Digitoxin

* Context.-The expression of galectin-3, a human lectin, has been shown to be highly associated with malignant behavior of thyroid lesions.

Design.-We studied the immunohistochemical expression pattern of galectin-3 in a variety of follicular-derived thyroid lesions (13 benign and 62 malignant), including Hurthle cell and follicular carcinoma, papillary carcinomas and variants, and anaplastic and poorly differentiated carcinomas.

Results.-Immunoreactivity was strongest in papillary thyroid carcinomas, whereas staining was less intense in Hurthle cell and anaplastic carcinomas, and even weaker in the follicular variant of papillary thyroid carcinoma. Staining was absent or weak in the 3 follicular thyroid car

cinomas and was negative in both insular carcinomas. In several tumors, staining was stronger at the advancing invasive edge of the lesion than in the central portion of the tumor. Galectin-3 was also expressed focally and weakly in reactive follicular epithelium and entrapped follicles in chronic lymphocytic thyroiditis. A variety of thyroid lesions showed prominent endogenous, biotin-like activity, which could cause flaws in interpretation if a biotin-detection system were used.

Conclusion.-We conclude that galectin-3 immunostaining, when used in biotin-free detection systems, may be useful as an adjunct to distinguish benign from malignant thyroid lesions.

(Arch Pathol Lab Med. 2002;126:710-713)

The frequently rendered diagnosis of follicular neoplasm on thyroid fine-needle aspiration biopsies does not distinguish benign from malignant lesions. Postsurgical analysis of tumor architecture, that is, invasion, provides the current criteria for definitive evidence of malignancy in follicular neoplasms.1,2 Likewise, the distinction of hyperplastic nodules from follicular variant papillary carcinoma can be difficult. Molecules that may be expressed during the crucial steps in the development of malignancy are expected to have strong effects on clinical management.3-5 Recently, attention has been drawn to galectins, which are human lectins involved in cell-cell and cell-matrix interactions in growth regulation, transformation, and metastasis in a variety of tumors.6,7 A few previous reports have addressed thyroid lesions; these studies have shown that galectin-3 is differentially expressed in malignant and benign thyroid lesions.8-13 With 1 exception (in which a polyclonal antihuman antibody was used), these immunohistochemical studies have used non-commercially available low-titer antisera, which reacted against non-human galectin-3 and cross-reacted with human galectin-3.8

We studied galectin-3 expression in a spectrum of malignant and benign thyroid lesions of follicular cell origin, using commercially available monoclonal antibody to human galectin-3. We also compared digoxigenin and biotin immunodetection systems to assess the endogenous biotin-like reactivity present in a variety of thyroid lesions.

METHODS

Thirteen benign and 62 malignant thyroid lesions were selected at the hospital of the University of Pennsylvania, at Loyola University Medical Center, or were contributed by one of the authors (V.A.L.) (Table 1). Five-micrometer sections of paraffinembedded tissues were dewaxed, and antigen was retrieved by microwaving in 0.1M Tris buffer containing 0.15M sodium chloride and 0.1M potassium chloride at pH 7.4. To reduce background staining, quenching with 4% peroxide in methanol and biotin with avidin-biotin pretreatment was performed. For immunostaining, a monoclonal antibody to human galectin-3 (Research Diagnostics, Flanders, NJ, dilution 1:2500) was used and detected with biotin amplification and DAB color development (Chemmate System, Vector Laboratories, Burlingame, Calif). A second detection system was used for all positive staining samples to overcome nonspecific staining by endogenous biotin. For this approach, a digoxigenin-labeled secondary antibody (antimouse FAB fragments, Chemicon, Temecula, Calif, dilution 1:500) was employed; reaction was detected by antidigoxigenin FAB fragments labeled with alkaline phosphatase (Boehringer, Mannheim, Germany, dilution 1:200) and naphthyl fast red (Research Genetics, Huntsville, Ala). For negative controls, a panel of benign and malignant thyroid lesions was treated in an identical manner to the samples, except the primary galectin-3 antibody was omitted. Negative controls showed only weak blush of colloid and no staining of epithelium in the digitoxin detection system. Erratic staining of some papillary thyroid carcinomas and reactive oncocytic cells was identified with the biotin detection system when the primary antibody was omitted. The antigen-- retrieval procedures and antibody dilutions were tested on benign breast tissue using breast epithelium as the positive contorl.14

