Fish Physiology and Biochemistry vol. 1 no. 3 pp 153-162 (1986) Kugler Publications, Amsterdam/Berkeley
Ultrastructural changes in the parrotfish thyroid after in vitro stimulation with bovine thyrotropin Carol Johnson Smith and E. Gordon Grau Department of Zoology and Hawaii Institute of Marine Biology University of Hawaii Honolulu, Hawaii 96822 Keywords: thyroid gland, teteost, ultrastructure, thyrotropin, thyroid hormone, organelles, morphology, parrotfish (Scarus dubius), stereology.
Abstract
Six groups o f thyroid glands of Scarus dubius were examined and compared by electron microscopy after an in vitro culture for 4h with graded doses o f bovine thyrotropin (bTSH). Five doses of bTSH were used
encompassing the full range of the dose-response curve developed for this tissue. Upon electron microscopic examination, micrographs were taken randomly and at the same magnification, and three intracellular inclusions were quantified. The relative surface density o f rough endoplasmic reticulum (rER) and the relative surface area of tysosomes and engulfed colloid droplets were recorded for each group. Three treatment groups, (1) control, no bTSH, (2) tissues exposed to 1 m l U / m l bTSH, and (3) tissues exposed to 2 m l U / m l bTSH, did not differ from each other in the quantified organelles nor in general appearance. Overall, these three groups were similar in appearance to the ultrastructure described in other teleosts except for a lack of flagellated cells. Compared to the first three groups, treatment with 5 m l U / m l bTSH, increased the density of rER, and the proportion of cell area occupied by lysosomes and engulfed colloid. This group also possessed either more microvilli or pseudopods at the lumenal surface of the follicular epithelium. After exposure to 10 m l U / m l bTSH there was an even greater increase in surface density of rER, and in surface area occupied by lysosomes and engulfed colloid droplets. The apical portion of this group was highly irregular, commonly displaying pseudopods. Group (6), (20 mlU/ml), showed a decline in cytoplasm in comparison to group (5) with many epithelial cells breaking apart. A few cells in this group were still intact but contained huge engulfed colloid droplets which extended from the basal to apical borders. This first detailed description suggests that the teleost thyroid gland undergoes ultrastructural changes with exogenous TSH stimulation in a manner similar to that seen in higher vertebrates.
Introduction
Thyroid hormones have been associated with a variety o f activities in teleostean fishes, including reproduction (Sage 1973; Hurlburt 1977; Chakraborti and Bhattacharya 1984), oxygen consumption (Pandey and Munshi 1976), and with parr-smolt transformation in certain migratory salmonids (Folmar and Dickhoff 1980; Nishioka et al. 1982).
However, the specific roles thyroid hormones play in these activities are often difficult to define. One reason that makes thyroid research difficult in teleosts, is that in many species, the thyroid follicles are scattered throughout the pharyngeal region. This diffuse nature of the gland impedes the removal of the tissue for experimentation. For the same reason, ultrastructural studies of teleost thyroids have also been infrequently at-
154 tempted. Fujita and Machino (1965), Fujita et al. (1966), and Suemasa et al. (1968), have described the fine structure of four teleosts, Seriola quinqueradiata, Anguilla japonica, Semicossyphus reticulatus, and Sebastiscus marrnoratus. A n ontogenetic approach has been taken by Takagoshi (1975), who described the development of the thyroid in the goby, Chaenogobius urotaenia, by electron microscopy. Nishioka et al. (1982), compared the thyroid ultrastructure of normal and growth-impaired coho salmon, Oncorhynchus kisutch, while Leatherland et al. (1978), described goitered thyroid tissue in the same fish. All of these studies involved description of the gland without exogenous stimulation. It is well known that the mammalian thyroid gland can be stimulated by the pituitary hormone, thyrotropin (TSH), to produce and secrete thyroid hormone (T4) (cf Fujita 1975, 1984). This stimulation results in ultrastructural changes which have been described in many mammals including rats (Wissig 1963; Wetzel et al. 1965), pigs (Miyagawa et al. 1983), mice (Payer et al. 1981) and guinea pigs (Kosanovic et al. 1968). In teleosts, pituitary control of thyroid function has also been established (Fontaine 1969; Gorbman 1969; Sage and Bern 1971; Grau and Stetson 1977; Ng et al. 1982). Under light microscopy, exposure of teleost (goldfish) thyroid to TSH elicits certain morphological changes indicative of stimulated activity (Gorbman 1940; Ortman and Billig 1966). Nevertheless, to our knowledge, the actions of TSH on ultrastructural features of the teleost thyroid have not been addressed in detail in the published literature (cf Lai et aL 1980). The ultrastructure of a gland provides an excellent assessment of its endocrine activity. In the case of the thyroid, an increase in the amount or volume of rough endoplasmic reticulum (rER), Golgi vesicles, and mitochondria has been suggested to indicate an increase in the synthesis of thyroglobulin, the inactive form of thyroid hormone that is stored in the lumen as colloid (Stein and Gross 1964; Fujita 1969). Likewise, an increase in the number of microvilli or the formation of pseudopods is believed to point to an increase in endocytotic and exocytotic activity (Miyagawa et al. 1983). Finally, an
increase in the number or size of large colloid droplets which have been phagocytosed from the lumen, and an increase in lysosomes is widely held to suggest an enhancement of hydrolysis of thyroglobulin to yield more free thyroid hormone for release into circulation (Sheldon et al. 1964; Stein and Gross 1964; Ekholm and Smeds 1966; Fujita 1969). All of these organelles or inclusions can be indicators then, of a particular function of endocrine activity (i.e. synthesis, transport, or hydrolysis). This experiment demonstrates the effects of five different doses of TSH on the ultrastructure of the thyroid gland of the Hawaiian parrotfish, Scarus dubius. The doses encompass the dose-response curve developed for these fish (Grau et al. 1986) in an attempt to characterize ultrastructural changes occurring in the thyroid following moderate and strong stimulation. Stereological principles were used to express some of the ultrastructural changes in an objective manner and to more accurately assess the morphological changes which would be difficult to illustrate with only a few micrographs. Scarus dubius was chosen because it's thyroid gland is not diffuse, but discrete and compact. This facilitates its quick removal and provides the assurance of a relatively homogeneous tissue.
