Logo of endoArchiveHomepageTES HomepageSubscriptionsSubmissionAbout
Endocrinology. 2008 Sep; 149(9): 4435–4451.
Published online 2008 May 22. doi:  10.1210/en.2008-0315
PMCID: PMC2553376

Graded Hedgehog and Fibroblast Growth Factor Signaling Independently Regulate Pituitary Cell Fates and Help Establish the Pars Distalis and Pars Intermedia of the Zebrafish Adenohypophysis


The vertebrate adenohypophysis forms as a placode at the anterior margin of the neural plate, requiring both hedgehog (Hh) and fibroblast growth factor (Fgf) mediated cell-cell signaling for induction and survival of endocrine cell types. Using small molecule inhibitors to modulate signaling levels during zebrafish development we show that graded Hh and Fgf signaling independently help establish the two subdomains of the adenohypophysis, the anteriorly located pars distalis (PD) and the posterior pars intermedia (PI). High levels of Hh signaling are required for formation of the PD and differentiation of anterior endocrine cell types, whereas lower levels of Hh signaling are required for formation of the PI and differentiation of posterior endocrine cell types. In contrast, high Fgf signaling levels are required for formation of the PI and posterior endocrine cell differentiation, whereas anterior regions require lower levels of Fgf signaling. Based on live observations and marker analyses, we show that the PD forms first at the midline closest to the central nervous system source of Sonic hedgehog. In contrast the PI appears to form from more lateral/posterior cells close to a central nervous system source of Fgf3. Together our data show that graded Hh and Fgf signaling independently direct induction of the PD and PI and help establish endocrine cell fates along the anterior/posterior axis of the zebrafish adenohypophysis. These data suggest that there are distinct origins and signaling requirements for the PD and PI.

THE PITUITARY GLAND consists of two major lobes: the epidermally derived endocrine lobe, the adenohypophysis, and the neurally derived regulatory lobe, the neurohypophysis. The organization and cellular composition of the pituitary is highly conserved across vertebrate species, from teleosts to mammals (1). During embryogenesis of all vertebrate species, the adenohypophysis forms as one of 10 cranial placodes from epidermal tissue adjacent to the developing brain (2). In mammals, oral ectoderm invaginates to form Rathke’s pouch, which in turn thickens to form the adenohypophysis (3), whereas in zebrafish, epidermal cells at the anterior margin of the neural plate thicken and coalesce at the midline to form the pituitary placode (4). Fate-mapping studies in both fish (5) and mammals (6) showed that the adenohypophyseal placode is derived from cells that reside at the anterior neural ridge and are continuous with a preplacodal region that contributes to all of the cranial placodes (2,7). In zebrafish, cells at the midline of the anterior neural ridge are induced to express pituitary markers and express hormone genes well before oral cavity formation (4,8), suggesting early signals from the central nervous system (CNS) are responsible for inducing distinct placodal identities from the preplacodal region.

The vertebrate adenohypophyseal placode is further divided into two distinct regions, the pars distalis (PD) anteriorly and the pars intermedia (PI) posteriorly (9,10,11,12). In both mammals and teleosts, pituitary precursor cells give rise to a variety of endocrine cell types within the adenohypophysis (4,13). During development, endocrine cells are organized along the dorsal-ventral axis in mammals and the anterior-posterior axis of the adenohypophysis in teleosts (4,13). Within the developing adenohypophysis of the teleosts, the most anteriorly located cells are ACTH-secreting corticotropes and prolactin (Prl)-secreting lactotropes (4,14). More medially positioned cells include GH-secreting somatotropes, TSH-secreting thyrotropes, and FSH- and LH-secreting gonadotropes (4,15). MSH-secreting melanotropes are located in the posterior adenohypophysis (3,14). The prohormone gene pro-opiomelanocortin (pomc) is expressed both in ACTH-secreting corticotropes of the PD and in MSH-secreting melanotropes of the PI, providing a convenient marker for both anterior and posterior endocrine cell differentiation (4,13,16). Importantly, this basic arrangement of endocrine cell types along the embryonic adenohypophyseal axis appears to be conserved between mammals and teleosts, implying conserved developmental mechanisms (4,8,9,14,17).

A number of signaling molecules, including bone morphogenetic proteins, fibroblast growth factors (Fgfs), Sonic hedgehog (Shh), Wnts, and Notch have now been implicated in pituitary induction and maintenance (1), but how these molecules affect functional patterning of the pituitary is largely unknown. Hedgehog (Hh) signaling is absolutely required for pituitary formation (3,8). Hh signaling is required in ectodermal cells, suggesting this requirement is direct (18). Using timed loss of Hh signaling in zebrafish, we previously demonstrated that Hh signaling from the neural ectoderm is required to induce expression of the pituitary marker lim3 at the anterior margin as early as 10–15 h post fertilization (hpf), a time when these cells are actively responding to Hh signals (8).

Gradients of secreted signaling molecules are known to provide positional information in the embryo and influence the patterning of embryonic tissues. One of the best studied morphogens is Shh (19). In the spinal cord, a gradient of Shh activity from the notochord and floor plate influences the expression of a large number of transcription factors in neural precursor cells. The combinations of these transcription factors then direct differentiation of distinct neuronal fates along the dorsal/ventral axis (20,21). We previously showed that Hh signaling, most likely from an anterior/dorsal source in the diencephalon, is required to maintain expression of the Hh target gene nkx2.2a in the anterior region of the adenohypophysis (8). These data suggested that there is a differential response to Hh in the developing placode along the anterior/posterior axis, but whether and how Hh signaling levels pattern the adenohypophysis and regulate endocrine cell differentiation is not known.

