than those

with no MRP expression (Fig. 2E; P < 0.05). As mentioned

above, the expression rate of the MRP antigen was significantly

higher in the T1 cases than it was in the T2 or T3 cases (Table 3).



MUC1 mucinMUCI-COREDF3MY. lEl2HMFG-l

MUC2 mucinMRP

4 4 0

36 36 0”

55” 32 23’

78 23 55


Table 3 Dep th of tumor invasion and expression of antigens”

Depth of tumor invasion”

1 (n 22) 12 (n = 94) 13 (n = 20)

Total + + + Total + + + Total + + +

16 13 3 10 5 5

58 23 35’ 60 45 15”

78” 26 52’ 75 25 50

76 30 46 80 35 45

41’ 23” 30’ 19’ II” 3ff1 15 15

-(; Values for turnorinvasion and staining intensity are percentages.,, T,, tumor invasion of submucosa; T2, tumor invasion of muscularis propria or subserosa: T�. tumor penetration of serosa (according to Ref. 9.

C p < 0.01 (significant difference in the expression).

‘I p < 0.05 (significant difference in the expression).

Antigen Expression and Nodal Status

The expression rates of MUCI-CORE (18%) and DF3

(62%) in the primary tumors of the patients with positive lymph

node metastasis (ii = 87) were significantly higher than those

(MUC1-CORE, 4%; DF3, 43%) in the patients with negative

lymph node metastasis (n 49; P < 0.05, respectively). In

contrast, the expression rate of MRP (28%) in the primary

tumors of the patients with positive lymph node metastasis was

significantly lower than that (49%) in the patients with negative

lymph node metastasis (P < 0.05). The expressions of

MY.lEl2 and HMFG-l showed no significant correlation with

nodal status.

Antigen Expression and Histological Type

There was no significant difference in patient survival

between intestinal-type cancer (n = 68) and diffuse-type cancer

(n 68) as analyzed by the Kaplan-Meier analysis and log-rank

test (data not shown), although the early-stage cases (T1 ) were

more numerous in intestinal-type cancer (25%) than in the

diffuse-type cancer (7%; P < 0.01). The expression rates of

DF3 (7 1 %), MY. 1El2 (84%), and HMFG- 1 (88%) in intestinal-

type cancer were significantly higher than those (DF3, 40%;

MY.lEl2, 63%; HMFG-l, 66%) in diffuse-type cancer (P <

0.01, respectively). The expressions of MUC1-CORE and MRP

showed no significant correlation with histological type.

The Degree of Positive Expression and Prognosis

Patients with + + expression of MY. lEl2 in the carcinoma

showed significantly poorer survival than those with + expres-

sion of MY.lEl2 (P < 0.01; Fig. 2C, inset). The other antigens

examined showed no significant difference in patient survival

between patients with + + expression and those with + expres-

sion (data not shown).

Survival Related to the Combined Status of MUC1 and

MUC2 Expression

The results of the combined evaluation of MUC1 and

MUC2 expression were shown in Fig. 3, A (MUCI-CORE and

MRP), B (DF3 and MRP), C (MY.lEl2 and MRP), and D

(HMFG- 1 and MRP). The patients with the MUC 1 + and

MUC2- expression pattern always showed the worst outcome.

In contrast, the patients with the MUCI - and MUC2+ expres-

sion pattern always showed the best outcome. This phenomenon

is seen in every MUC1 mucin examined (MUC1-CORE, DF3,

MY.IEl2, and HMFG-l: Fig. 3, A-D).

