Ultrastructural alterations in thyrocytes of zebrafish Danio rerio
1
after exposure to propylthiouracil and perchlorate
2 3
Florian Schmidt*,
1,3Raoul Wolf,
2,3Lisa Baumann,
3and Thomas Braunbeck
34 5
1
BASF Schweiz AG, Product Safety, CH-4057 Basel, Switzerland 6
2
Section for Aquatic Biology and Toxicology (AQUA), Department of Biosciences, University 7
of Oslo, NO-0316 Oslo, Norway 8
3
Aquatic Ecology and Toxicology, Centre for Organismal Studies, University of Heidelberg, 9
D-69120 Heidelberg, Germany 10
11
* Corresponding author
12
Running title
13
Thyroid ultrastructure zebrafish 14
Schmidt et al.
15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35
Florian Schmidt, BASF Schweiz AG, Product Safety, Klybeckstrasse 141, CH-4057 Basel, 36
Switzerland, florian.schmidt@basf.com
37
Keywords: Endocrine disruption, perchlorate, propylthiouracil, thyroid, ultrastructure, zebrafish
38
Abstract
39
Histopathology is a widely used, powerful and sensitive approach to evaluate effects of endocrine-active chemicals
40
in the thyroid. However, effects at an ultrastructural level have hardly been examined in fish thyroids.
41
In the present study, we exposed zebrafish Danio rerio to sublethal concentrations of the known goitrogens
42
propylthiouracil (PTU; 0–50 mg/L) and perchlorate (PER; 0–5000 µg/L) for five weeks in a modified early life-
43
stage test. None of the treatments caused mortality, so survival NOECs of ≥ 50 mg/L (PTU) and ≥ 5000 µg/L
44
(PER) were determined.
45
PTU induced dose-dependent alterations in the rough endoplasmic reticulum (rER) in all exposure groups, while
46
only the exposure groups with the two highest PER concentrations (500 and 5000 µg/L) resulted in alterations of
47
the rER. Both substances caused an increase in the numbers of lysosomes and mitochondria, with mitochondria
48
displaying distorted cristae. Increased mitochondrial diameters were only observed in the PTU treatment. PER-
49
exposed zebrafish thyrocytes displayed an increase in apical microvilli. In the exposure group with the highest
50
PTU concentration (50 mg/L), first signs of degeneration were visible.
51
Ultrastructural changes in zebrafish thyrocytes thus appear specific for different chemicals, most likely depending
52
on their specific mode of action. Additional knowledge of subcellular changes in thyrocytes can help to better
53
understand and interpret existing histological data in the future.
54
Introduction
55
The thyroid system of vertebrates is essential for controlling growth and development, as well as certain aspects
56
of reproduction (Cyr & Eales, 1988; Leatherland, 1994; Power et al. 2001; Brown et al., 2004). To detect adverse
57
effects of thyroid-disruptive substances, an OECD guideline has been established in 2009, which uses South
58
African clawed frog Xenopus laevis as test organism (OECD, 2009). One crucial endpoint in this guideline is
59
thyroid histology; a powerful and sensitive tool for evaluating adverse effects of chemicals on thyroid functionality
60
(Grim et al., 2009).
61
As test organism, Xenopus laevis is perfectly suited. However, the most versatile and heterogeneous vertebrate
62
group regarding anatomy, physiology, reproduction behavior, and ecology are fish (Lagler et al., 1977; Janz, 2000;
63
Damstra et al., 2002). For today’s chemical regulatory purposes, zebrafish Danio rerio is one of the most important
64
test organisms. Consequently, the number of studies dealing with histological alterations in the thyroid of zebrafish
65
is increasing (Jianjie et al., 2016; Pinto et al., 2012; Sharma et al., 2013, 2016; van der Ven et al., 2006). Compared
66
to Xenopus, the zebrafish thyroid is challenging to investigate due to the loose distribution of follicles around the
67
ventral aorta in the pharyngeal region. Follicle size and number can vary between individuals. Thus, small changes
68
in follicular cell height, colloid consistency, or other parameters caused by sectioning or histologic artifacts can
69
easily be over-interpreted as treatment-related effects.
70
In contrast to light microscopical analyses, subcellular effects have not been widely documented, even though they
71
usually precede the manifestation of histologically observable alterations; both in terms of sensitivity (Braunbeck
72
et al., 1989) and time (Braunbeck & Storch, 1992). The present study was conducted to investigate ultrastructural
73
alterations in the thyroid gland of zebrafish after exposure to propylthiouracil (PTU) and perchlorate (PER), two
74
well-characterized thyroid-disruptive substances. For both, histopathological alterations in zebrafish thyroids have
75
been described in our previous studies from the same experiments as presented here (Schmidt & Braunbeck, 2011;
76
Schmidt et al., 2012). This allows further comparisons between histopathological and ultrastructural effects.
77
In the present study, PTU and PER were used to induce alterations in zebrafish thyrocytes. PTU blocks thyroid
78
peroxidase, whereas PER is a competitive inhibitor of the sodium-iodide symporter (Elsalini & Rohr, 2003;
79
Cooper, 2005; Opitz et al., 2005; van der Ven et al., 2006; Tietge et al., 2010, Wolff, 1998). Due to its specific use
80
as pharmaceutical, PTU is a popular reference substance for thyroid disruption. PER is frequently detected in the
81
environment due to both natural and anthropogenic sources (Kumarathilaka et al., 2016). Both substances have
82
adverse effects on different biological processes regulated by thyroid hormones in fish, e.g. the immune system
83
(Quesada-Garcia et al., 2016), eye development and function (Baumann et al., 2016), swim bladder infiltration
84
(Thienpont, 2011), embryonic development and behavior (Jomaa et al. 2014; Zahao, et al. 2013, 2014), and gonad
85
development (Mukhi, 2007; van der Ven, 2006; Sharma et al., 2013; Petersen et al., 2015). At light microscopical
86
level, PTU induces hyperemia, proliferations in epithelial cell height and stratification, while PER causes severe
87
colloid depletion and proliferations of average-sized thyroid follicles in exposed zebrafish (Schmidt & Braunbeck,
88
2011; Schmidt et al., 2012). Ultrastructural studies in mammals showed that exposure to methimazole and PTU
89
leads to cuboidal and columnar epithelial cells, with numerous microvilli in the apical region of the cells.
90
Additionally, alterations in mitochondria, proliferations in the rough endoplasmic reticulum (rER), and
91
accumulation of moderately dense vesicles could be detected (Fujita et al., 1963; Tsujio et al., 2007). After PER
92
exposure, tadpoles of sand toad Bufo arenarum displayed conspicuous proliferations of the endoplasmic reticulum
93
and Golgi complex, as well as increasing numbers of mitochondria and colloid droplets (Miranda et al., 1996).
94
The primary objective of the present study was to determine ultrastructural alterations in zebrafish thyrocytes after
95
exposure to PTU and PER. Samples originated from our previous histological studies on PTU and PER exposure
96
of zebrafish (Schmidt & Braunbeck, 2011; Schmidt et al., 2012). This allowed further comparisons between
97
ultrastructure and histology. To the best of our knowledge, this is the first study exploring ultrastructural alterations
98
in thyroid follicles of fish after exposure to substances with known thyroid-disruptive capacities. The results are
99
expected to improve our understanding and facilitate the interpretation of existing histological data on thyroid
100
disruption in zebrafish.
101
Material and methods
102
Chemicals
103
Unless stated otherwise, all chemicals were purchased from Sigma-Aldrich (Deisenhofen, Germany).
