Annals of Surgical Oncology Sign the Guestbook
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
10.1245/s10434-007-9363-2
Annals of Surgical Oncology 14:1880-1889 (2007)
© 2007 Society of Surgical Oncology
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tsuboi, K.
Right arrow Articles by Yazawa, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsuboi, K.
Right arrow Articles by Yazawa, S.

Original Article

{alpha}1,2Fucosylation Is a Superior Predictor of Postoperative Prognosis for Colorectal Cancer Compared with Blood Group A, B, or Sialyl Lewis X Antigen Generated within Colorectal Tumor Tissues

Kaori Tsuboi, MD1, Takayuki Asao, MD, PhD1, Munenori Ide, MD, PhD1, Shinji Hashimoto, MD1, Kasumi Noguchi, PhD2, Yoshihiko Kominato, MD, PhD3, Abby R. Saniabadi, PhD4, Hiroyuki Kuwano, MD, PhD1 and Shin Yazawa, PhD1,5,6

1 Department of General Surgical Science, Gunma University Graduate School of Medicine, Maebashi, 371-8511, Japan
2 Criminal Investigation Laboratory, Gunma Prefectural Police HQ, Maebashi, 371-0846, Japan
3 Department of Legal Medicine and Molecular Genetics, Graduate School of Medicine, Gunma University, Maebashi, 371-8511, Japan
4 Japan Immunoresearch Laboratories, Takasaki, 370-0021, Japan
5 Tokushima Research Institute, Otsuka Pharmaceutical Co. Ltd., Tokushima, 771-0192, Japan
6 Innovation Center for Cooperative Research of Gunma University, Maebashi, 371-8511, Japan

Correspondence: Address correspondence and reprint requests to: Shin Yazawa, PhD; E-mail: syazawa{at}titan.ocn.ne.jp


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Background: We have previously demonstrated tumor-specific {alpha}1,2fucosylation, which is associated with resistance of tumor cells to anticancer treatment in human colorectal tumor tissues. By using the YB-2 monoclonal antibody, the resulting products have been identified as Y, Leb, and H type 2 antigens in colorectal tumor tissues.

Methods: Immunohistochemical analyses of colorectal cancer tissues (74 specimens) were performed with a newly established mouse monoclonal antibody, YB-3 specifically recognizing H disaccharide (Fuc{alpha}1,2Galß) structures, and anti-A, anti-B, YB-2, and anti–sialyl Lewis X (SLX) antibodies, together with the analyses of glycosyltransferases involved in the synthesis of ABH antigens in the same tissues.

Results: The YB-3 antibody enabled us to detect colorectal tumors, particularly tumors in the distal large intestine and the rectum, with high sensitivity (74.3%) and specificity (100%). From immunohistochemical and enzymatic analyses of colorectal tissues, we found that once {alpha}1,2fucosylation had proceeded in tumor tissues, blood group A or B antigen was also synthesized in approximately half of the tissues of A or B blood type, but not in their normal tissues. A correlation of survival rate with immunostaining of tissues was found only by YB-3 antibody and not by anti-A, anti-B, or anti-SLX antibody.

Conclusions: As a predictor of postoperative prognosis of patients with colorectal cancer, immunodetection of {alpha}1,2fucosylated antigens with the YB-3 antibody seemed to be superior to blood groups A, B, or SLX antigen in colorectal tumor tissues.

Key Words: Colorectal cancer • {alpha}1,2Fucosylation • Distal colon • Blood group A and B antigens • Sialyl Lewis X antigen • A and B enzyme • Survival rate • Diagnosis


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Altered expression of blood group antigens and their related glycoconjugates has been widely observed in a variety of cancers.1,2 In addition to their detailed structures, enzymatic and molecular genetic analyses of glycosyltransferases involved in the synthesis of each glycosidic linkage have made it possible to figure out such changes more clearly.1–5 The epithelial cells of the gastrointestinal wall express the ABH antigens depending on the secretor status of the tissue. However, beginning from the descending colon, the ABH activity gradually decreases and is not detected in the distal parts of the colon or in the rectum,6 whereas a series of fucosylated and sialylated antigens have been detected in association with presence of tumors.2,7,8 In colorectal tumor tissues, however, both aberrant accumulation and deletion of blood groups and their related antigens have been reported.7,9,10 In our previous studies,7,9,10 an aberrant {alpha}1,2fucosylation that relates to the synthesis of tumor-specific Y, Leb, and H type 2 antigens has been observed in colorectal tumor tissues, and enzymatic and molecular genetic analyses were performed.5,11 Increased activities of {alpha}1,2fucosyltransferase were also demonstrated to highly correlate with malignant progression of human colorectal adenocarcinoma.12 It was then thought that altered fucosylation in the colorectal tissues by the aberrant {alpha}1,2fucosyltransferase must induce some other glycosylations and produce tumor-associated antigens other than Y, Leb, and H antigens, which were demonstrated previously.7,9,10

Recently, we and others reported that the loss of A/ B antigens has been widely observed in A and B blood type patients with cancer, and that it could be attributed to the allelic loss of the ABO genes13 and the DNA hypermethylation of the ABO promoter region.14,15 More recently, some other factors accounting for loss of A/B antigens have also been suggested to be present in oral squamous cell carcinoma,16 whereas reduction or deletion of A/B antigens has been found to affect the survival of cancer patients1,2 and the tumorigenicity of carcinoma cells.17–20

