Papillary thyroid carcinoma overexpresses fully and underglycosylated mucins together with native...

Clinical Research

Papillary Thyroid Carcinoma Overexpresses Fully and Underglycosylated Mucins Together with Native and Sialylated Simple Mucin Antigens and Histo-Blood Group Antigens

Pedro Alves, 1 Paula Soares, 1 Elsa Fonseca 1.2 and Manuel Sobrinho-Sim6es, MD, PHD 1'2

Abstract We studied the immunohistochemical expression of mucins (MUCl, underglycosylated MUCl, MUC2, MUC5AC, and MUC6), simple mucin antigens (Tn, sialyl Tn, and T), and histo-blood group antigens (type 1--Lewis a and sialyl Lewis; a type 2--Lewis x and sialyl Lewis x) in a series of 26 papillary thyroid carcinomas (PTC), 6 follicular carcinomas, and a control group of 32 cases of "normal" thyroid parenchyma adjacent to the tumors. PTC expressed more often, more intensively, and more extensively every antigen but MUC6, which was not observed in any case. The expression of MUC5AC was also extremely rare. MUCl expression was related to the expression of underglycosylated MUCl, MUC2, Lewis, a and sialyl Lewis. a A trend toward an association between the expression of MUCl and that of type 2 histo-blood group antigens was also observed. Whenever there was a dissociation between the expression of type 1 and type 2 Lewis antigens, MUC1 appeared closely related to type 1 and independent from type 2 histo-blood group antigens. We conclude that MUC1 plays a pivotal, though not exclusive, role in the glycosylation features of well differentiated thyroid carcinomas. Despite the prominent expression of mucins and carbohydrate antigens in PTC, no significant differences were observed between PTC and follicular carcinoma thus ruling out the possibility of using the aforementioned antigens as diagnostic markers per se.

Key Words- Thyroid carcinoma; papillary thyroid carcinoma; mucins; simple mucin antigens; histo-blood group antigens.

qnstitute of Pathology and Molecular Immunology of the University of Porto (IPATIMUP), Porto, Portugal, and 2Department of Pathology, Medical Faculty of Porto, Porto, Portugal

Address correspondence to Manuel Sobrinho-Sim6es, MD, PhD, IPATIMUP, Rua Roberto Frias, s/n, 4200 Porto, Portugal Email: sobrinho.simoes @ipatimup.pt

Endocrine Pathology, vol. 10, no. 4, 315-324, Winter 1999 �9 Copyright 1999 by Humana Press Inc. All rights of any nature whatsoever reserved. 1046-3976/99/10:315-324/ $12.50

Introduction The abnormal glycosylation pattern of

glycolipids and glycoproteins at the surface of cells is a common and well-described phenomenon in neoplastic development [1,2]. It is known that sugars play an important role in a large variety of biologi- cal processes, including cell interactions, growth control, cell-substratum interactions, and cell homing [1,2]. Neoplastic cells, by achieving aberrant glycosylation features on their surfaces, may acquire selective advantages that can allow their

survival, invasion, and/or spreading to other tissues and organs [1,2].

Thyroid tumors, as many others, appear to fit into the aforementioned model. Pap- illary thyroid carcinomas (PTC), and to a lesser extent, follicular and medullary carcinomas, overexpress simple mucin antigens (namely, Tn and sTn) and type 1 and type 2 histo-blood group antigens (Lewis a and sialyl Lewis a of the former type, and Lewis X and sialyl Lewis x of the latter) [3-7]. The diagnostic importance of these find- ings has been emphasized by several groups.

315

316 Endocrine Pathology Volume 10, Number 4 Winter 1999

Bryne et al. [3] advanced that the immunodetection of sialyl Lewis ~ could help in the separation of PTC from both benign lesions and follicular carcinomas. Sialyl Lewis a and sialyl Tn were recognized, respectively, by Vierbuchen et al. [8] and Yonezawa et al. [7], as good tumor markers in thyroid pathology. And finally, we have recently observed [4,9] that some histo-blood group antigens like Lewis ", sialyl Lewis% Lewis ~, and sialyl Lewis x may be used to distinguish, in most instances, benign from malignant thyroid lesions, despite lacking the ability to separate PTC from follicular carcinomas.