104 105
Experimental context and set-up
106
The experiments in the present study have been partly described in our previous communications (Schmidt &
107
Braunbeck, 2011; Schmidt et al. 2012). Thus, the following comparison of histological and ultrastructural
108
investigations refers to fish from the same experiments.
109 110
Animal husbandry
111
Fertilized zebrafish (Danio rerio, Westaquarium strain) eggs were obtained from in-house breeding facilities of
112
the Aquatic Ecology and Toxicology Group at the Centre for Organismal Studies, University of Heidelberg. All
113
experiments were conducted in compliance with institutional guidelines for the care and use of animals, as well as
114
permission by the regional animal welfare commission (AZ 35-9185.81/G-144/07). Fertilized eggs were initially
115
raised in 20 cm Petri dishes in a KB 115 incubator (Binder, Tuttlingen, Germany) at a temperature of 27 ± 1 °C
116
and 12:12 h light:dark cycle, while under exposure to the different exposure solutions. Three days after
117
fertilization, eggs were transferred into 10 L flow-through tanks (triplicate water change per day, 27 ± 1 °C, 12:12
118
h light:dark cycle, oxygen saturation > 80%). Flow-through conditions guaranteed ammonia, nitrite, and nitrate
119
levels below detection limits (0–5, 0.025–1, and 0–140 mg/L, respectively). After hatching, larvae were fed twice
120
daily with Sera Micron (Sera, Heinsberg, Germany) for one week. Afterwards, larvae were fed freshly raised
121
Artemia nauplii (Sanders, Mountain Green, UT, USA) ad libitum. Excessive food and feces were regularly
122
removed from the aquaria.
123 124
Exposure to thyroid-disruptive chemicals
125
Exposures were carried out under flow-through conditions with daily triplicate water exchange over a period of
126
five weeks. Sixty embryos were placed in each of two replicate tanks per exposure concentration and per treatment
127
substance (PTU: 0, 2.5, 10, 25, and 50 mg/L; PER: 0, 62.5, 125, 250, 500, and 5000 µg/L). In total, 1320 fish were
128
used. Both substances are known to be stable over time, as shown in former experiments including an inter-
129
laboratory study on Xenopus laevis performed in the framework of the validation of the amphibian metamorphosis
130
assay (AMA; OECD, 2004; 2007). For this reason, verifications of nominal concentrations were not performed.
131
Throughout the exposure, tanks were inspected daily for dead embryos, which were removed immediately.
132
Mortality was below 10 % in all tanks, and no significant differences among tanks could be observed. After five
133
weeks, fish were euthanized in a saturated solution of 4-ethylaminobenzoate (benzocaine). Before further
134
processing, weight and length of each fish were measured.
135 136
Ultrastructure
137
For ultrastructural studies, samples of the pharyngeal region from ten arbitrarily chosen individuals of each tank
138
were fixed in a solution of 2.5 % glutardialdehyde in 0.1 M sodium cacodylate buffer (pH = 7.4) at 4 °C for a
139
minimum of 24 h, and postfixed with 1% osmium ferrocyanide for 2 h (Karnovsky, 1971). After triplicate rinsing
140
in 0.1 M sodium cacodylate buffer (pH = 7.4), tissues were stained en bloc with 1 % uranyl acetate in 0.05 M
141
maleic buffer (pH = 5.2) overnight at 4 °C, dehydrated in a graded series of ethanol, and embedded in Spurr’s
142
medium (Spurr, 1969). For localization of the thyroid area, semi-thin sections (> 5 µm) were prepared on a
143
Reichert-Jung Ultracut microtome (Leica Microsystems, Nussloch, Germany) and stained with methylene
144
blue/azur II (Richardson et al., 1960). After verification, ultrathin sections (60 nm) were counterstained with
145
alkaline lead citrate (Reynolds, 1963) and examined in a EM 10 transmission electron microscope (Carl Zeiss,
146
Oberkochen, Germany).
147 148 149
Qualitative and quantitative morphometric evaluation and data analysis
150
Ten fish of each treatment (chosen from one replicate tank, as histological investigations did not reveal any
151
differences between the replicates) were examined for ultrastructural alterations in an un-blinded manner. Only
152
follicles directly bordering the ventral aorta close to its most anterior branching were selected from semi-thin
153
sections to account for heterogeneity of follicles in the zebrafish thyroid gland. For both qualitative and
154
quantitative analyses, three follicles sectioned at the equatorial midline were selected. In each of these three
155
follicles, three centrally sectioned thyrocytes were picked for further ultrastructural analyses.
156
For qualitative analyses, the following parameters were evaluated: (1) irregular outline of the nucleus, (2) the
157
amount of nuclear heterochromatin, (3) lipofuscinogenesis in lysosomes, (4) colloid inclusions, (5) electron
158
density of the colloid, (6) appearance of apical vesicles protruding into the colloid, (7) proliferations, fenestrations
159
and dilations of the rER, (8) increased endo- or exocytotic activity, (9) the appearance of microvilli, (10) necrotic
160
thyrocytes, and (11) alterations in cell shape.
161
For quantification of the major alterations after exposure to PTU and PER, five parameters were measured: (1)
162
total number of mitochondria, (2) number of mitochondria with distorted cristae, (3) diameters of the mitochondria,
163
(4) total number of lysosomes, and (5) height of the apical part of the cell. The latter was chosen as sensitive
164
parameter for the proliferation of cell organelles, which are mostly located in the apical part of thyrocytes. The
165
diameters of mitochondria were calculated as the longest distance in ovoid and elongated mitochondria. Apical
166
cell height was defined as the direct distance from the apical border of the nucleus to the apical cell membrane.
167
All measurements were performed using the free software tool ImageJ (version 1.44; National Institutes of Health,
168
USA).
169
For statistical analyses of quantitative measurements, lysosome count data was normally distributed for both PTU
170
and PER exposure and analyzed, following significant ANOVA results, using pairwise t-test comparisons with
171
Holm’s P value adjustment. All other data sets were not normally distributed and analyzed, following significant
172
Kruskal-Wallis rank sum tests, using pairwise Wilcoxon rank sum tests with Holm’s P value adjustment. All
173
statistical analyses were performed using open source statistical software R (version 3.2.4; R Core Team, 2016).
174
Differences were considered significant at P < 0.05 (*), highly significant at P < 0.01 (**), and highest significant
175
at P < 0.001 (***).
176
Results
177
Controls
178
Thyrocytes of control zebrafish in both treatment groups were morphologically comparable to each other (Table
179
1, Fig. 1). The epithelium enclosed a homogeneously stained colloid without any inclusions (Fig. 1a) and the flat
180
to cuboidal thyrocytes displayed a basally located nucleus with evenly dense chromatin and regularly shaped
181
nucleoli (Fig. 1b). Overall, the number of organelles was small; most organelles were located in the apical part of
182
the cells with an average height of approximately 1 µm (Fig. 1c, SI-Table 1). At the border to the follicle lumen,
183
few short microvilli were present (Fig. 1e). Under higher magnification, most mitochondria appeared in spherical
184
shapes with occasional ovoid and elongated shapes. The average diameter of mitochondria was approximately 500
185
nm (Figs. 1e, d, SI-Table 1). In both experimental groups, cross sections of exposure control thyrocytes contained
186
approximately 2 mitochondria, with occasionally distorted cristae (SI-Table 1). The endoplasmic reticulum
187
comprised slightly branched cisternae and Golgi fields were regularly present (Figs. 1d, e). In addition to Golgi
188
vesicles, electron-dense vesicles (most probably lysosomes) were visible in the apical part of the cells (Figs. 1b,
189
d, e). An average of approximately 18 lysosomes could be detected (SI-Table 1).