DISCUSSION

In the current study, we examined the expression of dif-

ferent glycoforms of MUC 1 mucin antigens detected by four

MAbs (NCL-MUC-l-CORE, DF3, MY.lEl2, and HMFG-l) in

136 patients with gastric carcinomas invading the submucosal

layer or the deeper layer of the stomach. A comparison of the

survival of the antigen-positive patient group and the antigen-

negative patient group by using the Kaplan-Meier method mdi-

cated that the patients whose tumors were positive for NCL-

MUC-l-CORE, DF3, MY.lEl2, and HMFG-l antibodies

showed significantly worse survival compared to patients whose

tumors showed negative staining with these antibodies. Because

these four MAbs recognize either only the core peptide or

different glycoforms of MUC 1 mucin, these results indicate that

MUC 1 mucin antigen expression was associated with a poor

outcome in patients irrespective of its glycosylation status, and

MUC1 is thus considered to be a useful prognostic marker ftr

poor outcome in patients with gastric cancers.

The MUC1 mucin is a transmembrane glycoprotein with an

extracellular domain consisting of a variable number of highly

conserved tandem repeats of 20 amino acids, a transmembrane

domain, and a cytoplasmic tail of 69 amino acids (16-21).

Overexpression of MUC 1 by cultured cells inhibits their aggre-

gation, possibly because of its large, extended, and rigid struc-

ture (22). MUC1 expressed in tumors may function as an anti-

adhesion molecule that inhibits cell-cell adhesion, inducing the

release of cells from the tumor (22, 23). In these manners,

MUC1 mucin expression may be associated with the invasive or

metastatic properties of carcinoma cells, resulting in a poor

prognosis for patients with gastric cancers with MUC 1 expres-

sion. A recent experimental study by Suwa et a!. (24) showed

that cultured gastric cancer cells acquired increased motility and

invasive property when they were transfected with the MUCI

gene. These findings are consistent with our observations dem-

onstrating high expression of MUC1 mucin (MUCI-CORE and

DF3) in the primary gastric cancers of patients with lymph node



(%) (%)

100

MUC1-CORE negative (n=118)

C,

‘� 50

Cl)

0F3

D

p< 0.01

20 40 60 80 100 120 140 160

Months after surgery

(%)

��)] p< 0.01

++ (n=64)0

0

100

I60 120


C,

.� 50

C/)

HMFG-1 negative (n=31)

p< 0.05

20 40 60 80 100 120 140 160


20 40 60 80 100 120 140 160


(%)100�

C,

� 50

Ci)

p< 0.05

2610 MUCI and MUC2 Mucins in Gastric Carcinoma

A

C

E

C,

>

� 50

Co

MUd-CORE positive (n=18)

0 20 40 60 80 100 120 140 160


(%)

100

50

20 40 60 80 100 120 140 160


B

Fig. 2 Comparison of survival between the groups with positive and negative expression of the MUC1 mucins [MUd-CORE (A), DF3 (B),MY.lEl2 (C). and HMFG-l (D)l and the MUC2 mucin [MRP (E)] as determined by the Kaplan-Meier method. The patients with positive MUCI

expression in the carcinoma showed significantly poorer survival than did those with no MUC 1 expression (A-D). In contrast, the patients withpositive MUC2 expression in the carcinoma showed significantly better survival than did those with negative MUC2 expression (E). Statistical

analysis for the correlation of the degree of staining with the prognosis of the patients demonstrated that the patients with + + expression of MY. 1E12

in the carcinoma showed significantly poorer survival than did those with + expression of MY.lEl2 (C, inset). The other antigens examined showedno significant difference between the survival of the patients with + + expression and those with + expression. E and 0, patients who were aliveat the last follow-up.



(%)MUC1 (MUd-CORE) and MUC2 (MRP)

I

MUC1 (DF3) and MUC2 (MAP)

A B

C

C,

>

aCo

20 40 60 80 100 120


140 160

p< 0,01

60 80 100 120Months after surgery

D

MUC1 (MY. 1E12) and MUC2 (MAP) � MUC1 (HMFG-1) and MUC2 (MAP)(%)

100 -

50

0

C,>

aCl)

C,

>

aU)

50

0

00 1] U 0 00 �1

� p< ooiMUC1(-), MUC2(+) (n=9)

� UC1(+). MUC2(+) (n=39)

MUC1(+), MUC2(-) (n=66)

t . I

120 140 160160 0 20 40 60 80 100



MUC1(-), MUC2(+) (n=15)