Materials and methods Tissue culture
For each fish, the tail was severed and the blood was collected from the caudal artery for future assays. Thyroid lobes were then dissected and trimmed to equal sizes from juvenile parrotfishes and maintained under a gas mixture of 95o70 02/5°70 CO2 at 28°C in a Krebs bicarbonate Ringer's solution. Eagles MEM (GIBCO: 2 ml of 50×/100 ml), L-glutamine (29 ml/100 ml), glucose (50 ml/100 ml), and penicillin (6 mg/100 ml) were added to the Ringer's solution. Components of this media were modified to conform closely to the level of constituent ions (except bicarbonate) which were measured in the plasma of Scarus dubius. The concentrations of these components were as follows: NaC1
155 120.00 mM, KCI 2.35 mMM, CaC12 4.20 mM, MgSO4 1.40 mM, KH2PO4 1.25 mM, and NaHCO3 25.00 mM. Bovine thyrotropin was purchased from Sigma Chemical Company (St. Louis, MO), and dissolved in 0.9% NaCI to a concentration of 5 IU/ml. In this form, the bovine TSH was diluted in the culture medium to the desired concentrations. The tissues were divided into 6 groups consisting of 4 glands each. These tissues were incubated for 4h, group (1) without the addition of bTSH, group (2) with 1 mlU/ml bTSH, group (3) with 2 mlU/ml bTSH, group (4) with 5 mlU/ml bTSH, group (5) with I0 mlU/ml bTSH, and group (6) with 20 mlU/ml bTSH. In addition to these groups, thyroid tissue freshly excised, was fixed and examined to compare to the cultured controls.
reticulum (rER), large less-dense colloid inclusions, and lysosomes. These structures were chosen because quantification of rER provides an assessment of thyroglobulin synthesis while engulfed colloid and lysosomes provide an estimate of thyroglobulin transport and hydrolysis. Micrographs of approximately 50 cells in each group were used in the stereological analysis. Relative surface density of rER, contained within a cell profile, was measured by counting the intersections of rER membrane with an 18-line test system. Relative surface area of colloid inclusions and lysosomes within individual cells, was measured by counting the grid points of a 9:1 double lattice enclosed in the respective organelles. Because shrinkage or swelling of the tissue during preparation may distort the structures somewhat, the quantification is expressed in relative terms.
Electron microscopy Immediately after the 4h culture, the thyroid glands were fixed overnight in a 2% formaldehyde, 2.50/o glutaraldehyde fixative buffered with 0.1 M cacodylate at pH 7.4. The glands were post-fixed with 1% osmium tetroxide in 0.2 M cacodylate for 2 hours at O°C. They were then stained en bloc with 2% aqueous uranyl acetate overnight. After dehydration with increasing acetone/water mixtures, the tissues were infiltrated and embedded using Spurr's resin (Spurr t969). Thick sections (t/zm) were stained with Richardson's stain and examined under a light microscope. Ultrathin sections were cut on a Reichert Om U2 ultramicrotome, and placed on 200 mesh copper grids. The sections were stained with uranyl acetate and lead citrate and examined with a Phillips 201 electron microscope at 60 KV. Stereological e~;aluation Thrce int,racellular structures were quantified following the stereological principles of Weibel (1969). Stereological evaluation of cellular inclusions which is borrowed from geological methodology, provides a substantial improvement over previously available procedures (Weibel 1969; Williams 1977). The three structures included rough endoplasmic
Results
The thyroid gland in the parrotfish, Scarus dubius, consists of two lobes located along the ventral aorta. The anterior lobe is near the bifurcation of the first afferent branchial arteries while the posterior lobe is between the second and third afferent branchial arteries (Grau et al. 1986). In an earlier study (Smith, unpublished), there appeared to be no significant ultrastructural difference between the two lobes so the results for both lobes were pooled together. In this study, there was also no significant ultrastructural difference between groups (1), (2), and (3) in the organelles and structures which were quantified (Fig. 7, 8, 9) nor was there any apparent difference between freshly excised tissue and the cultured controls. Groups (1), (2), and (3) are therefore discussed together. Groups (1), (2), (3). Overall the follicular epithelium cells in these groups give the impression that they are moderately active. The apical border is fairly regular with very few microvilli extending into the lumen (Fig. 1). Junctional complexes are present in the apical-lateral boundaries between neighboring cells (Fig. 2). The basal membrane has a regular and otherwise unremarkable outline. The nuclei are located basally (Fig. 1). Their
156
Fig. 1. Follicular cells cultured with no bTSH. Rough endoplasmic reticulum surrounds the nucleus (N). The large tess dense droplets (c) are believed to be colloid inclusions. Note the scarcity of microvilli (my). L, lumen; m, mitochondria, 8800x.
Fig. 2. A follicular cell cultured without exogenous TSH. High magnification shows the junctional complexes in the apicallateral boundaries and the distribution of rough endoplasmic reticulum. Free ribosomes are scattered throughout the cytoplasm. L, lumen; JC, junctional complex; N, nucleus. 23,000x. Fig. 3. A follicular cell cultured without exogenous TSH. The moderately developed Golgi complex (G) and moderately-dense secretory vesicles can be seen in the subapical region. Some vesicles can be seen in the apical region as well. L, lumen. 23,000x.
157
Fig. 4. Two cells stimulated with 5 mlU/ml bTSH. There is an increase in rough endoplasmic reticulum and colloid inclusions (c). There also appears to be an increase in microvilli (mv). N, nucleus; L, lumen. 8800x. Fig. 5. Follicular cells cultured with 10 mlU/ml bTSH. The apical portion bulges into the lumen and is highly irregular. C, colloid inclusions; L, lumen; m, mitochondria; N, nucleus; ly, lysosome. 7800 x .
158 lobular shape, allowing for more surface area, suggests metabolically active cells. The nuclei are surrounded by rough endoplasmic reticulum which is moderately developed and located apically as well as basally and laterally (Figs 1 and 2). Ribosomal rosettes are scattered throughout the cytoplasm. The moderately developed Golgi apparatuses are found in the supranuclear region of the cell (Fig. 3). Nearby are many small moderately-dense vesicles. Large less-dense droplets are also present in the apical and subapical areas (Fig. 1). Many of the large less-dense droplets contain material similar to the colloid in density, along with dense inclusions a n d / o r membranous structures. Small dense granules which may be primary lysosomes are infrequently seen. Mitochondria are distributed evenly within the cells. Group (4). Group (4) was cultured with 5 m l U / ml bTSH and shows an augmentation of cytoplasm in comparison to groups (1), (2), and (3) (Fig. 4). The rough endoplasmic reticulum has significantly increased in relative surface density while the large less-dense colloid droplets, and lysosomes have significantly increased in relative surface area (Figs 7, 8, 9). Engulfed colloid droplets which are the same density as the lumenal colloid are more commonly found (Fig. 4). The apical border is less regular, having either more microvilli, or an occasional pseudopod present. Other ultrastructural structures show no apparent change from the controls. Group (5). This group, which was cultured with 10 mIU/ml bTSH, shows a striking difference in morphology from the previous groups. The cell height increases significantly to form columnar cells. In many of these cells, the apical portion of the cytoplasm bulges into the lumen (Fig. 5). The apical border is highly irregular (Fig. 5), and is commonly formed into pseudopods. The extent of the rER is greatly increased (Fig. 7). The small moderately-dense vesicles associated with the Golgi complex are located in the apical region, and appear to occupy a far greater area than in the previous two groups (Fig. 5). The large lessdense droplets are generally located more apically than before, and occupy a larger fractional area than the lower-dose groups (Fig. 8). The large droplets are heterogeneous and often contain
membrane-bound, dense vesicles which are probably lysosomes. The fractional surface area of these lysosomes is also augmented in this group (Fig. 9). Group (6). The thyroids of this group were cultured with 20 m I U / m l bTSH and again drastic changes are evident. Most of the cells are broken apart (Fig. 6). Rough endoplasmic reticulum is frequently seen in the follicular lumen. The membranes between neighboring cells are also broken, therefore, in most cases, the borders were not detectable. A few cells were found intact. These show a decrease in cell height in comparison to group (5). Huge less-dense colloid droplets are present in the intact cells which occupy almost the entire cell from basal to apical borders. This indicates that cell lysis from over-stimulation may result from the endocytose of more colloid than can be contained within the cell boundaries. Because most cells examined in this group were broken, stereological analysis was not considered.