Slides were screened by 3 pathologists for distribution and intensity of signal. Morphology and cytology was recorded, and intensity was scored from 0 to 3, where 1 indicates focal/weak staining; 2, moderate staining; and 3, strong staining. For photography, a Nikon digital camera (Coolpix 990) and Adobe Photoshop software version 5.5 were used.

RESULTS

Immunoreactivity with the monoclonal antibody against human galectin-3 showed strong staining in the majority of malignant lesions and only focal reactivity in a few benign adenomatous lesions (Figure 1). All 4 goiters and the single hyperplastic nodule were nonreactive.

The staining reaction was seen in the cytoplasm, although in some cases, additional nuclear staining was identified. No correlation between nuclear staining pattern and histologic subtype was seen. Strong reactivity was seen in papillary neoplasms as follows: tall cell variants, Warthin-like variant, columnar cell variant, and diffuse sclerosing variants scored 2 to 3. Usual papillary thyroid carcinomas, including 2 microcarcinomas and 3 metastases, showed positivity levels of 1 to 3 (17 of 19 cases). Five of 15 follicular variant papillary thyroid carcinomas stained with a score of 1 to 2.

Of the 3 true follicular carcinomas, 2 showed focal reactivity with a score of 1. Five of 8 Hurthle cell carcinomas showed highly variable reactivity with a score of 1 + focal to 3 diffuse. A single case of HUrthle cell carcinoma ineluded areas of transformation to anaplastic carcinoma, which reacted 3+ in contrast to focal 1+ staining in the Hurthle cell background. The second anaplastic and 2 insular carcinomas were nonreactive. One medullary thyroid carcinoma showed focal reactivity. Some of the papillary and Hurthle cell carcinomas showed increased intensity of staining at the invasive edge of the tumor and in foci outside the main tumor mass (Figure 2, A and B). Strong reactivity of 1 tall cell variant and 1 microcarcinoma is shown in Figure 2, C and D, respectively.

Endothelial cells showed moderate galectin-3 reactivity. Benign follicular epithelium neighboring the tumor nodules demonstrated weak reactivity in a few lesions. This reactivity was magnitudes weaker than that in the tumor nodule; dilution of the antibody to 1:5000 abrogated the reaction completely, while endothelium and tumor remained reactive.

When a biotin detection system was used, nonspecific staining was seen regularly in oxyphilic cells in lymphocytic thyroiditis. All Hurthle cell nodules and neoplasms scored strongly positive. Negative controls of these cases omitting the primary galectin-3 antibody demonstrated staining with similar reactivity. The reactivity could be weakened by pretreatment with avidin and biotin. However, the difference in staining with and without avidinbiotin pretreatment was negligible when a more sensitive biotin detection system, such as CSA (Vector), was used. The endogenous biotin-like reactivity prompted us to use a digoxigenin detection system. This detection system showed at most a weak pink blush in colloid in the negative control slides and was nonreactive in normal follicular epithelium.

COMMENT

Galectin-3 is a human lectin that binds to beta-galactosides.15 Differential expression between normal and neoplastic tissues has been demonstrated in diverse organ systems, such as breast, colon, salivary glands, and thyroid.6,16-18 Galectin-3 overexpression has been associated with invasive and metastatic properties.11,15 A breast carcinoma model suggested that overexpression of galectin3 protects cells from apoptosis induced by loss of anchorage, thereby promoting survival of anchorage-independent cells.19

