| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Endocrinology and Diabetes Unit (T.T.T., S.H, B.A.), Department of Medicine, University of Wuerzburg, D-97080 Wuerzburg, Germany; Max-Planck Institute for Immunobiology (G.N., M.H.), D-79108 Freiburg, Germany; Institute for Developmental Biology (K.B.R.), University of Cologne, D-50923 Cologne, Germany; and Department of Physiological Chemistry I (C.W.), University of Wuerzburg, D-97074 Wuerzburg, Germany
Address all correspondence and requests for reprints to: Professor Dr. B. Allolio, Endocrinology and Diabetes Unit, Department of Medicine, University of Wuerzburg, D-97080 Wuerzburg, Germany. E-mail: Allolio_b{at}medizin.uni-wuerzburg.de.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
After the initial generation of an adrenal primordium induced by a program of sequential gene expression and specific nuclear transcription factors such as steroidogenic factor 1 (SF1) and the orphan nuclear receptor DAX1, functional differentiation and growth of the adrenal gland are, to a large extent, guided by pituitary-derived signals regulating both steroidogenesis and adrenal proliferation. In higher vertebrates, the key pituitary regulators are peptides cleaved from proopiomelanocortin (POMC) and released from pituitary corticotrophs. Accordingly, adult mice homozygous for a Pomc null allele lack macroscopically distinct adrenals. In addition, they show obesity and pigmentation defects and are born in only a quarter of the expected frequency (2), indicating that the phenotype also reflects lack of extra pituitary (e.g. hypothalamic) Pomc expression.
In contrast to adult mice, newborn Pomc-null mutants have been reported to have adrenal glands of normal morphology. In these mice, adrenal hypoplasia becomes evident at only 1 wk of age with loss of zonation followed by progressive atrophy including the zona glomerulosa in later life (3). However, as early as 1 wk of age corticosterone secretion becomes undetectable. Transplantation of Pomc-null mutant adrenals to adrenalectomized wild-type littermates results in normal adrenal morphology (3).
ACTH, the principal hormone derived from POMC stimulates adrenal glucocorticoid synthesis and secretion. However, it remains uncertain whether ACTH is also essential for physiological adrenal growth and proliferation. In vitro studies suggest that ACTH acts mainly as a differentiation factor lacking mitogenic activity (4, 5, 6). Recently, it has also been demonstrated that exogenous ACTH even inhibits the growth of adrenal tumors in a mouse tumor model (5).
On the other hand, Coll et al. (7) treated Pomc –/– mice with high doses of exogenous ACTH and thereby restored hormone production and adrenal zonation. However, the adrenal phenotype generated by supraphysiological ACTH in these mice was mainly hypertrophic, suggesting that under physiological conditions other POMC-derived peptides may participate in adrenal growth (7). This view is further supported by in vitro and in vivo data (8, 9, 10, 11, 12), indicating that peptides generated from the N terminus of POMC possess mitogenic activity in adrenocortical cells.
In recent years the zebrafish (Danio rerio) has become a successful model organism for the study of early development because of its small size, short generation time, large numbers of offspring, transparency of embryos, and its potential for easy molecular manipulation (13, 14). In teleosts the steroidogenic cells, together with closely intermingled chromaffin cells, are embedded in the head kidney forming the interrenal organ, the homolog to the mammalian adrenal gland (15, 16). First studies have demonstrated that early development of zebrafish interrenal glands resembles adrenal development in higher vertebrates (17, 18, 19). It has been demonstrated that ff1b, the zebrafish homolog of SF1, is required for interrenal differentiation and activation of side chain cleavage enzyme cyp11a1 (17, 18, 20). Accordingly, knockdown of ff1b activity by antisense morpholino technique led to down-regulation of interrenal steroidogenic enzymes and loss of interrenal tissue (17, 18), similar to the phenotype in Sf-1 knockout mice (21). In addition, it has been demonstrated that zebrafish wt1 (Wilm’s tumor suppressor 1) is involved in interrenal development and ff1b expression (18), an observation in line with a similar role in mice, because wt1 knockout mice also lack adrenal glands (22).