Discussion
The thyroid of juvenile parrotfish appears moderately active in culture without exogenous stimulation. This is perhaps not surprising inasmuch as it is likely that these fish were actively growing, a process which seems to involve thyroid hormone (Pickford and Atz 1957; Narayansingh and Eales 1975). The cells of the unstimulated tissues are similar to the ultrastructural descriptions of other teleosts (Fujita and Machino 1965; Fujita et al. 1966), except that no flagellated cells or centrioles were seen. Nevertheless, the great increase in the density and extent of rER, engulfed colloid droplets and lysosomes in the tissues exposed to 10 mIU/ml bTSH has not been described in detail in teleosts until now, although Leatherland et al. (1978), described extensive rER in goitered coho salmon thyroid tissue. The cellular lysis seen in the tissue exposed to 20 m I U / m l bTSH has also not been reported previously. Two explanations of this phenomenon include: 1) the lysis may be an event similar to follicular aging in which older follicles are thought to break apart (cf Eales 1979), and 2) the lysis may be the
159 3 0 i" ' '
...
t 20
(/)
E
0%+..
0
1
I
1
I
In t'" I
I
I
I
'i'
I
0 1 234567891011
bTSH (mlU/ml) Fig. 6. A typical thyroid cell after culture with 20 mlU/ml bTSH. The apical portion has broken away with the apical contents scattered in the lumen (L). The boundaries between neighbouring cells has also broken, probably due to over-stimulation with bTSH. N, nucleus. 8800 x .
pathological consequence of overstimulation by TSH. We do not believe that the lysis is merely an artifact for two reasons. First, lysis occurred in tissues exposed to the highest dose of bTSH though all tissues were processed concurrently. Second, unlysed cells, though infrequent, showed a single large colloid droplet similar to the vacuolization seen in the lysed cells. This suggests that lysis may result from the endocytosis of more colloid than can be contained within cell boundaries. We cannot rule out the possibility, however, that tissue disruption after treatment with 20 m I U / m l bTSH occurred as an artifact o f processing consequent to culture. If true, this interpretation would suggest that thyroid tissues exposed to high dose of bTSH are extremely fragile compared with more moderately stimulated tissues. This is the first ultrastructural study in which a teleost thyroid gland has been stimulated in vitro.
Fig. 7. Surface density of rough endoplasmic reticulum measured with an 18 line test system (Weibel 1969) in Scarus dubius thyroid glands. Five groups of glands, each group cultured with a different dose of bTSH, were used in this analysis. Means are displayed with 95% confidence intervals,
Although it is probable that the Sigma bTSH may contain small quantities of contaminantes such as gonadotropins, and that these other glycoproteins may stimulate the parrotfish thyroid gland as well as thyrotropin, we assume that all glycoproteins which stimulate the thyroid gland do so by activating the adenylate cyclase system after binding with an appropriate receptor (cf Norris 1980). The intraceUular events from that point on, would not be dependent therefore on whether the receptor was activated by a thyrotropin or a gonadotropin. In an in vivo study on the coho salmon, Lai et aL (1980) report increases in rough endoplasmic reticulum, the Golgi complex, dense vesicles, and mitochondria upon stimulation with ovine thyrotropin. Although micrographs were not present in this publication, their stated results are similar to ours with the exception of increased mitochondria. In a previous study using doses of 0.2 m l U / m l bTSH, 10 mIU/ml bTSH, and a control group, mitochondria were quantified for the parrotfish thyroids. There
160
a2 o
6o
'~
10
°° ,o
I
!
tr~
0 me,
0 1 234567891011
bTSH (ml U/ml)
0 1 234567891011
brSH (ml U/ml)
Fig. 8. Surface area of engulfed colloid droplets measured with a 9:1 double lattice (Weibel 1969)in Scarus dubius thyroids. Five
Fig. 9. Surface area of lysosomes measured with a 9:1 double lattice (Weibel 1969) in Scarus dubius thyroids. Five groups of
groups of glands, each stimulated with a different dose of bTSH, were used in this analysis. Means are displayed with 95% confidence intervals.
glands, each cultured with a different dose of bovine TSH, were used in this analysis. Means are displayed with 95% confidence intervals.
was no significant difference in the number of mitochondria between the three groups (Smith, unpublished). In this earlier study, the glands cultured in 0.2 m l U / m l bTSH showed no ultrastructural difference from the controls while the group exposed to 10 m l U / m l bTSH appeared identical to the tissue cultured with the same dose in the present study. In the parrotfish thyroids, the marked increase in size of the cells and in amounts of rER and secretory granules from the Golgi complex suggests an increase in thyroid hormone production. The formation of pseudopods and the increase in the number of microvilli and fractional surface area of engulfed colloid found in the groups stimulated with 5 and 10 m l U / m l bTSH, indicate the stimulation of exocytosis and endocytosis of thyroglobulin to and from the lumen. These two groups also showed an increased surface area occupied by lysosomes, suggesting that bTSH also stimulates hydrolysis of thyroglobulin and subsequent release of thyroid hormone.
Other studies in this laboratory have shown that T4 release is enhanced in vitro in a dose-related manner by bTSH (Grau et al. 1986). The first increase in thyroid hormone release from the controls occurs when the tissue is cultured in 2 m l U / m l bTSH. At this same dose, there was no ultrastructural difference between this tissue and the controls, so there appears to be a lag between thyroid hormone release and organelle augmentation. The dose-response curve peaks at about 10 m l U / m l bTSH which is the level at which the ultrastructural organelles are also at their peak in this experiment. At 20 m l U / m l bTSH, the release of T4 shows a downward trend. The ultrastructural results suggests that this declining trend is due in part, to the cell lysis from overstimulation.
Acknowledgemenls This material is based upon work supported under
161 a National Science Foundation Graduate Fellowship to C.J. Smith, National Science Foundation Grant PCM-83-14294 to E.G. Grau, and U.H. Sea Grant NOAA NA 81AA-D-00070 to A. Fast and E.G. Grau. We also gratefully acknowledge Dr. F. Kazama and the U.H. Dept. o f Botany for their instruction in electron microscopy.
