ICANCERRESEARCH55.3654-3658.Augustis. iws]

Fucosyltransferase and «-L-FucosidaseActivities and Fucose Levels in Normal and

Malignant Endometrial TissueJen-Wei Wang, Robert A. Ambros, Peter B. Weber, and Thomas G. Rosano1

Departments of Pathology and Laboratory Medicine /J-W. W.. K. A. A., T. G. R./ und Biochemistry ¡P.B. WJ. Albany Medical Center, Albany, New York 12208-3478

ABSTRACT

Previous ¡miniino- and lectin-histochemical studies using mAbs andUlex europaeus lectin I, which recognize various fucose-containing blood

group antigens, have shown an increased expression of Lewis and H bloodgroup antigens in endometrial carcinoma. We investigated the biochemical basis of aberrant fucose-containing antigen expression by comparingthe activity of fucosyltransferases (FTase) and a-L-fucosidase in tissuebiopsies from normal (n = 18) and malignant (n = 20) endometrium.

Alteration of FTase activity in tumor tissue homogenates was evaluated byusing a panel of FTase substrates including /V-acetyllactosamine (type 2),lacto-N-biose I (type 1), and phenyl-ß-D-galactoside. Based on histológica!subtyping, the endometrioid group (n = 14) showed a significant

(P < 0.05) increase in tumor FTase activity with all three substrates, whileno significant increase was detected for the papillary serous group (n = 4).

Matched pair analysis of normal and tumor tissue from a subgroup(n = 5) of the patients with increased tumor enzyme activity also showedhigher FTase activity (/" < 0.05) in the tumor tissue when the type 1

substrate was used. Regression analysis showed a correlation between theFTase activities acting on type 2 or type 1 substrates (r = 0.821 andr = 0.722, respectively) and the endogenous fucose levels in tumor homogenates. Spectrophotometric analysis of o>l.-fucosidase activity using p-ni-trophenyl-a-L-fucoside revealed a higher activity in tumor homogenatesthan in normal homogenates (/' < 0.05) and, therefore, could not account

for the enhanced expression of fucose-containing antigens. The currentstudy suggests that aberrant expression of fucose-containing antigens,such as the H and the Lewis blood-group antigens, in endometrial carcinoma is consequential to the change in FTase rather than in a-L-fucosidase activity. In addition, the investigation suggests that different glyco-

sylation mechanisms are operative in different subtypes of endometrial

INTRODUCTION

Alteration in cellular glycosylation patterns is a feature routinelyshared by diverse types of neoplastia diseases (1-6). Such changes

may occur in the form of: («)incomplete synthesis of carbohydratestructures; (b) reexpression of oncofetal antigens that normally existduring the embryonic development but either disappear or becomeminimally expressed in the corresponding adult tissues; or (c) organizational changes, as shown by immuno- or lectin-histochemical

staining, due to demasking or increased expression of normallyexpressed carbohydrate structures (7).

Previous immuno- and lectin-histochemical comparison of normaland malignant uterine endometrium have demonstrated that fucose-containing blood group antigens (H and Le"/h/y) consistently showed

higher and differential expression (cytoplasmic versus apical or luminal surface) in malignant tumor cells (8-12). No correlations havebeen found, however, between the aberrant expression of fucose-

containing antigens and the age, ABO status, or the menstruationcycle of the patients (8-10). Recently, a positive correlation between

the Ulex europaeus lectin I staining and myometrial and vascular

Received 3/13/95; accepted 6/16/95.The costs of publication of this article were defrayed in part by the payment of page

charges. This article must therefore be hereby marked advertisement in accordance with18 U.S.C. Section 1734 solely to indicate this fact.

1To whom requests for reprints should be addressed, at Department of Pathology and

Laboratory Medicine A-22, Albany Medical Center, 47 New Scotland Avenue, Albany,NY 122U8-3478.

invasion has been reported in endometrial carcinoma (12). Collectively, these observations of malignant endometrial tissue provideindirect evidence for an abnormality in the activity of either theglycosyltransferases responsible for synthesis of the fucose-contain

ing antigens or a fucosidase that catalyzes the hydrolytic cleavage ofthe terminal fucose from the carbohydrate chain.