In this study, we established a new anti-fucosylated antigen mouse monoclonal antibody, YB-3, and used it, along with anti-A, -B, –sialyl Lewis X (SLX), and YB-2 antibodies, for immunohistochemical staining of the distal large intestine and the rectum (distal colorectal tissues) obtained from patients with distal colorectal cancer. The results were analyzed in association with their genuine ABO blood types and levels of their corresponding glycosyltransferases. Immunodetection of {alpha}1,2fucosylated antigens that could be easily detected by our newly established YB-3 monoclonal antibody showed superiority over A, B, or SLX antigen, which is eventually synthesized in tumor tissues of patients with A or B blood type for predicting postoperative patients who do not achieve full remission.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Materials
Guanosine diphosphate (GDP)-[3H]Fuc, uridine diphosphate (UDP)-[3H]GalNAc, UDP-[3H]Gal, and Aquasol-2 were purchased from Du Pont (Boston, MA). GDP-Fuc, UDP-GalNAc, UDP-Gal, phenyl ß-Gal, and Triton X-100 were from Sigma (St. Louis, MO). Sep-Pak plus C18 reverse-phase cartridges were obtained from Waters (Milford, CT). Fuc{alpha}1,2-GalßPNP21 and YB-2 and YB-3 mouse monoclonal antibodies7 were prepared as described previously. Anti-A and anti-B mouse monoclonal antibodies were obtained from Kainos Lab (Tokyo, Japan), and anti-SLX (KM93) was from Seikagaku (Tokyo, Japan). Biotinylated anti-mouse IgM and the ABC kit were from Vectorstain (Vector Laboratories, Burlingame, CA). Chemically synthesized oligosaccharides, such as Galßl,3GlcNAcß, Galßl,4GlcNAcß, Galßl,3Gal-NAc{alpha}, Fuc{alpha}1,2Galß, Fuc{alpha}1,2Galßl,3GlcNAcß, Fuc{alpha}1,2Galßl,4GlcNAcß, Fuc{alpha}1,2Galß1,3GalNAc{alpha}, Galßl,3[Fuc{alpha}l,4]GlcNAcß, Galßl,4[Fuc{alpha}1,3]Glc-NAcß, Fuc{alpha}1,2Galß1,3[Fuc{alpha}1,4]GlcNAcß, Fuc{alpha}1, 2Galß1,4[Fuc{alpha}1,3]GlcNAcß attached to bovine serum albumin, and Fuc{alpha}1,2Galß attached to silica beads (H disaccharide Synsorb) were obtained from Chembiomed (Edmonton, Canada). Tissue specimens from distal colon were obtained from patients with colorectal cancer admitted to the Department of General Surgical Science, Gunma University Hospital, Gunma, Japan, in accordance with the guidelines for informed consent. Surgically resected tissues were stored at –80°C until assays of glycosyltransferases were performed, or the tissues were fixed in 10% formalin and embedded in paraffin for immunohistochemical analysis. Classification of tumors was according to the tumor, node, metastasis system.22

Enzyme Preparation
One gram of distal colorectal tissue from tumors and normal part was solubilized in 2 mL of 0.01 M phosphate-buffered saline containing 2% Triton X-100 by sonication at 4°C, and the mixture was centrifuged at 10,000 rpm for 30 minutes. The supernatant was used for assay.

Glycosyltransferase Assay
The standard reaction mixture in a final volume of 100 µL contained 4 µmol of HEPES-NaOH, pH 7.0, l µmol of MnCl2, .5 µmol of ATP, 500 µg of Triton X-100, 10 µL of the enzyme supernatant, 5 nmol of GDP-[3H]Fuc (for {alpha}1,2fucosyltransferase), UDP-[3H]GalNAc (for A enzyme), or UDP-[3H]Gal (for B enzyme) and 20 nmol of phenyl ß-Gal (for {alpha}1,2fucosyltransfease) or Fuc{alpha}1,2GalßPNP (for A and B enzyme).21 After incubation at 37°C for 4 hours, the reaction mixture was terminated by addition of an equal volume of ethanol. The enzyme activities of the supernatant were measured with a Sep-Pak plus C18 cartridge as previously described.21

Immunostaining of Tissues
Paraffin was removed from sections of distal colon, and sections were stained by the indirect immunoperoxidase method. The sections were treated with 100% methanol containing.3% hydrogen peroxide and were preincubated in 10% normal horse serum for 30 minutes. The specimens were then incubated for 30 minutes with YB-2, YB-3, anti-A, anti-B, and anti-SLX antibodies, followed by staining with biotinylated anti-mouse IgM and treatment with the Vectastain ABC kit according to the manufacturer’ instructions. Color was developed with a solution containing .02% 3-3'-diaminobenzidine tetrahydrochloride and .01% hydrogen peroxide in .05 M of ammonium acetate-citrate buffer, pH 6.0. The sections were then counterstained with hematoxylin, washed with distilled water, and mounted. Slides were examined blindly without knowledge of clinicopathological background, and "positive" was defined as more than 15% of the tissue being positively stained with the antibody. Further, classification into three groups was based on the intensity of staining with YB-3 antibody as follows: grade 1, negative; grade 2, very weak but positive with intratumor heterogeneity; and grade 3, uniformly positive.