References cited Chakraborti, R. and Bhattacharya, S. 1984. Plasma thyroxine levels in freshwater perch: influence of season, gonadotropins, and gonadal hormones. Gen. Comp. Endocrinol. 53: 179186. Eales, J. 1979. Thyroid functions in cyclostomes and fishes, hi Hormones and Evolution, Vol. I. pp. 344-436. Edited by E.J. Barrington. Academic Press, London. Ekholm, R. and Smeds, S. 1966. On dense bodies and droplets in the follicular cells of the guinea pig thyroid. J. Ultrastruct. Res. 16: 71-82. Folmar, [.. and Dickhoff, W. 1980. The parr-smolt transformation (smoltification) and sea water adaptation in salmonids; a review of selected literature. Aquaculture 21: 1-37. Fontaine, Y. 1969. La specificitd zoologique des proteines hypophysaires capables de stimuler la thyroide. Acta Endocrinol. Suppl. 136: 1-154. Fujita, H. 1969. Studies on the iodine metabolism of the thyroid gland as revealed by electron microscopic autoradiography of ~251. Virch. Arch. Abt. B. Cellpathol. 2: 265-279. Fujita, H. 1975. Fine structure of the thyroid gland. Int. Rev. Cytol. 40: 197-280. Fujita, H. 1984. Fine structure of the thyroid gland. In Ultrastructure of Endocrine Cells and Tissues. pp. 265-275. Edited by P.M. Motta. Martinus Nijhoff, Boston. Fujita, H. and Machino, M. 1965. Electron microscopic studies on the thyroid gland of a teleost, Seriola quinqueradiata. Anat. Rec. 152: 81-98. Fujita, H., Suemasa, H. and Honma, Y. 1966. An electron microscopic study of the thyroid gland of the silver eel, A nguillajaponica, (a part of a phylogenetic studies of the fine structure of the thyroid). Arch. Distol. Jap. 27: 153-163. Gorbman, A. 1940. Suitability of the common goldfish for assay of thyrotropic hormone. Soc. Exptl. Biol. Meal. 45: 772-773. Gorbman, A. 1969. Thyroid function and its control in fishes. In Fish Physiology. Vol. 2. pp. 241-274. Edited by W.S. Hoar and D.J. Randall. Academic Press, New York. Grau, E.G. and Stetson, M. 1977. Pituitary autotransplants in Fttndtthts heteroclitus effect on thyroid function. Gen. Comp. Endocrinol. 32:427-43 I. Grau, E.G., Helms, L.M.H., Shimoda, S.K., Ford, C.A., LeGrand, J. and Yamauchi, K. 1986. The thyroid gland of the Hawaiian parrotfish and its use as an in vitro model system. Gen. Comp. Endocrinol. (in press).
Hurlburt, M. 1977. Role of the thyroid gland in ovarian maturation of the goldfish, Carassius auratus L. Can. J. Zoo[, 55: 1906-1913. Kosanovic, M., Ekholm, R., Strandberg, U. and Smeds, S. 1968. The effect of TSH on the acid phosphatase thyroglobulin hydrolyzing activities in the guinea pig thyroid. Exp. Cell Res. 52: 147-160. Lai, K., Grau, E,G., Nishioka, R. and Bern, H. 1980. Influence of TSH upon structure of thyroid gland in coho salmon. Am. Zool. 20: 857. Leatherland, J., Moccia, R. and Sonstegard, R. 1978. Ultrastructure of the thyroid gland in goitered coho salmon (Oncorhynchus kisutch). Cancer Res. 38: 149-158. Miyagawa, J., Yamashita, K. and Fujita, H. 1983. Fine structural aspects of phagocytotic activity in inverted follicle cells of cultured porcine thyroids. Cell Tiss. Res. 232: 327-334. Narayansingh, T. and Eales, J. 1975. The influence of physiological doses of thyroxine on the lipid reserves of starved and fed brook trout, Salvelinus./bntinalis(Mitchill). Comp. Biochem. Physiol., 52B: 407-412. Ng, T., Idler, D. and Eales, J. 1982. Pituitary hormones that stimulate the thyroidal system in teleost fishes. Gen. Comp. Endocrinol. 48: 372-389. Nishioka, R., Bern, H., Lai, K., Nagahama, Y. and Grau, E. 1982. Changes in the endocrine organs of coho salmon during normal and abnormal smoltification-an electron-microscopic study. Aquaculture 28:21-38. Norris, D. 1980. Vertebrate Endocrinology. Lea and Febiger, Philadelphia. Ortman, R. and Billig, R. 1966. A re-examination of the gold fish microhistometric assay method for thyrotropin. Gen. Comp. Endocrinol. 6: 362-370. Pandey, B. and Munshi, J. 1976. Role of the thyroid gland in regulation of metabolic rate in an air breathing siluroid fish, Heteropneustesfossilis (Bloch). J. Endocrinol. 69: 421-425. Payer, A., Battle, C. and Peake, R. 1981. Ultrastructural and cytochemical effects of trypan blue on TSH stimulation of thyroid follicular cells. Cell Tiss. Res. 218: 547-556. Pickford, G. and Atz, J. 1957. The Physiology of the Pituitary Gland of Fishes. New York Zoological Society, New York. Sage, M. 1973. The evolution of thyroid function in fishes. Am. Zool. 13: 899-905. Sage, M. and Bern, H. 1971. Cytophysiology of the teleost pituitary. Int. Rev. Cytol. 31: 339-376. Seljelid, R. 1967. Endocytosis in thyroid follicle cells. IV. On the acid phosphatase activity in thyroid follicle cells, with special reference to the quantitative aspects. J. Ultrastruct. Res. 18: 237-256. Sheldon, J., McKenzie, J. and Van Nimwegan, D. 1964. Electron microscopic autoradiography. The localization of 1~z5 in suppressed and thyrotropin stimulated mouse thyroid gland. J. Cell Biol. 23: 200-205. Spurr, A. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. Uhrastruct. Res. 26:31-43. Stein, O. and Gross, J. 1964. Metabolism of ~"51 in the thyroid gland studied with electron microscopic atttoradiography. En-
162 docrinology 75: 787-798. Suemasa, H., Yoshoharu, H. and Fujita, H. 1968. Electron microscopic observations on the thyroid gland of two species of teteost, Semicossyphus reticulatus, and Sebastiscus marmoratus. (A part of pbylogenetic studies of the fine structure of the thyroid). Arch. Histol. Jap. 29: 363-375, Takagoshi, T. 1975. Electron microscopic studies on the development of the thyroid function in a goby, Chaenogobius urotaenia. Bull. Fac. Fish. Hokkaido Univ. 25: 283-290. Weibel, E. 1969. Stereological principles for morpbometry in electron microscopic cytology. Int. Rev. Cytol. 26: 235-302.
Wetzel, B., Spicer, S. and Wollman, S. 1965. Changes in fine structure and acid phosphatase localization in rat thyroid cells following thyrotropin administration. J. Cell Biol. 25: 593-618. Williams, M. 1977. Quantilative methods in biology. In Practical Methods in Electron Microscopy, pp. t -80. Edited by A. Glauert. North Holland Publishing Company, Amsterdam. Wissig, S. 1963. The anatomy of secretion in the follicular cells of the thyroid gland. 11. The effect of acute thyrotropic hormone stimulation on the secretory apparatus. J. Cell Biol. 16: 93-117.