Mucins are major carriers in their sac- charide component of simple mucin and histo-blood antigens [10]. The overexpression of such antigens in many neoplastic conditions may reflect either an increase in the amount of carrier mucins (provided the appropriate glycosyltransferases are available) or an altered pattern of mucin glycosylation leading to an underglycosylated phenotype that enhances the recognition of antigens otherwise not immunodetectable [ 11,12].

The available data on MUC1 transcrip- tion and translation in thyroid tumors [14], and the results of the detection of epithelial membrane antigen (EMA), which grossly corresponds to MUC1 [15], in several types of thyroid neoplasms [16-18] support the assumption that the overexpression in PTC of simple mucin and histo-blood group antigens is related, partly at least, with MUC1 overexpression, and that MUC1 immunodetection may be diagnostically useful.

It remains to be found ifPTC overexpress underglycosylated MUC1 as well as other mucins (MUC2, MUCSAC, and MUC6) together with fully glycosylated MUC 1, and ifMUC1 serves as an equal carrier for both types ofhisto-blood group antigens.

To address these issues and to clarify the putative diagnostic importance of mucin immunodetection in thyroid oncology, we undertook the present study of serially sec- tioned samples of normal thyroid, PTC, and follicular carcinomas, using monoclonal antibodies for MUC1, underglycosylated MUC1, MUC2, MUCSAC, and MUC6, as well as for some simple mucin antigens, and type 1 (Lewis a, sialyl Lewis a) and type 2 (Lewis X, sialyl Lewis ~) histo- blood group antigens.

Materials and Methods

We studied 32 samples of"normal" thyroid parenchyma obtained from surgical specimens with tumors, 26 cases of PTC (there were 4 cases of the follicular variant of PTC and 9 of the remaining 22 PTC displayed a mixed papillary and follicular pattern) and 6 cases of follicular carcinoma. The carcinomas were classified according to the criteria of Hedinger et al. [19], LiVolsi [20] and Rosai [21] .

The expression of mucin antigens (MUC 1, underglycosylated form of MUC 1- SM3, MUC2, MUCSAC, and MUC6), simple mucin antigens (Tn, sialyl Tn), Lewis type 1 histo-blood group antigens (Lewis a, sialyl Lewisa), and Lewis type 2 histo-blood group antigens (Lewis x, sialyl Lewis X) were analyzed by immunohis- tochemistry using the monoclonal antibodies listed in Table 1.

Immunohistochemical analysis was per- formed using the avidin-biotin-peroxidase procedure. Briefly, sequential histologic sections (4 Jam of thickness) from paraf- fin embedded material of each case were mounted in gelatin-coated slides. The avidin-biotin-peroxidase complex staining method was applied. Sections designed for neuraminidase treatment were washed three times in phosphate-buffered saline

Mucins and Carbohydrate Antigens in Thyroid Carcinomas 317

Table 1. Monoclonal Antibodies and Dilutions Used in the Irnmunohistochemical Study

Specificity Antibody Dilution Reference

MUC I HMFG1 1/10 Immunotech MUC 1" SM3 1/2 Burchell et al. [22] MUC 2 LDQ10/PMH1 1/2000 Reis et al. [23] MUC 5AC** CLH2 1/5 Reis et al. [24] MUC6 CLH5 1/2 Reis et al. (unpublished results) Tn HB-Tn 1/5 DAKO sialyl Tn HB-STn 1/1 DAKO Lewis a CA3F4 1/5 Young er al. [26] sialyl Lewis a CA-19-9 1/5 Magnani et al. [27] Lewis x SH1 1/5 Fukushi et al. 1984 sialyl Lewis x FH6 1/5 Fukushi et al. 1984

*Underglycosylated MUC1 mucin;** Code number MAB2011 from Chemicon Interna- tional Inc. (Temecula, CA).