190 191
PTU exposure
192
Both weight and length of zebrafish showed a biphasic response pattern to PTU exposure. A significant decrease
193
in body length was observed in fish exposed to 50 mg/L, whereas a slight increase of weight was observed at 2.5
194
mg/L (Schmidt & Braunbeck, 2011).
195
At the ultrastructural level, thyrocytes displayed numerous dose-dependent effects (Tables 1 and 2, Figs. 7a, b, SI-
196
Table 1). Apical cell height increased with significant alterations in concentrations ≥ 10 mg/L (Table 2, Figs. 7a,
197
b, SI-Table 1). At 50 mg/L, condensed thyrocytes with electron-dense cytoplasm and shrunken nuclei were visible
198
(Fig. 2a), which could be regarded as a first sign of cell degeneration. The nuclei appeared irregular (Figs. 2a, b)
199
and the amount of heterochromatin was increased at 50 mg/L. Alterations of the rER included massive
200
proliferation, fenestration and dilation (Figs. 2a, 3a). This was evident in all exposure groups, and markedly
201
increased in exposure groups ≥ 25 mg/L (Figs. 2a, b). Mitochondria showed proliferations with significant increase
202
in exposure groups ≥ 25 mg/L (Table 2). Furthermore, the diameter of mitochondria was significantly increased
203
in all exposure groups; however, this effect did not seem to be dose-dependent (Figs. 3a, b, SI-Tables 1 and 2).
204
The mitochondria were characterized by extremely distorted cristae, associated with a strong inflation of the inter-
205
cristae space with significant increases in exposure groups ≥ 25 mg/L, where the majority of mitochondria were
206
affected (Table 2, Figs. 3a, b, Fig. 7b).
207
In apical cell regions, moderate proliferations of electron-dense bodies (most probably lysosomes) could be
208
detected (Fig. 2a, Table 1). These proliferations did not reveal a clear dose-dependency; however, the highest
209
exposure group (50 mg/L) showed a significant increase. In exposure groups ≥ 10 mg/L, the colloid was
210
interspersed with cytoplasmic inclusions (Fig. 4a). The electron density of the colloid remained intermediate, but
211
cloudy tinctorial properties were visible in the highest exposure group (50 mg/L). In exposure groups ≥ 2.5 mg/L,
212
protrusions of apical vesicles into the follicular lumen were observed, most prominently at 50 mg/L (Fig. 4b).
213
Together with the occurrence of apical vesicles, endo- or exocytosis was detected in exposure groups ≥ 10 mg/L
214
(Figs. 4b, c). Furthermore, the apical part of thyrocytes displayed short microvilli, which moderately increased
215
with PTU concentrations (Fig. 4b). Only at 10 mg/L, microvilli appeared slightly elongated.
216 217
PER exposure
218
The weight of fish did not show any significant changes. Only in exposure groups ≥ 500 µg/L, a slight decrease in
219
weight was observed, which was statistically not significant. The length of fish was significantly increased at 125
220
µg/L, whereas other exposure groups were not affected. However, the condition factor (mg/mm3) decreased
221
throughout exposure groups with significant alterations in exposure groups ≥ 125 µg/L (Schmidt et al., 2012).
222
Regarding ultrastructure, exposure to PER led to numerous, partially dose-dependent effects (Tables 1 and 2).
223
Epithelial cell height monotonously increased to columnar in the highest exposure group. Measurements of the
224
apical cell height revealed slight increases in exposure groups ≥ 62.5 µg/L, with the highest exposure group (5000
225
µg/L) showing a significant increase (SI-Table 1). The nuclei showed structural changes in exposure groups ≥ 125
226
µg/L, with an irregular outline and little increases of heterochromatin. These effects increased monotonously (Fig.
227
6b), but were less evident at the highest exposure group (5000 µg/L). The rER was only affected in exposure
228
groups ≥ 500 µg/L, showing moderate proliferations and fenestrations (Fig. 5d). Additionally, slight dilations of
229
the rER were observable in exposure groups of 250 and 5000 µg/L. In exposure groups ≥ 62.5 µg/L, approximately
230
50 % of the mitochondria showed distorted cristae (Fig. 5a, SI-Table 1). This effect was significant in exposure
231
groups ≥ 250 µg/L. Moreover, the total number of mitochondria monotonously increased with significant
232
differences in exposure groups ≥ 250 µg/L (SI-Tables 1 and 2). In contrast to thyrocytes of PTU-exposed zebrafish,
233
the diameter of mitochondria was not affected by PER exposure. Electron-dense bodies, most likely lysosomes,
234
showed a monotonous increase with significant differences in exposure groups ≥ 125 µg/L (Figs. 5b, c, d, Fig. 7c).
235
Above 62.5 µg/L, the number of lysosomes with lipofuscin agglomerations was increased (Fig. 6a). This effect
236
did not seem to be dose-dependent and was absent in the controls. In exposure groups ≥ 125 µg/L, large vesicles
237
appeared in apical locations. At 125 µg/L, these vesicles showed rod-like shapes, and at 250 µg/L, some fiber-like
238
inclusions were present. Exposure groups ≥ 62.5 µg/L showed a monotonous increase in the number and length
239
of apically located microvilli (Figs. 5b, c). Exposure ≥ 500 µg/L PER induced small amounts of colloid inclusions.
240
The electron density of the colloid dose-dependently decreased in exposure groups ≥ 62.5 µg/L (Fig. 5c). Apical
241
vesicles were not present.
242
Discussion
243
The aim of our study was to investigate ultrastructural changes in zebrafish thyrocytes after exposure to PTU and
244
PER. Histological alterations of thyroid follicles after exposure to both substances have already been addressed in
245
our previous communications (Schmidt & Braunbeck, 2011; Schmidt et al., 2012), and will be further commented
246
below, based on the findings described in this study.
247
Our two previous communications indicated that both substances lead to an overall activation of the thyroid gland.
248
This was likely due to enhanced stimulation by the thyroid signaling hormone (TSH), as a feedback to lowered
249
levels of thyroxin (T4; Schmidt & Braunbeck, 2011; Schmidt et al., 2012). Despite similar stimulation, distinct
250
differences were detected at the light microscopical level. More specifically, PTU exposure induced hyperemia,
251
proliferations in epithelial cell height, and stratification, while PER exposure resulted in severe colloid depletion
252
and proliferation of average-sized follicles (Schmidt & Braunbeck, 2011; Schmidt et al., 2012). Besides
253
stimulation by TSH, it may be expected that the molecular modes of action of the substances play an important
254
role. A recent study showed that exposure of zebrafish larvae to comparable levels of PTU (as in our previous
255
communication; Schmidt & Braunbeck, 2011) resulted in up-regulated mRNA expression of thyroperoxidase
256
(TPO), TSH and deiodinase 2 (dio2), while thyroid receptors (TR) alpha and beta and deiodinase 3 (dio3) were
257
down-regulated (Baumann et al., 2016). Exposure to PER lead to up-regulation of dio2 and sodium iodide
258
symporter (NIS), and down-regulation of dio1 and dio3 genes in rare minnow Gobiocypris rarus (Li et al., 2011),
259
which underlines the different modes of action of the two substances.