Ultrastructural changes in the parrotfish thyroid after in vitro stimulation with bovine thyrotropin Carol Johnson Smith and E. Gordon Grau Department of Zoology and Hawaii Institute of Marine Biology University of Hawaii Honolulu, Hawaii 96822 Keywords: thyroid gland, teteost, ultrastructure, thyrotropin, thyroid hormone, organelles, morphology, parrotfish (Scarus dubius), stereology.
Abstract
Six groups o f thyroid glands of Scarus dubius were examined and compared by electron microscopy after an in vitro culture for 4h with graded doses o f bovine thyrotropin (bTSH). Five doses of bTSH were used
encompassing the full range of the dose-response curve developed for this tissue. Upon electron microscopic examination, micrographs were taken randomly and at the same magnification, and three intracellular inclusions were quantified. The relative surface density o f rough endoplasmic reticulum (rER) and the relative surface area of tysosomes and engulfed colloid droplets were recorded for each group. Three treatment groups, (1) control, no bTSH, (2) tissues exposed to 1 m l U / m l bTSH, and (3) tissues exposed to 2 m l U / m l bTSH, did not differ from each other in the quantified organelles nor in general appearance. Overall, these three groups were similar in appearance to the ultrastructure described in other teleosts except for a lack of flagellated cells. Compared to the first three groups, treatment with 5 m l U / m l bTSH, increased the density of rER, and the proportion of cell area occupied by lysosomes and engulfed colloid. This group also possessed either more microvilli or pseudopods at the lumenal surface of the follicular epithelium. After exposure to 10 m l U / m l bTSH there was an even greater increase in surface density of rER, and in surface area occupied by lysosomes and engulfed colloid droplets. The apical portion of this group was highly irregular, commonly displaying pseudopods. Group (6), (20 mlU/ml), showed a decline in cytoplasm in comparison to group (5) with many epithelial cells breaking apart. A few cells in this group were still intact but contained huge engulfed colloid droplets which extended from the basal to apical borders. This first detailed description suggests that the teleost thyroid gland undergoes ultrastructural changes with exogenous TSH stimulation in a manner similar to that seen in higher vertebrates.
Introduction
Thyroid hormones have been associated with a variety o f activities in teleostean fishes, including reproduction (Sage 1973; Hurlburt 1977; Chakraborti and Bhattacharya 1984), oxygen consumption (Pandey and Munshi 1976), and with parr-smolt transformation in certain migratory salmonids (Folmar and Dickhoff 1980; Nishioka et al. 1982).
However, the specific roles thyroid hormones play in these activities are often difficult to define. One reason that makes thyroid research difficult in teleosts, is that in many species, the thyroid follicles are scattered throughout the pharyngeal region. This diffuse nature of the gland impedes the removal of the tissue for experimentation. For the same reason, ultrastructural studies of teleost thyroids have also been infrequently at-
154 tempted. Fujita and Machino (1965), Fujita et al. (1966), and Suemasa et al. (1968), have described the fine structure of four teleosts, Seriola quinqueradiata, Anguilla japonica, Semicossyphus reticulatus, and Sebastiscus marrnoratus. A n ontogenetic approach has been taken by Takagoshi (1975), who described the development of the thyroid in the goby, Chaenogobius urotaenia, by electron microscopy. Nishioka et al. (1982), compared the thyroid ultrastructure of normal and growth-impaired coho salmon, Oncorhynchus kisutch, while Leatherland et al. (1978), described goitered thyroid tissue in the same fish. All of these studies involved description of the gland without exogenous stimulation. It is well known that the mammalian thyroid gland can be stimulated by the pituitary hormone, thyrotropin (TSH), to produce and secrete thyroid hormone (T4) (cf Fujita 1975, 1984). This stimulation results in ultrastructural changes which have been described in many mammals including rats (Wissig 1963; Wetzel et al. 1965), pigs (Miyagawa et al. 1983), mice (Payer et al. 1981) and guinea pigs (Kosanovic et al. 1968). In teleosts, pituitary control of thyroid function has also been established (Fontaine 1969; Gorbman 1969; Sage and Bern 1971; Grau and Stetson 1977; Ng et al. 1982). Under light microscopy, exposure of teleost (goldfish) thyroid to TSH elicits certain morphological changes indicative of stimulated activity (Gorbman 1940; Ortman and Billig 1966). Nevertheless, to our knowledge, the actions of TSH on ultrastructural features of the teleost thyroid have not been addressed in detail in the published literature (cf Lai et aL 1980). The ultrastructure of a gland provides an excellent assessment of its endocrine activity. In the case of the thyroid, an increase in the amount or volume of rough endoplasmic reticulum (rER), Golgi vesicles, and mitochondria has been suggested to indicate an increase in the synthesis of thyroglobulin, the inactive form of thyroid hormone that is stored in the lumen as colloid (Stein and Gross 1964; Fujita 1969). Likewise, an increase in the number of microvilli or the formation of pseudopods is believed to point to an increase in endocytotic and exocytotic activity (Miyagawa et al. 1983). Finally, an
increase in the number or size of large colloid droplets which have been phagocytosed from the lumen, and an increase in lysosomes is widely held to suggest an enhancement of hydrolysis of thyroglobulin to yield more free thyroid hormone for release into circulation (Sheldon et al. 1964; Stein and Gross 1964; Ekholm and Smeds 1966; Fujita 1969). All of these organelles or inclusions can be indicators then, of a particular function of endocrine activity (i.e. synthesis, transport, or hydrolysis). This experiment demonstrates the effects of five different doses of TSH on the ultrastructure of the thyroid gland of the Hawaiian parrotfish, Scarus dubius. The doses encompass the dose-response curve developed for these fish (Grau et al. 1986) in an attempt to characterize ultrastructural changes occurring in the thyroid following moderate and strong stimulation. Stereological principles were used to express some of the ultrastructural changes in an objective manner and to more accurately assess the morphological changes which would be difficult to illustrate with only a few micrographs. Scarus dubius was chosen because it's thyroid gland is not diffuse, but discrete and compact. This facilitates its quick removal and provides the assurance of a relatively homogeneous tissue.