(PBS) and incubated with neuraminidase from Clostridium perfringens type VI (Sigma) diluted in a 0.1 Msodium acetate buffer, pH 5.5, to the final concentration of 0.1 U/mL. The incubation, carried out for 2 h at 37~ was followed by three washings in ice-cold water. All the sections were treated with 0.3% v/v H202 in meth- anol for 30 rain to block endogenous peroxidase. The sections were incubated for 20 min with normal immune serum with 1:5 dilution on TBS containing 25% of bovine serum albumin (BSA) to eliminate nonspecific staining. Excess of normal serum was removed and replaced by the monoclonal antibodies specified in Table 1. After overnight incubation at 4~ slides were incubated with a 1:200 dilution of biotin-labeled secondary antibody for 30 min. Sections were incubated with avidin-biotin-peroxidase complex (10 mg/ mL of biotin-labeled peroxidase) for 30 min, followed by staining with 0.05% diaminobenzidine, freshly prepared in 0.05% TRIS buffer, pH 7.6, containing 0.01% of hydrogen peroxide. Finally, sections were counterstained with hema- toxylin, dehydrated, and mounted. Dilu-

tion of primary antibodies, biotin-labeled secondary antibodies, and avidin-biotin- peroxidase complex was made with TBS containing 12.5% BSA.

Negative controls of the immunostaining were carried out by omission of the primary antibody. As a positive control, sections from previously studied cases of other tissues known to be positive for each of the primary antibodies (MUC1, SM3, MUC2, MUC5AC, MUC6, Tn, sialyl Tn, Lewis a, sialyl Lewis a, Lewis x, and sialyl Lewis x) were used.

Immunostained sections were classified as positive or negative; the percentage of positive cells was semiquantitatively evalu- ated into the following categories: 0-5%, 5-25%, 25-50%, 50-75%, and more than 75% of positive cells. The intensity of the staining was classified as weak, medium, or strong. The cellular localization of the staining was classified as membranous (apical and/or lateral membranes) and/or cytoplasmatic (granular and/or diffuse).

Statistical Analysis

Results are expressed as a percentage or mean (+ SE). For statistical analysis, the chi-square test with the Yates correction, paired and unpaired Student's t-test, and the ANOVA test were used. Two values were considered significantly different whenp < 0.05.

Results

An overview of the results is presented in Table 2.

Normal Thyroid

Mucins

MUC1 was focally detected in the majority of the surgical specimens resected for PTC, namely, in areas localized in the


Table 2. Overview of the Imrnunohistochemical Results for Each Antigen in PTC a

% Histotype positive displaying

Antigens cases Distribution positivity Intensity Localization

MUC 1 73.1 focal pap medium to strong apical SM3 42.3 focal pap medium cytopl MUC2 15.4 focal pap weak cytopl MUC5AC 0.0* focal pap weak* cytopl* MUC6 0.0 . . . . Tn 61.5 focal pap weak cytopl s Tn 100.0 homog pap/foil weak cytopl Lewis a 80.8 focal pap medium to strong apical s Lewis a 38.5 focal pap medium to strong apical Lewis x 40.0 focal pap medium to strong apical s Lewis x 84,6 focal pap/foil medium to strong apical

*In two cases of PTC immunoreactivity for MUC5AC cells.

was detected in less than 5% of the

aA case was considered as positive whenever the number of postive cells was higher than 5% of the total number of cells. The immunoreactivity was classified as focal when the staining was restricted to particular areas of the slide and as homogeneous (homog) when the whole or almost the whole slide was stained. The histotypes were classified as papillary (pap) or follicular (foll). The staining was also classified in relation to its main localization within the cells as membranous (apical) or cytoplasm (cytopl).

neighborhood of the carcinomas. No immunoreactivity for MUC1 was seen in the normal thyroid adjacent to follicular carcinomas. The staining was of medium to strong intensity, and its pattern was of the diffuse cytoplasmatic type. There was no immunoreactivity for underglycosylated MUC1 (SM3 antibody), MUC2, MUC5AC, and MUC6.