MU�(+), Muc2(+)�=3j P< 0.01

MUC1(+), MUC2(-) (n=67)

, i’i’I’Jf’J’

0 20 40 60 80 100 120 140


Fig. 3 Combined evaluation of MUCI and MUC2 expression. Patient survival was compared between the group with the MUCI + and MUC2-expression pattern and the group with the MUC1 - and MUC2+ expression pattern in each MUCI mucin and MUC2 mucin IMUCI-CORE and MRP(A), DF3 and MRP (B), MY.lEl2 and MRP (C), and HMFG-l and MRP (D)]. The patients with the MUC1 + and MUC2- expression pattern alwaysshowed the worst outcome. In contrast, the patients with the MUCI - and MUC2+ expression pattern always showed the best outcome. �. 0. A.and 0 . patients who were alive at the last follow-up.

metastasis and poor prognosis for patients with positive MUC I

expression (all MUC1 mucins examined) in the gastric cancers.

Generally, intestinal-type cancers behave more passively

than diffuse-type cancers (25-27). However, in the present

study, there was no significant difference in patient survival

between intestinal-type and diffuse-type cancers, despite the

prevalence of the early-stage cases in the intestinal-type cancer.

Therefore, the expression ofMUCl mucin (DF3. MY.lEl2, and

HMFG-l) in patients with intestinal-type cancer is considered to

affect the survival of patients with this type cancer to the same

level as patients with diffuse-type cancer.

The exposure of cryptic polypeptide epitopes recognized

by T and B cells (28-31) may result in carcinoma-specific

cytotoxic lymphocytes that recognize MUC 1 mucin core pep-

tides ( 16). These epitopes serve as antigenic determinants for

many anti-MUCI mucin MAbs (3, 4, 6, 32-37). On the other

hand, tumor cells expressing sialomucin were shown to be less

sensitive to cytolysis by human lymphokine-activated killer

lymphocytes (38-41). Thus, high levels of cell surface sialo-

mucin may be related to an escape from immune killing, result-

ing in an increased metastatic colonization of carcinoma cells

(41, 42). In the present study, the expression rate of sialylated

MUC I mucin detected by MAlI MY. 1E 1 2 was significantly

higher in T, than in T1. The ++ expression rate of the DF3

antigen, whose epitope seems to have sialyl oligosaccharides on

the core peptide (2), was also significantly higher in T, and T3

than in T1 (Table 3). Therefore, sialylation of the MUC 1 mucin

might be related to the invasiveness of gastric cancer cells and.

consequently, to the poor survival of the patients whose tumors

express the sialylated MUC 1 epitope. In particular. the apparent

poor survival of patients with + + expression of MY. I El 2 in

the carcinoma (Fig. 2C, inset) may indicate that a dominant

expression of sialylated MUC1 mucin detected by this MAb is

related to cancer cell invasion, inducing poor outcome in the

patients.

On the other hand, the expression of the other MUCI

mucin antigens (MUC1-CORE and HMFG-l) was not corre-

lated with the depth of tumor invasion (Table 3). Because the

glycosylation levels of MUd-CORE and HMFG-l antigens are

different, the production of MUC I mucin peptide itself may also




be related to poor prognosis of patients with gastric cancer, in

the different mechanism from the sialylated-MUC 1 mentioned

above. Overexpression of MUC 1 on the membrane of cultured

cells may also inhibit interaction between cytotoxic lympho-

cytes and tumor cells (43). Cells with overexpression of MUC1

showed an inhibition of integrin-mediated cell adhesion to the

extracellular matrix (44). A recent report by Agrawal et a!. (45)

showed that cancer-associated MUC I mucin and synthetic tan-

dem repeats of MUC 1 mucin core peptide can suppress human

T-cell-proliferative responses, and high levels of MUC 1 mucin

are correlated with immunosuppression in adenocarcinoma pa-

tients. The immune suppression by MUC1 mucin may also

result in the poor prognosis of patients with gastric cancer. The

survival of patients with positive expression of nonglycosylated

MUC1-CORE in the carcinoma was particularly poor (Fig. 2A)

compared with those with positive expression of the other

glycosylated MUC1 mucin antigens (DF3, MY.lEl2, and

HMFG-l ; Fig. 2, B-D). The suppression of human T-cell-

proliferative responses by the MUC1 mucin core peptide (45)

might be related to the particularly poor prognosis of MUCI-

CORE-positive patients.