260
Our results document that both PTU- and PER-exposure lead to clear ultrastructural alterations, however with
261
distinct qualitative discrepancies. The effects were observed in both treatment groups, i.e., increased numbers of
262
electron-dense bodies, microvilli, apically located luminal vesicles, and significant exo- and endocytotic activity,
263
are well-defined indicators of colloid reabsorption and thyroidal activation, likely due to increased TSH levels
264
(French & Hodges, 1977; Fujita, 1975; Henderson & Gorbman, 1971; Olen, 1969). The electron-dense bodies
265
observed were also reported in other studies (Fujita & Machino, 1965; Henderson & Gorbman, 1971; Leatherland
266
et al., 1978; Leatherland & Sonstegard, 1980), but are still not fully characterized. Usually, smaller electron-dense
267
droplets located in the apical or subapical regions are regarded as lysosomes, especially with the appearance of
268
lipofuscin. Larger droplets are regarded as colloid reabsorption from the follicle lumen. Occurrences of fiber-like
269
inclusions in some droplets of PER-exposed fish have also been described for Japanese amberjack Seriola
270
quinqueradiata, where they are believed to be associated with old or altered droplets (Fujita & Machino, 1965).
271
Lysosomal architecture and microvilli appearance differed between fish of the two treatment groups; PER
272
exposure caused stronger lysosomal alterations than PTU exposure. The number of follicles with electron-lucent
273
colloid (i.e. colloid depletion) was elevated after PER exposure. The occurrence of lysosomes is important for the
274
digestion of reabsorbed colloid, and TSH leads to increased colloid reabsorption by endocytosis (Eales & Brown,
275
1993). Studies from Scranton & Halm (1965) and Surks (1967) revealed that PER exposure leads to iodide efflux
276
from thyroid glands in rats. In mice, PER exposure rapidly increased the secretory response of the thyroid to TSH,
277
including both iodide and iodothyronines (Rousset et al., 1977). We conclude that the elevated number of
278
lysosomes detected in PER-exposed thyrocytes is likely due to an interplay of three processes: (1) the ongoing
279
stimulation by TSH, (2) the increasing the secretory response to TSH and (3) the efflux of iodide and
280
iodothyronines. Hence, the increase of lysosomes could lead to decreased levels of colloid in the follicular lumen;
281
a phenomenon we reported in our previous communication (Schmidt et al. 2012).
282
In contrast to PER, PTU exposure did not increase the number of lysosomes. The abovementioned increase of
283
TSH-producing cells, which was detected in PTU-exposed zebrafish (Schmidt & Braunbeck, 2011), could be
284
responsible for this. A comparable depletion of colloid storage deposits could not be detected, only alterations in
285
tinctorial properties. Studies by Anderberg et al. (1981; 1980) on human thyroids revealed that the colloid is
286
composed of 19S thyroglobulin, larger iodoproteins, and smaller protein fractions (an albumin-like protein and a
287
pre-albumin fraction). In these studies, exposure to carbimazole (which, together with PTU and methimazole,
288
belongs to the group of thioamides), lead to a decrease of larger thyroglobulin aggregates; likely due to an
289
insufficient capacity to iodinate thyroglobulin. This could provide an explanation for the different tinctorial
290
properties of the colloid observed in our histological and ultrastructural studies.
291
Interestingly, the staining properties of the colloid coincide with the appearance of apical microvilli. The latter
292
showed proliferation and elongation, especially under PER exposure. This proliferation supports colloid depletion
293
due to surface multiplication. Additionally, PER exposure groups ≥ 250 µg/L revealed enhanced endo- or
294
exocytotic activity, which could also be found in PTU exposure groups ≥ 2.5 mg/L. The temporal responses to
295
TSH seemed to be dependent on the mode of action of the test substance, since PTU induces endo- and exocytotic
296
activity prior to microvilli elongation, whereas in PER-exposed samples, microvilli elongation occurred first.
297
In addition to the mere increase of lysosomes, PER exposure led to pronounced lipofuscin agglomerations within
298
lysosomes. Lipofuscin is known as “age pigment”, progressively accumulating within lysosomes in long-lived
299
post-mitotic cells (Brizzee et al., 1969; Donato & Sohal, 1981; Strehler, 1964b; a). In zebrafish, thyroid cell
300
turnover times are not known; but eventually the turnover times might be long enough to allow accumulations of
301
lipofuscin inside lysosomes of zebrafish thyrocytes. On the other hand, lipofuscin is formed within secondary
302
lysosomes (i.e., primary lysosomes fused with colloid droplets) due to the interplay of two processes: (1) the
303
production of partially reduced oxygen species by mitochondria and (2) the autophagocytotic degradation within
304
lysosomes (Brunk et al., 1992). Although lipofuscin agglomerations could only be detected in PER-exposed fish,
305
both exposure groups revealed mitochondrial alterations, such as dose-dependent increases and proliferations of
306
mitochondria with distorted cristae. PTU-exposed fish displayed increased mitochondrial diameters. Similar
307
effects were observed in white leghorn chicken after exposure to PTU and methimazole (Handa & Chiasson, 1980)
308
and in iodide-treated BB/W rats (Li & Boyages, 1994). Based on reviews by Hotchkiss et al. (2009), Skulachev
309
(2006), Tsujimoto & Shimizu (2007), and Ulivieri (2010), the observed effects could be interpreted as first signs
310
of necrosis, at least degeneration, which was most evident in the highest PTU exposure group (50 mg/L).
311
Interestingly, PER-exposed fish revealed around 50 % of mitochondria with distorted cristae. This proportion was
312
independent of exposure groups, as the numbers of both normal and affected mitochondria increased
313
simultaneously. On the other hand, PTU-exposed fish revealed increasing fractions of affected mitochondria with
314
rising exposure groups. Throughout, the percentage of affected mitochondria rose from 20 % to 93 %.
315
Mitochondrial damage is regarded as an indicator of oxidative stress. This could overcharge the capacity of the
316
antioxidant defense system, thus leading to damaged cellular functions (Gille et al., 1989). PER is known to be
317
actively taken up by thyrocytes (Dohan et al., 2007; Tran et al., 2008), but not to be metabolized (Anbar et al.,
318
1959). In rat thyrocytes, PER causes an iodide efflux (Scranton & Halm, 1965; Surks, 1967). It might be
319
hypothesized that PER accumulation could interfere with the intracellular antioxidant defense system, which could
320
cause mitochondrial distortion. It is also known that PER inhibits iodide uptake by the sodium iodide symporter
321
(NIS), resulting in decreased TH levels (Schmidt & Braunbeck, 2011; Schmidt et al., 2012).
322
Similar to the mitochondrial damages in PER-exposed fish, the PTU-induced effects could be an indicator of
323
oxidative stress. The mode of action of PTU is the blockade of thyroid peroxidases, which prevents thyroid
324
hormone synthesis. The subsequent activation of thyroidal tissue via TSH could lead to an increased influx of
325
iodide and sodium into thyrocytes via the NIS. As shown by Li & Boyages (1994), both iodide excess and PTU
326
exposure affected mitochondria. It is known that mitochondria show a sudden increase in permeability of the inner
327
mitochondrial membrane for solutes smaller than 1500 Da, which can result in mitochondrial swelling (Halestrap
328
et al., 2002; Zoratti & Szabo, 1995). Under these stress conditions, it is likely that mitochondria were not capable
329
of producing sufficient amounts of ATP; which would eventually guide the cell towards necrotic death (Crompton,
330
1999; Halestrap et al., 2002).