Simple Mucin Antigen and Histo-blood Group Antigens

The results were identical to those previously reported--no immunoreactivity for Tn, Lewis a, sialyl Lewis a, Lewis x, and sialyl Lewis x [4J--but for the widespread positivity for sialyl Tn in every sample of normal thyroid (weak or even very weak diffuse cytoplasmatic staining) and for the frequent positivity for sialyl Lewis x in the stromal cells and connective tissue under- lying neoplastic follicles or papillae.

Thyroid Carcinomas (PTC and Follicular Carcinoma)

Mucins

MUC1 was observed in every PTC and in two follicular carcinomas. In 19 cases of PTC (73%) MUC 1 immunoreactivity was observed in more than 5% of the neoplastic cells. In most cases the immunoreactivity was focal. All the histotypes of PTC displayed foci positive for MUC1. PTC predominantly or exclusively composed of papillary structures were particularly immunoreactive (Fig. 1). The staining was more intense and more widespread in the invasive front of PTC. The percentage of positive cells ranged in most cases from 5% to 25% and the intensity ranged from medium to strong. The staining was mainly seen in the apical membrane of the cells (namely, in the papillae of PTC) but it also appeared in their lateral membranes. Diffuse cytoplasmatic staining was also observed, particularly when there was a strong staining.

Immunoreactivity for SM3 was observed in 11 cases of PTC (42.3%). The immu- noreact iv i ty was focal, restricted in most cases to cells lining the papillae of PTC. The number of positive cells was very small and the intensity of the staining was usually classified as medium (Fig. 1). The pattern of staining was cytoplasmatic and often displayed a granular supranuclear pattern.

MUC2 was observed in four cases of PTC (15.4%). The immunoreactivity was focal and almost exclusively restricted to the papillary structures of PTC. Rare cells were observed in positive cases. The staining was weak and had a diffuse cytoplasmatic pattern.

MUC5AC was observed in rare cells of two PTC. Whenever present, the staining was weak and had a diffuse cytoplasmatic pattern.


O

Fig. 1, Immunoreactivity for MUC1 (A and B), underglycosylated MUC1 (C), and sialyl Lewis x (D) in PTC. Note the prominence of the immunostaining in the apical membrane of the neoplastic cells forming the papillae and the differences of the MUCl staining pattern in two PTC (widespread immunoreactivity in the case depicted in A and focal immunoreactivity in the case depicted in B). (A,B, MUCl staining, original magnification x200; C, SM3 staining, original magnification xl00; D, sialyl Lewisxstaining, original magnification x200.)

MUC6 was not detected in any PTC. There was a significative correlation

between the expression of MUC1 and those of underglycosylated MUC 1 (SM3) (p = 0.0003) and MUC2 (p = 0.006). The two cases expressing MUC5AC also over- expressed MUC 1.

No immunoreactivity for underglycosylated MUC1, MUC2, MUC5AC, and MUC6 was found in follicular carcinomas.

Simple Mucin Antigens and Histo-blood Antigens

The results obtained in the study of simple mucin antigens are similar to those previously reported: Tn was detected in about 60% of the cases of PTC and folli-

cular carcinoma (diffuse cytoplasmatic staining in more than 75% of the neoplastic cells) and sialyl Tn was detected in every case (weak cytoplasmatic staining) [4].

The results obtained in the study of histo-blood group antigens also resemble those previously reported [4]: Lewis a was observed in about 80% of the cases, sialyl Lewis a and Lewis x in about 40% of the cases, and sialyl Lewis x in about 85% of the cases. The pattern of staining for the sialylated antigens was apical membranous (occasionally the lateral membranes were also stained), and for the native antigens was both cytoplasmic and membranous (Figs. 1 and 2).