MUC2 is a secretory mucin that forms large polymers as a

result of disulfate bonding. MUC2 has a high viscosity in

solution (46). Our results indicate that in contrast to MUC1

mucin antigen expression, MUC2 antigen expression seems to

be a prognostic factor associated with a favorable outcome in

patients with gastric cancers (Fig. 2E). Our previous studies on

tumors of the pancreas and intrahepatic bile duct showed that

MUC2 was expressed mainly in the noninvasive-type tumors

( 14, 15). In the present study on gastric carcinomas, MUC2

expression was seen in many cases of early-stage gastric carci-

noma (T1). The significantly high expression of MUC2 (MRP

antigen) in early-stage cancers (T1; Table 3) and the signifi-

cantly low expression of MUC2 in primary gastric cancers with

lymph node metastasis are consistent with the favorable survival

of patients with MUC2 expression observed in this study. The

cysteine-rich domains of MUC2 (47, 48) may play a role in

regulating cell proliferation (48), which may in turn contribute

to the low malignant potential of gastric cancers with positive

MUC2 expression.

Mucin glycoproteins are heavily glycosylated, with ap-

proximately 50-85% ofthe total molecular weight composed of

carbohydrates (49). MUC2 expression in gastric cancers was

usually intracytoplasmic and was not seen in the secreted mucin

fractions (Fig. I C). The observed pattern of the intracytoplasmic

expression suggests the following possibilities: (a) precursor

forms of the MUC2 polypeptide backbone are not heavily

glycosylated in early compartments of the Golgi; therefore,

these forms are readily detected by anti-MRP; and (b) heavily

glycosylated MUC2 mucin is present in secreted fractions and is

difficult to detect with antibodies due to the blocking of epitope

binding on the polypeptide backbone by carbohydrate side

chains. Further investigation of carbohydrate structures linked to

the MUC2 polypeptide backbone in the secreted mucin may be

important to our understanding of the biological role of secreted

and extracellular mucin in gastric cancers. The anti-MRP anti-

body recognizes the MUC2 core protein (50, 5 1). No antibody

is available for specific glycoforms of MUC2. The relationship

between the degree of MUC2 glycosylation and prognosis

would also be an important area for future study.

The combined evaluation of MUC1 and MUC2 expression,

as shown in Fig. 3, A-D, could show very clear differences in

outcome between the group of patients with MUC1 + and

MUC2- expression and the group of patients with MUC1 - and

MUC2+ expression. The combined evaluation of MUC1 and

MUC2 mucin staining may be useful clinically to predict the

outcome in patients with gastric cancer. Recently, Ho et al. (52)

reported that gastric cancers contain apomucin antigens other

than MUC1 and MUC2 mucins, and that more advanced stage

gastric cancers express greater numbers of apomucin antigens.

The relationship between the other apomucin antigens and the

outcome of patients would also be an interesting future area of

study.

ACKNOWLEDGMENTS

We are grateful to Dr. Joyce Taylor-Papadimitriou (Imperial Can-

cer Research Fund Laboratories, London, United Kingdom) for kindly

providing the MAb HMFG-l used in this study and also to Yoshiharu

Atsuji and Yoshiko Arimura for excellent technical assistance.

REFERENCES

1 . Sakamoto, H., Yonezawa, S., Tanaka, S., Kim, Y. S., and Sato, E.

Mucin antigen expression in gastric carcinomas of young and old adults.

Hum. Pathol.,28: 1056-1065, 1997.