Preparation of Immunogen and Monoclonal Antibody
Preparations of the immunogen from human saliva samples and mouse monoclonal antibodies raised against fucosylated antigens were performed as described previously.7 Characterization of each antibody was carried out by enzyme-linked immunosorbent assay that used an oligosaccharide attached to bovine serum albumin as a standard antigen.7 YB-2 and YB-3 antibodies were purified by affinity chromatography on a column of H disaccharide Synsorb beads.9

Determination of Protein
Protein was determined with a DC protein assay kit (Bio-Rad) with bovine serum albumin used as a standard.

Statistical Analysis
Survival curves were obtained according to the Kaplan-Meier method, and the differences in survival rates of each group were calculated by the Cox-Mantel test. Other data were analyzed by the {chi}2 test.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Specificity of Anti-Fucosylated Antigen Antibody, YB-3
YB-3, one of the anti-fucosylated antigen antibodies obtained from culture supernatant of clones, showed cross-reactivity with H disaccharide (Fuc{alpha}1,2Galß), H type 1 (Fuc{alpha}1,2Galßl,3GlcNAcß), H type 2 (Fuc{alpha}1,2Galßl,4GlcNAcß), H type 3 (Fuc{alpha}1,2Galßl, 3GalNAc{alpha}), Leb (Fuc{alpha}1,2Galßl,3[Fuc{alpha}l,4]GlcNAcß), and Y (Fuc{alpha}1,2Galßl,4[Fuc{alpha}1,3]GlcNAcß) structures. No cross-reactivity with other fucosylated structures, including Lea (Galßl,3[Fuc{alpha}l,4]GlcNAcß), X (Galßl, 4[Fuc{alpha}1,3]GlcNAcß), or nonfucosylated ones, was observed (Fig. 1Go). It was therefore concluded that YB-3 antibody could specifically recognize Fuc{alpha}1, 2Galß linkages regardless of their backbone structures.


Figure 1
View larger version (25K):
[in this window]
[in a new window]

  		FIG. 1. Specificity of YB-3 mouse monoclonal antibody measured by enzyme-linked immunosorbent assay with various oligosaccharides attached to bovine serum albumin. The relative activity was calculated on the basis of the reactivity against the Fuc{alpha}1,2Galß antigen as 100%.

 
{alpha}1,2Fucosyltransferase, {alpha}1,3N-Acetylgalactosaminyl-transferase (A Enzyme), and {alpha}1,3galactosyltransferase (B Enzyme) in Distal Colorectal Tissues
Colorectal tumor tissues (n = 9) were obtained from patients with colon cancer at distal region together with adjacent tissues that were pathologically found to be free of tumor. {alpha}1,2Fucosyl-, {alpha}1,3N-acetylgalactosaminyl (A enzyme), and {alpha}1,3galactosyl (B enzyme) transferase activities were measured both in the tumor and normal tissues, along with typing of ABO blood groups from their red blood cells. In normal tissues, no {alpha}1,2fucosyltransferase activity was detected in all cases, but in tumor tissues, the presence of the enzyme activity was demonstrated irrespective of their ABO blood types (Table 1Go), whereas in normal tissues, A and B enzymes were detected depending on their ABO blood types, but no inconsistency was found. Further, the levels of A and B enzyme activities were fairly high when compared with those of {alpha}1,2fucosyltransferase.


View this table:
[in this window]
[in a new window]

  		TABLE 1. Levels of glycosyltransferase activities in the distal large intestine and rectum tissues and their immunohistochemical staining with YB-3, anti-A, and anti-B antibodies
 
Immunohistochemical Staining of Distal Colon with YB-2, YB-3, Anti-A, Anti-B, and Anti-SLX Antibodies
Immunohistochemical staining of distal colorectal tissues was performed by using YB-3, YB-2, anti-A, anti-B, and anti-SLX antibodies. As shown in Fig. 2Go, tumor-specific staining with YB-3 as well as YB-2 antibody was demonstrated in distal colorectal tissues irrespective of ABO blood type. Further, positive staining with anti-A and anti-B antibodies was observed in tumor tissues from approximately half of the cases according to their ABO blood types. Most of the positive staining with anti-A or anti-B antibody was found to be restricted within those with YB-3 antibody. Positive staining with anti-SLX antibody was also observed in tumor tissues irrespective of ABO blood type, but the expression of SLX antigen on tumors seemed to be markedly different from that of fucosylated antigens detected by YB-3 antibody; 21% (16 of 74) of the specimens showed positive staining with YB-3 antibody but negative staining with anti-SLX antibody (Fig. 3aGo), whereas 25.7% (19 of 74) of the specimens showed the opposite (Fig. 3bGo). It was noteworthy that not a few tumors stained with both antibodies (16.2%, 12 of 74) showed strikingly different staining patterns (Fig. 3cGo). The incidences of positive staining with YB-3 antibody in colorectal cancer tissues from all patients and the patients with A, B, and AB blood types were 74.3% (55 of 74) and 71.2% (42 of 59), respectively; those in normal tissues were 0% in both groups (Table 2Go). In patient 3 (Table 1Go, blood type A), no clear staining with anti-A was seen in the tumor tissues, although strikingly positive staining was observed with YB-3 as well as YB-2 (data not presented) antibodies. In patient 6 (Table 1Go, blood type AB), a strong positive staining was seen with anti-A antibody but not with anti-B antibody. The incidence of positive staining of tumors with anti-SLX antibody was fairly low (41.9%, 31 of 74) compared with that with YB-3 antibody (Table 2Go).