32.0 Endocrine Pathology Vo lume 10, N u m b e r 4 W i n t e r 1999

Fig. 2. In these two serial sections of a PTC with a mixed pattern of growth one can see a sort of a mirror image. In A there is intense (membranous and cytoplasm) immunoreactivity for Lewis x in the cribriform component, whereas in B, stained for MUC1, only the papillary component is intensely immunoreactive. The striking contrast between the two components regarding Lewis x immunoreactivity is high- lighted in the inset of A. (A, Lewisxstaining, original magnification x40; B, MUC1 staining, original magnification x40. Inset original magnification x200.)

PTC tended to display more intense immunoreactivity for all the antigens than follicular carcinomas. Within the PTC group, cases predominantly or exclusively composed of papillae were particularly immunoreactive for histo-blood group antigens although immunoreactivity was also seen in some follicular variants of PTC. The invasive front of PTC in close contact with fibrous connective tissue always displayed a more intense and more widespread immunoreactivity for all the histo-blood group antigens than the bulk of the tumors.

The study of serial sections revealed that there were fewer positive cells and less intensity of the staining for sialyl Lewis a than for Lewis a. The opposite was observed in the comparison of sialyl Lewis x and Lewis X immunoreactivity: there was a higher number of positive cells and more intense staining in the former than in the latter. Neuroaminidase treatment of slides did not cause any major changes in the amount ofimmunoreactive cells, nor in the intensity of the staining for sialyl Lewis a and sialyl Lewis X.

Mucins versus Simple Mucin Antigens and Histo-blood Group Antigens

No correlation was found between the expression of MUC1 and Tn.

The expression o fMUC 1 was correlated with the expression of type 1 histo-blood group antigens (Lewis", p = 0.02, sialyl Lewis", p = 0.1). There were, however, a few PTC with less than 5% MUC1 immunoreactive neoplastic cells that were positive for Lewis a and sialyl Lewis" in more than 25% of the neoplastic cells.

At variance with type 1 histo-blood antigens, no correlation was observed between MUC1 overexpression and immunoreactivity for Lewis X and sialyl Lewis x, despite the existence of a trend toward the con- current expression in the same cases of MUC1 and type 2 histo-blood group antigens.

The observation of serial sections showed that, in some cases, type 1 and type 2 histo- blood group antigens are expressed in different cells, either within the same area or in different areas of the tumors. Whenever this phenomenon was present, there was a


negative association between the presence of type 2 histo-blood antigens and MUC1 immunoreactivity.

In one case of PTC displaying two different histological patterns--a cribriform/ solid pattern and a papillary pattern--we observed a totally different staining pattern for MUC 1 and LewisX: the cribriform,'solid histotype displayed intense cytoplasmatic immunoreactivity for Lewis x and negativity for MUC 1, whereas the papillary component displayed a strong apical membranous immunoreactivity for MUC1 and was completely negative for Lewis x (Fig. 2). All the other antigens were either absent or detected only in few cells in both components without any apparent restriction pattern.

Discussion

In accordance with the results of previ- ous studies focusing on the immunodetection of EMA, as well as with the data recently reported by Bieche et al. [14], MUC1 expression was detected in every PTC of our series. Despite this, MUC1 cannot be used as an indisputable molecular "marker" of PTC because it is also present in follicular carcinomas, as well as in the "normal" thyroid parenchyma adjacent to some PTC. A similar conclusion holds true regarding other carbohydrate "markers," namel> histo-blood group antigens, in general, and sialyl Lewis x, in particular [4, 29]. However, this does not belittle the diagnostic importance of such "markers" whenever used together with relevant clinicopathological and histological data in the appropriate settings [29].

We have shown, for the first time, that PTC express underglycosylated MUC1 and, exceptionally, also MUC5AC. The expression of these mucins is focal and tends to concentrate on the apical membrane of the papillae of PTC. These obser-

vations and the strong apical staining for MUC1 in most PTC fit with the his- tochemical demonstration of alcian blue mucinous material in many PTC [16].

The close correlation we have observed between the immunoreactivity for MUC 1 and SM3 reveals the existence of a defi- cient glycosylation pathway of MUC1, either because the amount of substrate is over the action threshold of glycosyltransferases, thus leading to the accumulation of underglycosylated MUC1, or because there is an intrinsic insufficient action of glycosyltransferses.