331
PTU exposure resulted in alterations of the rER. Proliferation, fenestration, and dilation are common reactions of
332
thyrocytes exposed to thyroid peroxidase-inhibitors. Similar reactions were observed in Wistar rats after exposure
333
to methimazole (Tsujio et al., 2007) and white leghorn chicken after exposure to PTU (Handa & Chiasson, 1980).
334
In thyroid hormone production, the rER plays an important role, as it is essential in the synthesis of thyroglobulin
335
and thyroid peroxidases. PTU’s mode of action is the inhibition of peroxidases, resulting in lowered thyroid
336
hormone levels. This could lead to a higher demand to synthesize these peroxidases, in order to adequately
337
maintain TH production. This would lead to a proliferation of the rER, as observed in PTU-exposed fish. As only
338
PTU-exposed fish showed proliferations of the rER, the specific mode of action seems to be of great importance.
339
We observed an increase in apical cell height revealed in PTU-exposed fish. Apical cell height represents an
340
indirect parameter to detect proliferations of organelle contents, whose main location is the apical part of the cell.
341
Thus, the observed proliferations of lysosomes, mitochondria and the rER could contribute to the increase in apical
342
cell height. This correlates with our previous histological findings (Schmidt & Braunbeck, 2011). PER exposure
343
caused only a slight increase of apical cell height at exposure groups ≥ 62.5 µg/L, and a significant increase in the
344
highest exposure group (5000 µg/L). Again, this correlates with our previous histological observations (Schmidt
345
et al., 2012). Although a marked proliferation of mitochondria and lysosomes could be detected, the effect on
346
overall apical cell height was moderate. This suggests that the proliferation of the rER (as observed in PTU-
347
exposed fish) is the major contributor to the increased apical cell height. Consequently, this increase in epithelial
348
cell height would also be histologically detectable.
349
Alterations in nuclear morphology cannot be considered as suitable endpoint due to its low sensitivity in PTU-
350
exposed fish. Thyrocytes of PER-exposed zebrafish showed an increasingly irregular outline of the nucleus in
351
exposure groups ≥ 125 µg/L. The amount of heterochromatin moderately increased in higher exposure groups of
352
both substances. Despite the relatively moderate extent, the nuclear changes could be interpreted as first reactions
353
of the thyrocytes leading to cellular death seen at the highest PTU concentration. The appearance of degenerated
354
cells could also be a first sign of cellular narcosis by the test substances (due to high intra-cellular concentrations)
355
and should not be overestimated. No signs of general toxicity were observed at light-microscopical level (Schmidt
356
& Braunbeck, 2011; Schmidt et al., 2012) and the applied substance concentrations in the exposure groups were
357
far from known lethal ranges (Jomaa et al., 2014; Park et al., 2006). This further supports that ultrastructural
358
investigations represent a very sensitive tool for the detection of effects of thyroid-disrupting substances.
359
When dealing with effects of thyroid-disrupting substances, the feedback mechanism responsible for thyroid
360
homeostasis has to be taken into account. The thyroid is regulated by a negative feedback loop with the pituitary
361
acting as the main control organ, excreting TSH. If the concentration of thyroid hormones decreases due to
362
inhibiting substances, elevated concentrations of TSH are responsible for activating the TH synthesis. Numerous
363
studies have addressed this aspect in, e.g., tadpoles (Neuenschwander, 1972), chicken (Fujita, 1963), and rats
364
(Fujita & Suemasa et al., 1968; Lupulescu et al., 1968; Roos, 1960; Seljelid, 1965; 1967a, b, c, d, e; Wetzel et al.,
365
1965; Wissig, 1963). These studies show remarkable similarities to the effects observed in this study, e.g.,
366
increases in cell height, dilation of rER cisternae, and alterations of microvilli. The influence of TSH on the
367
ultrastructural appearance of thyrocytes is significant, but this influence alone cannot explain differences of effects
368
between the two test substances. As mentioned above, both PER and PTU lead to increased stimulation by TSH
369
due to lowered TH levels. Thus, alterations of thyrocytes should not deviate much. The fact that thyrocytes of
370
zebrafish exposed to PER differ from their PTU-exposed counter parts, can only be explained with the specific
371
modes of action of the test substances.
372
In the context of endocrine disruption, histopathology provides a powerful and sensitive tool for the detection of
373
thyroid-disrupting substances (Grim et al., 2009). However, histopathological endpoints in the thyroid are usually
374
limited to rather general endpoints such as hyperplasia and hypertrophy, which does not exploit the full
375
opportunities that histology can offer. In contrast, ultrastructural investigations not only confirm histopathological
376
observations, but also provide additional and more detailed information about specific cellular changes induced
377
by chemical exposure. Closer inspection of morphological processes revealed that fundamental cellular parameters
378
and functions displayed distinct differences between the two substances, likely due to the unique underlying modes
379
of action. Thus, histopathology and ultrastructural investigations are two methods which should complement each
380
other in toxicological studies.
381
Conclusions
382
In conclusion, the present study highlights the value of zebrafish as sensitive test organism for thyroid disruption.
383
This provides a great opportunity to improve our understanding of (fish) thyroid function and histopathological
384
alterations. For the first time, goitrogen-induced alterations in zebrafish thyrocyte ultrastructure were evaluated to
385
further our understanding of existing histopathological data. Proliferation of microvilli, large electron-dense
386
droplets, apically located luminal vesicles and significant exo- and endocytotic activity are common features of
387
thyroidal activation via TSH. Different modes of action of goitrogens lead to different changes in thyrocyte
388
architecture. Beside massive proliferation of the rER in PTU-exposed zebrafish thyrocytes, the most striking effect
389
was strong alteration in mitochondrial morphology in both PTU- and PER-exposed zebrafish. Distinct differences
390
in mitochondrial morphology, likely due to the mode of action of the two substances, were also observed. In the
391
future, the combination of cytopathological observations and histopathological investigations can be of crucial
392
importance. It can document effects on thyroid development in fish larvae, which is essential for embryonic, larval,
393
and juvenile development. The list of ultrastructural effects under different thyroid-disrupting exposure conditions
394
provides a valuable resource for further comparative studies. Together, specific sub-cellular aspects of thyroid
395
pathology and the evaluation and interpretation of existing histopathological endpoints can greatly improve the
396
detection of substances with thyroid-disrupting properties in fish.
397
References
398
Anbar, M., Guttmann, S. and Lewitus, Z. (1959). The mode of action of perchlorate ions on the iodine uptake of
399
the thyroid gland. Int J Appl Radiat Isot 7: 87-96.
400
Anderberg, B., Enestrom, S., Gillquist, J., Kagedal, B., Mansson, J. C. and Smeds, S. (1981). Protein composition
401
in single follicles, homogenates and fine-needle aspiration biopsies from normal and diseased human
402
thyroid. Acta Endocrinol (Copenh) 96(3): 328-334.
403
Anderberg, B., Enestrom, S., Gillquist, J. and Smeds, S. (1980). Protein composition of the thyroid colloid in
404
patients with hyperthyroidism. J Endocrinol 86(3): 443-449.
405
Baumann, L., Ros, A., Rehberger, K., Neuhauss, S. and Segner, H. (2016), Thyroid disruption in zebrafish (Danio
406
rerio) larvae: different molecular response patterns lead to impaired eye development and visual functions.
407
Aquatic Toxicology, DOI: 10.1016/j.aquatox.2015.12.015.
