Path. Res. Pract. 187, 629-631 (1991)
Lectin-Bindings in Pituitary Adenomas and Normal Pituitaries A. Behncken and W. Saeger Department of Pathology of the Marienkrankenhaus Hamburg, Hamburg, FRG
SUMMARY
The first study was undertaken for detecting hormone-receptors for estrogen and progesteron with immunohistochemical methods in frozen sections of 58 pituitary adenomas of different types. In all cases the result was negative. In the second study we tested the binding pattern for the lectins Peanut lectin (PNA), Ulex europaeus agglutinin (UEA-1) and the Concanavalin agglutinin (Con-A) in 144 pituitary adenomas ofdifferent types and in 20 post-mortem pituitaries using the two-layer indirect technique. All normal pituitaries reacted positive for all three lectins, elsewhere the binding feature for pituitary adenomas was different. Via statistical analysis, correlating lectin-bindings with clinical hyperfunction and immunohistochemical hormone content, we found an association of acromegaly and PNA and ConA and of Cushing's disease and PNA and UEA -1; furthermore of GHand UEA-1 and Con-A, ofTSH and UEA-1 and of ACTH and PNA and UEA-1. Using serial sections (AB-method for demonstrating hormones and two layer indirect technique for lectin-binding) and the double-staining technique (labelled StreptavidinBiotin-method for demonstration of hormones and the two-layer indirect technique for lectins) we found Con-A reacting with GH and PRL cells and with a part of FSH-cells, UEA-1 reacting with most of the ACTH-cells and PNA reacting with FSH-cells.
Introduction
It is well-known that it is possible to show hormone receptors for estrogen and progesteron with immunohistochemical methods in frozen sections of breast cancer tissue. Furthermore we know that estrogen-receptors are detected in pituitary adenomas by using biochemical methods l . Our aim was to detect the receptors for estrogen and progesteron in frozen sections of pituitary adenomas by immunohistochemistry. All cases showed negative results. This negative result prompted our studies of the pituitary adenomas with regard to a lectin-binding scheme. Lectins are sugar binding proteins and glycoproteins of non-immune origin that agglutinate cells and precipitate glycoconjugates having saccharides of appropriate complementarity2,3. In the literature it is reported that there is a © 1991 by Gustav Fischer Verlag, Stuttgart
70% correlation between the presence of estrogen receptors and the binding of the Peanut lectin 4 • We tested the Peanut lectin (PNA) from Arachis hypogea and the sugar specification D Gal-~ (1-3) Gal-Nac, Ulex europaeus agglutinin (UEA-l) from gorse, with specificity for a-L fucose, and the Concanavalia ensiformis agglutinin (Con-A) from the Jack bean with the specification a-D-Glc, a-D-Man.
Material and Methods For immunohistochemical hormone-receptor analyses we studied frozen sections of 58 pituitary adenomas: 13 highly differentiated GH-cell adenomas with acromegaly, 34 undifferentiated acidophilic adenomas (32 with acromegaly, 2 without hyperfunction), 2 small cell chromophobic adenomas (with hyperprolacti0344-0338/91/0187-0629$3.50/0
630 . A. Behncken and W. Saeger nemia), 4 large cell chromophobic adenomas (2 with hyperprolactinemia, 1 with acromegaly, 1 without hyperfunction), 4 oncocytomas without hyperfunction, 1 ACTH-cell adenoma with Cushing's disease. For the lectin-binding we studied 144 pituitary adenomas: - 13 highly differentiated GH-cell adenomas with acromegaly - 35 undifferentiated acidophilic adenomas (33 with acromegaly, 2 without hyperfunction) - 35 small cell chromophobic adenomas (4 with hyperprolactinemia, 31 without hyperfunction) - 13 large cell chromophobic adenomas (2 with hyperprolactinaemia, 1 with acromegaly, 10 without hyperfunction) - 36 oncocytomas (without hyperfunction) - 12 ACTH-cells adenomas (9 with Cushing's disease, 3 with Nelson's syndrome). Furthermore, we studied 20 post-mortem pituitaries with no morphological or endocrine abnormalities. For serial sections and double-staining technique we used 20 post-mortem pituitaries. For the receptor analyses we used frozen sections and the estrogen-receptor-immunocytochemical-assay (ER-ICA monoclonal-Kit; Abbot-Laboratories) for demonstrating the estrogenreceptor; for demonstrating the progesteron-receptor we used the Avidin-Biotin-method (anti-progesteron receptor, Dianova; Elite ABC-Kit, Vectastain). The staining followed with DAB, counterstaining with Harris haematoxylin. For the lectin-binding study we used paraffin sections and tested the lectins PNA, UEA-1 and Con-A (dilution 1: 500, Dakopatts, FRG). We used the two-layer indirect technique (peroxidase-conjugated rabbit anti-lectin 1: 50, Dakopatts, Hamburg, FRG). We stained with DAB and counterstained with Harris-haematoxylin. For serial sections we cut 2 flm paraffin sections of normal pituitaries and demonstrated the hormone first and the lectins on the next three slides. We tested GH (rabbit-anti GH-antibody 1: 200), PRL (1: 300), ACTH (1: 300), FSH (1: 200), LH (1: 800), and TSH (1: 1000). All antibodies from Dakopatts, Hamburg, FRG. We used the ABC-method: biotinilated antirabbit IgG (1: 50, Dakopatts, Hamburg, FRG), AB-complex and stained with DAB, counterstaining with Harris-haematoxylin. For the lectin binding we used the two-layer indirect technique again. For the double-staining technique in paraffin sections we used the labelled Steptavidin-Biotin-method (Dako LSAB TM Kit, Dakopatts, Hamburg, FRG) to demonstrate the hormones, and we stained with AEC (red product). For the lectins we used the two-layer indirect technique and stained with 4-Chlor-Naphtol (blue product).