Our observation of MUC 1 immunoreactivity in the "normal" thyroid parenchyma adjacent to most PTC had not been previously reported to the best of our knowledge. Since we did not observe any immunoreactivity in the neighborhood of follicular carcinomas, nor in the vicinity of benign tumors (Alves et al., unpublished results), it is tempting to advance a patho- genic relationship between PTC and MUC 1 expression in the adjacent "normal" thyroid. It remains to be clarified whether such expression points to the existence of a sort of field cancerization in thyroid glands with PTC, or merely reflects the response of the follicle cells to putative fac- tors produced by PTC.

It also remains to be found if the overexpression of simple mucin antigens and histo-blood group antigens in most PTC reflects mainly the overexpression of MUC 1, which may serve as a carrier of those antigens, or the existence of an aberrant/defi- cient action ofglycosyltransferases, which specifically add and build the side chains that branch from the protein core of the mucins. Our results indicate that both of the aforementioned mechanisms are prob- ably operative in PTC. (We decided to focus this discussion on PTC because the size of the sample of PTC was much larger


than that of follicular carcinomas and also because the expression of mucins and carbohydrate antigens tends to be much more impressive in PTC than in follicular carcinomas.) The significant correlation between MUC1 expression and the expression of other mucins and several carbohydrate antigens, as well as the close topographic relationship disclosed by the study of serial sections, support the con- cept that MUC1 overexpression plays a pivotal role as a carrier protein for many carbohydrate antigens (e.g., type 1 histo- blood group antigens). On the other hand, we also have evidence indicating that Lewis x can (and often does) use as its presenting structure molecule(s) other than MUC1 (glycosyltransferases can act not only in glycoproteins but also in glycolipids).

With regard to the expression of both types of histo-blood group antigens, our results do not fit with those ofLarena et al. [30], who reported a complementary distribution of Lewis a and Lewis x in different tumor areas. Such a clear-cut distribution was not observed in our series. However, we observed a distinct association of Lewis type 1 and Lewis type 2 antigens with MUC 1, and did inverse ratios of native and sialylated antigens of histo-blood groups: sialyl Lewis a was present in less numerous cases and displaced a less intense staining than Lewis a, in contrast to the results observed in the comparison ofsialyl Lewis ~ with Lewis • .

Our data and those on record [3, 4, 8, 9, 30] support the assumption that the expression of native and sialylated carbohydrate antigens of both simple mucin and histo-blood group types is finely regulated in PTC, as in many other tumor models [2]. Such regulation reflects not only the amount and efficiency of glycosyl and sialyltransferases but also the availability and specificity of the substrates. This latter aspect appears to be organ and tumor

regulated: in colon cancer, for instance, MUC1 is the carrier glycoprotein for sialyl Lewis x [31], whereas in PTC MUC1, it serves mainly as a carrier for type 1 histo- blood group antigens.

Taking into consideration the biologi- cal roles in cell recognition, cell adhesion, cell mobility, and immunosurveilance of many of the molecules overpressed by PTC (fully and underglycosylated MUC1, histo-blood group antigens, and sialic acid, among others), it is tempting to state that the florid and variable glycosylation pattern exhibited by PTC cells may contrib- ute to explain the peculiar invasiveness and nodal metastatic ability of most PTC.

Acknowledgments

The authors thank the following col- leagues who have kindly provided most of the monoclonal antibodies: J. Burchell and J. Taylor-Papadimitriou (SM3), C. Bol6s (LDQ10), H. Clausen (CA3F4, CA- 19-9, SH 1, and FH6), and C. Reis (CLH2 and CLHS). The technical collaboration of F~itima Magalh~es, Dina Leit~o, Maria Jose Bento, and Paula Silva is also gratefully acknowledged. This study was supported by PRAXIS XXI (Project BIA/100/94) and by a technical grant from The Norwegian Radium Hospital.