2. Siddiqui, J., Abe, M., Hayes, D., Shani, E., Yunis, E., and Kufe, D.

Isolation and sequencing of a cDNA coding for the human DF3 breast

carcinoma-associated antigen. Proc. Natl. Acad. Sci. USA, 85: 2320-

2323, 1988.

3. Burchell, J., Gendler, S., Taylor-Papadimitriou, J., Girling, A.,Lewis, A., Millis, R., and Lamport, D. Development and characteriza-

tion of breast cancer reactive monoclonal antibodies directed to the core

protein of the human milk mucin. Cancer Res., 47: 5476-5482, 1987.

4. Burchell, J., and Taylor-Papadimitriou, J. Effect of modification ofcarbohydrate side chains on the reactive of antibodies with core-protein

epitopes of the MUCI gene product. Epithelial Cell Biol., 2: 155-162,

1993.

5. Girling, A., Bartkova, J.. Burchell, J., Gendler, S., Gillett, C., and

Taylor-Papadimitriou, J. A core protein epitope of the polymorphic

epithelial mucin detected by the monoclonal antibody SM-3 is selec-

tively exposed in a range of primary carcinomas. Int. J. Cancer, 43:

1072-1076, 1989.

6. Burchell, J., Taylor-Papadimitriou, J., Boshell, M., Gendler, S., and

Duhig, T. A short sequence within the amino acid tandem repeat of a

cancer-associated mucin contains immunodominant epitopes. Int. J.

Cancer. 44: 691-696, 1989.

7. Hull, S. R., Bright, A., Carraway, K. L., Abe, M., Hayes, D. F., and

Kufe, D. W. Oligosaccharide differences in the DF3 sialomucin antigenfrom normal human milk and the BT-20 human breast carcinoma cellline. Cancer Commun., 1: 261-267, 1989.

8. Nakamori, S., Ota, D. M., Cleary, K. R., Shirotani, K., and Irimura,

1. MUC 1 mucin expression as a marker of progression and metastasis

of human colorectal carcinoma. Gastroenterology, 106: 353-361, 1994.

9. Japanese Research Society for Gastric Cancer. Japanese Classifica-

tion of Gastric Carcinoma, First English Edition. Tokyo, Japan: Kane-

hara & Co., Ltd., 1995.

10. Yamamoto, M., Bhavanandan, V. P., Nakamori, S., and Irimura, T.A novel monoclonal antibody specific for sialylated MUCI mucin. Jpn.

J. Cancer Res., 87: 488-496, 1996.

I 1 . Taylor-Papadimitriou, J., Peterson, J. A., Arklie, J., Burchell, J.,

Ceriani, R. L., and Bodmer, W. S. Monoclonal antibodies to epithelium-

specific components of the human milk fat globule membrane: produc-

tion and reaction with cells in culture. Int. J. Cancer, 28: 17-21, 1981.




12. Burchell, J., Durbin, H., and Taylor-Papadimitriou, J. Complexity

of expression of antigenic determinants recognized by monoclonal an-tibodies HMFG-l and HMFG-2 in normal and malignant human mam-mary epithelial cells. J. Immunol., 131: 508-513, 1983.

13. Hsu, S. M., Game, L., and Fanger, H. Use of avidin-biotin-perox-idase complex (ABC) in immunoperoxidase techniques: a comparison

between ABC and unlabeled antibody (PAP) procedures. J. Histochem.Cytochem., 29: 577-580, 1981.

14. Yamashita, K., Yonezawa, S., Tanaka, S., Shirahama, H., Sakoda,K., Imai, K., Xing, P. X., McKenzie, I. F., Hilkens, J., Kim, Y. S., andSato, E. Immunohistochemical study of mucin carbohydrates and coreprotein in hepatolithiasis and cholangiocarcinoma. Int. J. Cancer, 55:

82-91, 1993.

15. Osako, M., Yonezawa, S., Siddiki, B., Huang, J., Ho, J. J. L., Kim,Y. S., and Sato, E. Immunohistochemical study of mucin carbohydratesand core protein in human pancreatic tumors. Cancer (Phila.), 71:

2191-2199, 1993.