Figure 2
View larger version (202K):
[in this window]
[in a new window]

  		FIG. 2. Immunohistochemical staining of the distal colon with YB-3, YB-2, anti-A, and anti-B antibodies. Blood types of the specimens were A (a), B (b), O (c), and AB (d) (original magnification, x100).

 

Figure 3
View larger version (143K):
[in this window]
[in a new window]

  		FIG. 3. Immunohistochemical staining of the distal colon with YB-3 and anti-SLX antibody. (a) YB-3(+)/SLX(–), (b) YB-3(–)/ SLX(+), (c) YB-3(+)/SLX(–) (original magnification x100).

 

View this table:
[in this window]
[in a new window]

  		TABLE 2. Incidence of staining with YB-3, anti-A, anti-B, and anti-SLX antibodies in distal colorectal tissues
 
Clinicopathological Features of Specimens According to Immunostaining with YB-3 Antibody and Survival Rates According to Immunostaining with YB-3, Anti-A, Anti-B, and Anti-SLX Antibodies
There was no statistically significant correlation between YB-3 expression and patients’ age (data not shown), sex, tumor location, lymph node metastasis, liver metastasis, or lymphatic invasion (Table 3Go). However, there were statistically significant correlations between YB-3 expression and depth of tissue invasion, blood vessel invasion, and tumor stages (P < .001, P < .01, P < .02, respectively). The expression of A or B antigen in patients with A and AB, or B and AB blood types, respectively, did not show correlation with clinicopathological features (data not shown). The overall 5-year survival rates were investigated according to the immunostaining with YB-3, anti-A, anti-B, and anti-SLX antibodies (Fig. 4Go). When the survival rates were followed up by using the YB-3 antibody, the difference between patients with negative and very weak staining and with positive staining was significant both in patients with A, B, O, and AB blood types (P = .0023) and with A, B, and AB blood types (P = .0038). However, no such difference between positive and negative staining in blood type A and AB individuals, in blood type B and AB individuals, or in all individuals was found when tissues were stained with anti-A (P = .5079), anti-B (P = .3998), or anti-SLX (P = .0787) antibody, respectively, even though survival rates of patients with positive staining seemed to be lower than those with negative staining.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Relationship between clinicopathological features of specimens and grades of staining with YB-3 antibody
 

Figure 4
View larger version (17K):
[in this window]
[in a new window]

  		FIG. 4. Overall 5-year survival rates according to the immunostaining with YB-3, anti-A, anti-B, and anti-SLX antibodies, respectively. Overall survival is indicated for patients with positive (solid lines) and negative (dashed lines) staining.

 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Biosynthetic pathways of blood group A, B, Lewis, sialylated Lewis, and their related antigens in human colorectal tissues were illustrated mainly on the basis of our previous studies (Fig. 5Go). The presence of aberrant {alpha}1,2fucosyltransferase ({alpha}1,2FT), the highly increased activities of {alpha}1,3/4fucosyltransferase ({alpha}1,3/ 4FT), and {alpha}2,3sialyltransferase ({alpha}2,3NeuAcT) in human colorectal cancer has been demonstrated. From substrate specificities of glycosyltransferases involved in the synthesis of these antigens, it is believed that the H active structure could be a precursor for not only blood groups A and B, but also Leb and Y active species. Further, both A and B active structures could be {alpha}1,4/3fucosylated to form A·Leb/Y (GalNAc{alpha}1,3[Fuc{alpha}1,2]Galß1,3/4[Fuc{alpha}l,4/3]GlcNAcß) and B·Leb /Y (Gal{alpha}1,3 [Fuc{alpha}1,2]Galßl,3/4[Fuc{alpha}l,4/ 3]GlcNAcß) active species, but neither Leb nor Y could be converted to A·Leb/Y or B·Leb/Y.6


Figure 5
View larger version (27K):
[in this window]
[in a new window]

  		FIG. 5. Biosynthesis of ABH, Lewis, and sialyl-Lewis and their related antigens in colorectal tissues. FT, fucosyltransferase; Gal(NAc)T, galactosyltransferase (N-acetylgalactosaminyl-transferase); NeuAcT, sialyltransferase. References are as follows: (a) 5, 11; (b) 4, 24–26; (c) this study; (d) 8.