In this study, we have used the zebrafish to analyze the role of the pituitary in regulating growth and functional differentiation of the interrenal organ. To this end we compared wild-type zebrafish interrenal development with that of aal/eya1 and lia/fgf3 mutants lacking various pituitary cell types including corticotrophs and with that of pit1 mutant having corticotrophs but lacking other pituitary cell types. In addition, we studied the effects of acth receptor (mc2r) knockdown and exogenous dexamethasone (dex) on interrenal development and feedback regulation at the pituitary level.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
).
|
), ventral to the third somite. These two cell domains then fuse to a single cell cluster, slightly asymmetrically located to the right of the notochord at around 24 hpf (Fig. 1C). Expression of the key steroidogenic genes star (Fig. 1, C and D) and cyp11a1, the gene encoding cytochrome p450 cholesterol side chain cleavage enzyme (data not shown), is initiated shortly thereafter. From 4 dpf onward, the steroidogenic component of the interrenal primordium develops further into a distinct bilobed organ lateral to the notochord with the right lobe in general larger than the left lobe (Fig. 1, G–M). To analyze the effect on the protein level, enzymatic activity of 3-ß-hydroxysteroid dehydrogenase (3ß-Hsd) was also studied using a chromogenic reaction with etiocholan-3ß-ol-17-one as 3ß-Hsd substrate (data not shown).
The sequential onset of gene expression is summarized in Table 1 with star and cyp11a1 expression preceding 3ß-Hsd activity and melanocortin-2 receptor (mc2r) expression.
|
).
|
, A and B). The expanding interrenal steroidogenic primordium then covers also the left dßh-expressing cell cluster (Fig. 2C). At 3 dpf, the two dßh-expressing domains fuse to a single domain (Fig. 2, E and F) and develop further as a bilobed domain lateral to the notochord (Fig. 2, G–M) consistently in close contact to the steroidogenic cells. At 5–7 dpf the chromaffin cells are located closer to the trunk midline than the steroidogenic cells and appear enveloped by the steroidogenic compartment of the interrenal organ (Fig. 2, I–M).
Mutants that Lack Pituitary Cells Including Corticotrophs Show Normal Early Interrenal Development
To study the role of the pituitary in regulating early development of zebrafish interrenal organ, we analyzed the expression of zebrafish steroidogenic genes cyp11a1, mc2r, star, and of dßh specific for chromaffin cells in aal/eya1 and lia/fgf3 mutants. aal/eya1 and lia/fgf3 are zebrafish mutants from a zebrafish ENU mutagenesis screen for mutations affecting adenohypophysis development (26). Of the different pituitary cell types (somatotrophs, lactotrophs, thyrotrophs, melanotrophs, corticotrophs, and gonadotrophs), aal/eya1 lack melanotrophs, corticotrophs, and gonadotrophs (26, 27), whereas lia/fgf3 mutants lack all pituitary cell types (26, 28). Thus both mutants lack corticotrophs and therefore, expression of pituitary pomc is missing (Fig. 3, B, D, F, H, K, and M).
|
). Similarly, at 2 dpf, no changes were observed in the morphogenetic movement of steroidogenic and chromaffin cells compared with those of wild-type embryos (Fig. 3, I-M). Enzyme activity of 3ß-Hsd was not analyzed in pituitary mutants for methodological reasons, because at 2 dpf phenotypic differentiation between wild-type and mutant embryos is not possible without concomitant analysis of pomc expression.
Mutants without Pituitary Cells (Including Corticotrophs) Exhibit Impaired Interrenal Steroidogenic Function at 5 dpf
At 5 dpf a significant reduction of mRNA expression of cyp11a1, mc2r, and star, as assessed by ISH, was evident in both aal/eya1 and lia/fgf3 mutants (Fig. 4, A–M), indicating that at this stage these genes are partly controlled by the pituitary gland. Densitometric analysis revealed both area and density of cyp11a1 were significantly reduced in 5 dpf aal/eya1 mutants (n = 10) (density, 45.8 ± 7.6%; and area, 45.5 ± 6.8%;
± SEM, P < 0.01, compared with wild-type embryos (n = 17) (Fig. 4R). Both intensity and area of the staining for 3ß-Hsd enzyme activity were weaker in the mutants compared with wild type (Fig. 4, N–Q), indicating a role of the pituitary at 5 dpf not only at the transcriptional level but also at the level of enzyme function. Furthermore, expression of the chromaffin gene dßh was also reduced in aal/eya1 and lia/fgf3 mutants, indicating a role of steroidogenic cells for function of chromaffin cells (data not shown).