408
Braunbeck, T., Storch, V. and Nagel, R. (1989) Sex-specific reaction of liver ultrastructure in zebrafish
409
(Brachydanio rerio) after prolonged sublethal exposure to 4-nitrophenol. Aquat. Toxicol. 14: 185-202.
410
Braunbeck, T. and Storch, V. (1992) Senescence of hepatocytes isolated from rainbow trout (Oncorhynchus
411
mykiss) in primary culture - An ultrastructural study. Protoplasma 170: 138-159.
412
Brizzee, K. R., Cancilla, P. A., Sherwood, N. and Timiras, P. S. (1969). The amount and distribution of pigments
413
in neurons and glia of the cerebral cortex. Autofluorescent and ultrastructural studies. J Gerontol 24(2):
414
127-135.
415
Brown, S. B., Adams, B. A., Cyr, D. G. and Eales, J. G. (2004). Contaminant effects on the teleost fish thyroid.
416
Environ Toxicol Chem 23(7): 1680-1701.
417
Brunk, U. T., Jones, C. B. and Sohal, R. S. (1992). A novel hypothesis of lipofuscinogenesis and cellular aging
418
based on interactions between oxidative stress and autophagocytosis. Mutat Res 275(3-6): 395-403.
419
Cooper, D. S. (2005). Antithyroid drugs. N Engl J Med 352(9): 905-917.
420
Crompton, M. (1999). The mitochondrial permeability transition pore and its role in cell death. Biochem J 341(2):
421
233-249.
422
Cyr, D. G. and Eales, J. G. (1988). Influence of Thyroidal Status on Ovarian Function in Rainbow Trout, Salmo
423
gairdneri. J Exp Zool 248: 81-87.
424
Damstra, T., Barlow, S., Bergman, A., Kavlock, R. and Van der Kraak, G. (2002). Global assessment of the state-
425
of-the-science of endocrine disruptors, International Programme on Chemical Safety, prepared by an expert
426
group on behalf of the World Health Organisation, the International Labour Organisation, and the United
427
Nations Environment Programme.
428
Dohan, O., Portulano, C., Basquin, C., Reyna-Neyra, A., Amzel, L. M. and Carrasco, N. (2007). The Na+/I
429
symporter (NIS) mediates electroneutral active transport of the environmental pollutant perchlorate. Proc
430
Natl Acad Sci USA 104(51): 20250-20255.
431
Donato, H. J. and Sohal, R. S. (1981). Lipofuscin. In: J. Florini (Ed.). Handbook of Biochemistry in Aging, CRC
432
Press, Boca Raton, FL, pp. 221-227.
433
Dumont, J. E., Malone, J. F. and van Herle, A. J. (1980). Irradiation and Thyroid Disease. Euratom 67.13. Office
434
for Publications of European Community, Luxembourg.
435
Eales, J. G. and Brown, S. B. (1993). Measurement and regulation of thyroidal status in teleost fish. Rev Fish Biol
436
Fisher 3: 299-347.
437
Elsalini, O. A. and Rohr, K. B. (2003). Phenylthiourea disrupts thyroid function in developing zebrafish. Dev
438
Genes Evol 212(12): 593-598.
439
French, E. I. and Hodges, R. D. (1977). Fine structural studies on the thyroid gland of the normal domestic fowl.
440
Cell Tissue Res 178(3): 397-410.
441
Fujita, H., Machino, M. and Nakagami, K. (1963). Electron Microscopic Studies on the Rat Thyroid Gland
442
Following Administration of Propylthiouracil and Thyradin, with Special Reference to the Inclusion Body
443
in the Follicular Cell. Okajimas Folia Anat Jap 39: 157-177.
444
Fujita, H. (1963). Electron Microscopic Studies on the Thyroid Gland of Domestic Fowl, with Special Reference
445
to the Mode of Secretion and the Occurrence of a Central Flagellum in the Follicular Cell. Z Zellforsch
446
Mikrosk Anat 60: 615-632.
447
Fujita, H. and Machino, M. (1965). Electron microscopic studies on the thyroid gland of a teleost, Seriola
448
Quinqueradiata. Anat Rec 152: 81-97.
449
Fujita, H. and Suemasa, H. (1968). Cytological effects of TSH on the thyroid of hypophysectomized rats with and
450
without previous administration of actinomycin D. An electron microscope study. Arch Histol Jap 30(1):
451
45-59.
452
Fujita, H. (1975). Fine structure of the thyroid gland. Int Rev Cytol 40: 197-280.
453
Galand, P. (1967). Comparison of 2 autoradiographic methods based on the use of tritiated thymidine to measure
454
the duration of the S phase (phase of deoxyribonucleic acid synthesis) and interphase of cells from different
455
mouse tissues. Arch Biol 78(2): 167-191.
456
Gille, J. J., van Berkel, C. G., Mullaart, E., Vijg, J. and Joenje, H. (1989). Effects of lethal exposure to hyperoxia
457
and to hydrogen peroxide on NAD(H) and ATP pools in Chinese hamster ovary cells. Mutat Res 214(1):
458
89-96.
459
Grim, K. C., Wolfe, M., Braunbeck, T., Iguchi, T., Ohta, Y., Tooi, O., Touart, L., Wolf, D. C. and Tietge, J. (2009).
460
Thyroid histopathology assessments for the amphibian metamorphosis assay to detect thyroid-active
461
substances. Toxicol Pathol 37(4): 415-424.
462
Halestrap, A. P., McStay, G. P. and Clarke, S. J. (2002). The permeability transition pore complex: another view.
463
Biochimie 84(2-3): 153-166.
464
Handa, R. J. and Chiasson, R. B. (1980). Comparative effects of three goitrogenic treatments on White Leghorn
465
chickens. Avian Dis 24(4): 916-929.
466
Henderson, N. E. and Gorbman, A. (1971). Fine structure of the thyroid follicle of the Pacific hagfish, Eptatretus
467
stouti. Gen Comp Endocrinol 16(3): 409-429.
468
Hotchkiss, R. S., Strasser, A., McDunn, J. E. and Swanson, P. E. (2009). Cell death. N Engl J Med 361(16): 1570-
469
1583.
470
Janz, D. M. (2000). Endocrine System. (Chapter 25) In G. K. Ostrander (ed.). The Laboratory Fish, Academic
471
Press, San Diego, CA.
472
Jianjie, C., Wenjuan, X., Jinling, C., S., Ruhui, J. and Meiyan, L. (2016). Fluoride caused thyroid endocrine
473
disruption in male zebrafish (Danio rerio). Aquat Toxicol 171: 48–58.
474
Jomaa, B., 2014. Developmental toxicity of thyroid-active compounds in a zebrafish embryotoxicity test. ALTEX
475
31, 303–317.
476
Karnovsky, M. J. (1971). Use of ferrocyanide-reduced osmium tetroxide in electron microscopy. J Cell Biol 51,
477
146A.
478
Kumarathilaka, P., Oze, C., Indraratne, S. and Vithanage, M. (2016). Perchlorate as an emerging contaminant in
479
soil, water and food. Chemosphere 150: 667-677.
480
Lagler, K., Bardach, J., Miller, R. and Passino, D. (1977). Ichthyology 2nd ed. New York: John Wiley & Sons.
481
Leatherland, J. F., Moccia, R. and Sonstegard, R. (1978). Ultrastructure of the thyroid gland in goitered coho
482
salmon (Oncorhynchus kisutch). Cancer Res 38(1): 149-158.