Since the first description of increased expression of a beta-galactoside binding protein in papillary thyroid carcinomas, only a few studies have been published.20 Most of these studies used a low-titer supernatant from monoclonal antibodies raised against a mammalian galectin-3 with cross-reactivity to human galectin-3.9-13 One study used a polyclonal antibody at a dilution of 1:1000 against human galectin-3.8 To our knowledge, our study is the first to use a monoclonal antibody against human galectin-3. We also addressed nonspecific, endogenous, biotin-- like activity in thyroid epithelium by using a digoxigenin detection system for the analysis of galectin-3 expression in the thyroid. The previously published studies demonstrated high-intensity immunohistochemical expression of galectin-3 in the majority of papillary carcinomas, including microcarcinomas and metastases.11,13 Galectin-3 expression was confirmed by Northern or Western blotting of a few immunohistochemically strongly positive cases.8,10,13 Interestingly, in our study the subtypes of thyroid carcinoma with papillary morphology scored highest for galectin-3 expression (Table). No data are currently available to provide an explanation for this association.

We confirmed the observation of Fernandez et al,8 who detected reactivity in endothelial cells and described a predominantly cytoplasmic distribution with nuclear reactivity in some tumor cells. Determination of whether a change in cellular distribution of galectin-3 plays a role in thyroid carcinoma biology requires further investigation. Our study also noted the reactivity of endothelial cells for galectin-3.8

Our findings differ from previous studies in several areas: (1) reactivity of anaplastic carcinomas, (2) reactivity of medullary carcinomas, and (3) reactivity in benign follicular lesions.8,10, 11,13 The number of anaplastic and medullary carcinoma cases is small, and galectin-3 expression appears not to be observed consistently, making a clinical application for galectin-3 in these rare subgroups unlikely.

A few studies have reported negative results in all benign lesions and high frequencies of galectin-3 expression in the common malignant thyroid neoplasm, advocating a potential use of this stain for fine needle aspirates.110,13 However, Kawachi et al,ll observed focal positivity in a minority of follicular adenomas. We also noted weak staining in benign lesions.

The previous studies did not mention whether cases with chronic lymphocytic thyroiditis were included.8-13 The differences may be attributed to geographic or dietary variation, to the selection of case material, or to methodological differences. All previously published studies used a biotin-based detection method for the immunohistochemical analysis.8-13 In our experience, the use of a biotin detection system was hampered by inconsistent nonspecific biotin-like antigenicity in reactive and malignant thyroid follicular epithelium.21

A few cases in the present study showed increased galectin-3 reactivity at the advancing edge of the tumor. Although this observation is interesting, additional studies are needed to define whether galectin-3 expression is associated with the invasive capacity of the tumor.

References

1. LiVolsi V. Follicular lesions of the thyroid. In: Zorab R, ed. Surgical Pathology of the Thyroid. Philadelphia, Pa: WB Saunders Co; 1990:173-212.

2. Rosai J, Carcangiu ML, DeLellis RA. Tumors of the Thyroid Gland. Washington, DC: Armed Forces Institute of Pathology; 1993. Atlas of Tumor Pathology; 3rd series, fascicle 5.

3. Herrmann ME, Talpos GB, Mohamed AN, et al. Genetic markers in thyroid tumors. Surgery. 1991;110:941-948.

4. Fusco A, Grieco M, Santoro M, et al. A new oncogene in human thyroid papillary carcinomas and their lymph-nodal metastases. Nature. 1987;328:170172.

5. Kroll T, Sarraf P, Pecciarini L, et al. PAX8-PPARgamma1 fusion oncogene in human thyroid carcinoma. Science. 2000;289:1357-1360.

6. Andre S, Kojima S, Yamazaki N, et al. Galectins-1 and -3 and their ligands in tumor biology: non-uniform properties in cell-surface presentation and modulation of adhesion to matrix glycoproteins for various tumor cell lines, in biodistribution of free and liposome-bound galectins and in their expression by breast and colorectal carcinomas with/without metastatic propensity. J Cancer Res Clin Oncol. 1999;125:461-474.