Table 1. Lectin-bindings in different cells of normal pituitaries Cell types by immunohistology GH cells PRL cells ACTH cells TSH cells LH cells FSH cells
Lectins PNA
UEA
+ +
ConA
++ ++
(+)
+ + = all cells positive, + = most of the cells positive, (+) = small part of the cells positive, - = negative. Comparing lectin-binding of normal pltUltary and immunohistochemical hormone-content using serial sections and double staining technique we found correlations between lectins and hormones. The comparison was difficult because of the fact that Concanavalin A binds to more than one cell type (Table 1).
Results Using immunohistochemical methods we did not find receptor bindings for estrogen and progesteron. By lectin histochemistry all normal pituitaries were positive for all three lectins. The lectin binding scheme for pituitary adenomas was different. Comparing lectin bindings in pituitary adenomas and clinical hyperfunction by statistical analyses we found a correlation of acromegaly and PNA and Con-A and of Cushing's disease and PNA and UEA-l. Using the same statistical analyses when comparing lectin-binding and immunohistochemical hormone content in pituitary adenomas, we found associations of GH content with UEA-1 and Con-A, of TSH content and UEA-1 and of ACTH content and PNA and UEA-l.
Fig. 1. Normal pituitary: Peanut lectin (PNA) binding in about 20% of cells. Two-layer indirect technique, immunostaining with DAB, counterstaining with haematoxylin. Magnification x 280.
Lectin-Bindings in Pituitary Adenomas and Normal Pituitaries· 631
gates is not specific for one type of cell and varies greatly in different tissues. Using double-staining technique and serial sections of normal human pituitary it was possible to show that PNA positive cells belong to a part of the cells that react basophilic or chromophobic when using PAS-orange G staining5 • The basophilic PNA positive cells conform to the Proopiomelanocortinderivates y-MSH, ACTH, aMSH, ~-Endorphin, Met-Enkephalin and Leu-Enkephallll.
In chromophobic pituitary adenomas the PNA-positive cells correlated with Prolactin or TSH producing cells or functional inactive cells 6• The lectin binding in pituitary adenomas is a good supplement to the immunohistochemical demonstration of hormones, but it cannot replace immunohistochemistry. It would seem to be a further component for the comprehension of adenoma biology. Lectin bindings to special cells are likely to depend on the biochemistry of hormones. The lectin-binding takes place in the cytoplasm, but it is not possible to define the lectin receptors exactly, probably they bind in more than only one cell organelle. This should be the field for future biochemical and ultrastructural research. From the lectin-binding we can make no conclusions as to the amount of receptors. In corticotrophic adenomas there was no difference between the reaction of ACTH-cells and Crooke's cells. By comparing the binding features of normal pituitary cells and their neoplastic counterparts with immunohistochemical hormone content a lectin-map may result which may identify special cells. References Fig. 2. Highly differentiated GH-cell adenoma: Concanavalin-A (Con-A) in about 50% of cells. Two-layer indirect technique, immunostaining with DAB, counterstaining with haematoxylin. Magnification x 450.