 
In our previous studies on immunohistochemical staining of human colorectal tumor tissues, the accumulation of fucosylated antigens including the Y, Leb, and H type 2 was detected with the YB-2 antibody.7,9,10 All these fucosylated antigens possess Fuc{alpha}1,2Galß linkages probably synthesized with an aberrant {alpha}1,2fucosyltransferase, which was demonstrated to be expressed only in tumors of colorectal tissues.10,11 The strongly increased levels of the mRNA transcript for the {alpha}1,2fucosyltransferase were also observed in human colorectal tumors and cell lines.12 Recently, the expression of Leb antigen in distal colorectal cancer was demonstrated to be caused by upregulation of the FUT-2–encoded {alpha}1,2fucosyltransferase and partly by the FUT-1–encoded {alpha}1,2fucosyltransferase.26 Further, the expression of the H antigens in normal colon and in colon cancer tissues was thought to be dependent on the Se gene (FUT2)-encoded {alpha}1,2fucosyltransferase.27 A newly established mouse monoclonal antibody, YB-3, was characterized to react with the so-called H disaccharide structure (Fuc{alpha}1,2Galß), which is present in H type 1, H type 2, H type 3, Leb, and Y antigens. From immunohistochemical staining of distal colorectal tissues, it was shown that the YB-3 antibody could react with high sensitivity (74.3%) and specificity (100%) with distal colorectal tumors from all patients. The incidence of positive staining was also very high (71.2%) in patients from A, B, and AB blood types. Therefore, as demonstrated in our previous studies,7,9–11 tumor-specific occurrence and accumulation of {alpha}1,2fucosylated antigens were shown in this study by using completely different specimens from our previous ones7,9–11 and could readily be detected by the YB-3 antibody irrespective of their backbone structures. Molecular mechanisms involving in such tumor-specific occurrence and accumulation of {alpha}1,2fucosylated antigens in colorectal tissues need further investigation.

Glycosyltransferases responsible for the synthesis of all the aforementioned fucosylated antigens have been demonstrated in colorectal tumors in our previous studies.7,9–11 In this study, blood group A–specific {alpha}1,3N-acetylgalactosaminyltransfease (A enzyme) and blood group B–specific {alpha}1,3galactosyltransferase (B enzyme) activities could be clearly detected in normal tissues from blood types A and B, respectively, and then tumor-specific {alpha}1,2fucosylation and simultaneous action of the A or B enzyme on these {alpha}1,2fucoylated antigens must occur even though they were inert without the expression of their precursor, the H disaccharide. Indeed, immunohistochemical staining of colorectal tissues with anti-A and anti-B antibodies showed not only that these antigens were expressed in tumor tissues with low incidences of positive staining (61.9% and 46.2%, respectively), but also that their expressions were restricted within positive staining with the YB-3 antibody.

There was a statistically significant correlation between immunoreactivity with the YB-3 antibody and clinicopathological features. It was of particular interest that an overall 5-year survival rate was correlated with the immunoreactivity with YB-3 antibody. However, there was no correlation between immunoreactivity with anti-A, anti-B, or anti-SLX antibody and clinicopathological features or survival rates. Formation of A/B and A/B·Leb/Y antigens via the H disaccharide and A/B antigens, respectively, in distal colorectal tumors seemed not to contribute to the prediction of patients’ survival time. Alteration, deletion, and incompatible expression of ABH blood group antigens on the cell surface have been reported in several human cancers with enhanced malignancy.1,2,28 Deletion or reduction of A/B antigen in tumors of blood type A or B individuals is known to be correlated with the degree of malignancy and metastatic tendencies in colorectal,18 lung,20,29 and oral30 carcinomas. Therefore, it has been suggested that the expression of A/B antigen in tumor cells is a positive prognostic factor, but on the contrary, deletion of A/B antigen is a negative prognostic factor for survival. Contrary results have also been reported that the expression of A, ABH, or H type 1 antigen in carcinoma tissues correlates with a tendency to malignancy.31–34

Increased expression of sialyl-Lex and sialyl-Lea antigens has also been reported in colorectal tumors, but the present study indicates that immunohistochemical staining of distal colorectal tumors with anti-SLX antibody (KM93), which could define sialyl Lewis X antigen associated with mucin-type glycoproteins,35,36 showed relatively low sensitivity and specificity and no statistically significant correlation with survival rate. Taken together, it is plausible that the expressions of H disaccharide and Leb/Y antigens without further glycosylation to proceed to A/B or A/ B·Leb/Y antigens is the most powerful negative predictor of survival in distal colorectal cancer among ABH, Lewis, sialyl-Lewis, and their related antigens synthesized from type 1/2 precursor chains (Fig. 5Go). Therefore, we conclude that immunohistochemical studies with YB-3 antibody are potentially useful in predicting the prognosis of distal colorectal cancer and an independent factor of clinicopathological value in estimating malignant potential. It is of particular interest that both SLX and {alpha}1,2fucosylated antigens, which were detected by anti-SLX antibody and YB-3, respectively, were found to be expressed differently in some tumors.

Levels of the A and B enzyme activities detected in the distal colorectal tissues were observed to be extremely high when compared with those of {alpha}1,2fucosyltransferase detected in tumor tissues. Two cases of immunostaining with anti-A and anti-B antibodies in Table 1Go (patients 3 and 6) showed differently than other cases. No clear, positive staining with anti-A antibody was seen in patient 3 (blood type A), even though activities of the A enzyme and {alpha}1,2fucosyltransferase were detected in the normal and tumor tissues. It was suggested that in patient 3, {alpha}1,2fucosylations of type 1 (Galßl,3GlcNAcß) and type 2 (Galßl,4GlcNAcß) structures must proceed faster than the conversion into A or B active structure once {alpha}1,2fucosylation occurred, resulting in the formation of Leb and/or Y antigens, which was a typical event in colorectal tumors.7 In fact, tumor tissue of patient 3 was strongly stained with YB-2, which could detect Leb and Y antigens as well as H type 2 antigen (data not shown). The tumor tissue of patient 6 (blood group AB) was stained strongly with anti-A antibody but hardly stained with anti-B antibody. Activity of the A enzyme more than 70 times higher was detected in tissues when compared with the B enzyme activity when the same acceptor substrate was used. {alpha}1,3N-Acetylgalactosaminylation of H active structures might therefore occur faster than {alpha}1,3galactosylation in this tissues.