7. Perillo N, Marcus M, Baum L. Galectins: versatile modulators of cell adhesion, cell proliferation, and cell death. J Mol Med. 1998;76:402-412.

8. Fernandez PL, Merino MI, Gomex M, et al. Galectin-3 and laminin expression in neoplastic and non-neoplastic thyroid tissue. J Pathology. 1997;181:8086.

9. Orlandi F, Saggiorato E, Pivano G, et al. Galectin-3 is a presurgical marker of human thyroid carcinoma. Cancer Res. 1998;58:3015-3020.

10. Xu XC, EI-Naggar AK, Lotan R. Differential expression of galectin-1 and galectin-3 in thyroid tumors. Am j Pathol. 1995;147:815-822.

11. Kawachi K, Matsushita Y, Yonezawa S, et al. Galectin-3 in various thyroid neoplasms and its possible role in metastasis formation. Hum PathoL 2000;31: 428-433.

12. Cvejic D, Savin S, Paunovic I, Tatic S, Havelka M, Sinadinovic J. Immunohistochemical localization of galectin-3 in malignant and benign human thyroid tissue. Anticancer Res. 1998;18:2637-2641.

13. Inohara H, Honjo Y, Yoshii T, et al. Expression of galectin-3 in fine needle aspirates as a diagnostic marker differentiating benign from malignant thyroid neoplasms. Cancer. 1999;85:2475-2484.

14. Castronovo V, Van Den Brule F, jackers P, et al. Decreased expression of galectin-3 is associated with progression of human breast cancer. Pathology. 1996; 179:43-48.

15. Barondes SH, Cooper DN, Gitt MA, Leffler H. Galectins (minireview). J Biol Chem. 1994;269:20807-20810.

16. ldikio H. Galectin-3 expression in human breast carcinoma: correlation with cancer histologic grade. Intl Oncol. 1998;12:1287-1290.

17. Xu XC, Sola Gallego 1, Lotan R, El-Naggar A. Differential expression of galectin-1 and galectin-3 in benign and malignant salivary gland neoplasms. Int j Oncol. 2000;17:271-276.

18. van den Brule F, Waltregny D, Liu FT, Castronovo V. Alteration of the cytoplasmic/nuclear expression pattern of galectin-3 correlates with prostate carcinoma progression. Intl Cancer. 2000;89:361-367.

19. Kim H, Lin H, Biliran H, Raz A. Cell cycle arrest and inhibition of anoikis by galectin-3 in human breast epithelial cells. Cancer Res. 1999;59:4148-4154. 20. Chiariotti L, Berlingieri MT, De Rosa P, et al. Increased expression of the

negative growth factor, galactoside-binding protein, gene in transformed thyroid cells and in human thyroid carcinomas. Oncogene. 1992;7:2507-2511.

21. Kashima K, Yokoyama S, Daa T, Nakayama 1, Nickerson P, Noguchi S. Cytoplasmic biotin-like activity interferes with immunohistochemical analysis of thyroid lesions: a comparison of antigen retrieval methods. Mod Pathol. 1997;10: 515-519.

Marille E. Herrmann, MD, PhD; Virginia A. LiVolsi, MD; Theresa L. Pasha, MT; Shelley A. Roberts, MS; Eva M. Wojcik, MD; Zubair W. Baloch, MD, PhD

Accepted for publication February 14, 2002.

From the Department of Pathology, University of Pennsylvania, Philadelphia (Drs Herrmann, LiVolsi, and Baloch; Ms Pasha; and Ms Roberts); and the Department of Pathology, Loyola University Chicago, Maywood, III (Dr Wojcik). Dr Herrmann is now with the Armed Forces Institute of Pathology, Washington, DC.

Reprints: Marille E. Herrmann, MD, PhD, Department of Pathology, 6 Founders, University of Pennsylvania, 3400 Spruce St, Philadelphia, PA 19104 (e-mail: herrmannm@afip.osd.mil).

Copyright College of American Pathologists Jun 2002
Provided by ProQuest Information and Learning Company. All rights Reserved