Discussion Immunohistochemistry as used for hormone receptors in breast cancers did not reveal any receptor bindings in frozen sections of pituitary adenomas in acromegaly and hyperprolactinemia. In a higher concentration of the antibodies the demonstration of the receptors may be possible (Lloyd, personal communication at 4th meeting of international pituitary pathology club, Vaison la romaine, 1990). With lectins it is possible to demonstrate the distribution of glycoconjugates in cells and tissue. Therefore they are of importance in cellular biology and experimental pathology. Though the use of lectins is limited, tissue conservation and fixation are important for the binding power of lectins. Lectins have a high affinity but no absolute specificity for special glycoconjugates. The distribution of glycoconjuReceived December 15, 1990 . Accepted January 16, 1991
1 Pichon M-F, Bression D, Peillon F, Milgrom E (1980) Estrogen Receptors in Human Pituitary Adenomas. Journal of Clinical Endocrinology and Metabolism, pp 897-902 2 Alroy J, Ucci AA, Pereira, MEA (1984) Lectins: histochemical probes for specific carbohydrate residues. In: De Lellis RA (Ed): Diagnostic Immunohistochemistry, Vol 2: 66-88. Masson, New York 3 Goldstein IJ, Poretz RD (1986) Isolation, physiochemical characterization, and carbohydrate-binding specificity of lectins. In: The Lectins, Properties, Functions and Applications in Biology and Medicine, Liener IE, Sharon N, Goldstein IJ (Eds.) pp 33-247. Academic Press, Orlando, FL 4 Remmele W, Hildebrandt U, Hienz HA, Klein PJ, Vierbuchen M, Behnken LJ, Heicke B, Scheidt E (1986) Comparative histological, histochemical, immunohistochemical and biochemical studies on estrogen receptors, lectin receptors and Barr bodies in human breast cancer. Virchow's Archiv, A Pathol Anat 409: 127-147 5 Kern P (1984) Funktionelle Charakterisierung der PeanutLektin-positiven Zellpopulation in der menschlichen Adenohypophyse und in Hypophysenadenomen. Eine lichtmikroskopischeimmunocytochemische Untersuchung. Inauguraldissertation, Heidelberg 6 Kern P (1984) Funktionelle Charakterisierung der PeanutLektin-positiven Zellpopulation in der menschlichen Adenohypophyse und in Hypophysenadenomen. Eine lichtmikroskopischeimmunocytochemische Untersuchung. Inauguraldissertation, Heidelberg
Key words: Lectin-binding - Pituitary - Pituitary adenomas Prof. Dr. W. Saeger, Abteilung fur Pathologie des Marienkrankenhauses Hamburg, Alfredstr. 9, D-W 2000 Hamburg 76, FRG
Lectin-Bindings in Pituitary Adenomas and Normal Pituitaries A. Behncken and W. Saeger Department of Pathology of the Marienkrankenhaus Hamburg, Hamburg, FRG
SUMMARY
The first study was undertaken for detecting hormone-receptors for estrogen and progesteron with immunohistochemical methods in frozen sections of 58 pituitary adenomas of different types. In all cases the result was negative. In the second study we tested the binding pattern for the lectins Peanut lectin (PNA), Ulex europaeus agglutinin (UEA-1) and the Concanavalin agglutinin (Con-A) in 144 pituitary adenomas ofdifferent types and in 20 post-mortem pituitaries using the two-layer indirect technique. All normal pituitaries reacted positive for all three lectins, elsewhere the binding feature for pituitary adenomas was different. Via statistical analysis, correlating lectin-bindings with clinical hyperfunction and immunohistochemical hormone content, we found an association of acromegaly and PNA and ConA and of Cushing's disease and PNA and UEA -1; furthermore of GHand UEA-1 and Con-A, ofTSH and UEA-1 and of ACTH and PNA and UEA-1. Using serial sections (AB-method for demonstrating hormones and two layer indirect technique for lectin-binding) and the double-staining technique (labelled StreptavidinBiotin-method for demonstration of hormones and the two-layer indirect technique for lectins) we found Con-A reacting with GH and PRL cells and with a part of FSH-cells, UEA-1 reacting with most of the ACTH-cells and PNA reacting with FSH-cells.
Introduction
It is well-known that it is possible to show hormone receptors for estrogen and progesteron with immunohistochemical methods in frozen sections of breast cancer tissue. Furthermore we know that estrogen-receptors are detected in pituitary adenomas by using biochemical methods l . Our aim was to detect the receptors for estrogen and progesteron in frozen sections of pituitary adenomas by immunohistochemistry. All cases showed negative results. This negative result prompted our studies of the pituitary adenomas with regard to a lectin-binding scheme. Lectins are sugar binding proteins and glycoproteins of non-immune origin that agglutinate cells and precipitate glycoconjugates having saccharides of appropriate complementarity2,3. In the literature it is reported that there is a © 1991 by Gustav Fischer Verlag, Stuttgart
70% correlation between the presence of estrogen receptors and the binding of the Peanut lectin 4 • We tested the Peanut lectin (PNA) from Arachis hypogea and the sugar specification D Gal-~ (1-3) Gal-Nac, Ulex europaeus agglutinin (UEA-l) from gorse, with specificity for a-L fucose, and the Concanavalia ensiformis agglutinin (Con-A) from the Jack bean with the specification a-D-Glc, a-D-Man.
Material and Methods For immunohistochemical hormone-receptor analyses we studied frozen sections of 58 pituitary adenomas: 13 highly differentiated GH-cell adenomas with acromegaly, 34 undifferentiated acidophilic adenomas (32 with acromegaly, 2 without hyperfunction), 2 small cell chromophobic adenomas (with hyperprolacti0344-0338/91/0187-0629$3.50/0
630 . A. Behncken and W. Saeger nemia), 4 large cell chromophobic adenomas (2 with hyperprolactinemia, 1 with acromegaly, 1 without hyperfunction), 4 oncocytomas without hyperfunction, 1 ACTH-cell adenoma with Cushing's disease. For the lectin-binding we studied 144 pituitary adenomas: - 13 highly differentiated GH-cell adenomas with acromegaly - 35 undifferentiated acidophilic adenomas (33 with acromegaly, 2 without hyperfunction) - 35 small cell chromophobic adenomas (4 with hyperprolactinemia, 31 without hyperfunction) - 13 large cell chromophobic adenomas (2 with hyperprolactinaemia, 1 with acromegaly, 10 without hyperfunction) - 36 oncocytomas (without hyperfunction) - 12 ACTH-cells adenomas (9 with Cushing's disease, 3 with Nelson's syndrome). Furthermore, we studied 20 post-mortem pituitaries with no morphological or endocrine abnormalities. For serial sections and double-staining technique we used 20 post-mortem pituitaries. For the receptor analyses we used frozen sections and the estrogen-receptor-immunocytochemical-assay (ER-ICA monoclonal-Kit; Abbot-Laboratories) for demonstrating the estrogenreceptor; for demonstrating the progesteron-receptor we used the Avidin-Biotin-method (anti-progesteron receptor, Dianova; Elite ABC-Kit, Vectastain). The staining followed with DAB, counterstaining with Harris haematoxylin. For the lectin-binding study we used paraffin sections and tested the lectins PNA, UEA-1 and Con-A (dilution 1: 500, Dakopatts, FRG). We used the two-layer indirect technique (peroxidase-conjugated rabbit anti-lectin 1: 50, Dakopatts, Hamburg, FRG). We stained with DAB and counterstained with Harris-haematoxylin. For serial sections we cut 2 flm paraffin sections of normal pituitaries and demonstrated the hormone first and the lectins on the next three slides. We tested GH (rabbit-anti GH-antibody 1: 200), PRL (1: 300), ACTH (1: 300), FSH (1: 200), LH (1: 800), and TSH (1: 1000). All antibodies from Dakopatts, Hamburg, FRG. We used the ABC-method: biotinilated antirabbit IgG (1: 50, Dakopatts, Hamburg, FRG), AB-complex and stained with DAB, counterstaining with Harris-haematoxylin. For the lectin binding we used the two-layer indirect technique again. For the double-staining technique in paraffin sections we used the labelled Steptavidin-Biotin-method (Dako LSAB TM Kit, Dakopatts, Hamburg, FRG) to demonstrate the hormones, and we stained with AEC (red product). For the lectins we used the two-layer indirect technique and stained with 4-Chlor-Naphtol (blue product).
Table 1. Lectin-bindings in different cells of normal pituitaries Cell types by immunohistology GH cells PRL cells ACTH cells TSH cells LH cells FSH cells
Lectins PNA
UEA
+ +
ConA
++ ++
(+)
+ + = all cells positive, + = most of the cells positive, (+) = small part of the cells positive, - = negative. Comparing lectin-binding of normal pltUltary and immunohistochemical hormone-content using serial sections and double staining technique we found correlations between lectins and hormones. The comparison was difficult because of the fact that Concanavalin A binds to more than one cell type (Table 1).
Results Using immunohistochemical methods we did not find receptor bindings for estrogen and progesteron. By lectin histochemistry all normal pituitaries were positive for all three lectins. The lectin binding scheme for pituitary adenomas was different. Comparing lectin bindings in pituitary adenomas and clinical hyperfunction by statistical analyses we found a correlation of acromegaly and PNA and Con-A and of Cushing's disease and PNA and UEA-l. Using the same statistical analyses when comparing lectin-binding and immunohistochemical hormone content in pituitary adenomas, we found associations of GH content with UEA-1 and Con-A, of TSH content and UEA-1 and of ACTH content and PNA and UEA-l.