Materials and methods Tissue culture
For each fish, the tail was severed and the blood was collected from the caudal artery for future assays. Thyroid lobes were then dissected and trimmed to equal sizes from juvenile parrotfishes and maintained under a gas mixture of 95o70 02/5°70 CO2 at 28°C in a Krebs bicarbonate Ringer's solution. Eagles MEM (GIBCO: 2 ml of 50×/100 ml), L-glutamine (29 ml/100 ml), glucose (50 ml/100 ml), and penicillin (6 mg/100 ml) were added to the Ringer's solution. Components of this media were modified to conform closely to the level of constituent ions (except bicarbonate) which were measured in the plasma of Scarus dubius. The concentrations of these components were as follows: NaC1
155 120.00 mM, KCI 2.35 mMM, CaC12 4.20 mM, MgSO4 1.40 mM, KH2PO4 1.25 mM, and NaHCO3 25.00 mM. Bovine thyrotropin was purchased from Sigma Chemical Company (St. Louis, MO), and dissolved in 0.9% NaCI to a concentration of 5 IU/ml. In this form, the bovine TSH was diluted in the culture medium to the desired concentrations. The tissues were divided into 6 groups consisting of 4 glands each. These tissues were incubated for 4h, group (1) without the addition of bTSH, group (2) with 1 mlU/ml bTSH, group (3) with 2 mlU/ml bTSH, group (4) with 5 mlU/ml bTSH, group (5) with I0 mlU/ml bTSH, and group (6) with 20 mlU/ml bTSH. In addition to these groups, thyroid tissue freshly excised, was fixed and examined to compare to the cultured controls.
reticulum (rER), large less-dense colloid inclusions, and lysosomes. These structures were chosen because quantification of rER provides an assessment of thyroglobulin synthesis while engulfed colloid and lysosomes provide an estimate of thyroglobulin transport and hydrolysis. Micrographs of approximately 50 cells in each group were used in the stereological analysis. Relative surface density of rER, contained within a cell profile, was measured by counting the intersections of rER membrane with an 18-line test system. Relative surface area of colloid inclusions and lysosomes within individual cells, was measured by counting the grid points of a 9:1 double lattice enclosed in the respective organelles. Because shrinkage or swelling of the tissue during preparation may distort the structures somewhat, the quantification is expressed in relative terms.
Electron microscopy Immediately after the 4h culture, the thyroid glands were fixed overnight in a 2% formaldehyde, 2.50/o glutaraldehyde fixative buffered with 0.1 M cacodylate at pH 7.4. The glands were post-fixed with 1% osmium tetroxide in 0.2 M cacodylate for 2 hours at O°C. They were then stained en bloc with 2% aqueous uranyl acetate overnight. After dehydration with increasing acetone/water mixtures, the tissues were infiltrated and embedded using Spurr's resin (Spurr t969). Thick sections (t/zm) were stained with Richardson's stain and examined under a light microscope. Ultrathin sections were cut on a Reichert Om U2 ultramicrotome, and placed on 200 mesh copper grids. The sections were stained with uranyl acetate and lead citrate and examined with a Phillips 201 electron microscope at 60 KV. Stereological e~;aluation Thrce int,racellular structures were quantified following the stereological principles of Weibel (1969). Stereological evaluation of cellular inclusions which is borrowed from geological methodology, provides a substantial improvement over previously available procedures (Weibel 1969; Williams 1977). The three structures included rough endoplasmic
Results
The thyroid gland in the parrotfish, Scarus dubius, consists of two lobes located along the ventral aorta. The anterior lobe is near the bifurcation of the first afferent branchial arteries while the posterior lobe is between the second and third afferent branchial arteries (Grau et al. 1986). In an earlier study (Smith, unpublished), there appeared to be no significant ultrastructural difference between the two lobes so the results for both lobes were pooled together. In this study, there was also no significant ultrastructural difference between groups (1), (2), and (3) in the organelles and structures which were quantified (Fig. 7, 8, 9) nor was there any apparent difference between freshly excised tissue and the cultured controls. Groups (1), (2), and (3) are therefore discussed together. Groups (1), (2), (3). Overall the follicular epithelium cells in these groups give the impression that they are moderately active. The apical border is fairly regular with very few microvilli extending into the lumen (Fig. 1). Junctional complexes are present in the apical-lateral boundaries between neighboring cells (Fig. 2). The basal membrane has a regular and otherwise unremarkable outline. The nuclei are located basally (Fig. 1). Their
156
Fig. 1. Follicular cells cultured with no bTSH. Rough endoplasmic reticulum surrounds the nucleus (N). The large tess dense droplets (c) are believed to be colloid inclusions. Note the scarcity of microvilli (my). L, lumen; m, mitochondria, 8800x.
Fig. 2. A follicular cell cultured without exogenous TSH. High magnification shows the junctional complexes in the apicallateral boundaries and the distribution of rough endoplasmic reticulum. Free ribosomes are scattered throughout the cytoplasm. L, lumen; JC, junctional complex; N, nucleus. 23,000x. Fig. 3. A follicular cell cultured without exogenous TSH. The moderately developed Golgi complex (G) and moderately-dense secretory vesicles can be seen in the subapical region. Some vesicles can be seen in the apical region as well. L, lumen. 23,000x.
157
Fig. 4. Two cells stimulated with 5 mlU/ml bTSH. There is an increase in rough endoplasmic reticulum and colloid inclusions (c). There also appears to be an increase in microvilli (mv). N, nucleus; L, lumen. 8800x. Fig. 5. Follicular cells cultured with 10 mlU/ml bTSH. The apical portion bulges into the lumen and is highly irregular. C, colloid inclusions; L, lumen; m, mitochondria; N, nucleus; ly, lysosome. 7800 x .
158 lobular shape, allowing for more surface area, suggests metabolically active cells. The nuclei are surrounded by rough endoplasmic reticulum which is moderately developed and located apically as well as basally and laterally (Figs 1 and 2). Ribosomal rosettes are scattered throughout the cytoplasm. The moderately developed Golgi apparatuses are found in the supranuclear region of the cell (Fig. 3). Nearby are many small moderately-dense vesicles. Large less-dense droplets are also present in the apical and subapical areas (Fig. 1). Many of the large less-dense droplets contain material similar to the colloid in density, along with dense inclusions a n d / o r membranous structures. Small dense granules which may be primary lysosomes are infrequently seen. Mitochondria are distributed evenly within the cells. Group (4). Group (4) was cultured with 5 m l U / ml bTSH and shows an augmentation of cytoplasm in comparison to groups (1), (2), and (3) (Fig. 4). The rough endoplasmic reticulum has significantly increased in relative surface density while the large less-dense colloid droplets, and lysosomes have significantly increased in relative surface area (Figs 7, 8, 9). Engulfed colloid droplets which are the same density as the lumenal colloid are more commonly found (Fig. 4). The apical border is less regular, having either more microvilli, or an occasional pseudopod present. Other ultrastructural structures show no apparent change from the controls. Group (5). This group, which was cultured with 10 mIU/ml bTSH, shows a striking difference in morphology from the previous groups. The cell height increases significantly to form columnar cells. In many of these cells, the apical portion of the cytoplasm bulges into the lumen (Fig. 5). The apical border is highly irregular (Fig. 5), and is commonly formed into pseudopods. The extent of the rER is greatly increased (Fig. 7). The small moderately-dense vesicles associated with the Golgi complex are located in the apical region, and appear to occupy a far greater area than in the previous two groups (Fig. 5). The large lessdense droplets are generally located more apically than before, and occupy a larger fractional area than the lower-dose groups (Fig. 8). The large droplets are heterogeneous and often contain
membrane-bound, dense vesicles which are probably lysosomes. The fractional surface area of these lysosomes is also augmented in this group (Fig. 9). Group (6). The thyroids of this group were cultured with 20 m I U / m l bTSH and again drastic changes are evident. Most of the cells are broken apart (Fig. 6). Rough endoplasmic reticulum is frequently seen in the follicular lumen. The membranes between neighboring cells are also broken, therefore, in most cases, the borders were not detectable. A few cells were found intact. These show a decrease in cell height in comparison to group (5). Huge less-dense colloid droplets are present in the intact cells which occupy almost the entire cell from basal to apical borders. This indicates that cell lysis from over-stimulation may result from the endocytose of more colloid than can be contained within the cell boundaries. Because most cells examined in this group were broken, stereological analysis was not considered.