483
Leatherland, J. F. and Sonstegard, R. A. (1980). Structure of thyroid and adrenal glands in rats fed diets of Great
484
Lakes coho salmon (Oncorhynchus kisutch). Environ Res 23(1): 77-86.
485
Leatherland, J. F. (1994). Reflections on the thyroidology of fishes: from molecules to humankind. Guelph
486
Ichthyol Rev 2: 3-67.
487
Li, M. and Boyages, S. C. (1994). Iodide induced lymphocytic thyroiditis in the BB/W rat: evidence of direct toxic
488
effects of iodide on thyroid subcellular structure. Autoimmunity 18(1): 31-40.
489
Li, W., Zha, J., L., Li, Z. and Wang, Z. (2011). Regulation of iodothyronine deiodinases and sodium
490
iodide symporter mRNA expression by perchlorate in larvae and adult Chinese rare minnow (Gobiocypris
491
rarus). Marine Pollution Bulletin 63: 350–355.
492
Lupulescu, A., Petrovici, A., Pop, A. and Heitmanek, C. (1968). Electron microscopic observations on the
493
parathyroid gland in experimental hypoparathyroidism. Experientia 24(1): 62-63.
494
Miranda, L. A., Pisano, A. and Casco, V. (1996). Ultrastructural study on thyroid glands of Bufo arenarum larvae
495
kept in potassium perchlorate solution. Biocell 20(2): 147-153.
496
Mukhi, S. and Patino, R. (2007). Effects of prolonged exposure to perchlorate on thyroid and reproductive function
497
in zebrafish. Toxicol Sci 96(2), 246–254.
498
Neuenschwander, P. (1972). Ultrastructure and iodine uptake of the thyroid gland in larvae of Xenopus laevis
499
Daud. Z Zellforsch Mikrosk Anat 130(4): 553-574.
500
OECD (2004). Report of the Validation of the Amphibian Metamorphosis Assay for the detection of thyroid active
501
substances: Phase 1 – Optimisation of the Test Protocol. Series on Testing and Assessment. In
502
Environmental Health and Safety Publications, Paris, France.
503
OECD (2007). Final Report of the Validation of the Amphibian Metamorphosis Assay: Phase 2 – Multi-chemical
504
Interlaboratory Study. In Series on Testing and Assessment. Environmental Health and Safety Publications,
505
Paris, France.
506
OECD (2009). OECD Guideline for the Testing of Chemicals - The Amphibian Metamorphosis Assay. In Effects
507
on Biotic Systems Environmental Health and Safety Publications, Paris, France.
508
Olen, E. (1969). The fine structure of experimentally induced hyperplastic and colloid goiter in the hamster. Lab
509
Invest 21(4): 336-346.
510
Opitz, R., Braunbeck, T., Bogi, C., Pickford, D. B., Nentwig, G., Oehlmann, J., Tooi, O., Lutz, I. and Kloas, W.
511
(2005). Description and initial evaluation of a Xenopus metamorphosis assay for detection of thyroid
512
system-disrupting activities of environmental compounds. Environ Toxicol Chem 24(3): 653-664.
513
Park, J., Rinchard, J., Liu, F., Anderson, T., Kendall, R. and Theodorakis, C. (2006). The thyroid endocrine
514
disruptor perchlorate affects reproduction, growth, and survival of mosquitofish. Ecotoxicology and
515
Environmental Safety 63 (2006) 343–352.
516
Petersen, A., Dillon, D., Bernhardt, R., Torunsky, R., Postlethwait, J., von Hippel, F., Buck, C. and Cresko, W.
517
(2015) Perchlorate disrupts embryonic androgen synthesis and reproductive development in threespine
518
stickleback without changing whole-body levels of thyroid hormone. General and Comparative
519
Endocrinology 210 (2015) 130–144.
520
Pinto, P.I.S., Guerreiro, E.M., Power, D.M., 2013. Triclosan interferes with the thyroid axis in the zebrafish (Danio
521
rerio). Toxicol. Res. 2, 60–69. doi:10.1039/C2TX20005H
522
Power, D. M., Llewellyn, L., Faustino, M., Nowell, M. A., Bjornsson, B. T., Einarsdottir, I. E., Canario, A. V. and
523
Sweeney, G. E. (2001). Thyroid hormones in growth and development of fish. Comp Biochem Physiol C
524
Toxicol Pharmacol 130(4): 447-459.
525
Quesada-Garcia, A., Encinas, P., Valdehita, A., Baumann, L., Segner, H., Coll, J. and Navas, J. (2016), Thyroid
526
active agents T3 and PTU differentially affect immune gene transcripts in the head kidney of rainbow trout
527
(Oncorhynchus mykiss). Aquatic Toxicology, DOI: 10.1016/j.aquatox.2016.02.016.
528
R Core Team (2016). R: A language and environment for statistical computing. R Foundation for Statistical
529
Computing, Vienna, Austria. URL https://www.R-project.org/.
530
Reynolds, E. S. (1963). The use of lead citrate at high pH as an electron-opaque stain in electron microscopy. J
531
Cell Biol 17: 208-212.
532
Richardson, K. C., Jarett, L. and Finke, E. H. (1960). Embedding in epoxy resins for ultrathin sectioning in electron
533
microscopy. Stain Technol 35: 313-323.
534
Roos, B. (1960). The submicroscopic structure of the rat thyroid gland. Its modification by high doses of
535
thyrotropic hormone. A contribution to the study of the secretory processes in the thyroid gland. Pathologia
536
et microbiologia 23: 129-157.
537
Rousset, B., Orgiazzi, J. and Mornex, R. (1977). Perchlorate ion enhances mouse thyroid responsiveness to
538
thyrotropin, human chorionic gonadotropin and long acting thyroid stimulator. Endocrinology 100(6):
539
1628-1635.
540
Schmidt, F. and Braunbeck, T. (2011). Alterations along the Hypothalamic-Pituitary-Thyroid Axis of the Zebrafish
541
(Danio rerio) after Exposure to Propylthiouracil. Journal of Thyroid Research 2011: 376243.
542
Schmidt, F., Schnurr, S., Wolf, R. and Braunbeck, T. (2012). Effects of the anti-thyroidal compound potassium-
543
perchlorate on the thyroid system of the zebrafish. Aquatic Toxicology 109: 47-58.
544
Scranton, J. R. and Halm, N. S. (1965). Thyroidal Iodide Accumulation and Loss in Vitro. Endocrinology 76: 441-
545
453.
546
Seljelid, R. (1965). Electron microscopic localization of acid phosphatase in rat thyroid follicle cells after
547
stimulation with thyrotropic hormone. The journal of histochemistry and cytochemistry: official journal of
548
the Histochemistry Society 13(8): 687-690.
549
Seljelid, R. (1967a). Endocytosis in thyroid follicle cells. III. An electron microscopic study of the cell surface and
550
related structures. Journal of ultrastructure research 18(1): 1-24.
551
Seljelid, R. (1967b). Endocytosis in thyroid follicle cells. I. Structure and significance of different types of single
552
membrane-limited vacuoles and bodies. Journal of ultrastructure research 17(3): 195-219.
553
Seljelid, R. (1967c). Endocytosis in thyroid follicle cells. II. A microinjection study of the origin of colloid
554
droplets. Journal of ultrastructure research 17(5): 401-420.
555
Seljelid, R. (1967d). Endocytosis in thyroid follicle cells. IV. On the acid phosphatase activity in thyroid follicle
556
cells, with special reference to the quantitative aspects. Journal of ultrastructure research 18(3): 237-256.