|
pit1 mutants showed normal interrenal development at 5 dpf with normal expression of all analyzed interrenal genes (Fig. 5), indicating that pomc-expressing pituitary cells are fully sufficient to maintain normal development of the steroidogenic interrenal compartment.
|
Efficacy of the mc2r antisense morpholino to block its target was confirmed by green fluorescent protein (GFP)-based experiments. mc2r antisense morpholino blocked GFP translation of mc2-r-GFP constructs. No GFP expression was observed in the 20 embryos injected with antisense mc2r morpholino in combination with the mc2r-GFP RNA, whereas strong GFP expression was observed in 19 of 21 embryos injected with RNA alone (supplemental Fig. 1, published as supplemental data on The Endocrine Society’s Journals Online web site at http://mend.endojournals.org).
In the mc2r antisense morphant embryos, similar results were obtained as in pituitary mutants lacking corticotrophs, with no change in expression of cyp11a1, star, and dßh until 2 dpf (Fig. 6). At 2 dpf the transcripts of pomc in the mc2r knockdown embryos also remained unchanged (Fig. 6, red arrowheads in A–C), suggesting no change in glucocorticoid feedback at pituitary corticotrophs at this stage of development.
|
, B and E) compared with those of wild-type and mc2r mismatch morphants (Fig. 7, A, D, C, and F), suggesting that lack of ACTH has a similar effect as the lack of all pituitary POMC-derived peptides. Reduction of steroidogenic protein activity is expected to cause glucocorticoid deficiency in the 5 dpf-mc2r knockdown larvae. To investigate evidence for feedback regulation, we analyzed pomc expression in mc2r knockdown larvae. In contrast to 2 dpf embryos, pomc expression in mc2r antisense morphants at 5 dpf was strongly increased in the anterior domain of the adenohypophysis [Fig. 8, B (red arrowhead) and D], compared with wild-type and mismatch morphants, indicating that the anterior domain of pomc-expressing cells is involved in glucocorticoid feedback regulation. By contrast, the posterior domain of pituitary pomc-expressing cells [Fig. 8, B (black arrowhead) and D] is not affected by mc2r knockdown.
|
|
).
|
). At 3 dpf, endogenous pomc expression in this domain is completely suppressed (red arrowhead in Fig. 9D) and also in 5 dpf embryos (red arrowhead in Fig. 9F). Again, the expression of cyp11a1 is unaffected by dex at 2 dpf (black arrowhead in Fig. 9B) but slightly down-regulated at 3 dpf (black arrowhead in Fig. 9D). At 5 dpf dex-treated embryos show only weak expression of cyp11a1 (black arrowhead in Fig. 9F).
At the protein level, 3ß-Hsd enzyme activity is also significantly reduced in the 5 dpf dex-treated embryos (Fig. 9H), compared with that of wild-type embryos (Fig. 9G).
These results indicate that glucocorticoid feedback at the pituitary level precedes pituitary-dependent steroidogenesis.
Interrenal Cell Proliferation Is Affected by dex
To investigate further whether deficiency in Pomc affects interrenal cell proliferation, we assessed cell counts in wild-type and dex-treated embryos after staining for 3ß-Hsd activity. The number of cells positive for 3ß-Hsd staining in wild-type embryos increased from a median of 17 (range 13–22) cells at 3 dpf (n = 19) to 34.5 (range 25–41) at 5dpf (n = 20) (P < 0.001). In contrast, after dex treatment we found an increase from 15 (range 11–22) (n = 25) to only 25 cells at 5 dpf (range 21–35, n =16, P < 0.001 vs. wild-type embryos). Staining for phosphorylated histone H3, which is present in the M phase of the mitotic cell cycle (51), revealed a significant decrease in the number of dividing cells in the interrenal region of the dex-treated embryos (n = 12) at 75 hpf: 0 (range 0–2) cells per embryo vs. three (range 2–6) cells per embryo in wild-type embryos (n = 5), P = 0.001 (supplemental Fig. 2, published as supplemental data on The Endocrine Society’s Journals Online web site) but not in two series at later stages [median, 0 cells (range 0–1 cells) for both wild-type and dex-treated embryos; n > 10 for each series].