The fucose-containing blood group antigens (H and Lea/b/x/y)

are synthesized from two carbohydrate precursors: type 1(Galßl^3GlcNAc-R) and type 2 (Galßl^4GlcNAc-R) precursors(13—16).Distinctive FTases2 are involved in substituting the two

precursors with fucose in either al-* 2 linkage to terminal Gal and/orin a 1—>3/4linkage to subterminal GlcNAc and are, therefore, designated as a-2-, a-3-, and a-4-L-FTase, respectively (16-19). Cancer-

associated increase of these FTases have been observed frequently ¡nserum or tumor tissues from patients with various types of carcinomas(20-31). In addition, it has been shown that, in patients with breast,

colon, or gastric cancers, the removal of tumors was followed by asignificant reduction in serum activity of FTases, while recurrence ofcancer or metastasis tended to occur in patients who displayed increased activity of the FTases (20, 28, 31). Similar to the observationswith the FTases, abnormal levels of a-L-fucosidase have also been

found in tissue extracts of ovarian, cervical, endometrial, colon, andliver carcinomas (32-35).

The present study has been undertaken to determine the activity ofFTases in normal and endometrial tumor tissues. Three differentfucose-acceptors with the precursor structures of type 1, type 2, and

pGal were used to detect FTases with different acceptor substratepreferences (18, 36, 37). In addition, a-L-fucosidase activity wasmeasured using /j-nitrophenyl-a-L-fucoside. Total fucose concentra

tion in tissue homogenate was assessed by HPAEC. Our data show anincrease in activity of FTases as well as a-L-fucosidase in the endo

metrioid subtype of carcinomas. Furthermore, the enhanced FTaseactivities are associated with an increase in tissue fucose content.

MATERIALS AND METHODS

Reagents. pGal was obtained from Fluka Chemical Corp. (Ronkonkoma,NY). Two shipments of GDP-[3H]fucose (6700 and 8100 Ci/mol) were pur

chased from New England Nuclear (Wilmington, DE). All other chemicalswere purchased from Sigma Chemical Co. (St. Louis, MO) unless statedotherwise.

Patient Material. Eighteen normal and 20 tumor biopsies of endometriumwere obtained from the Albany Medical Center Tumor Bank based upon arandom sampling of endometrial tissue collected from surgical cases at theAlbany Medical Center. Demographic and clinical information regarding theage of the patients, stage of the disease, as well as tumor grade and thehistological type were obtained after biochemical analysis was completed(Table 1). The endometrioid subtype predominated in the sample populationdue to the relatively higher frequency of occurrence for this histologicalsubtype of tumors ¡nendometrial carcinoma.

Tissue Processing. Hematoxylin and eosin-stained sections from OCT-

embedded (Baxter Diagnostics Inc., McGaw Park, IL) tissue samples were

2 The abbreviations used are: FTase, fucosyltransfcrase; Fuc, fucose; Gal, galactose;GlcNAc, /V-acetylglucosamine; Le", Leb, Le", Ley, Lewis a, b, x, and y antigen, respectively; type 1 precursor, Gal/31->3GlcNAc; type 2 precursor, Gal/31 >4GlcNAc; pGal,phenyl-ß-n-galactoside; HPAEC, high-performance aniónexchange chromatography.

3654

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

1-U(()SYI.[ RANNI l.KASf-: IN 1-NDOMLTRIAL TISSUE

Table 1 Clinical dala oj patienis willi/withtmt emfainefriitl cancer

NormalPatient

no.Nl"N2°N3-N4"ter6789101112131415161718Mean

age.Age52774689595751314745625775866542388659±17Patienl

no.TIT2T3T4T5192021222324252627282930313233MeanAge527746895952676280HS73658774625577476984age,

68 ±TumorSlageIBIBme1CIBHi1CIB1CIBmeIBmeÜBIBmIBIBIB1C13Grade222211222NA1'NA'3NA''222323NA'Tumor

typeEndometrioidEndometrioidEndomelrioidEndomelrioidMMMT''EndometrioidEndometrioidEndometrioidEndometrioidPap-serousPap-serousPap-serousPap-serousEndomelrioidEndometrioidEndometrioidEndometrioidEndomctrioidEndometrioidMMMT

" Normal and tumor matched pairs from five patients were available for the eurrent

investigation. The specimens are designated as Nl to N5 for normal tissue and Tl to T5for corresponding tumor tissue.

'' MMMT. malignant mixed mesoderma! tumor: pap-serous, papillary serous tumor.' NA. not available.

prepared and reviewed microscopically to confirm the normal or neoplasticcharacteristics of the tissue used for biochemical analysis. OCT compound wasremoved by placing the tissue in ice-cold, double-distilled water before the

homogenization procedure. All subsequent steps were performed on ice.Each specimen was finely minced with surgical scissors, and the tissue was

suspended in 50 mM HEPES buffer (pH 7.U) at a concentration of 1(K)mg oftissue/3 ml of buffer. The tissue was homogenized with a two-speed BiospecHomogenizer (Biospec Products, Inc., Bartlesville, OK) for two and four 10-sbursts at low and high speed, respectively. Triton X-KM) was added to the

homogenate at a final concentration of 0.1% (v/v). The homogenate wascentrifuged for 30 s in a Sero-fuge bench-top centrifuge (Clay-Adams, NewYork. NY), and the supernatant was aliquoted and stored at —70°Cin the

freezer until analyzed. The bicinchoninic acid assay (Pierce Chemical,Rockford, IL) was used to determine protein concentrations.

Fucosyltransferase Assays. A modification of a previously described assay was used for the detection of FTases in the homogenates (38). The FTasereaction mixture (80 /A!) consisted of 50 mM HEPES (pH 7.0). 10 min ATP,12.5 mM MnCI2, 12.5 mM NaN3, 6.25 mM acceptor substrate (type 1,Galßl »3GlcNAc;type 2, Galßl *4GlcNAc; pGal), 1.8 JXMGDP-[3H]fucose,and 20 fd of enzyme extract. Prior to the experiment, GDP-['H]fucose wasdiluted 6-fold with unlabcled GDP-ß-L-1'ucoseof equimolar concentration. In

addition, prior to analysis, the homogenates were diluted 2- or 4-fold for

normal and tumor tissues, respectively, with the homogenization buffer [50mM HEPES (pH 7.0)-0.1% Triton X-100]. After equilibration at 37°C,thereaction was initiated by the addition of GDP-|'H]fucose. For validation ofzero-order kinetics throughout the 37°Cincubation period, the reaction product

was monitored at 0, 30, and 60 min. At the end of the appropriate incubationtime, the reaction was stopped by placing the tubes in an ice bath. Controlassays for each sample were run in the absence of acceptor substrate.

Separation of the radiolabeled products from unused GDP-['Hjfucose was

achieved by using anión exchange column chromatography. Bond Elute columns (1.5ml capacity; Analytichem International, Harbor City, CA) containeda 20 fitn pore frit and were packed with a 0.5-ml bed volume of aniónexchange resin (Dowex 1 X 2-400, formate form). The total reaction mixturevolume was applied to the column, and the flow-through fraction was collected

directly into a scintillation vial. The column was washed four times with 500fil of distilled water. Fifteen milliliters of Scintisol (Krackeler Scientific, Inc.,Albany, NY) were then added to each scintillation vial to obtain a clearsolution. The radioactivity was counted for 10 min using a Beckman LS 5000TD counter (Beckman Instrument, Inc., Fullcrton, CA). The specific activity offucosyltransferase was expressed as the microunits/mg protein. A microunit of

enzyme activity is the amount of enzyme that will catalyze the conversion ofone pmol of substrate to product/min.

or-i.-Fucosidase Assay. The fucosidase activity was determined according

to the method of Wicderschain et cil. (39). In short, the final reaction mixture

(250 fil) consisted of 0.05 N acetic buffer (205 ¿¿I),pH 5.5, sample homogenate(20 /J.1),and 10 mM />-nitrophenyl-a-i.-fucoside (25 fj.1).At the end of 0, 30, or60 min incubation at 37°C,2.5% ZnSO4 (0.1 ml) and 0.15 N NaOH (0.1 ml)

were added to terminate the reaction. The samples were centrifuged for 3 min.and a 300-fxl aliquot of each supernatant was transferred into a cuvettecontaining 300 fil of 0.4 Mglycine-NaOH buffer (pH 10.5). The absorbance ofliberated /)-nitrophenol was measured at 410 nm (Beckman DU-70 Spectro-

photomcter; Beckman Instruments, Inc., Fullerton, CA). One unit of enzymeactivity is defined as the amount of enzyme that will convert 1 fimolp-nitrophenol/min. The specific activity of a-i,-fucosidasc is expressed in

units/mg protein.Total Fucose Content of Tissue Homogenates. Concentration of fucosc

in the tissue was determined after resin-hydrolysis of vacuum-dried tissuehomogenate (500 fig protein) with 2 ml of a 25% v/v Dowex 50 X 2-400 [H +]suspension in 0.03 M trifluoroacetic acid for 5 h at 100°Cin 8-ml screw cap

tubes. The reaction mixtures were applied to anión exchanger columns containing 0.5 ml of Dowex 1 X 2 (200-400 meshes), acetate form, and the

columns were washed three times with 0.5 ml 50% mcthanol containing 1%octanol as an antifoaming agent. The filtrates were evaporated (Evapo-Mix;

Buchler Instruments, Fort Lee, NJ) to dryness and were reconstituted with 250fil of distilled water. The samples were then transferred to 500 fd Eppendorfmicrofuge tubes and centrifuged for 5 min. Aliquots of the aqueous solutionwere either analyzed immediately by HPAEC or stored at -20°C in the freezer

with recentrifugation prior to the analysis.For analysis of fucose, 20 fil of sample were injected onto a C'arhpak

column (Dionex Corp., Sunnyvale, CA) using a HPAEC sugar analyzer (40).Elution was performed with 10 mM aqueous NaOH at a flow rate of 40 ml/h.A pulsed amperometric detector with dual gold electrodes (EG&G PrincetonApplied Research, Princeton. NJ) was used lor detection of sugar in the eluant.The optimized settings for the electrochemical analysis were: 2 fiA full scaleresponse, El: 50 mV, Tl: 633 ms (38 cycle); E2: 650 mV, T2: 250 ms (15cycle); E3: -800 mV, T3: 133 ms (8 cycle); and total, 1016 ms. The reference

electrode was filled with 3 M NaCI in 0.4 N NaOH saturated with AgCl. Thepeak identified as fucose typically eluted from the column between 6 and 8min. The analysis for each sample was performed at least twice. The replicateconcentrations determined for each tissue homogenate were averaged and usedto determine the fucose concentration based on calibration runs with a mixtureof pure sugars including fucose.

The precision of the resin-hydrolysis protocol was validated by treating aknown concentration of orosomucoid (human a,-acid glycoprotein) in the

same way as the sample homogenates. The total fucose recovered under thecurrent assay condition was within 90% of the published fucose contents forthis glycoprotein (data not shown; Ref. 41).

RESULTS

Fucosyltransferase Activities in Normal and Malignant UterineEndometrium. Biochemical assessment of FTase activity in normal(ii = 18) and tumor (n = 20) tissue homogenates revealed a signif

icant increase in tumor tissue activity with all three acceptor substrates(Fig. 1). Statistical analysis of tissue activity was also performedbased on the histological subtype of the tumors. The results showed asignificant increase in FTase activity with all three substrates for theendometrioid type (n = 14) of carcinoma (Fig. 1). Furthermore, with

type 1 acceptor substrate, the mean FTase activity in endometrioidcarcinoma homogenates was 11-fold higher than in the normal tissue

homogenates. In contrast, no significant difference in enzyme activitywas observed between the normal homogenates and papillary serouscarcinoma homogenates independent of the substrate used (Fig. 1).

To determine whether the increase in FTase activities in endome-

trial tissues is a general phenomenon in both normal and tumor tissuesfrom patients with endometrial cancer, normal and tumor matchedpairs of tissue were obtained from different sites of surgically resected

3655

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

FUCOSYLTRANSFERASE IN ENDOMtTRIAL TISSUE

1?23.00-1If

Z-50'Sa

3z-00-'"3

1.5O

g»1.00-Ha

0.50-10.00-E3

Normalf^Tnmnrf1pJÌ

^1?:Jl BìrÌiji ISrflTyp* 2 Typ. I pG.1

All Tumor

Typ. Z Typ. l pO.1

Endometrioid

Type 2 Type 1 pG«l

Pap-serous

Typ. 2 Typ.1 pG.1

Matched Pairs

Fig. 1. Statistical analyses of the FTase activities (columns, mean; bars, SEM) innormal (n = 18) and tumor (n ~ 20) tissues from uterine endometrium. Tissues werehomogenized, and the FTase activities ¡nhomogenates were measured with three differentfucose-acceptors (see "Materials and Methods"). Four types of statistical analyses were

performed: Normal versus All Tumor (n = 20); Normal versus Endomelrioitl-lypecarcinoma (n = 14); Normal versus papillary serous (Pap-serous)-lyjK carcinoma(n = 4); and Matched Pairs (n = 5). *, P < 0.05.

endometrium from five patients. Again, mean FTase activity washigher in tumor homogenates with all three substrates, but the difference reached statistical significance only with the type 1 acceptor(Fig. 1).

In addition to the increase in FTase activity, tumor tissue alsodisplayed differential patterns of activity between the three FTasesubstrate assays. As shown in Fig. 2, for example, tumor tissues T2and 21 displayed higher FTase activity with type 2 acceptor ascompared to the type 1 precursor. In sharp contrast, samples T4, 22,

and 27 showed a prominent increase in FTase activity with type 1precursor. In a number of tumor preparations, including T3 and 32,FTase activities increased proportionately in the type 1 and 2 acceptorassays. Therefore, the activity patterns varied significantly amongpatients with the endometrioid carcinoma. In contrast, patients whohad papillary serous carcinoma (patients 23 to 26) showed no significant change in FTase activity nor any differential change in enzymeactivity patterns for the three substrates (Fig. 2).

Total Fucose Measurement and Comparison to FTase Activities. The direct measurement of fucose in tissue from patients withand without endometrial carcinoma showed an increased fucose concentration in homogenates from patients with endometrial cancer (Fig.3). A significant increase in fucose levels in tumor homogenates wasalso observed in matched pair comparisons of normal and endometrialcancer sites within the same patient. In contrast, the mean fucose levelin the papillary serous carcinoma homogenates was comparable to thenormal controls. Linear regression analyses of data for all tumorhomogenates showed a strong correlation between total fucose levelsand FTase activity with either type 2 or type 1 acceptor substrate(r = 0.82 and r = 0.72, respectively). Furthermore, total FTase

activity as determined by summation of activity for the three substrateassays showed the highest correlation with fucose level in tumortissue (r = 0.88).

a-L-Fucosidase Activities in Normal and Malignant Uterine

Endometrium. A comparison of fucosidase activity in normal andtumor tissue showed an activity increase in the tumor group (Fig. 4).Similar to FTase findings, histológica! subtyping showed a significantactivity increase in the endometrioid but not the papillary serous typeof endometrial carcinoma. The increase in fucosidase activity inendometrioid-type tumor homogenates, however, averaged only2-fold higher than control tissue. A weak positive correlation(r = 0.57) was observed between the tissue fucose levels and enzyme

activity for the patients with endometrial carcinoma.

Fig. 2. Assessment of FT activities inendometrium tissues using type 2, type 1,and pGal acceptor. The results for normaland tumor homogenates are shown in upperand lower panels, respectively. The fivenormal-tumor matched pairs are denotedwith prefix Wand 7"to represent normal and

tumor homogenates, respectively. Clinicalstage is listed above the third coluiitn ofeach patient sample. D, type 2 acceptor(Galßl^4GlcNAc); • pGal; ®,type 1acceptor (Galßl^3GlcNAc).

5̂.5-•|

5.0-M

4.0-!„.3

3.0-»1•s

2-5-•S2.0-<:„

1.5-i

^^0.5-_

5.5-•|

5.0-Ê

4.5-o.M

4.0-B$

3.5-g-S

3.0-|

2.5-'•§2.0-„

1.5-è

1-°-E0.5-0.0-Normal1

L FLB n_ nja t-L^. r-w. FU, FU, FU. n_ PL, fLH n_n_n_n_NI

N2 N3 N4 N5 6 7 8 9 10 11 12 13 14 15 16 1718TumorIBnj3ÃœBIB

nic

1H

H(1ncIcil i àíl'lH

Hini i iBíllU.CIBJnlnl .njTI T2 T3 T4 T5 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33

Patient Assigned Number

3656

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

FUCOSYLTRANSFERASE IN ENDOMETRIAL TISSUE

DISCUSSION

A number of studies have found an enhanced expression of Lea/h/y

and H-antigens, which are fucose-containing blood group antigens, inendometrial cancer cells (8-11). It was further suggested that in

creased Lewis antigen expression results from increased fucose transfer onto type 1 precursor substrates (10). The present biochemicalfindings provide direct evidence in support of the hypothesis that theincreases in fucose-containing antigen profiles in endometrial cancer

are, at least in part, related to alteration in FTase activities. Theincreased expression of tumor-associated fucose-containing antigens

in endometrial cancer cells cannot be explained by a decrease ina-L-fucosidase activity since this degradative enzyme also showed

enhanced activity in tumor homogenates. This observation is in agreement with the result of Vesce and Biondi (35), who also found anincreased a-L-fucosidase activity in endometrial cancer tissues. Since

certain glycosidases may act only on natural oligosaccharides but noton synthetic glycosides such as p-nitrophenyl-glycosides, natural oligosaccharides with different a-fucosyl linkages may be required todetermine the possible substrate specificity of the tumor-associateda-L-fucosidase (42).

Although the use of three acceptors in our FTase assay allowed theassessment of changes in a-2-, a-3-, and a-[3/4]-L-FTase (Lewis-type

FTase), only the assay using pGal as acceptor was specific since thisprecursor allows only a single fucose substitution by a-2-L-FTase, the

FTase responsible for synthesizing H antigen (37). Nevertheless,the results with pGal precursor, which showed very low a-2-L-FTase

activity in most of the tumor homogenates, indicate that the majorincrease in FTase activity is in FTase with a-3- or a-[3/4]-FTaseactivity. The predominate increases in a-3- or a-[3/4]-FTase activities

are consistent with enhanced expression of Lewis antigens by endometrial cancer cells as previously observed in histochemical studies(8-11).

The present study also provides evidence indicating that aberrantFTase activity in endometrial cancer cells is a specific, rather than ageneral, process in that differential patterns of enzyme activitiestoward type 1 and type 2 acceptors were observed in a number oftumor homogenates (Fig. 2). These differential patterns of enzymeactivities may be explained by either a potential loss in substratespecificity in transformed tumor cells or by the possible co-expression

l&OO

4U.U-i?

35.0-gO,

JO.O-OC-S"5S

20.11-•

"0315.0-Fucose

Levs

is8 ???Õ7H

Normal*DTumorTJL¡1—

l—ÉJ

Endometrioid Pap-serous Matched Pairs

Fig. 3. Measurement of total fucose (columns, mean; bars. SEM) in normal (n = 18)and tumor (n = 20) homogenates prepared from tissues of endometrium. Tissues werehomogenized, and the total fucose levels were measured by HPAEC (see "Materials andMethods"). Four types of statistical analyses were performed: Normal versus All Tumor

(n = 20); Normal VÕT.YH.YEndometrioid type carcinoma (n = 14); Normal versus Papillaryserous (Pap-serous) type carcinoma (n = 4); and Matched Pairs (n —5). *, P < 0.05.

AllTumor Endometrioid Pap-serous Matched Pain

Fig. 4. Statistical analysis of a-L-fucos idase activity (columns, mean; bars, SEM) innormal (n = 18) and tumor (n —20) homogenates prepared from endometriul tissues.Tissues were homogenized, and the en/yme activity was determined using/>-nitropnenyl-a-L-fucoside as reaction substrate (see "Materials and Methods"). Four types of statistical

analyses were performed: Normal versus AH Tumor (n = 20); Normal versus Endomelrioid type carcinoma (n = 14); Normal versus Papillary serous (Pap-serous) typecarcinoma (n = 4); and Matched Pairs (n = 5). *, P < 0.05.

of multiple FTases within the tumor mass (43, 44). Specific fucosy-

lation mechanisms in endometrial cancer cells are further supportedby the observations that increases in FTase activity was measuredonly in specific histological subtypes of endometrial carcinoma (Fig.2). Higher FTase activities were found in the endometrioid subtype,which is associated with hyperestrinism and arises on a background ofhyperplasia. In contrast, levels of FTase activity in papillary serouscarcinoma, which are not associated with hyperestrinism or endometrial hyperplasia, were comparable to levels in normal homogenates.

In conclusion, the current investigation provides direct evidencethat increased FTase activities are associated with endometrioid carcinoma and thus provide the biochemical basis for the increasedexpression of fucose-containing antigens observed previously byother investigators (8-11). Future molecular and biochemical characterization of the tumor-associated FTases (18, 38) in malignant uterine

endometrium will be necessary to elucidate the potential significanceof enhanced FTase activities and their differential pattern of activity inendometrial cancer genesis and subsequent metastasis.

REFERENCES

1. Hakomori, S. Tumor-associated carbohydrate antigens. Annu. Rev. Immunol., 2:103-126, 1984.

2. Feizi, T. Carbohydrate antigens in human cancer. Cancer Surv., 4: 245-269, 1985.3. Feizi, T. Demonstration by monoclonal antibodies that carbohydrate structures of

glycoproteins and glycolipid are onco-developmental antigens. Nature (Lond.), 314:53-57, 1985.

4. Hakomori, S. Aberrant glycosylation in tumors and tumor-associated carbohydrateantigens. Adv. Cancer Res., 52: 257-331, 1989.

5. Coon, J. S., and Weinstein, R. S. Blood group-related antigens as markers ofmalignant potential and heterogeneity in human carcinomas. Hum. Palhol., 17:1089-1106, 1986.

6. Sell, S. Cancer-associated carbohydrates identified by monoclonal antibodies. Hum.Pathol., 21: 1003-1019, 1990.

7. Hakomori, S. Tumor-associated carbohydrate markers. In: S. Sell (ed.), SerologicalCancer Markers, pp. 207-232. Totowa: The Human Press, 1992.

8. Inoue, M., Sasagawa. T., Saito, J., Shimizu, H., Ueda. G., Tanizawa, O., andNakayama, M. Expression of blood group antigens A, B, H, Lcwis-a, and Lewis-b infetal, normal, and malignant tissues of the uterine endometrium. Cancer (Phila.), 60:2985-2993, 1987.

9. West, K. P., and Cope, J. L. The binding of peroxidase-labelled Icctins to humanendometrium in normal cyclical endometrium and endometrial adcnocarcinoma. J.Clin. Pathol.. 42: 140-147, 1989.

H). Inoue, M. Nakayama, M., and Tanizawa, O. Altered expression of Lewis blood groupand related antigens in fetal, normal adult and malignant tissues of the uterineendometrium. Virchows Arch. A Pathol. Ana!., 416: 221-228, 1990.

3657

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

FUCOSYLTRANSFERASE IN ENDOMETRIAL TISSUE

11. Memon, R. S., Yonezawa, S., Hasui, K., and Sato, E. Expression of blood groupantigens and lectin binding in adenocareinomas of the endometrium. Acta Pathol.Jpn., 40: 509-516, 1990.

12. Ambras, R. A., and Kurman, R. 3. Association of Ulex europaeus agglutinin I bindingwith invasion in endometrial carcinoma. Int. J. Gynecol. Pathol., 12: 301-306, 1993.

13. Watkins, W. M. Biochemistry and genetics of the A, B, O, Lewis, and P blood groupsystems. Adv. Hum. Genet., 10: 1-136, 1980.

14. Watkins, W. M., Greenwell. P., Yates, A. D., and Johnson, P. H. Regulation ofexpression of carbohydrate blood group antigens. Biochimie, 70: 1597-1611, 1988.

15. Oriol, R., Le Pendu, J., and Mollicone, R. Genetics of A, B, O, H, Lewis, X, andrelated antigens. Vox Sang., 51: 161-171, 1986.

16. Oriol, R. Genelic control of the fucosylation of ABH precursor chains. Evidence fornew epistatic interactions in different cells and tissues. J. Immunol., 17: 235-245,

1990.17. lizuka, S., and Yoshida, A. Tissue-dependent expression of two (a, .2)-fucosyltrans-

ferases specified by the H and Se genes. Enzyme, 37: 159-163, 1987.18. Lowe, J. B. Molecular cloning, expression, and uses of mammalian glycosyltrans-

ferases. Semin. Cell Biol., 2: 289-307, 1991.

19. Macher, B. A., Holmes, E. H., Swiedler, S. J., Stults. C. L. M., and Srnka, C. A.Human al-3 fucosyltransferases. Glycobiology, 1: 577—584,1991.

20. Bauer, C., Reutter, W. G., Erhart, K. P., Kotigen, E., and Gerok, W. Decrease ofhuman serum fucosyltransferase as an indicator of successful tumor therapy. Science(Washington DC), 201: 1232-1233, 1978.

21. Thompson, S., Cantwell, B. M. J., Matta, K. L., and Turner, G. A. Parallel changesin the blood levels of abnormally fucosylated haptoglobin and a-l,3-fucosyltrans-

ferase in relationship to tumour burden: more evidence for a disturbance of fucoscmetabolism in cancer. Cancer Lett., 65: 115-121, 1992.

22. Ronquist, G., and Nou, E. Serum sialyltransferase and fucosyltransferase activities inpatients with bronchial carcinoma. Cancer (Phila.), 52: 1679-1683, 1983.

23. Yazawa, S., Madiyalakan, R., Izawa, H., Asao, T., Furukawa, K.. and Malta, K. L.Cancer-associated elevation of a(l —>3)-t.-fucosyltransferase activity in human serum.Cancer (Phila.), 62: 516-520, 1988.

24. Itzkowitz, S. H., Dahiya, R., Byrd, J. C., and Kim, Y. S. Blood group antigensynthesis and degradation in normal and cancerous colonie tissues. Gastroenterology,99: 431-442, 1990.

25. 0rntoft, T. F., Greenwell, P., Clausen, H., and Watkins, W. M. Regulation of theoncodevelopmenlal expression of type 1 chain ABH and Lewis'1 blood group antigens

in human colon by a-2-L-fucosylation. Gut, 32: 287-293, 1991.

26. Kessel, D., Sykes, E., and Henderson, M. Glycosyltransferase levels in tumors metastaticto liver and in uninvolved liver tissue. J. Nail. Cancer Inst., 59; 29-32, 1977.

27. Hutchinson, W. L., Du, M-Q., Johnson, P. J., and Williams, R. Fucosyltransferases:differential plasma and tissue alterations in hepatocellular carcinoma and cirrhosis.Hepatology. 13: 683-688, 1991.

28. Asao, T., Yazawa, S., Nagamachi, Y., Abbas, S. A., and Matta, K. L. Seruma(l^»3)-L-fucosyltransferase, carcinoembryonic antigen, and sialyl Lewis X-i antigenlevels in lung cancer. Cancer (Phila.), M: 2541-2545, 1989.

29. Yazawa, S., Madiyalakan, R., Piver, M. S., and Matta, K. L. Elevated activities of

blood group Le gene dependent a(l >3)-L-fucosyllransferase in human saliva ofLewis negative patients with epithelial ovarian cancer. Cancer Lett., 32: 165-169,1986.

30. Chandrasekaran, E. V., Jain, R. K., and Matta, K. L. Ovarian cancer al,3-L-fucosyl-transferase. J. Biol. Chem., 267: 23806-23814, 1992.

31. Yazawa, S., Asao, T., Nagamachi, Y., Abbas, S. A., and Malta, K. L. Tumor-relatedelevation of serum («1—>3)-L-fucosyltransferaseactivity in gastric cancer. J. CancerRes. Clin. Oncol., 115: 451-455, 1989.

32. Itzkowitz, S. H., Dahiya, R., Byrd, J. C., and Kim, Y. S. Blood group antigensynthesis and degradation in normal and cancerous colonie tissues. Gastroenterology,99: 431-442, 1990.

33. Leray, G., Deugnier, Y., Jouanolle, A. M., Lehry, D., Bretagne, J. F., Campion, J. P.,Brissot, P., and Le Treut, A. Biochemical aspects of a-L-fucosidase in hepatocellularcarcinoma. Hepatology, 9: 249-252, 1989.

34. Hutchinson, W. L., Johnson, P. J., Du, M-Q., and Williams, R. Serum and tissuea-i.-fucosidase activity in the pre-clinical and clinical stages of hepatocellular carcinoma. Clin. Sci., 81: 177-182, 1991.

35. Vesce, F., and Biondi. C. a-t.-Fucosidasc activity in endometrial, cervical and ovariancancer. Eur. J. Gynaecol. Oncol., 4: 135-138, 1983.

36. Mollicone, R., Gibaud, A., Francois, A., Ratcliffe, M., and Oriol, R. Acceptorspecificity and tissue distribution of the three human a-3-fucosyttransferases. Eur. J.Biochem., 191: 169-176, 1990.

37. Chester, M. A., Yates, A. D., and Watkins, W. M. Phenyl /3-D-galactopyranoside asan acceptor substrate for the blood-group H gene-associated guanosine diphosphateL-fucose:ß-D-galactosyl a-2-L-fucosyltransferase. Eur. J. Biochem., 69: 583-592,

1976.38. Yazawa, S., Madiyalakan, R., Jain, R. K., Shimoda, N.. and Matta, K. L. Use of

benzyl 2-acetamido-2-deoxy-3-O-(2-O-methyl-ß-D-galactosyl)-ß-D-glucopyranoside[2'-O-methyllacto-A'-biose I ßBn]as a specific acceptor for GDP-fucose: N-acetyl-

glucosaminide a(l +4)-L-fucosyltransferase. Anal. Biochem., 187: 374-378, 1990.39. Wiederschain, G. Y., Rosenfeld, E. L., Brusilovsky, A. I., and Kolibaba, L. G.

a-l.-Fucosidase and other glycosidases in human placenta fetus liver and amnioticfluid at various stages of gestation. Clin. Chim. Acta, 35: 99-107, 1971.

40. Hardy, M. R., Townsend, R. R., and Lee, Y. C. Monosaccharide analysis of glyco-conjugates by aniónexchange chromatography with pulsed amperometric detection.Anal. Biochem., 170: 54-62, 1988.

41. Winzler, R. J. Glycoproteins, human plasma. In: H. A. Sober (ed.), Handbook ofBiochemistry, pp. C-40. Cleveland, OH: Chemical Rubber Co., 1968.

42. Kobata, A., and Takasaki, S. Structural characterization of oligosaccharides fromglycoproteins. In: M. Fukuda and A. Kobata (eds.), Glycobiology: A PracticalApproach, pp. 165-185. New York: Oxford University Press, 1993.

43. Stroup, G. B., Anumula, K. R., Kline, T. F.. and Caltabiano, M. M. Identification andcharacterization of two distinct a-( l-3)-i.-fucosyltransferase activities in human coloncarcinoma. Cancer Res., 50: 6787-6792, 1990.

44. Holmes, E. H., and Levery, S. B. Biosynthesis of fucose containing lacto-series

glycolipid in human colonie adenocarcinoma Colo 205 cells. Arch. Biochem.Biophys., 274: 633-647, 1989.

3658

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from

1995;55:3654-3658. Cancer Res   Jen-Wei Wang, Robert A. Ambros, Peter B. Weber, et al.   Levels in Normal and Malignant Endometrial Tissue

-l-Fucosidase Activities and FucoseαFucosyltransferase and

  Updated version

  http://cancerres.aacrjournals.org/content/55/16/3654

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

  .pubs@aacr.orgDepartment 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://cancerres.aacrjournals.org/content/55/16/3654To request permission to re-use all or part of this article, use this link

Research. on January 22, 2020. © 1995 American Association for Cancercancerres.aacrjournals.org Downloaded from