An increased level of {alpha}1,2fucosyltransferase that may result in an increased expression of H type 2 antigen in the rat colorectal carcinoma cells is known to be involved in the increased tumorigenicity and resistance to apoptosis and treatment with anticancer drugs.37–41 Accumulation of {alpha}1,2fucosylated antigens in human colorectal carcinoma cells has also been observed to play a major role in resistance to anti-cancer treatments.11 The priming effect on the suppression of {alpha}1,2fucosylation on the various cancer cells by using a series of primers was suggested to be reduced, depending on the presence of additional sugars to the terminal {alpha}1,2fucosyl residues (our unpublished results). Recently, it has been reported that transgene expression of {alpha}1,2fucosyltransferase in tumor cells inhibits expression of sialyl Lex antigen, which is one of the most important ligands for selectins resulting in the prevention of the hematogenous metastatic spread of cancer.36,42 Aberrant glycosylation at the terminal of each chain in glycoconjugates on the cell should affect not only the phenotypic determination of the cells, but also cell recognition, cell-cell interaction, and responses against anticancer treatments. Because the suppression of the {alpha}1,2fucosylated antigen expression on colorectal cancer cells induced clear susceptibility to such treatments, it is important to investigate whether additional sugars are carrying on the tumor-associated {alpha}1,2fucosylated antigens in colorectal tumors.

Recently, Gao et al.16 examined 30 oral squamous cell carcinomas for the expression of the A and B antigens and A/B glycosyltransferase, together with loss of heterozygosity at the ABO locus and hypermethylation of the ABO gene promoters. Loss of A or B antigen expression was found in 21 (84%) of 25 patients with tumors, whereas the expression of the glycosyltransferase was absent in all of these tumors, showing negative expression of A or B antigen. Loss of the A or B allele was found in 3 (15%) of 20 tumors heterozygous for the ABO locus and hypermethylation of the promoter region in 10 (33.3%) of 30 tumors. In addition, the same conclusion has been reported by analysis of 81 patients with bladder cancer,43 in which the expression of the A antigen was lost in 40 samples; the loss of the A antigen was caused mostly by allelic loss and/or hypermethylation of the ABO gene proximal promoter. However, we report here that A and/or B antigen was expressed in the colorectal tumors of patients 1, 2, 4, 5, and 6, whereas these antigens were not detected in the normal tissues. Furthermore, our preliminary results from bisulfite genomic sequencing showed that no distinct difference in the methylation pattern of the CpG island of the promoter region of the ABO glycosyltransferase was found in both normal and tumor tissues of five cases randomly selected from present samples, and that hypomethylation seemed to occur in that region (data not shown). Thus, loss of blood group antigen expression was not observed, which coincided with the evidence indicating the occurrence of {alpha}1,2fucosylated, H-like antigens in colorectal tumors with high frequencies and the absence of hypermethylation of the promoter region in the A and B alleles.

In conclusion, the expression of blood group A and B antigens was found to be induced in distal colorectal tumor tissues of patients with blood types A and B, which was specifically associated with occurrence of {alpha}1,2fucosylation. The immunodetection of {alpha}1,2fucosylated antigens with YB-3 showed superiority over that with anti-A, anti-B, or anti-SLX antibody as a predictor of postoperative prognosis in patients with distal colorectal cancer.


   ACKNOWLEDGMENTS
 
This study was supported in part by a Grant-in-Aid (No. 40212469) from the Ministry of Education, Culture, Sports, Science and Technology, Japan.