These findings suggest that the absence of Pomc from the anterior pituitary domain leads to interrenal hypoplasia via reduced cell proliferation.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Zebrafish interrenal primordia are first visualized as bilateral clusters of cells expressing ff1b, the teleost homolog of mammalian SF1. These clusters are derived from the lateral intermediate mesoderm ventral to the third somite and migrate medially to form a single cell mass, consistent with previous reports (17, 18). In wild-type zebrafish, expression of steroidogenic genes such as cyp11a1 and star became detectable only after fusion of these two clusters. However, it has been previously shown in zebrafish mutants with midline defects that fusion of the interrenal primordia is not a prerequisite for initiation of steroidogenesis (18). The temporal pattern of expression of steroidogenic markers in interrenal cells indicates stepwise maturation of steroidogenic cells. The sequence of gene expression resembles the findings in mammals: in mice the adrenocortical primordium is earliest visualized via expression of SF1 at embryonic d 9 (E9) followed by expression of steroidogenic enzymes that become detectable at E11 (24). Intriguingly, in zebrafish, it appears that expression of the mc2r gene is a late step in interrenal maturation, because mc2r transcripts are detectable only after steroidogenic enzyme expression has been initiated. Whether this is also the case in rodents is not known.
In our study we have extended the limited data on the development of the chromaffin component of the zebrafish interrenal organ (17). Intriguingly dopamine ß hydroxylase (dßh) (16, 31) expressing chromaffin cells initially overlap with the steroidogenic primordium only on one side. Chromaffin cells then converge to the midline and fuse at 3 dpf. From this stage onward, steroidogenic and chromaffin cells remain in close contact and accordingly expand together bilateral to the notochord. At 5–7 dpf, chromaffin cells appear to form a central compartment of the interrenal organ. Again, the codevelopment of steroidogenic and chromaffin cells resembles the organogenesis of the adrenal gland in mammals. In mice neural crest-derived chromaffin cells migrate into the adrenocortical primordium only at E12–E14 to form cell clusters that later coalesce to a distinct layer in the center of the organ (24). However, no such distinct layers are present in zebrafish.
Our results document that early interrenal development is fully independent of any pituitary influence. In fact, steroidogenic enzyme expression, as assessed by 3ß-Hsd enzyme activity, precedes expression of the mc2r, giving indirect evidence of early autonomous steroidogenesis. At 2 dpf, expression of interrenal markers is not affected in pituitary mutants lacking pomc-expressing cells, or in mc2r knockdown embryos, or embryos treated with exogenous dex. Thus, although pomc expression in wild-type zebrafish is first detectable at 18 hpf (32, 33) and, therefore, precedes interrenal ff1b expression, its effects on the interrenal steroidogenic component are delayed, and control of steroid hormone production by pituitary Pomc requires further maturation of the interrenal tissue. These findings are in agreement with adrenal development in Pomc-null mice, because these mice are born with adrenal glands that are morphologically indistinguishable from those of their wild-type littermates (3). Only postnatally, in Pomc-null mice, adrenal cells fail to proliferate and gradually develop atrophy (3). Moreover, our results of pituitary-independent early interrenal development resemble findings in human anencephalic fetuses that do not have a pituitary: in early gestation (before wk 10–15), adrenal development of anencephalic fetuses is normal. Only thereafter the fetal zone fails to develop and does not exhibit its characteristic growth and steroidogenic activity (34).
However, in contrast to our mutants, Pomc-null mice lack all Pomc transcripts, and anencephalic fetuses also lack the hypothalamus. Thus, with regard to adrenal development, hypophysectomized animals or mice lacking the transcription factor TPIT (pituitary cell-restricted T-box factor) essential for development of pituitary corticotrophs seem to be more comparable to zebrafish aal/eya1 and lia/fgf3 mutants. In Tpit (–/–) mice, adrenals are detectable but hypoplastic, with significant loss in the glucocorticoid-producing zona fasciculata. Similar to Pomc-null mice, corticosterone is undetectable, suggesting dependence on pituitary POMC of both adrenal growth and corticosterone secretion (35, 36). However, the age of the mice in these studies was not given, and it is likely that these data were gained in adult Tpit (–/–) animals. Thus, data on fetal and neonatal Tpit (–/–) mice are needed to assess the specific influence of pituitary corticotrophs on prenatal development of the adrenal gland. Results obtained in hypophysectomized fetal sheep and pigs have demonstrated that hypophysectomy inhibits the intrauterine growth of the adrenal cortex, particularly of the zona fasciculata (37, 38, 39). However, in these models, analysis is restricted to later stages of gestation, thereby precluding analysis of early loss of corticotroph function.