16. Gendler, S., Taylor-Papadimitriou, J., Duhig, T., Rothbard, J., andBurchell, J. A highly immunogenic region of a human polymorphicepithelial mucin expressed by carcinomas is made up of tandem repeats.J. Biol.Chem., 263: 12820-12823, 1988.

17. Gendler, S. J., Lancaster, C. A., Taylor-Papadimitriou, J., Duhig, T.,Peat, N., Burchell, J., Pemberton, L., Lalani, E. N., and Wilson, D.Molecular cloning and expression of human tumor-associated polymor-phic epithelial mucin. J. Biol. Chem., 265: 15286-15293, 1990.

I 8. Wreschner, D. H., Tsarfaty, I., Hareuveni, M., Zaretsky, J., Smoro-

dinsky, N., Weiss, M., Horev, J., Kotkes, P., Zrihan, S., Jeltsch, J. M.,

Green, S., Lathe, R., and Keydar, I. Isolation and characterization of fulllength cDNA coding for the H23 breast tumor associated antigen. In:M. H. Rich, J. A. Hager, and I. Kevdar (eds.), Breast Cancer: Progressin Biology, Clinical Management and Prevention, pp. 41-60. Amster-

dam, the Netherlands: Kiuwer Academic Publishers, 1989.

19. Wreschner, D. H., Hareuveni, M., Tsarfaty, I., Smorodinsky, N.,Horev, J., Zaretsky, J., Kotkes, P., Weiss, M., Lathe, R., Dion, A., andKeydar, I. Human epithelial tumor antigen cDNA sequences. Differen-tial splicing may generate multiple protein forms. Eur. J. Biochem., 189:

463-473, 1990.

20. Ligtenberg, M. J. L., Vos, H. L., Gennisen, A. M. C., and Hilkens,J. Episialin, a carcinoma-associated mucin, is generated by a polymor-

phic gene encoding splice variants with alternative amino termini.J. Biol. Chem., 265: 5573-5578, 1990.

21. Lan, M. S., Batra, S. K., Qi, W., Metzgar, R. S., and Hollingsworth,

M. A. Cloning and sequencing of a human pancreatic mucin cDNA. J.

Biol. Chem., 265: 15294-15299, 1990.

22. Ligtenberg, M. J. L., Buijs, F., Vos, H. L., and Hilkens, J. Suppres-sion of cellular aggregation by high level of episialin. Cancer Res., 52:

2318-2324, 1992.

23. Makiguchi, Y., Hinoda, Y., and Imai, K. Effect of MUC1 mucin

and anti-adhesion molecule on tumor cell growth. Jpn. J. Cancer Res.,87: 505-511, 1996.

24. Suwa, 1., Hinoda, Y., Makiguchi, Y., Takahashi, T., Itoh, F.,

Adachi, M., Hareyama, M., and Imai, K. Increased invasiveness of

MUCI cDNA-transfected human gastric cancer MKN74 cells. mt. J.Cancer, 76: 377-382, 1998.

25. Davessar, K., Pezzulo, J. C., Kessimian, N., Hale, J. H., andJauregui, H. 0. Gastric adenocarcinoma: prognostic significance ofseveral pathologic parameters and histologic classifications. Hum.Pathol., 21: 325-332, 1990.

26. Moutinho, R. M., Sarmento, J. A., Sobrinho, S. M. A., and Bastos,J. Prognostic significance of Lauren and Ming classifications and otherpathologic parameters in gastric carcinoma. Cancer (Phila.), 47: 780-

784, 1981.

27. Stemmermann, G. N., and Brown, C. A survival study of intestinal

and diffuse types ofgastric carcinoma. Cancer (Phila.), 33: 1 190-1 195,1974.

28. Band, D. L., Lan, M., Metzgar, R. S., and Finn, 0. J. Specific.major histocompatibility complex-unrestricted recognition of tumor-

associated mucins by human cytotoxic T cells. Proc. NatI. Acad. Sci.