Discussion
The thyroid of juvenile parrotfish appears moderately active in culture without exogenous stimulation. This is perhaps not surprising inasmuch as it is likely that these fish were actively growing, a process which seems to involve thyroid hormone (Pickford and Atz 1957; Narayansingh and Eales 1975). The cells of the unstimulated tissues are similar to the ultrastructural descriptions of other teleosts (Fujita and Machino 1965; Fujita et al. 1966), except that no flagellated cells or centrioles were seen. Nevertheless, the great increase in the density and extent of rER, engulfed colloid droplets and lysosomes in the tissues exposed to 10 mIU/ml bTSH has not been described in detail in teleosts until now, although Leatherland et al. (1978), described extensive rER in goitered coho salmon thyroid tissue. The cellular lysis seen in the tissue exposed to 20 m I U / m l bTSH has also not been reported previously. Two explanations of this phenomenon include: 1) the lysis may be an event similar to follicular aging in which older follicles are thought to break apart (cf Eales 1979), and 2) the lysis may be the
159 3 0 i" ' '
...
t 20
(/)
E
0%+..
0
1
I
1
I
In t'" I
I
I
I
'i'
I
0 1 234567891011
bTSH (mlU/ml) Fig. 6. A typical thyroid cell after culture with 20 mlU/ml bTSH. The apical portion has broken away with the apical contents scattered in the lumen (L). The boundaries between neighbouring cells has also broken, probably due to over-stimulation with bTSH. N, nucleus. 8800 x .
pathological consequence of overstimulation by TSH. We do not believe that the lysis is merely an artifact for two reasons. First, lysis occurred in tissues exposed to the highest dose of bTSH though all tissues were processed concurrently. Second, unlysed cells, though infrequent, showed a single large colloid droplet similar to the vacuolization seen in the lysed cells. This suggests that lysis may result from the endocytosis of more colloid than can be contained within cell boundaries. We cannot rule out the possibility, however, that tissue disruption after treatment with 20 m I U / m l bTSH occurred as an artifact o f processing consequent to culture. If true, this interpretation would suggest that thyroid tissues exposed to high dose of bTSH are extremely fragile compared with more moderately stimulated tissues. This is the first ultrastructural study in which a teleost thyroid gland has been stimulated in vitro.
Fig. 7. Surface density of rough endoplasmic reticulum measured with an 18 line test system (Weibel 1969) in Scarus dubius thyroid glands. Five groups of glands, each group cultured with a different dose of bTSH, were used in this analysis. Means are displayed with 95% confidence intervals,
Although it is probable that the Sigma bTSH may contain small quantities of contaminantes such as gonadotropins, and that these other glycoproteins may stimulate the parrotfish thyroid gland as well as thyrotropin, we assume that all glycoproteins which stimulate the thyroid gland do so by activating the adenylate cyclase system after binding with an appropriate receptor (cf Norris 1980). The intraceUular events from that point on, would not be dependent therefore on whether the receptor was activated by a thyrotropin or a gonadotropin. In an in vivo study on the coho salmon, Lai et aL (1980) report increases in rough endoplasmic reticulum, the Golgi complex, dense vesicles, and mitochondria upon stimulation with ovine thyrotropin. Although micrographs were not present in this publication, their stated results are similar to ours with the exception of increased mitochondria. In a previous study using doses of 0.2 m l U / m l bTSH, 10 mIU/ml bTSH, and a control group, mitochondria were quantified for the parrotfish thyroids. There
160
a2 o
6o
'~
10
°° ,o
I
!
tr~
0 me,
0 1 234567891011
bTSH (ml U/ml)
0 1 234567891011
brSH (ml U/ml)
Fig. 8. Surface area of engulfed colloid droplets measured with a 9:1 double lattice (Weibel 1969)in Scarus dubius thyroids. Five
Fig. 9. Surface area of lysosomes measured with a 9:1 double lattice (Weibel 1969) in Scarus dubius thyroids. Five groups of
groups of glands, each stimulated with a different dose of bTSH, were used in this analysis. Means are displayed with 95% confidence intervals.
glands, each cultured with a different dose of bovine TSH, were used in this analysis. Means are displayed with 95% confidence intervals.
was no significant difference in the number of mitochondria between the three groups (Smith, unpublished). In this earlier study, the glands cultured in 0.2 m l U / m l bTSH showed no ultrastructural difference from the controls while the group exposed to 10 m l U / m l bTSH appeared identical to the tissue cultured with the same dose in the present study. In the parrotfish thyroids, the marked increase in size of the cells and in amounts of rER and secretory granules from the Golgi complex suggests an increase in thyroid hormone production. The formation of pseudopods and the increase in the number of microvilli and fractional surface area of engulfed colloid found in the groups stimulated with 5 and 10 m l U / m l bTSH, indicate the stimulation of exocytosis and endocytosis of thyroglobulin to and from the lumen. These two groups also showed an increased surface area occupied by lysosomes, suggesting that bTSH also stimulates hydrolysis of thyroglobulin and subsequent release of thyroid hormone.
Other studies in this laboratory have shown that T4 release is enhanced in vitro in a dose-related manner by bTSH (Grau et al. 1986). The first increase in thyroid hormone release from the controls occurs when the tissue is cultured in 2 m l U / m l bTSH. At this same dose, there was no ultrastructural difference between this tissue and the controls, so there appears to be a lag between thyroid hormone release and organelle augmentation. The dose-response curve peaks at about 10 m l U / m l bTSH which is the level at which the ultrastructural organelles are also at their peak in this experiment. At 20 m l U / m l bTSH, the release of T4 shows a downward trend. The ultrastructural results suggests that this declining trend is due in part, to the cell lysis from overstimulation.
Acknowledgemenls This material is based upon work supported under
161 a National Science Foundation Graduate Fellowship to C.J. Smith, National Science Foundation Grant PCM-83-14294 to E.G. Grau, and U.H. Sea Grant NOAA NA 81AA-D-00070 to A. Fast and E.G. Grau. We also gratefully acknowledge Dr. F. Kazama and the U.H. Dept. o f Botany for their instruction in electron microscopy.