The understanding of pituitary-independent early steroidogenesis is incomplete but may be of clinical relevance, as it is the hallmark of adrenal Cushing’s syndrome. Thus, autonomous POMC-independent cortisol production in adrenal tumors may be the result of adrenal reprogramming toward an early developmental phenotype. In zebrafish ff1b is clearly required for early interrenal steroidogenesis, because ff1b knockdown leads not only to down-regulation of cyp11a1 and 3ß-Hsd but eventually also to loss of steroidogenic tissue (17, 18). More recently, an important role for the interaction of the transcription factor Prox1 with Ff1b has been reported, as prox1 morphants display loss of ff1b expression and 3ß-Hsd activity (19). In addition, the transcription factor Wt1 has been shown to be involved in early zebrafish interrenal development, because reduced wt1 levels in knockdown experiments led to smaller interrenal primordia and decreased ff1b expression (18). The pivotal role of ff1b for early pituitary-independent steroidogenesis is also evident from experiments demonstrating direct activation of cyp11a1 transcription similar to its mammalian counterpart SF1 (18). On the other hand, expression of cyp11a1 has been described most recently in the extraembryonic yolk syncytial layer of zebrafish embryos, converting cholesterol to pregnenolone and playing a major role in embryonic cell movement and stabilization of microtubules (40). Intriguingly, this expression is seemingly independent of ff1b, which suggests that ff1b is not an invariable prerequisite for steroidogenic activity of cyp11a1 in zebrafish.
Our investigations in zebrafish at 5 dpf clearly indicate that at this stage interrenal development and function have become dependent on pituitary signals. In both aal/eya1 and lia/fgf3 mutants, not only the expression of steroidogenic markers was decreased, indicating lower functional activity, but also the area of expression was reduced, suggesting interrenal hypoplasia. Hammerschmidt and co-workers (29) have suggested that, at 5 dpf, zebrafish development largely resembles the developmental stage at birth in mammals. In human anencephalic fetuses at late gestation, adrenal hypoplasia with a strongly reduced fetal zone has been described (41, 42), suggesting that in humans regulation of the adrenal by the pituitary gland is, at least in part, established before birth.
Because aal/eya1 and lia/fgf3 mutants lack multiple pituitary cell types, the interrenal phenotype in zebrafish could be the result of multiple hormonal deficiencies. However, our findings in pit1 mutants clearly suggest that the presence of corticotrophs is sufficient for normal interrenal development at 5 dpf, indicating that pituitary Pomc secretion by corticotrophs is the essential signal. Furthermore, mc2r morphants exhibit a similar phenotype as mutants lacking pituitary corticotrophs, suggesting that Acth signaling is crucial for the action of pituitary corticotrophs on interrenal development at 5 dpf. This is in keeping with the observation that the adrenal phenotype in patients with ACTH resistance due to inactivating MC2R mutations, familial glucocorticoid deficiency type 1, resembles the findings in anencephalic fetuses (43). Because Mc2r-null mouse mutants have not yet been generated, mc2r-zebrafish morphants provide a unique tool to dissect the action of Acth out of the combined activity of Pomc-derived peptides. The role of other POMC-derived peptides for the development of steroidogenic cells remains a matter of debate. There is substantial evidence that N-terminal POMC-derived peptides possess mitogenic activity in the adrenal cortex and may be involved in adrenal proliferation (8, 9, 10, 11, 12, 44). However, although we have demonstrated that 1–28 N-POMC induces cell proliferation in adrenal cells in vitro, administration of 1–28 N-POMC in Pomc-null mice and Tpit (–/–) mice has failed so far to affect adrenal growth (45). Even so, it is possible that different dosing or other N-POMC-derived peptides, including glycosylated forms, may result in more biological activity of the exogenously administered N-POMC. Also, zebrafish Pomc contains a highly conserved homolog of the N terminus of N-Pomc (32). In addition, it has been reported that N-Pomc-derived peptides can slightly enhance Acth-induced cortisol release in teleosts (46). Thus, a role of N-Pomc-derived peptides for growth of interrenal cells in zebrafish cannot be fully excluded. Nevertheless, our findings clearly indicate that intact Mc2r signaling is a prerequisite for such a role, and any role of other Pomc peptides for adrenal development remains to be demonstrated.