Received for publication November 14, 2006. Accepted for publication January 9, 2007.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Hakomori S. Antigen structure and genetic basis of histoblood groups A, B and O: their changes associated with human cancer Biochim Biophys Acta 1999; 1473:247–66.[Medline]
  2. Le Pendu J, Marionneau S, Cailleau-Thomas A, et al. ABH and Lewis histoblood group antigens in cancer. APMIS 2001; 109:9–31.[Medline]
  3. Varki A. (1999) Glycosylation changes in cancers. In: Varki A, Cummings R, Esko J, Freeze H, Hart G, Marth J (edsEssentials of Glycobiology, Cold Spring Harbor Press, Cold Spring Harbor NY, pp 537–49.
  4. Yazawa S, Nishihara S, Iwasaki H, et al. Genetic and enzymatic evidence for Lewis enzyme expression in Lewis-negative cancer patients. Cancer Res 1995; 55:1473–8.[Abstract/Free Full Text]
  5. Nakamura J, Mogi A, Asao T, Nagamachi Y, Yazawa S. Evidence that the aberrant {alpha}l -> 2fucosyltransferase found in colorectal carcinoma may be encoded by Fuc-TIII (Le) gene. Anticancer Res 1997; 17:4563–9.[Medline]
  6. Ørntoft TF, Greenwell P, Clausen H, Watkins WM. Regulation of the oncodevelopmental expression of type 1 chain ABH and Lewis(b) blood group antigens in human colon by alpha-2-L-fucosylation. Gut 1991; 32:287–93.[Abstract/Free Full Text]
  7. Yazawa S, Akamatsu S, Tachikawa T, et al. Development and characterization of a novel anti-fucosylated antigen monoclonal antibody YB-2 and its usefulness in the immunohistochemical diagnosis of colorectal cancer. Jpn J Cancer Res 1993; 84:641–8.[CrossRef]
  8. Akamatsu S, Yazawa S, Tachikawa T, et al. {alpha}2 -> 3sialyltransferase associated with the synthesis of CA 19-9 in colorectal tumors. Cancer 1996; 77:1694–700.[Medline]
  9. Naitoh H, Nakajima T, Nagamachi Y, Yazawa S. A clinicopathological evaluation of anti-fucosylated antigens antibody (YB-2) in colorectal carcinoma. Glycosylation Dis 1994; 1:31–6.[CrossRef]
  10. Yazawa S, Nakamura J, Asao T, et al. Aberrant {alpha}1 -> 2 fucosyltransferases found in human colorectal carcinoma involved in the accumulation of Leb and Y antigens in colorectal tumors. Jpn J Cancer Res 1993; 84:989–95.[CrossRef]
  11. Yazawa S, Nishimura T, Ide M, et al. Tumor-related expression of {alpha} 1,2fucosylated antigens on colorectal carcinoma cells and its suppression by cell-mediated priming using sugar acceptors for {alpha}1,2fucosyltransferase. Glycobiology 2002; 12:545–53.[Abstract/Free Full Text]
  12. Sun J, Thurin J, Cooper HS, et al. Elevated expression of H type GDP-L-fucose: ß-D-galactoside {alpha}-2-L-fucosyltransferase is associated with human colon adenocarcinoma progression. Proc Natl Acad Sci U S A 1995; 92:5724–8.[Abstract/Free Full Text]
  13. Orlow I, Lacombe L, Pellicer I, et al. Genotype and phenotypic characterization of the histoblood groups ABO(H) in primary bladder tumors. Int J Cancer 1998; 75:819–24.[CrossRef][Medline]
  14. Iwamoto S, Withers DA, Handa K, Hakomori S. Deletion of A-antigen in a human cancer cell line is associated with reduced promoter activity of CBF/NF-Y binding region, and possibly with enhanced DNA methylation of A transferase promoter. Glycoconjugate J 1999; 16:659–66.[CrossRef][Medline]
  15. Kominato Y, Hata Y, Takizawa H, Tsuchiya T, Tsukada J, Yamamoto F. Expression of human histoblood group ABO genes is dependent upon DNA methylation of the promoter region. J Biol Chem 1999; 274:37240–50.[Abstract/Free Full Text]
  16. Gao S, Worm J, Guldberd P, et al. Genetic and epigenetic alteration of the blood group ABO gene in oral squamous cell carcinoma. Int J Cancer 2004; 109:230–7.[CrossRef][Medline]
  17. Ichikawa D, Handa K, Withers DA, Hakomori S. Histo-blood group A/B versus H status of human carcinoma cells as correlated with haptotactic cell motility: approach with A and B gene transfection. Cancer Res 1997; 57:3092–6.[Abstract/Free Full Text]
  18. Ichikawa D, Handa K, Hakomori S. Histo-blood group A/B antigen deletion/ reduction vs. continuous expression in human tumor cells as correlated with their malignancy. Int J Cancer 1998; 76:284–9.[CrossRef][Medline]
  19. Gao S, Bennett EP, Reibel J, et al. Histo-blood group ABO antigen in oral potentially malignant lesions and squamous cell carcinoma-genotypic and phenotypic characterization. APMIS 2004; 112:11–20.[CrossRef][Medline]
  20. Graziano SL, Tatum AH, Gonchoroff NJ, Newman NB, Kohman LJ. Blood group antigen A and flow cytometric analysis in resected early-stage non–small cell lung cancer. Clin Cancer Res 1997; 3:87–93.[Abstract]
  21. Yazawa S, Takeya A, Hosomi O, et al. Use of synthetic H disaccharides as acceptors for detecting activities of UDP-GalNAc:Fuc {alpha}l -> 2Gal ß-R {alpha}1 -> 3-N-acetylgalactosaminyl-transferase in plasma samples from blood group A subgroups. Clin Chem 1992; 38:2392–5.[Abstract/Free Full Text]