USA, 86: 7159-7163, 1989.

29. Jerome, K. R., Bamd, D. L., Bendt, K. M., Boyer, C. M., Taylor-

Papadimitriou, J., McKenzie, I. F., Bast, R. C., Jr., and Finn, 0. J.

Cytotoxic T-lymphocytes derived from patients with breast adenocarci-noma recognize an epitope present of the protein core of a mucinmolecule preferentially expressed by malignant cells. Cancer Res., Si:

2908-2916, 1991.

30. Jerome, K. R., Bu, D., and Finn, 0. J. Expression of tumor-associated epitopes of Epstein-Barr virus-immortalized B-cells and Bur-kitts’s lymphomas transfected with epithelial mucin complementary

DNA. Cancer Res., 52: 5985-5990, 1992.

31. Jerome, K. R., Domenech, N., and Finn, 0. J. Tumor-specificcytotoxic T cell clones from patients with breast and pancreatic adeno-

carcinoma recognize EBV-immortalized B cells transfected with poly-morphic epithelial mucin complementary DNA. J. Immunol., iSi:

1654-1662, 1993.

32. Ban, T., Imai, K., and Yachi, A. Immunohistological and immuno-

chemical characterization of a novel pancreatic cancer-associated anti-gen, MUSEI 1. Cancer Res., 49: 7141-7146, 1989.

33. Hinoda, Y., Nakagawa, N., Ohe, Y., Kakiuchi, H., Tsujisaki, M.,

Imai, K., and Yachi, A. Recognition of the polypeptide core of mucin by

monoclonal antibody MUSEI I against an adenocarcinoma-associated

antigen. Jpn. J. Cancer Res., 8i: 1206-1209, 1990.

34. Hilkens, J., Buijs, F., Hilgers, J., Hageman, P., Calafat. J., Sonnen-berg, A., and Van Der Valk, M. Monoclonal antibodies against humanmilk-fat globule membranes detecting differentiation antigens of the

mammary gland and its tumors. Int. J. Cancer, 34: 197-206, 1984.

35. Hilkens, J., Kroezen, V., Bonfrer, J. M. G., De Jong-Bakker, M.,and Bruning, P. F. MAM-6 antigen, a new serum marker for breastcancer monitoring. Cancer Res., 46: 2582-2587, 1986.

36. Hilkens, J., and Buijs, F. Biosynthesis of MAM-6, an epithelial

sialomucin. J. Biol. Chem., 263: 4215-4222, 1988.

37. Hilkens, J., Kroezen, V., Buijs, F., Hilgers, J., Van Der Vilet, M.,De Voogd, W., Bonfrer, J., and Bruning, P. F. MAM-6, a carcinomaassociated marker: preliminary characterization and detection in sera ofbreast cancer patients. In: R. L. Ceriani and M. Nijhoff (eds.). Mono-clonal Antibodies and Breast Cancer, pp. 28-42. Hague, the Nether-lands: Martinus Nijhoff, 1985.

38. Rinsum, J. V., Smets, L. A., Rooy, H. V., and Van Den Eijnden,D. H. Specific inhibition of human natural killer cell-mediated cytotox-icity by sialic acid and sialooligosaccharides. Int. J. Cancer, 38: 915-922, 1986.

39. Moriarty, J., Skelly, C. M., Bharathan, S., Moody, C. E., andSherblom, A. P. Sialomucin and lytic susceptibility of rat mammary

tumor ascites cells. Cancer Res., SO: 6800-6805, 1990.

40. Ogata, S., Maimonis, P. J., and Itzkowitz, S. H. Mucins bearing thecancer-associated sialosyl-Tn antigen mediated inhibition of naturalkiller cell cytotoxicity. Cancer Res., 52: 4741-4746, 1992.

41. Irimura, 1., Mclsaac, A. M., Carlson, D. A., Yagita, M., Grimm,

E. A., Menter, D., Ota, D. M., and Cleary, K. R. Soluble factor in normaltissues that stimulates high molecular weight sialoglycoprotein pro-

duced by human colon carcinoma cells. Cancer Res., 50: 3331-3338,

1990.