The suppression of endogenous Pomc secretion by exogenous dex in wild-type zebrafish indicates that a pituitary influence is initiated as early as at 3 dpf, because at this stage a reduction in steroidogenic markers commences and becomes progressively more pronounced until 5 dpf. Thus, in contrast to what has been described for Pomc(–/–) mice, interrenal proliferation in zebrafish is affected early by loss of Pomc. Moreover, lack of endogenous Pomc is associated with reduced proliferation markers and number of steroidogenic cells.
mc2r morphants facilitated analysis of feedback control of reduced steroidogenesis at the pituitary level. As anticipated, no effect of mc2r knockdown on pituitary pomc expression is found at 2 dpf, because steroidogenesis is not yet affected. However, in response to decreased Mc2r-dependent endogenous steroidogenesis, pituitary pomc expression was strongly up-regulated at 5 dpf. This increase in pomc expression is restricted to the anterior domain of the pituitary gland, indicating that only this compartment of pituitary pomc-expressing cells is involved in the control of interrenal steroidogenesis. Accordingly, exogenous dex reduces pomc expression only in this domain, whereas otherwise pomc expression remains unchanged. These findings are in agreement with the report by Liu et al. (33) using transgenic zebrafish expressing GFP driven by pomc promoter.
Our studies using exogenous dex clearly demonstrate glucocorticoid feedback at 2 dpf and, therefore, indicate that negative feedback at the pituitary level precedes initiation of the control of interrenal steroidogenesis by the anterior domain of pituitary corticotrophs. It is not known whether a similar sequence of events is also operating in rodents. However, very recent data on early cortisol synthesis in humans also suggest early inhibition of fetal pituitary corticotrophs, providing a rationale for treatment with dex in congenital adrenal hyperplasia (47).
From our findings it is evident that interrenal development in the zebrafish shares many conserved molecular and developmental mechanisms with higher vertebrates. Moreover, the zebrafish morphants can be successfully used to manipulate hormonal signaling. Thus zebrafish provide a highly suitable model organism with which to further investigate the roles of transcription factors involved in human and mouse adrenal development. Moreover, with the availability of robust and specific markers of steroidogenic cells such as star and cyp11a1, zebrafish mutagenesis screens can be used to detect mutations associated with early interrenal hypoplasia or even agenesis. Such mutants may eventually allow identification of new genes involved in early adrenal development in higher vertebrates and humans.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
dex Treatment
To analyze feedback regulation of pituitary pomc expression by glucocorticoids and expression of steroidogenic genes, wild-type embryos were continuously treated with 40 µM dex, starting at 1 dpf, using water-soluble dex (Sigma). dex Stock solution (1 mM) was added to a dish containing 30–40 1-phenyl-2-thiourea-treated embryos in 24 ml embryo medium. Medium and dex were changed daily, and embryos were collected in daily intervals from 2–5 dpf.
Morpholino Injection
mc2r Antisense morpholino was injected into the yolk of 1–2 cell embryos with an optimum concentration of 16.7µg/µl. Five-nucleotide-mismatch mc2r morpholino was also injected as control using the identical concentration. Morpholino oligonucleotides were designed and synthesized by Gene Tools (Philomath, OR). Their sequences are as follows: mc2r antisense: ATCACTCTTAATTGTAGATCAGTTG, corresponding to nucleotides –12 to –37 in the 5'-untranslated region of the mc2r cDNA; mc2r -mismatch: ATgACTgTTAATTcTAcATgAGTTG.
To test for morpholino efficiency, a GFP-based approach was used as described earlier (50). Briefly, the 25-nucleotide target sequence for the mc2r morpholino was cloned in front of GFP into the CS2+GFP expression vector using annealed oligos containing BamHI overhangs. Capped mRNA was transcribed in vitro using the mMessage mMachine Kit from Ambion, Inc. (Austin, TX) and injected into zebrafish embryos at the one- to two-cell stage. Half of the injected embryos were injected with 8 ng of the mc2r morpholino directly afterward. Efficiency of morpholino-induced knockdown of translation was tested by GFP expression analysis.