References Longenecker BM. Introduction: Glycosyl- ation changes associated to malignancy. Sere Cancer Biol 2:355-356, 1991.

2. Hakomory S. Aberrant glycosylation in tumors and tumor-associated carbohydrate antigens. Adv Cancer Res 52:257-331, 1992.

3. Bryne M, Dabelsten E, Nesland JM. The glycosylation pattern in follicular and papillary thyroid carcinomas. Glycosyl Dis 1:119- 125, 1995.

4. Fonseca E, Castanhas S, Sobrinho-Sim6es M. Expression of simple mucin type antigens and


Lewis type 1 and type 2 chain antigens in the thyroid gland: an Immunohistochemical study of normal thyroid tissues, benign lesions, and malignant tumors. Endocrine Pathol 7:291- 301, 1996.

5. Vierbuchen M, Larena A, Schr6der S, Han- isch E Ortmann M, Larena A, Uhlenbruck G, Fischer R. Blood group antigen expression in medullary carcinoma of the thyroid. An immunohistochemical study on the occur- rence of type 1 chain - derived antigens. Virchows Archiv B Cell Pathol 62:79-88, 1992.

6. Vierbuchen M, Schr6der S, LarenaA, Uhlen- bruck G, Fischer R. Native and sialic acid masked Lewis(a) antigen reactivity in medullary thyroid carcinoma. Distinct tumor-associated and prognostic relevant antigens. Virchows Arch 424:205-211, 1994.

7. Yonezawa S, Tachikawa T, Shin S, Sato E. Sialosyl-Tn antigen. Its distribution in normal human tissues and expression in adeno- carcinomas. Am J Clin Pathol 98:167-174, 1992.

8. Vierbuchen M, Shr6der S, Uhlenbruck G, Ortmann M, Fischer R. CA 50 and CA 19-9 antigen expression in normal, hyperplastic, and neoplastic thyroid tissue. Lab Invest 60:726-732, 1989.

9. Fonseca E, Castanhas S, Sobrinho-Sim6es M. Carbohydrate antigens as oncofetal antigens in papillary carcinoma of the thyroid gland. Endocrine Pathol 8:301-303, 1997.

10. Gendler S.J., Spicer AP. Epithelial Mucin Genes. Annu Rev Physio157:607-634, 1995.

l 1. Girling A, Bartkova J, Burchell J, Gendler S, Gillet C, Taylor-Papadimitriou J. A core protein epitope of the polymorphic epithelial mucin detected by the monoclonal antibody SM-3 is selectively exposed in a range of primary carcinomas. Int J Cancer 43:1072- 1076, 1989.

12. Ho SB, Niehans GA, Lyftogt C, Yan PS, Cherwitz DL, Gum ET, Dahiya R, Kim YS. Heterogeneity of mucin gene expression in normal and neoplastic tissues. Cancer Res 53:641-651, 1993.

13. Ho JJL, Kim YS. Do mucins promote tumor cell metastasis? Int Oncology 7:913-926, 1995.

14. Bieche I, Ruffet E, Zweibaum A, Vilde F, Lidereau R, Franc B. MUC1 mucin gene, transcripts, and protein in adenomas and pap-

illary carcinomas of the thyroid. Thyroid 7:725-731, 1997.

15. Gendler SJ, Spicer AP, Lalani EN, Duhig T, Peat N, Burchell J, Pemberto L, Boshell M, Taylor-Papadimitrou J. Structure and biology of a carcinoma-associated mucin, MUC1. Am Rev Respir Dis 144:$42-$47, 1991.

16. Damiani S, Fratamico F, Lapertosa G, Dina R, Eusebi V., Alcian Blue and epithelial membrane antigen are useful markers in differen- tiating benign from malignat papillae in thyroid lesions. Virchows Archiv A Pathol Anat 419:131-135, 1991.

17. Yamamoto Y, Izumi K, Otsuka H. An immunohistochemical study of epithelial membrane antigen, cytokeratin, and vimentin in papillary thyroid carcinoma. Recognition of lethal and favorable prognostic types. Cancer 70:2326-2333, 1992.