References cited Chakraborti, R. and Bhattacharya, S. 1984. Plasma thyroxine levels in freshwater perch: influence of season, gonadotropins, and gonadal hormones. Gen. Comp. Endocrinol. 53: 179186. Eales, J. 1979. Thyroid functions in cyclostomes and fishes, hi Hormones and Evolution, Vol. I. pp. 344-436. Edited by E.J. Barrington. Academic Press, London. Ekholm, R. and Smeds, S. 1966. On dense bodies and droplets in the follicular cells of the guinea pig thyroid. J. Ultrastruct. Res. 16: 71-82. Folmar, [.. and Dickhoff, W. 1980. The parr-smolt transformation (smoltification) and sea water adaptation in salmonids; a review of selected literature. Aquaculture 21: 1-37. Fontaine, Y. 1969. La specificitd zoologique des proteines hypophysaires capables de stimuler la thyroide. Acta Endocrinol. Suppl. 136: 1-154. Fujita, H. 1969. Studies on the iodine metabolism of the thyroid gland as revealed by electron microscopic autoradiography of ~251. Virch. Arch. Abt. B. Cellpathol. 2: 265-279. Fujita, H. 1975. Fine structure of the thyroid gland. Int. Rev. Cytol. 40: 197-280. Fujita, H. 1984. Fine structure of the thyroid gland. In Ultrastructure of Endocrine Cells and Tissues. pp. 265-275. Edited by P.M. Motta. Martinus Nijhoff, Boston. Fujita, H. and Machino, M. 1965. Electron microscopic studies on the thyroid gland of a teleost, Seriola quinqueradiata. Anat. Rec. 152: 81-98. Fujita, H., Suemasa, H. and Honma, Y. 1966. An electron microscopic study of the thyroid gland of the silver eel, A nguillajaponica, (a part of a phylogenetic studies of the fine structure of the thyroid). Arch. Distol. Jap. 27: 153-163. Gorbman, A. 1940. Suitability of the common goldfish for assay of thyrotropic hormone. Soc. Exptl. Biol. Meal. 45: 772-773. Gorbman, A. 1969. Thyroid function and its control in fishes. In Fish Physiology. Vol. 2. pp. 241-274. Edited by W.S. Hoar and D.J. Randall. Academic Press, New York. Grau, E.G. and Stetson, M. 1977. Pituitary autotransplants in Fttndtthts heteroclitus effect on thyroid function. Gen. Comp. Endocrinol. 32:427-43 I. Grau, E.G., Helms, L.M.H., Shimoda, S.K., Ford, C.A., LeGrand, J. and Yamauchi, K. 1986. The thyroid gland of the Hawaiian parrotfish and its use as an in vitro model system. Gen. Comp. Endocrinol. (in press).
Hurlburt, M. 1977. Role of the thyroid gland in ovarian maturation of the goldfish, Carassius auratus L. Can. J. Zoo[, 55: 1906-1913. Kosanovic, M., Ekholm, R., Strandberg, U. and Smeds, S. 1968. The effect of TSH on the acid phosphatase thyroglobulin hydrolyzing activities in the guinea pig thyroid. Exp. Cell Res. 52: 147-160. Lai, K., Grau, E,G., Nishioka, R. and Bern, H. 1980. Influence of TSH upon structure of thyroid gland in coho salmon. Am. Zool. 20: 857. Leatherland, J., Moccia, R. and Sonstegard, R. 1978. Ultrastructure of the thyroid gland in goitered coho salmon (Oncorhynchus kisutch). Cancer Res. 38: 149-158. Miyagawa, J., Yamashita, K. and Fujita, H. 1983. Fine structural aspects of phagocytotic activity in inverted follicle cells of cultured porcine thyroids. Cell Tiss. Res. 232: 327-334. Narayansingh, T. and Eales, J. 1975. The influence of physiological doses of thyroxine on the lipid reserves of starved and fed brook trout, Salvelinus./bntinalis(Mitchill). Comp. Biochem. Physiol., 52B: 407-412. Ng, T., Idler, D. and Eales, J. 1982. Pituitary hormones that stimulate the thyroidal system in teleost fishes. Gen. Comp. Endocrinol. 48: 372-389. Nishioka, R., Bern, H., Lai, K., Nagahama, Y. and Grau, E. 1982. Changes in the endocrine organs of coho salmon during normal and abnormal smoltification-an electron-microscopic study. Aquaculture 28:21-38. Norris, D. 1980. Vertebrate Endocrinology. Lea and Febiger, Philadelphia. Ortman, R. and Billig, R. 1966. A re-examination of the gold fish microhistometric assay method for thyrotropin. Gen. Comp. Endocrinol. 6: 362-370. Pandey, B. and Munshi, J. 1976. Role of the thyroid gland in regulation of metabolic rate in an air breathing siluroid fish, Heteropneustesfossilis (Bloch). J. Endocrinol. 69: 421-425. Payer, A., Battle, C. and Peake, R. 1981. Ultrastructural and cytochemical effects of trypan blue on TSH stimulation of thyroid follicular cells. Cell Tiss. Res. 218: 547-556. Pickford, G. and Atz, J. 1957. The Physiology of the Pituitary Gland of Fishes. New York Zoological Society, New York. Sage, M. 1973. The evolution of thyroid function in fishes. Am. Zool. 13: 899-905. Sage, M. and Bern, H. 1971. Cytophysiology of the teleost pituitary. Int. Rev. Cytol. 31: 339-376. Seljelid, R. 1967. Endocytosis in thyroid follicle cells. IV. On the acid phosphatase activity in thyroid follicle cells, with special reference to the quantitative aspects. J. Ultrastruct. Res. 18: 237-256. Sheldon, J., McKenzie, J. and Van Nimwegan, D. 1964. Electron microscopic autoradiography. The localization of 1~z5 in suppressed and thyrotropin stimulated mouse thyroid gland. J. Cell Biol. 23: 200-205. Spurr, A. 1969. A low viscosity epoxy resin embedding medium for electron microscopy. J. Uhrastruct. Res. 26:31-43. Stein, O. and Gross, J. 1964. Metabolism of ~"51 in the thyroid gland studied with electron microscopic atttoradiography. En-
162 docrinology 75: 787-798. Suemasa, H., Yoshoharu, H. and Fujita, H. 1968. Electron microscopic observations on the thyroid gland of two species of teteost, Semicossyphus reticulatus, and Sebastiscus marmoratus. (A part of pbylogenetic studies of the fine structure of the thyroid). Arch. Histol. Jap. 29: 363-375, Takagoshi, T. 1975. Electron microscopic studies on the development of the thyroid function in a goby, Chaenogobius urotaenia. Bull. Fac. Fish. Hokkaido Univ. 25: 283-290. Weibel, E. 1969. Stereological principles for morpbometry in electron microscopic cytology. Int. Rev. Cytol. 26: 235-302.
Wetzel, B., Spicer, S. and Wollman, S. 1965. Changes in fine structure and acid phosphatase localization in rat thyroid cells following thyrotropin administration. J. Cell Biol. 25: 593-618. Williams, M. 1977. Quantilative methods in biology. In Practical Methods in Electron Microscopy, pp. t -80. Edited by A. Glauert. North Holland Publishing Company, Amsterdam. Wissig, S. 1963. The anatomy of secretion in the follicular cells of the thyroid gland. 11. The effect of acute thyrotropic hormone stimulation on the secretory apparatus. J. Cell Biol. 16: 93-117.