Preparation of Antisense RNA Probes and Whole-Mount RNA ISH
Partial cDNAs of zebrafish ff1b, cyp11a1, star, and mc2r were amplified by RT-PCR with total RNA from adult zebrafish using the one-step RT-PCR kit (QIAGEN, Chatsworth, CA). Primers were designed based on the sequences in GenBank for ff1b (accession no. AF1980868), cyp11a1 (AF5277558), star (BC075967), and mc2r (AY1618489), and are described in Table 2. PCR fragments were cloned into pCRII-TOPO vector. The dßh plasmid was kindly provided by Professor Korzh (Institute of Molecular and Cell Biology, Singapore). pit1 plasmid was generated as described previously (29). Plasmids were linearized and transcribed to prepare digoxigenin or fluorescein-labeled antisense RNA probes: ff1b (XhoI/Sp6), cyp11a1 (HindIII/T7), star (XhoI/Sp6), mc2r (HindIII/T7), dßh (NcoI/Sp6), and pit1 (NotI/Sp6) using an RNA labeling kit (Roche Molecular Biochemicals, Mannheim, Germany). Probes were maintained in 50% formamide and stored at –20 C.
|
For double ISH, a mix of digoxigenin-labeled and fluorescein-labeled probes was used for hybridization. After incubation with the first antibody (antidigoxigenin AP) and detection of the first signal in blue color by BM purple substrate, embryos were incubated two times, 15 min each in 0.1 M glycine/HCl (pH 2.2)/0.1% Tween to completely remove AP activity, followed by four washes, 5 min each with 0.1% Tween in PBS (PBST), and 2 h incubation with second antibody (antifluorescein AP, diluted to 1:3000 in 1% blocking reagent in PBST). After 2 h washing with PBST and three washes in AP-reaction buffer, 10 min each, the second AP staining substrate giving red color, INT/BCIP (2-[4-iodophenyl]-3-[4-nitrophenyl]-5-phenyl-tetrazolium chloride)/(5-bromo-4-chloro-3-indolyl phosphate) substrate solution (Roche), was added to detect the second signal. Double-stained embryos were kept in 80% glycerol in PBST. The embryo was mounted in benzyl alcohol-glycerol (5:1) to take photos. In this mounting solution, only the blue stain is stable, whereas the red color rapidly fades out. To analyze whether individual cells are positive for one probe only (red) or for both probes (red and blue), photos were taken immediately after mounting and after the red color had vanished.
Chromogenic Histochemical Staining for 3ß-Hsd
Whole embryos were histochemically stained for 3ß-Hsd enzymatic activity using a protocol based on Levy’s method as previously described (16, 17). In the presence of etiocholan-3b-ol-17-one, 3ß-Hsd produces an insoluble diformazan precipitate by transferring protons to a proton acceptor such as tetrazolium salts. Therefore, nitroblue tetrazolium, a color substrate, was used for the specific detection of 3ß-Hsd activity in adrenal/interrenal tissue (16, 17). After overnight fixation in 4% paraformaldehyde in PBS (4% PFA/PBS), embryos were washed twice with PBST. The chromogenic reaction was perfomed at 37 C in a substrate of 2.0 mg of etiocholan-3b-ol-17-one (Sigma E-5251) dissolved in 15 ml 0.1 M phosphate buffer, pH 7.2, and 5 ml 50% polyvinylpyrrolidone in the same phosphate buffer, 30.0 mg ß-nicotinamide adenine dinucleotide (catalog no. N-1511; Sigma), 20.0 mg nitroblue tetrazolium (catalog no. 74030; Fluka Chemical Co., Buchs, Switzerland), 50.0 mg EDTA (Merck, Darmstadt, Germany), and 0.2 ml N,N-dimethyl formamide (Merck). Reactions were monitored until sufficient signal intensities were obtained (3–8 h, depending on embryonic/larval stage). Staining reactions were terminated by washing in PBST followed by fixation in 4% PFA/PBS for 1 h. In control embryos, the substrate was omitted.