  22. Sobin LH, Wittekind C, eds. TNM: Classification of Malignant Tumours. 6th ed. New York: Wiley-Liss, 2002.
  23. Yazawa S, Madiyalakan R, Izawa H, et al. Cancer-associated elevation of {alpha}(l -> 3)-L-fucosyltransferase activity in human serum. Cancer 1988; 62:516–520.[CrossRef][Medline]
  24. Yazawa S, Asao T, Izawa H, et al. The presence of CA19-9 in serum and saliva from Lewis blood-group negative cancer patients. Jpn J Cancer Res 1988; 79:538–43.[CrossRef][Medline]
  25. Akamatsu S, Yazawa S, Zenita K, et al. Elevation of an {alpha}(l,3)fucosyltransferase activity correlated with apoptosis in human colon adenocarcinoma cell line, HT-29. Glycoconj J 1996; 13:985–93.[CrossRef][Medline]
  26. Nishihara S, Hiraga T, Ikehara Y, et al. Molecular mechanisms of expression of Lewis b antigen and other type 1 Lewis antigens in human colorectal cancer. Glycobiology 1999; 9:607–16.[Abstract/Free Full Text]
  27. Fujitani N, Liu Y, Toda S, et al. Expression of H type 1 antigen of ABO histo-blood group in normal colon and aberrant expressions of H type 2 and H type 3/4 antigens in colon cancer. Glycoconjugate J 2000; 17:331–8.[CrossRef][Medline]
  28. Hakomori S. Tumor malignancy of defined by aberrant glycosylation and sphingo-(glyco)lipid metabolism. Cancer Res 1996; 56:5309–18.[Abstract/Free Full Text]
  29. Lee JS, Ro JY, Sahin AA, et al. Expression of blood-group antigen-A—a favorable prognostic factor in non–small-cell lung cancer. N Engl J Med 1991; 324:1084–90.[Abstract]
  30. Dabelsteen E. ABO blood group antigens in oral mucosa. What is new? J Oral Pathol Med 2002; 31:65–70.[Medline]
  31. Marionneau S, Le Moullac-Vaidve B, Le Pendu J. Expression of histo-blood group A antigen increases resistance to apoptosis and facilitates escape from immune control of rat colon carcinoma cells. Glycobiology 2002; 12:851–6.[Abstract/Free Full Text]
  32. Nakagoe T, Nanashima A, Sawai T, et al. Expression of blood group antigens A, B and H in carcinoma tissue correlates with a poor prognosis for colorectal cancer patients. J Cancer Res Clin Oncol 2000; 126:375–82.[CrossRef][Medline]
  33. Nakagoe T, Fukushima K, Nanashima A, et al. Comparison of the expression of ABH/Lewis-related antigens in polypoid and non-polypoid growth types of colorectal carcinoma. J Gastroenterol Hepatol 2001; 16:176–83.[CrossRef][Medline]
  34. Sakuma Y, Kubishiro K, Tsukazaki K, Nozawa S. Involvement of H type 1 carbohydrate antigen in cell adhesion to vascular endothelial cells of human endometrial cancer. Anti-cancer Res 2003; 23:1277–82.[Medline]
  35. Nakamura M, Kudo T, Narimatsu H, et al. Single glycosyltransferase, core 2 ß1 -> 6-N-actylglucosaminyltransferase, regulates cell surface sialyl-Lex expression level in human pre-B lymphocytic leukemia cell line KM3 treated with phorbolester. J Biol Chem 1998; 273:26779–89.[Abstract/Free Full Text]
  36. Mathieu S, Prorol M, Benoliel A, et al. Transgene expression of {alpha}(l,2)-fucosyltransfease-I (FUT1) in tumor cells selectively inhibits sialyl-Lewis x expression and binding to E-selectin without affecting synthesis of sialyl-Lewis or binding to P-selecin. Am J Pathol 2004; 164:371–83.[Abstract/Free Full Text]
  37. Labarrie` re N, Pian JP, Otry C, et al. H blood group antigen carried by CD44V modulates tumorigenicity of rat colon carcinoma cells. Cancer Res 1994; 54:6275–81.[Abstract/Free Full Text]
  38. Goupille C, Hallouin F, Meflah K, Le Pendu J. Increase of rat colon carcinoma cells tumorigenicity by {alpha}(l-2)fucosyltransferase gene transfection. Glycobiology 1997; 7:221–9.[Abstract/Free Full Text]
  39. Hallouin F, Goupille C, Bureau V, Meflah K, Le Pendu J. Increased tumorigenicity of rat colon carcinoma cells after 2-fucosyltranferase FTA anti-sense cDNA transfection. Int J Cancer 1999; 80:601–11.
  40. Cordel S, Goupille C, Hallouin F, et al. Role of {alpha}1,2fucosyltransferase and histo-blood group H antigen type 2 in resistance of rat colon carcinoma cells to 5-fluorouracil. Int J Cancer 2000; 85:142–8.[CrossRef][Medline]
  41. Goupille C, Marionneau S, Bureau V, et al. {alpha}1,2Fucosyltransferase increases resistance to apoptosis of rat colon carcinoma cells. Glycobiology 2000; 10:375–82.[Abstract/Free Full Text]
  42. Zerfaoui M, Fukuda M, Sbarra V, Lombardo D, El-Battari A. {alpha}(l,2)-Fucosylation prevents sialyl Lewis x expression and E-selectin–mediated adhesion of fucosyltransferase VII-transfected cells. Eur J Biochem 2000; 267:53–60.[Medline]
  43. Chihara Y, Sugano K, Kobayashi A, et al. Loss of blood group A antigen expression in bladder cancer caused by allelic loss and/or methylation of the ABO gene. Lab Invest 2005; 85:895–907.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tsuboi, K.
Right arrow Articles by Yazawa, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsuboi, K.
Right arrow Articles by Yazawa, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS