P/i~i?,rtrr (19Y2), 13, 3-19-355
Cellular Localization of Metallothionein Human Term Placenta
in
R. A. GOYER, M. D. HAUST & M. G. CHERIAN Dqartment of Pathology, The Unizersir,) of W&tern Ontario, London, Ontario, Canada. N6.A 5C1 Paper accepted 17.12.1991
SUMh4ARY Cellular localization of metallothionein (MT) in placenta may provide information on its function as a metal binding protein. Rabbit antibodies to rat liver MT crossreacted with human MT and were used to localize MT in human term placenta by aridin-biotin peroxidase technique, Serial sections (5 p) were cut from parafinembedded placentae obtained at term from fiue normal women and incubated with rabbit antibodies to MT. Normal rabbit serum was used as a negatitle control. The slides were incubated with biotinylated swine anti-rabbit IgG (linking antibody) then with avidin-biotin horseradish peroxidase complex and daleloped with diaminobenzidine in hydrogen peroxide (0.03 per cent) substrate. The optimum staining ofMT was obtained at a I:800 antibody dilution. MT was identrjied in fetal amniotic cells, symytial trophoblasts and oillous interstitial cells, and in maternal decidual cells. The presence ofMTat specific cellular sites suggests that it may regulate the transplacental transport of metals such as zinc, copper and cadmium. Since the level of cadmium is lower and that of zinc and copper higher in fetal than in maternal blood, this may suggest that placental MT may restrict cadmium while enhancing zinc and copper transport. INTRODUCTION nletallothionein (MT) is a low molecular weight cysteine-rich protein which can bind with essential metals like zinc and copper and toxic metals like cadmium and mercury, Kagi and Kojima (1987). This protein may play a role in homeostasis of zinc and copper and may also provide protection from toxic effects of cadmium and mercury, Cherian and Goyer (1978). During mammalian development, the placenta plays a major role in the transfer of essential nutrients from the mother to the embryo and also protecting the embryo from toxic compounds. Although both zinc and copper are readily transferred to the embryo through the placenta, there is a limited transfer of cadmium across the placenta (Barlow and Sullivan, 1982; Dencker et al, 1983). Little is known about the mechanism of selective transfer of essential metals through the placenta. Cadmium is known to accumulate in the placenta and cause cellular damage which can result in teratogenic or embryotoxic effects (Parizek, 1964; \Yeir et al, 1990). It has been postulated from animal experiments that placental MT may play (1143-4004/92/040319
+ 07 $05.00/O
0
1992 Baillike
Tindall Ltd
350
Phcmto (I 992), Ld. 1.7
a role in cadmium retention and cause a decrease in transfer of zinc to embryo (Samawickrama and Webb, 1979). The presence of MT has been shown in human placenta and fetal membranes (Waalkes et al, 1984; Wier and Miller, 1987). The induced synthesis of MT in cultured human trophoblasts by both cadmium and zinc salts is also reported. In a more recent study, De et al (1989) demonstrated high levels of IMT mRNA levels in the syncytiotrophoblasts of mouse placenta by in situ hybridization. These levels were further increased following injection of cadmium and zinc. The consistently high levels of MT and mRNA were also detected in the decidua following early implantation. The present study was undertaken to localize the MT in full-term human placenta in order to elucidate its physiological role in the transplacental transport of essential and toxic metals.
MATERIALS
AND
METHODS
This study was based on examination of tissue blocks from placentae obtained from live, fullterm healthy infants. There were no medical complications during any of the pregnancies and the mothers were primiparous and ranged in age from 21 to 32 years old. All were nonsmokers and received no medications during pregnancy apart from iron and vitamin supplements. There was no history of previous occupational or environmental exposure to cadmium. The placentae were collected immediately after each delivery and were perfused for 5 min with normal saline. Two blocks of tissue were removed from each placenta from the paracentral area. The samples included the entire thickness of the organ, i.e., the basal (maternal) and the chorionic plates. Care was taken to include the thin amniotic layer superimposed upon the chorion. A portion of the sample was analysed for Cd, Zn, Cu by atomic absorption spectroscopy and metallothionein by a silver saturation method (Scheuhammer and Cherian, 1986). Parallel blocks of tissues were fixed in 10 per cent buffered formalin for 24 h, trimmed to a thickness of approximately 0.5 cm and subsequently fixed in the same but fresh solution for another 24-h period. The blocks of tissues were processed for paraffin embedding by the tetrahydrofuran (THF) method, Haust (1958). Three-micron thick sections were stained for orientation, and cellular and vascular identification with the following stains: hamalumphloxine-saffron (HPS); Masson’s trichrome; resorcin fuchsin, nuclear fast red; metanil yellow; and alcian blue; periodic acid Schiff (PAS), hematopxylin, orange G. Additional sections were cut at 5 ,U for localization of MT by the following method. Antibodies to MT were raised in rabbits after repeated injection of rat liver MT polymerized with glutaraldehyde (Banerjee et al, 1982). This polyclonal antibody cross reacted with both monomer and polymer of MT in various species including human (Nartey, Banerjee and Cherian, 1987a; Nartey, Cherian and Banerjee, 1987b). The unstained sections were incubated with 20 per cent normal swine serum at room temperature for 30 min incubated at room to block the non-specific binding sites. They were subsequently temperature for 60 min with rabbit anti-MT serum at four different dilutions (1: 200; 1: 400; 1: 800; 1: 1000). Sections incubated with normal rabbit serum served as (negative) controls. After washing, the slides were incubated with biotinylated swine-anti-rabbit IgG (linking antibody) for 30 min at room temperature, exposed to avidin-biotin horseradish peroxidase complex (as directed: ABC kit, Vector Laboratories, Inc., Burlington, CA.), and developed with diaminobenzidine (DAB) in 0.03 per cent hydrogen peroxide substrate according to the
C;r~yrrYIrrl: Localizatiot~ i,l~lfetallothione
351
method of Hsu, Raine and Fanger (1981). Selected sections were counterstained with hematoxylin for nuclear identification. Because of lack of documentation in the literature of the presence of MT in amniotic fluid and the need for this information to support the immunohistochemistry observations amniotic fluid was collected from six additional full-term, uncompleted deliveries from nonsmoking primiparous mothers. The amniotic fluid was centrifuged to remove cellular and other debris, and Cd was analysed by atomic absorption spectroscopy. Metallothionein was measured by immunoassay using peroxidase-antiperoxidase with a sensitivity of 1 ng MT.
RESULTS The intensity of positive staining with rabbit anti-MT antibody was related to concentration of antibody. Sections incubate with anti-MT serum at 1:200, 1:400 dilutions yielded in final preparations ‘heavy’ staining which largely obscured the cellular details and the mode of the cytoplasmic MT-distribution. On the other hand, the still positive but faint reaction in slides incubated at 1:lOOO dilution, rendered these unsuitable for study and photographic documentation. Thus, slides incubated at 1800 dilution were stained optimally for the purpose of the present study and photography (Figure la-d). Control sections incubated with non-immune rabbit serum did not stain. The reaction product had a yellow-brown staining quality and was located entirely within cytoplasm of cells. Staining for MT was demonstrated in syncytiotrophoblasts lining the villi [Figure 1 (a)], villous interstitial cells [Figure l(b)], in maternal decidual (interstitial) cells [Figure l(c)], and in amniotic epithelium [Figure 1 (d)]. Positive staining for MT was moderately consistent in these cells in all sections. However, intensity of staining of syncytiotrophoblast was variable, hardly visible in some cells but moderately intense in others. In some trophoblasts, staining was not perceptible. Staining of the villus interstitial cells increased with breadth of the villus. Small villi sparse in interstitial cells stained lightly, but density of staining increased with breadth of v-illus and amount ofinterstitial cells. The villus in Figure 1 (b) is of a larger calibre than that in Figure 1 (a) (terminal villus) and contains many more interstitial cells. Moreover, the blood v-essel in Figure 1 (b) is a small arteriole, whereas the blood vessel in Figure 1 (a) is a capillary. The intense staining of the amniotic epithelium was consistent in all sections. Metallothionein and metal content of five placentae used for immunocytochemistry are shown in Table 1. The levels of metals are similar to those reported by others, Korpela et al (1986); Kuhnert, Kuhnert and Zarlingo (1988); Brophy (1985); Manci and Blackburn (1987). Levels of Cd and MT in amniotic fluid from six additional deliveries is shown in Table 2. Comparable measurements of MT in amniotic fluid are not available for comparison.
DISCUSSION The present investigation demonstrates the localization of MT in both maternal and fetal cells in human placenta. The presence of MT within specific cells suggests that the protein is either synthesized within these cells or is absorbed from surrounding media. Synthesis of MT has been demonstrated in isolated trophoblasts and in decidual cells, but has not been studied specifically in other types of cells in placenta. Metallothionein has been mainly
Figure 1. Localization of metallothionein (MT) using immunoperoxidase method and anti-MT antibodies. 1Metallothionein stains yellow-brown in colour. (a) Metallothionein is in trophoblasts in outer layer of a terminal villous containing thin-walled capillaries (arrows), x 200. (b) Metallothionein is in cells in interstitium of villous (Hofbauer cells), x65. Arrow indicates a small arteriole. (c) Metallothionein is in cytoplasm of decidual cells in basilar area ofplacenta, x90. (d) Metallothionein is in cells lining amniotic membrane, x90.
(b)
2
2
$3 ?
Gnyrr et ~1:Localization of.Metallothione
353
Table 1. LMetallothionein
IMT ( P!3@ 30.0 k4.5
1.9 to.3
and metal placentae
content
of human
Zn
CU
(/G/P)
(Pupk)
12.9 24.5
0.8 kO.3
Five placentas were collected from five full-term deliveries. .Metals were analysed by atomic absorption spectrophotometry. Metallothionein was measured by immunoassay. Results are expressed as mean f s.d.
localized in epithelial cells in human fetal liver and in neoplastic cells by immunohistochemistry (Nartey, Banerjee and Cherian, 1987a; Nartey, Cherian and Banerjee, 1987b). The availability of a polyclonal antibody that readily cross-reacted with human MT enabled us to demonstrate the presence of MT in human placenta. The large dilution of the antibody to obtain the optimum condition of localization of MT in the present study reflects the measurable amount of MT present in human placenta. The trophoblast in the human has a wide range of metabolic and endocrinologic functions essential to the development of the placenta and support of the embryo-fetal development, Yeh and Kurman (1989). It facilitates transport of substances entering the placenta from maternal blood. These may include substances from ions to various growth factors, including MT for the metabolism of the essential metals zinc and copper. As in other cells that synthesize MT, mRNA for MT is increased in trophoblasts on exposure to cadmium and zinc demonstrating the transcriptional control of MT synthesis in these cells (De et al, 1989). Both zinc and copper are essential metals for the growth and development of the fetus and they are found in high concentrations in human and rodent fetal liver bound to MT during a certain stage in perinatal development. Normally no cadmium is detected in fetal or newborn mammalian liver (Chung, Nartey and Cherian, 1986), and placenta accumulates cadmium bound to MT. Mas and Sarkar (1988) also found Cd bound to high molecular weight proteins in the rat placenta. Since MT is a storage protein for metals, the binding of metals to MT in placenta can be considered as an intermediate step in the regulation of the transport from maternal to fetal blood. However, the mechanism of specific transport of the essential metals such as zinc and copper through the villous to fetal blood with little transport of cadmium which is also partly bound to placental MT, is not yet understood. It should be noted that the concentration of zinc @g) in placenta is about 1000 times more than cadmium (ng). Since cadmium has higher affinity for MT than zinc, the difference in magnitude of zinc
Table 2. Cadmium
Cd (ng/ml) 26.4 k 4.1
and metallothionein amniotic fluid
content
of
MT (&ml) 1.7 f 0.2
Amniotic fluid was collected from six full-term deliveries. Cd was analysed by atomic absorption spectrophotometry. Metallothionein was measured by immunoassay. Results are expressed as mean + s.d.
354
Placenta (1992), Vol. 13
concentration in placenta may result in a greater affinity for cadmium to placental MT and may facilitate increased transfer of zinc. Hofbauer cells are, in essence, motile macrophages capable of phagocytosis and protein ingestion. Whether MT in these cells reflects primary synthesis of absorption enroute from trophoblast to the villous capillary cannot be determined from this study, but increase in staining density with increase in size ofvilli suggests accumulation from transvillus transport from maternal blood to the fetal capillary. Amniotic epithelial cells are fetal derivatives in origin but, again, whether localization of MT in these cells reflects accumulation by absorption from amniotic fluid (presumably from fetal excretion) or from primary synthesis, is not known. Our data (Table 2) show the presence of MT in amniotic fluid, analysed from six different deliveries. Decidual cells are endometrial stromal cells that have been transformed under hormonal influence into large pale cells, rich in glycogen. The decidual cells identified as containing MT in this study are those of placenta basalis. The role of MT in these cells is not known. In summary, the results of the present study demonstrate the presence of MT in various cell types in full-term human placenta. Its presence in both maternal and fetal components of the placenta suggest that it may have a regulatory role in the specific transport and storage of essential metals such as zinc and copper.
ACKNOWLEDGEMENTS Linda Veinot provided technical assistance with immunohistochemical grant-in-aid from the Medical Research Council of Canada.
staining.
This research
is sponsored
by a
REFERENCES
Banerjee, D., Onosaka,
S. & Cherian, M. G. (1982) Immunochemical localization of metallothionein in cell nucleus and cytoplasm of rat liver and kidney-. Toxicolo~, 24, 95-105. Barlow, S. M. & Sullivan, F. M. (1982) Cadmium and its compounds. In Repro&&~ Hazards of Industrial Chemicals. (Ed.) Barlow, S. M. & Sullivan, F. M. pp. 136-177. London: Academic Press. Brophy, M. H., Harris, N. F. & Crawford, I. L. (1985) Elevated copper and lowered zinc in the placenta ofpreeclamptics. Clinica Chimica Acta, 145, 107-l 12. Cherian, M. G. & Goyer, R. A. (1978) Metallothioneins and their role in the metabolism and toxicity of metals. Mini-review. Clinical Science, 23, l-10. Chung, J., Nartey, N. 0. & Cherian, M. G. (1986) Metallothionein levels in liver and kidney of Canadians-A potential indicator of environmental exposure to cadmium. Archiw ofEnaironmenta1 Health, 41, 3 19-323. De, S. K., McMaster, M. T., Dey, S. K. & Andrews, G. K. (1989) Cell-specific metallothionein gene expression in mouse deciduo and placentae. Dtwlopment, 107, 611-62 1. Dencker, L., Danielsson, B., Khayat, A. & Lindgren, A. (1983) Disposition ofmetals in the embryo and fetus. In Reprodudizvand Developmental Toxici[y ofXetals. (Ed.) Clarkson, T. QT., Nordberg, G. F. & Sager, P. R. pp. 607631. New York: Plenum Publishers. Haust, M. D. (1958) Tetrahydrofuran (THF) for dehydration and infiltration. Laboratory Investigation, 7, 58-67. Hsu, M., Raine, L. & Fanger, H. (1981) A comparative study of the peroxidase antiperoxidase method and an avidin-biotin complex method for studying pollieptide ho&ones with radioimmundassay antibodies. American ?ournal of Clinical Patholonv. 75. 734-738. K&i, J. H.P. & Kojima, Yy(1987) Chemistry and biochemistry ofmetallothionein. Experientiu (SuppI.), 52,25-61. Korpela, H., Loueniva, R., Yxjantieikki, E. & Kauppila, A. (1986) Lead and cadmium concentrations in maternal and umbilical cord blood, amniotic fluid, placenta and amniotic membranes. American ?oollmal of Obstetrzcs and Gynecoloal, 155, 1086-1089. Kuhnert. B. R.. Kuhnert P. M. & Zarlineo. T. I. (1988) Associations between olacental cadmium and zinc and age and parity in pregnant women in smoke.’ Obitetkcs and Gynecology, 71, 67-76. Manci, E. A. & Blackbum, W. R. (1987) Regional variations in the levels of zinc, iron, copper and calcium in the term human placenta. Placenta, 8, 497-502.
CCIW et di: l.o~~rrl~zriiiim ~~f’.1I~~trrllf~thio,2e
355
Xl;s, A. & Sarkar, B. (1988) The metabolism of metals in rat placenta. Biolo&ul Truce E/ertmtResenrch, 18, lOl190. Nartey, U. O., Banerjee, D. & Cherian, M. G. (1987a) Immunohistochemical localisation of metallothionein in cell nucleus and cytoplasm of fetal human liver and kidney. Pathology, 19,233-240. Xartey, U. O., Cherian, M. G. & Banerjee, D. (198713) Immunohistochemical localisation of metallothionein in human thyroid tumours.:lmt~rirun.~oounzal ofPathology, 129, 177-182. Parizek, J. (1964) \.ascular changes at sites of estrogen biosynthesis produced by parenteral injection of cadmium salts: The destruction of the placenta by cadmium salt.3ooumal ofReprodz&m and Teratology, 7, 263-265. %marawickrama, G. P. & Webb, M. (1979) Acute effects of cadmium on the pregnant rat and embryo-fetal dc\elopment. Erliir~~nnrr~t~trrI Health Perspectiws, 9, 245-249. Scheuhammer, A. M. 8i Cherian, M. G. (1986) Quantification of metallothioneins by a silver-saturation method. fixir~//qfl’ ~11ti.lpplied Phnmrum/o~: 82, 417-425. \vadlkes, M. P., Poisner, A. M., Wood, G. W. & Klaassen, C. D. (1984) .2letallothionein-like proteins in human placenta and fetal membranes. Toximlo~~ andApplied Pharmacology, 74, 179-18-l. n’ier, P. J. & Miller, R. K. (1987) The pharmacokinetics of cadmium in the dually perfused human placenta Trqhdh~stRe.~r~~h, 2. 35 7-366. \\.ier,P. J., Miller, R. Ii., Maulik, D. & DiSant’Agnese, P. A. (1990) Toxicit) ofcadmium in the perfused human placenta. To.~-iu~b~~q~ uml.-lpp/ied Phormamlo~~, 105, 156-l 71. I eh, I-T. i& Kurman, R. J. (1989) Editorial. Functional and morphologic evpressions of trophoblasts. Ldorrrtoq~ /rmst/:‘ntro,~, 6 I( I -.3,
Cellular Localization of Metallothionein Human Term Placenta
in
R. A. GOYER, M. D. HAUST & M. G. CHERIAN Dqartment of Pathology, The Unizersir,) of W&tern Ontario, London, Ontario, Canada. N6.A 5C1 Paper accepted 17.12.1991
SUMh4ARY Cellular localization of metallothionein (MT) in placenta may provide information on its function as a metal binding protein. Rabbit antibodies to rat liver MT crossreacted with human MT and were used to localize MT in human term placenta by aridin-biotin peroxidase technique, Serial sections (5 p) were cut from parafinembedded placentae obtained at term from fiue normal women and incubated with rabbit antibodies to MT. Normal rabbit serum was used as a negatitle control. The slides were incubated with biotinylated swine anti-rabbit IgG (linking antibody) then with avidin-biotin horseradish peroxidase complex and daleloped with diaminobenzidine in hydrogen peroxide (0.03 per cent) substrate. The optimum staining ofMT was obtained at a I:800 antibody dilution. MT was identrjied in fetal amniotic cells, symytial trophoblasts and oillous interstitial cells, and in maternal decidual cells. The presence ofMTat specific cellular sites suggests that it may regulate the transplacental transport of metals such as zinc, copper and cadmium. Since the level of cadmium is lower and that of zinc and copper higher in fetal than in maternal blood, this may suggest that placental MT may restrict cadmium while enhancing zinc and copper transport. INTRODUCTION nletallothionein (MT) is a low molecular weight cysteine-rich protein which can bind with essential metals like zinc and copper and toxic metals like cadmium and mercury, Kagi and Kojima (1987). This protein may play a role in homeostasis of zinc and copper and may also provide protection from toxic effects of cadmium and mercury, Cherian and Goyer (1978). During mammalian development, the placenta plays a major role in the transfer of essential nutrients from the mother to the embryo and also protecting the embryo from toxic compounds. Although both zinc and copper are readily transferred to the embryo through the placenta, there is a limited transfer of cadmium across the placenta (Barlow and Sullivan, 1982; Dencker et al, 1983). Little is known about the mechanism of selective transfer of essential metals through the placenta. Cadmium is known to accumulate in the placenta and cause cellular damage which can result in teratogenic or embryotoxic effects (Parizek, 1964; \Yeir et al, 1990). It has been postulated from animal experiments that placental MT may play (1143-4004/92/040319
+ 07 $05.00/O
0
1992 Baillike
Tindall Ltd
350
Phcmto (I 992), Ld. 1.7
a role in cadmium retention and cause a decrease in transfer of zinc to embryo (Samawickrama and Webb, 1979). The presence of MT has been shown in human placenta and fetal membranes (Waalkes et al, 1984; Wier and Miller, 1987). The induced synthesis of MT in cultured human trophoblasts by both cadmium and zinc salts is also reported. In a more recent study, De et al (1989) demonstrated high levels of IMT mRNA levels in the syncytiotrophoblasts of mouse placenta by in situ hybridization. These levels were further increased following injection of cadmium and zinc. The consistently high levels of MT and mRNA were also detected in the decidua following early implantation. The present study was undertaken to localize the MT in full-term human placenta in order to elucidate its physiological role in the transplacental transport of essential and toxic metals.
MATERIALS
AND
METHODS
This study was based on examination of tissue blocks from placentae obtained from live, fullterm healthy infants. There were no medical complications during any of the pregnancies and the mothers were primiparous and ranged in age from 21 to 32 years old. All were nonsmokers and received no medications during pregnancy apart from iron and vitamin supplements. There was no history of previous occupational or environmental exposure to cadmium. The placentae were collected immediately after each delivery and were perfused for 5 min with normal saline. Two blocks of tissue were removed from each placenta from the paracentral area. The samples included the entire thickness of the organ, i.e., the basal (maternal) and the chorionic plates. Care was taken to include the thin amniotic layer superimposed upon the chorion. A portion of the sample was analysed for Cd, Zn, Cu by atomic absorption spectroscopy and metallothionein by a silver saturation method (Scheuhammer and Cherian, 1986). Parallel blocks of tissues were fixed in 10 per cent buffered formalin for 24 h, trimmed to a thickness of approximately 0.5 cm and subsequently fixed in the same but fresh solution for another 24-h period. The blocks of tissues were processed for paraffin embedding by the tetrahydrofuran (THF) method, Haust (1958). Three-micron thick sections were stained for orientation, and cellular and vascular identification with the following stains: hamalumphloxine-saffron (HPS); Masson’s trichrome; resorcin fuchsin, nuclear fast red; metanil yellow; and alcian blue; periodic acid Schiff (PAS), hematopxylin, orange G. Additional sections were cut at 5 ,U for localization of MT by the following method. Antibodies to MT were raised in rabbits after repeated injection of rat liver MT polymerized with glutaraldehyde (Banerjee et al, 1982). This polyclonal antibody cross reacted with both monomer and polymer of MT in various species including human (Nartey, Banerjee and Cherian, 1987a; Nartey, Cherian and Banerjee, 1987b). The unstained sections were incubated with 20 per cent normal swine serum at room temperature for 30 min incubated at room to block the non-specific binding sites. They were subsequently temperature for 60 min with rabbit anti-MT serum at four different dilutions (1: 200; 1: 400; 1: 800; 1: 1000). Sections incubated with normal rabbit serum served as (negative) controls. After washing, the slides were incubated with biotinylated swine-anti-rabbit IgG (linking antibody) for 30 min at room temperature, exposed to avidin-biotin horseradish peroxidase complex (as directed: ABC kit, Vector Laboratories, Inc., Burlington, CA.), and developed with diaminobenzidine (DAB) in 0.03 per cent hydrogen peroxide substrate according to the
C;r~yrrYIrrl: Localizatiot~ i,l~lfetallothione
351
method of Hsu, Raine and Fanger (1981). Selected sections were counterstained with hematoxylin for nuclear identification. Because of lack of documentation in the literature of the presence of MT in amniotic fluid and the need for this information to support the immunohistochemistry observations amniotic fluid was collected from six additional full-term, uncompleted deliveries from nonsmoking primiparous mothers. The amniotic fluid was centrifuged to remove cellular and other debris, and Cd was analysed by atomic absorption spectroscopy. Metallothionein was measured by immunoassay using peroxidase-antiperoxidase with a sensitivity of 1 ng MT.
RESULTS The intensity of positive staining with rabbit anti-MT antibody was related to concentration of antibody. Sections incubate with anti-MT serum at 1:200, 1:400 dilutions yielded in final preparations ‘heavy’ staining which largely obscured the cellular details and the mode of the cytoplasmic MT-distribution. On the other hand, the still positive but faint reaction in slides incubated at 1:lOOO dilution, rendered these unsuitable for study and photographic documentation. Thus, slides incubated at 1800 dilution were stained optimally for the purpose of the present study and photography (Figure la-d). Control sections incubated with non-immune rabbit serum did not stain. The reaction product had a yellow-brown staining quality and was located entirely within cytoplasm of cells. Staining for MT was demonstrated in syncytiotrophoblasts lining the villi [Figure 1 (a)], villous interstitial cells [Figure l(b)], in maternal decidual (interstitial) cells [Figure l(c)], and in amniotic epithelium [Figure 1 (d)]. Positive staining for MT was moderately consistent in these cells in all sections. However, intensity of staining of syncytiotrophoblast was variable, hardly visible in some cells but moderately intense in others. In some trophoblasts, staining was not perceptible. Staining of the villus interstitial cells increased with breadth of the villus. Small villi sparse in interstitial cells stained lightly, but density of staining increased with breadth of v-illus and amount ofinterstitial cells. The villus in Figure 1 (b) is of a larger calibre than that in Figure 1 (a) (terminal villus) and contains many more interstitial cells. Moreover, the blood v-essel in Figure 1 (b) is a small arteriole, whereas the blood vessel in Figure 1 (a) is a capillary. The intense staining of the amniotic epithelium was consistent in all sections. Metallothionein and metal content of five placentae used for immunocytochemistry are shown in Table 1. The levels of metals are similar to those reported by others, Korpela et al (1986); Kuhnert, Kuhnert and Zarlingo (1988); Brophy (1985); Manci and Blackburn (1987). Levels of Cd and MT in amniotic fluid from six additional deliveries is shown in Table 2. Comparable measurements of MT in amniotic fluid are not available for comparison.
DISCUSSION The present investigation demonstrates the localization of MT in both maternal and fetal cells in human placenta. The presence of MT within specific cells suggests that the protein is either synthesized within these cells or is absorbed from surrounding media. Synthesis of MT has been demonstrated in isolated trophoblasts and in decidual cells, but has not been studied specifically in other types of cells in placenta. Metallothionein has been mainly
Figure 1. Localization of metallothionein (MT) using immunoperoxidase method and anti-MT antibodies. 1Metallothionein stains yellow-brown in colour. (a) Metallothionein is in trophoblasts in outer layer of a terminal villous containing thin-walled capillaries (arrows), x 200. (b) Metallothionein is in cells in interstitium of villous (Hofbauer cells), x65. Arrow indicates a small arteriole. (c) Metallothionein is in cytoplasm of decidual cells in basilar area ofplacenta, x90. (d) Metallothionein is in cells lining amniotic membrane, x90.
(b)
2
2
$3 ?
Gnyrr et ~1:Localization of.Metallothione
353
Table 1. LMetallothionein
IMT ( P!3@ 30.0 k4.5
1.9 to.3
and metal placentae
content
of human
Zn
CU
(/G/P)
(Pupk)
12.9 24.5
0.8 kO.3
Five placentas were collected from five full-term deliveries. .Metals were analysed by atomic absorption spectrophotometry. Metallothionein was measured by immunoassay. Results are expressed as mean f s.d.
localized in epithelial cells in human fetal liver and in neoplastic cells by immunohistochemistry (Nartey, Banerjee and Cherian, 1987a; Nartey, Cherian and Banerjee, 1987b). The availability of a polyclonal antibody that readily cross-reacted with human MT enabled us to demonstrate the presence of MT in human placenta. The large dilution of the antibody to obtain the optimum condition of localization of MT in the present study reflects the measurable amount of MT present in human placenta. The trophoblast in the human has a wide range of metabolic and endocrinologic functions essential to the development of the placenta and support of the embryo-fetal development, Yeh and Kurman (1989). It facilitates transport of substances entering the placenta from maternal blood. These may include substances from ions to various growth factors, including MT for the metabolism of the essential metals zinc and copper. As in other cells that synthesize MT, mRNA for MT is increased in trophoblasts on exposure to cadmium and zinc demonstrating the transcriptional control of MT synthesis in these cells (De et al, 1989). Both zinc and copper are essential metals for the growth and development of the fetus and they are found in high concentrations in human and rodent fetal liver bound to MT during a certain stage in perinatal development. Normally no cadmium is detected in fetal or newborn mammalian liver (Chung, Nartey and Cherian, 1986), and placenta accumulates cadmium bound to MT. Mas and Sarkar (1988) also found Cd bound to high molecular weight proteins in the rat placenta. Since MT is a storage protein for metals, the binding of metals to MT in placenta can be considered as an intermediate step in the regulation of the transport from maternal to fetal blood. However, the mechanism of specific transport of the essential metals such as zinc and copper through the villous to fetal blood with little transport of cadmium which is also partly bound to placental MT, is not yet understood. It should be noted that the concentration of zinc @g) in placenta is about 1000 times more than cadmium (ng). Since cadmium has higher affinity for MT than zinc, the difference in magnitude of zinc
Table 2. Cadmium
Cd (ng/ml) 26.4 k 4.1
and metallothionein amniotic fluid
content
of
MT (&ml) 1.7 f 0.2
Amniotic fluid was collected from six full-term deliveries. Cd was analysed by atomic absorption spectrophotometry. Metallothionein was measured by immunoassay. Results are expressed as mean + s.d.
354
Placenta (1992), Vol. 13
concentration in placenta may result in a greater affinity for cadmium to placental MT and may facilitate increased transfer of zinc. Hofbauer cells are, in essence, motile macrophages capable of phagocytosis and protein ingestion. Whether MT in these cells reflects primary synthesis of absorption enroute from trophoblast to the villous capillary cannot be determined from this study, but increase in staining density with increase in size ofvilli suggests accumulation from transvillus transport from maternal blood to the fetal capillary. Amniotic epithelial cells are fetal derivatives in origin but, again, whether localization of MT in these cells reflects accumulation by absorption from amniotic fluid (presumably from fetal excretion) or from primary synthesis, is not known. Our data (Table 2) show the presence of MT in amniotic fluid, analysed from six different deliveries. Decidual cells are endometrial stromal cells that have been transformed under hormonal influence into large pale cells, rich in glycogen. The decidual cells identified as containing MT in this study are those of placenta basalis. The role of MT in these cells is not known. In summary, the results of the present study demonstrate the presence of MT in various cell types in full-term human placenta. Its presence in both maternal and fetal components of the placenta suggest that it may have a regulatory role in the specific transport and storage of essential metals such as zinc and copper.
ACKNOWLEDGEMENTS Linda Veinot provided technical assistance with immunohistochemical grant-in-aid from the Medical Research Council of Canada.
staining.
This research
is sponsored
by a
REFERENCES
Banerjee, D., Onosaka,
S. & Cherian, M. G. (1982) Immunochemical localization of metallothionein in cell nucleus and cytoplasm of rat liver and kidney-. Toxicolo~, 24, 95-105. Barlow, S. M. & Sullivan, F. M. (1982) Cadmium and its compounds. In Repro&&~ Hazards of Industrial Chemicals. (Ed.) Barlow, S. M. & Sullivan, F. M. pp. 136-177. London: Academic Press. Brophy, M. H., Harris, N. F. & Crawford, I. L. (1985) Elevated copper and lowered zinc in the placenta ofpreeclamptics. Clinica Chimica Acta, 145, 107-l 12. Cherian, M. G. & Goyer, R. A. (1978) Metallothioneins and their role in the metabolism and toxicity of metals. Mini-review. Clinical Science, 23, l-10. Chung, J., Nartey, N. 0. & Cherian, M. G. (1986) Metallothionein levels in liver and kidney of Canadians-A potential indicator of environmental exposure to cadmium. Archiw ofEnaironmenta1 Health, 41, 3 19-323. De, S. K., McMaster, M. T., Dey, S. K. & Andrews, G. K. (1989) Cell-specific metallothionein gene expression in mouse deciduo and placentae. Dtwlopment, 107, 611-62 1. Dencker, L., Danielsson, B., Khayat, A. & Lindgren, A. (1983) Disposition ofmetals in the embryo and fetus. In Reprodudizvand Developmental Toxici[y ofXetals. (Ed.) Clarkson, T. QT., Nordberg, G. F. & Sager, P. R. pp. 607631. New York: Plenum Publishers. Haust, M. D. (1958) Tetrahydrofuran (THF) for dehydration and infiltration. Laboratory Investigation, 7, 58-67. Hsu, M., Raine, L. & Fanger, H. (1981) A comparative study of the peroxidase antiperoxidase method and an avidin-biotin complex method for studying pollieptide ho&ones with radioimmundassay antibodies. American ?ournal of Clinical Patholonv. 75. 734-738. K&i, J. H.P. & Kojima, Yy(1987) Chemistry and biochemistry ofmetallothionein. Experientiu (SuppI.), 52,25-61. Korpela, H., Loueniva, R., Yxjantieikki, E. & Kauppila, A. (1986) Lead and cadmium concentrations in maternal and umbilical cord blood, amniotic fluid, placenta and amniotic membranes. American ?oollmal of Obstetrzcs and Gynecoloal, 155, 1086-1089. Kuhnert. B. R.. Kuhnert P. M. & Zarlineo. T. I. (1988) Associations between olacental cadmium and zinc and age and parity in pregnant women in smoke.’ Obitetkcs and Gynecology, 71, 67-76. Manci, E. A. & Blackbum, W. R. (1987) Regional variations in the levels of zinc, iron, copper and calcium in the term human placenta. Placenta, 8, 497-502.
CCIW et di: l.o~~rrl~zriiiim ~~f’.1I~~trrllf~thio,2e
355
Xl;s, A. & Sarkar, B. (1988) The metabolism of metals in rat placenta. Biolo&ul Truce E/ertmtResenrch, 18, lOl190. Nartey, U. O., Banerjee, D. & Cherian, M. G. (1987a) Immunohistochemical localisation of metallothionein in cell nucleus and cytoplasm of fetal human liver and kidney. Pathology, 19,233-240. Xartey, U. O., Cherian, M. G. & Banerjee, D. (198713) Immunohistochemical localisation of metallothionein in human thyroid tumours.:lmt~rirun.~oounzal ofPathology, 129, 177-182. Parizek, J. (1964) \.ascular changes at sites of estrogen biosynthesis produced by parenteral injection of cadmium salts: The destruction of the placenta by cadmium salt.3ooumal ofReprodz&m and Teratology, 7, 263-265. %marawickrama, G. P. & Webb, M. (1979) Acute effects of cadmium on the pregnant rat and embryo-fetal dc\elopment. Erliir~~nnrr~t~trrI Health Perspectiws, 9, 245-249. Scheuhammer, A. M. 8i Cherian, M. G. (1986) Quantification of metallothioneins by a silver-saturation method. fixir~//qfl’ ~11ti.lpplied Phnmrum/o~: 82, 417-425. \vadlkes, M. P., Poisner, A. M., Wood, G. W. & Klaassen, C. D. (1984) .2letallothionein-like proteins in human placenta and fetal membranes. Toximlo~~ andApplied Pharmacology, 74, 179-18-l. n’ier, P. J. & Miller, R. K. (1987) The pharmacokinetics of cadmium in the dually perfused human placenta Trqhdh~stRe.~r~~h, 2. 35 7-366. \\.ier,P. J., Miller, R. Ii., Maulik, D. & DiSant’Agnese, P. A. (1990) Toxicit) ofcadmium in the perfused human placenta. To.~-iu~b~~q~ uml.-lpp/ied Phormamlo~~, 105, 156-l 71. I eh, I-T. i& Kurman, R. J. (1989) Editorial. Functional and morphologic evpressions of trophoblasts. Ldorrrtoq~ /rmst/:‘ntro,~, 6 I( I -.3,