Fig. 1. Normal pituitary: Peanut lectin (PNA) binding in about 20% of cells. Two-layer indirect technique, immunostaining with DAB, counterstaining with haematoxylin. Magnification x 280.
Lectin-Bindings in Pituitary Adenomas and Normal Pituitaries· 631
gates is not specific for one type of cell and varies greatly in different tissues. Using double-staining technique and serial sections of normal human pituitary it was possible to show that PNA positive cells belong to a part of the cells that react basophilic or chromophobic when using PAS-orange G staining5 • The basophilic PNA positive cells conform to the Proopiomelanocortinderivates y-MSH, ACTH, aMSH, ~-Endorphin, Met-Enkephalin and Leu-Enkephallll.
In chromophobic pituitary adenomas the PNA-positive cells correlated with Prolactin or TSH producing cells or functional inactive cells 6• The lectin binding in pituitary adenomas is a good supplement to the immunohistochemical demonstration of hormones, but it cannot replace immunohistochemistry. It would seem to be a further component for the comprehension of adenoma biology. Lectin bindings to special cells are likely to depend on the biochemistry of hormones. The lectin-binding takes place in the cytoplasm, but it is not possible to define the lectin receptors exactly, probably they bind in more than only one cell organelle. This should be the field for future biochemical and ultrastructural research. From the lectin-binding we can make no conclusions as to the amount of receptors. In corticotrophic adenomas there was no difference between the reaction of ACTH-cells and Crooke's cells. By comparing the binding features of normal pituitary cells and their neoplastic counterparts with immunohistochemical hormone content a lectin-map may result which may identify special cells. References Fig. 2. Highly differentiated GH-cell adenoma: Concanavalin-A (Con-A) in about 50% of cells. Two-layer indirect technique, immunostaining with DAB, counterstaining with haematoxylin. Magnification x 450.
Discussion Immunohistochemistry as used for hormone receptors in breast cancers did not reveal any receptor bindings in frozen sections of pituitary adenomas in acromegaly and hyperprolactinemia. In a higher concentration of the antibodies the demonstration of the receptors may be possible (Lloyd, personal communication at 4th meeting of international pituitary pathology club, Vaison la romaine, 1990). With lectins it is possible to demonstrate the distribution of glycoconjugates in cells and tissue. Therefore they are of importance in cellular biology and experimental pathology. Though the use of lectins is limited, tissue conservation and fixation are important for the binding power of lectins. Lectins have a high affinity but no absolute specificity for special glycoconjugates. The distribution of glycoconjuReceived December 15, 1990 . Accepted January 16, 1991
1 Pichon M-F, Bression D, Peillon F, Milgrom E (1980) Estrogen Receptors in Human Pituitary Adenomas. Journal of Clinical Endocrinology and Metabolism, pp 897-902 2 Alroy J, Ucci AA, Pereira, MEA (1984) Lectins: histochemical probes for specific carbohydrate residues. In: De Lellis RA (Ed): Diagnostic Immunohistochemistry, Vol 2: 66-88. Masson, New York 3 Goldstein IJ, Poretz RD (1986) Isolation, physiochemical characterization, and carbohydrate-binding specificity of lectins. In: The Lectins, Properties, Functions and Applications in Biology and Medicine, Liener IE, Sharon N, Goldstein IJ (Eds.) pp 33-247. Academic Press, Orlando, FL 4 Remmele W, Hildebrandt U, Hienz HA, Klein PJ, Vierbuchen M, Behnken LJ, Heicke B, Scheidt E (1986) Comparative histological, histochemical, immunohistochemical and biochemical studies on estrogen receptors, lectin receptors and Barr bodies in human breast cancer. Virchow's Archiv, A Pathol Anat 409: 127-147 5 Kern P (1984) Funktionelle Charakterisierung der PeanutLektin-positiven Zellpopulation in der menschlichen Adenohypophyse und in Hypophysenadenomen. Eine lichtmikroskopischeimmunocytochemische Untersuchung. Inauguraldissertation, Heidelberg 6 Kern P (1984) Funktionelle Charakterisierung der PeanutLektin-positiven Zellpopulation in der menschlichen Adenohypophyse und in Hypophysenadenomen. Eine lichtmikroskopischeimmunocytochemische Untersuchung. Inauguraldissertation, Heidelberg
Key words: Lectin-binding - Pituitary - Pituitary adenomas Prof. Dr. W. Saeger, Abteilung fur Pathologie des Marienkrankenhauses Hamburg, Alfredstr. 9, D-W 2000 Hamburg 76, FRG