557
Seljelid, R. (1967e). Endocytosis in thyroid follicle cells. V. On the redistribution of cytosomes following
558
stimulation with thyrotropic hormone. Journal of ultrastructure research 18(5): 479-488.
559
Sharma, P. and Patino, R. (2013). Regulation of gonadal sex ratios and pubertal development by the thyroid
560
endocrine system in zebrafish (Danio rerio). General and Comparative Endocrinology 184: 111–119.
561
Sharma, P., Grabowski, T. and Patino, R. (2016). Thyroid endocrine disruption and external body morphology of
562
Zebrafish. General and Comparative Endocrinology 226: 42–49.
563
Skulachev, V. P. (2006). Bioenergetic aspects of apoptosis, necrosis and mitoptosis. Apoptosis 11(4): 473-485.
564
Spurr, A. R. (1969). A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res
565
26(1): 31-43.
566
Strehler, B. L. (1964a). On the Histochemistry and Ultrastructure of Age Pigment. Adv Gerontol Res 21: 343-384.
567
Strehler, B. L. (1964b). On the Histochemistry and Ultrastructure of Age Pigment. Adv Gerontol Res 18: 343-384.
568
Surks, M. I. (1967). Determination of iodide clearance and exit rate constants in incubated thyroid lobes.
569
Endocrinology 80(6): 1020-1027.
570
Thienpont, B., Tingaud-Sequeira, A., Prats, E., Barata, C., Babin, P.J. and Raldúa, D., (2011). Zebrafish
571
Eleutheroembryos Provide a Suitable Vertebrate Model for Screening Chemicals that Impair Thyroid
572
Hormone Synthesis. Environ Sci Technol 45, 7525–7532.
573
Tietge, J. E., Butterworth, B. C., Haselman, J. T., Holcombe, G. W., Hornung, M. W., Korte, J. J., Kosian, P. A.,
574
Wolfe, M. and Degitz, S. J. (2010). Early temporal effects of three thyroid hormone synthesis inhibitors in
575
Xenopus laevis. Aquat Toxicol 98(1): 44-50.
576
Tran, N., Valentin-Blasini, L., Blount, B. C., McCuistion, C. G., Fenton, M. S., Gin, E., Salem, A. and Hershman,
577
J. M. (2008). Thyroid-stimulating hormone increases active transport of perchlorate into thyroid cells. Am
578
J Physiol Endocrinol Metab 294(4): E802-806.
579
Tsujio, M., Watahiki, Y., Yoshioka, K. and Mutoh, K. (2007). Morphology of thyroid follicular cells of
580
methimazole-treated rats. Anat Histol Embryol 36(4): 290-294.
581
Tsujimoto, Y. and Shimizu, S. (2007). Role of the mitochondrial membrane permeability transition in cell death.
582
Apoptosis 12(5): 835-840.
583
Ulivieri, C. (2010). Cell death: Insights into the ultrastructure of mitochondria. Tissue Cell.
584
van der Ven, L. T., van den Brandhof, E. J., Vos, J. H., Power, D. M. and Wester, P. W. (2006). Effects of the
585
antithyroid agent propylthiouracil in a partial life cycle assay with zebrafish. Environ Sci Technol 40(1):
586
74-81.
587
Wetzel, B. K., Spicer, S. S. and Wollman, S. H. (1965). Changes in fine structure and acid phosphatase localization
588
in rat thyroid cells following thyrotropin administration. The J Cell Biol 25(3): 593-618.
589
Wissig, S. L. (1963). The anatomy of secretion in the follicular cells of the thyroid gland. II. The effect of acute
590
thyrotrophic hormone stimulation on the secretory apparatus. The J Cell Biol 16: 93-117.
591
Wolff, J. (1998). Perchlorate and the thyroid gland. Pharmacol Rev 50(1): 89-105.
592
Zhao, X., Wang, S., Li, D., You, H. and Ren, X. (2013). Effects of perchlorate on BDE-47-induced alteration
593
thyroid hormone and gene expression of in the hypothalamus–pituitary–thyroid axis in zebrafish larvae.
594
Environ Toxicol Pharmacol 36: 1176–1185.
595
Zhao, J., Xu, T. and Yin, D. ( 2014) Locomotor activity changes on zebrafish larvae with different 2,20,4,40-
596
tetrabromodiphenyl ether (PBDE-47) embryonic exposure modes. Chemosphere 94: 53–61.
597
Zoratti, M. and Szabo, I. (1995). The mitochondrial permeability transition. Biochim Biophys Acta 1241(2): 139-
598 599
176600
.Figure legends
601
Fig. 1: Ultrastructure of thyroidal tissue in control zebrafish (Danio rerio). The epithelium encloses an evenly
602
stained colloid devoid of inclusions (a). The nucleus is basally located and most organelles can be found in apical
603
position (b, e). Mitochondria appear spherically to ovally shaped, the rough endoplasmic reticulum and Golgi
604
fields (*) are of cistern-like appearance (d, e). At the apical pole of thyrocytes, few electron-dense lysosomes (►),
605
and, at the border to the colloid, some microvilli are detectable (b, c, d, e). Magnifications: a: 2,000×; b: 10,000×;
606
c: 12,500×; d: 40,000×; e: 31,500×.
607 608
Fig. 2: Ultrastructure of PTU-exposed zebrafish thyroids. At 50 mg/L, an electron-dense cytoplasm and shrunken
609
nuclei present first symptoms of degeneration (a). Increased amounts of heterochromatin are visible (a, b). Marked
610
proliferation, dilation and fenestration in the rough endoplasmic reticulum (►) are further alterations (a). The
611
apical regions display proliferations of lysosomes (*; a). Magnifications: a: 10,000×; b: 4,000×.
612 613
Fig. 3: Mitochondrial alterations in zebrafish exposed to PTU. Already at 2.5 mg/L, mitochondria showed irregular
614
swelling of the intercristae space (c). Higher exposure groups displayed proliferations and extensive swellings (*;
615
a, b). Furthermore, dilation of the rough endoplasmic reticulum is visible (a). Magnifications: a: 10,000×; b and c:
616
40,000×.
617 618
Fig. 4: Apical alterations in zebrafish thyrocytes caused by PTU exposure. At concentrations ≥10 mg/L,
619
cytoplasmic inclusions were evident (a). At concentrations ≥ 2.5 mg/L, numerous apical vesicles were seen
620
protruding into the follicular lumen (b). Bleb-like structures indicate endo- or exocytotic processes at
621
concentrations ≥ 10 mg/L (b, c). Moderate proliferation of microvilli can be observed at concentrations ≥ 10 mg/L
622
(►; b). Magnifications: a: 12,500×; b: 8,000×; c: 20,000×.
623 624
Fig. 5: PER-induced ultrastructural alterations in zebrafish thyrocytes. At concentrations ≥62.5 µg/L, mitochondria
625
are swollen and display irregular swellings of the intercristae space (*; a). At concentrations of 5000 µg/L, a
626
marked increase of lysosomes mostly located in the apical part of thyrocytes was visible (b, c, d). The rough
627
endoplasmic reticulum showed moderate proliferation and some fenestration (d). The electron density of the
628
colloid markedly decreased in higher concentration groups (c). Proliferations of microvilli are observable in
629
concentrations ≥ 62.5 µg/L (►; b, c). Magnification: a: 31,500×; b: 20,000×; c: 10,000×; d: 16,000×.