42. Irimura, T., Nakamori, S., Matsushita, Y., Taniuchi, Y., Todoroki,

N., Tsuji, T., Izumi, Y., Kawamura, Y., Hoff, S. D., Cleary, K. R., andOta, D. M. Colorectal cancer metastasis determined by carbohydrate-mediated cell adhesion: role of sialyl-Lex antigens. Semin. Cancer Biol.,

4: 319-324, 1993.

43. van de Wiel-van Kemenade, E., Ligtenberg, M. J. L., de Beer, A. J.,Buijs, F., Vos, H. K., Melief, C. J., Hilkens, J., and Figdor, C. G.Episialin (MUC1 ) inhibits cytotoxic lymphocyte-target cell interaction.J. Immunol., iSi: 767-776, 1993.

44. Wesseling, J., Van Der Valk, S. W., Vos, H. L., Sonnenberg, A..and Hilkens, J. Episialin (MUCI ) over-expression inhibits integrin-mediated cell adhesion to extracellular matrix components. J. Cell Biol.,

i29: 255-265, 1995.



2614 MUCI and MUC2 Mucins in Gastric Carcinoma

45. Agrawal, B., Krants, M. J., Reddish, M., and Longenecked, B. M.Cancer-associated MUC1 mucin inhibits human T-cell proliferation,which is reversible by IL-2. Nat. Med., 4: 43-49, 1998.

46. Gum, J. R., Hicks, J. W., Toribara, N. W., Siddiki, B., and Kim,Y. S. Molecular cloning of human intestinal mucin (MUC2) cDNA:identification of the amino terminus and overall sequence similarity to

Pre-Pro-von Willebrand factor. J. Biol. Chem., 269: 2440-2446, 1994.

47. Toribara, N. W., Gum, J. R., Culhane, P. J., Lagace, R. E., Hicks,

J. W., Petersen, G. M., and Kim, Y. S. MUC-2 human small intestinalmucin gene structure: repeated arrays and polymorphism. J. Clin. In-vestig., 88: 1005-1013, 1991.

48. Carraway, K. L., and Fregien, N. Mucin structure and function:insights from molecular biology. Trends in Glycoscience and Gly-cotechnology, 33: 31-44, 1995.

49. Kim, Y. S. Mucin glycoproteins in gastrointestinal malignanciesand metastasis. Eur. J. Gastroenterol. Hepatol., 5: 219-225, 1993.

50. Gum, J. R., Byrd, J. C., Hicks, J. W., Toribara, N. W., Lamport,D. T. A., and Kim, Y. S. Molecular cloning of human intestinal mucincDNAs: sequence analysis and evidence for genetic polymorphism.J. Biol. Chem., 264: 6480-6487, 1989.

51. Ho, S. B., Niehans, G. A., Lyftogt, C., Yan, P. S., Cherwitz, D. L.,Gum, E. T., Dahiya, R., and Kim. Y. S. Heterogeneity of mucin geneexpression in normal and neoplastic tissues. Cancer Res., 53: 641-651,

1993.

52. Ho, S. B., Shekels, L. L., Toribara, N. W., Kim, Y. S., Lyftogt, C.,Cherwitz, D. L., and Niehans, G. A. Mucin gene expression in normal,preneoplastic, and neoplastic human gastric epithelium. Cancer Res.,

55: 2681-2690, 1995.



1998;4:2605-2614. Clin Cancer Res T Utsunomiya, S Yonezawa, H Sakamoto, et al. its relationship with the prognosis of the patients.Expression of MUC1 and MUC2 mucins in gastric carcinomas:

Updated version

http://clincancerres.aacrjournals.org/content/4/11/2605

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/4/11/2605To request permission to re-use all or part of this article, use this link



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

mailto:[email protected]



expression of muc1 and muc2 mucins in gastric carcinomas ... · vol. 4, 2605-2614, november 1998...

Documents