18. Ostrowski ML, Merino MJ. Tall cell variant of papillary thyroid carcinoma: a reassessment and immunohistochemical study with comparison to the usual type of papillary carcinoma of the thyroid. Am J Surg Pathol 20:964-74, 1996.

19. Hedinger C, Williams E, Sobin L. Histologi- cal typing of thyroid tumors. 2nd ed. WHO International histological classification of tumors. Springer-Verlag, Berlin, 1988.

20. LiVolsi V. Surgical pathology of the thyroid. WB Saunders, Philadelphia, 1990.

21. Rosai J. Thyroid gland. In: Ackerman's surgical pathology. 8th ed. Mosby, St. Louis, 1995; pp 493-567.

22. Burchell J, Gendler S, Taylor-Papadimitriou J, Girling A, Lewis A, Millis R, Lamport D. Development and characterization of breast cancer reactive monoclonal antibodies directed to the core protein of the human milk mucin. Cancer Res 47:5476-5482, 1987.

23. Reis CA, Sorensen T, Mandel U, David L, Mirgorodskaya K, Roepstorff P, Kihlberg J, Stig-Hansen J-E, Clausen H. Development and characterization of an antibody directed to an 0~-N-acetyl-D-galactosamine glycosylated MUC2 peptide. Glycoconj ugated J 15:51-62, 1998.

24. Reis CA, David L, Nielsen PA, Clausen H, Mirgorodskaya K, Roepstorff P, Sobrinho- Sim6es M. Immunohistochemical study of MUC 5AC expression in human gastric carcinomas using a novel monoclonal antibody. Int J Cancer 74:112-121,1997.


25. Taylor-Papadimitriou J, Peterson JJ, Arklie J, Burchell J, Ceriani RL, Bodmer WE Mono- clonal antibodies to epithelium-specific components of the human milk fat globule membrane: production and reaction with cells in culture. Int J Cancer 28:17-21, 1981.

26. Young WW, Johnson HS, Tamura Y, Karlsson KA, Larson G, Parker JMR, Khare DE Baker DA, Hindsgaul O, Lemieux RU. Character- ization of monoclonal antibodies specific for the Lewis A human blood group determinant. J Biol Chem 258:4890-4894, 1983.

27. Magnani JL, Nilsson BL, Brockhaus, Steple- wsky Z, Kropowski H, Ginsburg V. A monoclonal antibody-defined antigen associated with gastrointestinal cancer is a ganglioside containing sialylated lacto-N-fucopentose II. J Biol Chem 257:14365-14369, 1982.

28. Fukushi Y, Hakamori S, Nudelman E, Coch- ran N. Novel fucolipids accumulating in human adenocarcinoma. II. Selective isolation of

hybridoma antibodies that differentially rec- ognize mono-, di-, and trifucosylated type 2 chain. J Biol Chem 259:4681-4685.

29. Fonseca E, Sobrinho-Sim6es M. Expression of Lewis antigens in papillary carcinoma of the thyroid gland (Letter). Virchows Arch 432:483, 1998.

30. LarenaA, Vierbuchen M, Schroder S, Larena- Avellaneda A, Hadshiew I, Fischer R. Blood group antigen expression in papillary carcinoma of the thyroid gland. An immunohistochemical and clinical study of expression of Lewis, ABO and related antigens. Langen- becks Arch Chir 318:102-113, 1996.

31. Hanski C, Drechsler K, Hanisch FG, Sheehan J, Manske M, Ogorek D, Klussmann E, Hanski ML, Blank M, Xing PX, et al. Altered glycosation of the MUC-1 protein core con- tributes to the colon carcinoma-associated increase ofmucin-bound sialyl-Lewis(x) expression. Cancer Res 53:4082-4088, 1993.

Papillary thyroid carcinoma overexpresses fully and underglycosylated mucins together with native...

Documents

Transcript of Papillary thyroid carcinoma overexpresses fully and underglycosylated mucins together with native...