Densitometry and Statistics for ISH Signals
In situ hybridized embryos were manually dissected from the yolk and flat mounted in benzyl alcohol-benzyl benzoate (2:1) for one-color ISH or in benzyl alcohol-glycerol (5:1) for the two-color ISH for taking photographs. For comparison, photos of embryos from the respective groups were always taken in the identical orientation and illumination using the same magnification. Areas and density of the respective signal were measured by the Image Gauge program, version 3.4 (Fuji, Duesseldorf, Germany).
Significance of differences was evaluated by ANOVA using the statistical software program Stat View 4.51. A value of P < 0.05 was considered statistically significant with post hoc analysis carried out by Fisher-projected least significant difference test. All results are expressed as means ± SEM.
Cell Proliferation and Cell Counting
Proliferating cells were detected by immunohistochemistry using antibodies against phosphorylated histone H3 (Chemicon, Hampshire, UK) according to the method of Saka and Smith (51). To localize interrenal tissue, embryos were first stained for 3ß-Hsd activity as described above. 3ß-Hsd-stained embryos were then washed several times in PBST, fixed in 4% PFA/PBS for 20 min, again washed with PBST, and kept in PBST at 4 C until further treatment.
Embryos were then treated with proteinase K (40 µg/ml with an incubation time of 40 min). After proteinase K treatment, embryos were fixed again in PFA for 20 min, washed several times with PBST and 3 x 15 min in PBS-0.5% Triton X-100, followed by several washes in 3–4 h in pure water and incubation for 2 h in blocking solution [1% (vol/vol) dimethylsulfoxide, 1% (wt/vol) BSA, 1% (vol/vol) goat serum, 0.5% Triton X-100]. The embryos were then incubated overnight at 4 C in antiphospho histone H3 antibody solution diluted 1:1000 in blocking solution. The following day embryos were thoroughly washed several times and then incubated with antirabbit, biotin-conjugated secondary antibody (Vector Laboratories, Peterborough, UK) for 2 h at room temperature. The embryos were then rinsed with blocking solution followed by an overnight wash in PBST with 0.1% Triton at 4 C. The next day embryos were further washed two times, 15 min each, in PBST with 0.1% Triton at room temperature and then stained using the avidin-biotinylated enzyme complex (ABC) kit and 3,3'-diaminobenzidine substrate (Vector Laboratories).
Cell number of interrenal tissue was repeatedly counted in 3ß-Hsd-stained embryos under the microscope in a treatment-blinded fashion. Counting of cells had been trained previously using images of 3ß-Hsd-stained embryos at different stages of development. The coefficient of variation for the cell count was less than 15% for the stages 3 dpf–5 dpf. Significance of differences was calculated by Mann-Whitney U test.
|
ACKNOWLEDGMENTS |
|---|
|
FOOTNOTES |
|---|
Disclosure Statement: The authors have nothing to disclose.
First Published Online November 2, 2006
Abbreviations: AP, Alkaline phosphatase; dßh, dopamine ß-hydroxylase; dex, dexamethasone; dpf, days post fertilization; E, embryonic day; GFP, green fluorescent protein; hpf, hours post fertilization; 3ß-Hsd, 3ß-hydroxysteroid dehydrogenase; ISH, in situ hybridization; mc2r, acth receptor; PBST, Tween in PBS; 4% PFA/PBS, 4% paraformaldehyde in PBS; POMC, proopiomelanocortin; SF1, steroidogenic factor 1; star, steroidogenic acute regulatory protein; TPIT, pituitary cell-restricted T-box factor.
Received for publication May 22, 2006. Accepted for publication October 26, 2006.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
-melanocyte-stimulating hormone cleavage in adrenal gland undergoing compensatory growth. Nature 306:70–73[CrossRef][Medline]This article has been cited by other articles:
 |
|
 |
|
  J. M. Hillegass, C. M. Villano, K. R. Cooper, and L. A. White Matrix Metalloproteinase-13 Is Required for Zebra fish (Danio rerio) Development and Is a Target for Glucocorticoids Toxicol. Sci., November 1, 2007; 100(1): 168 - 179. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Endocrinology | Endocrine Reviews | J. Clin. End. & Metab. |
| Molecular Endocrinology | Recent Prog. Horm. Res. | All Endocrine Journals |