Like Hh, Fgf signaling is required for pituitary formation, in which it is thought to act as a mitogen based on mammalian studies (22,23) and a cell survival factor based on studies in both mammals and fish (1,24). In zebrafish, loss of Fgf3 signaling results in apoptosis of adenohypophyseal cells similar to the loss of Fgf10 in mammals (25,26,27). fgf3 is expressed adjacent to the developing pituitary and is maintained posterior to the placode (25). A recent study showed that Fgf signaling is required for proper development of the adenohypophysis between 18–22 hpf in zebrafish, suggesting that Fgf signaling may be necessary for both induction and survival of adenohypophyseal cells at the anterior margin (25). The posterior source of fgf3 is consistent with a role for graded Fgf signaling in patterning the placode along the anterior/posterior (A/P) axis, but whether and how this occurs is also unknown.

Given the complementary expression of Fgf and Shh near the developing adenohypophyseal placode and the known role for both of these molecules in overall formation of the adenohypophysis, we examined the relationship between Hh and Fgf signaling in early pituitary induction and anterior/posterior patterning. We chose to do these studies in zebrafish embryos because early developmental stages are highly accessible to experimental analysis and live imaging. Specifically, we tested whether levels of signaling play a role in patterning the placode along the anterior/posterior axis and guiding specific endocrine cell fates. By manipulating levels of signaling with small molecule inhibitors, we found that graded Fgf and Hh signaling help induce and pattern the adenohypophysis in a dose-dependent manner. We show that a subset of Hh-responsive transcription factors is expressed in the placode in a manner consistent with an anterior/posterior gradient of Hh signaling. Graded loss of Hh, but not Fgf, led to ectopic midline lens formation, with either one or two lenses forming depending on the level of Hh signaling. These two lenses appear to originate from two sublobes of the adenohypophysis that form early in embryogenesis and correspond to the PD and PI. Our results suggest that Hh and Fgf signaling act independently to induce and pattern the adenohypophysis, with high doses of Hh signaling being required for the induction of the anteriorly located PD and high doses of Fgf signaling being required for the induction of the posterior PI. Together, these data demonstrate a role for Hh and Fgf dosage in the functional patterning of the pituitary and suggest that the PD and PI arise from distinct origins with distinct signaling requirements.

Materials and Methods

Fish lines

Wild-type and mutant fish were maintained at the University of Massachusetts, Amherst zebrafish facility. The embryos were collected and incubated at 28 C and staged (according to Ref. 28). Then the embryos were fixed with 4% paraformaldehyde at the desired stage. The wild-type line used was TL and the mutant lines used were detour (dtrts269), a loss of function Gli1 allele (29), and you-too (yotty17), which encodes a dominant repressor form of Gli2 (30).

In situ hybridization and antibody labeling

Whole-mount in situ hybridization was performed as previously described (30). Probes used were lim3 (31), nkx2.2a (32), pax6a (33), pax7 (34), prl, pomc, tsh, gh (4), and α-gsu (15). Twenty-five-micrometer cryosections of larval and adult zebrafish heads were prepared and hybridized as reported elsewhere (35). The Pax7 monoclonal antibody was obtained from Developmental Studies Hybridoma Bank. Antibody labeling after in situ labeling was performed as previously described (36) with the anti-Pax7 primary antibody being added with the antidigoxigenin antibody in the in situ protocol. After the 4-nitro blue tetrazolium chloride/5-bromo-4-chloro-3-indoyl-phosphate, 4-toluidine salt (NBT/BCIP)(Roche, Mannheim, Germany) color reaction, biotin-avidin-horseradish peroxidase labeling was performed using a diaminobenzidine substrate (Sigma-Aldrich, St. Louis, MO). Labeled embryos were postfixed with 4% paraformaldehyde and then cleared in 75% glycerol and photographed using differential interference contrast (DIC) optics.

mRNA injections

Shh mRNA was in vitro transcribed using T7 polymerase (mMessage mMachine-T7 kit, Ambion, Austin, TX). The Shh/T7TS plasmid (37) was linearized with BamHI. Wild-type embryos were injected with approximately 100 pg of Shh RNA at the one- to two-cell stage. Injected embryos were incubated at 28 C until 24–25 hpf, fixed with 4% paraformaldehyde, and processed for in situ hybridization. This experiment was repeated twice and a minimum of 15 embryos were analyzed for each marker.

Hh signaling inhibition: cyclopamine treatments

Wild-type embryos were treated with 10, 20, 40, 60, and 100 μm cyclopamine (Toronto Research Chemicals, North York, Ontario, Canada) in 1 ml of embryo medium (2 m MgSO4, 2 m KCl, 2 m CaCl2, 0.5 m NaH2PO4, 5 m NaCl) from a stock of 10 mm cyclopamine, starting at shield stage or tailbud stage. The embryos were incubated in a 12-well plate at 28 C. As a control, embryos were treated with equal volumes of ethanol as described (8). The embryos were fixed at 24 hpt with 4% paraformaldehyde at 4 C overnight and later processed for in situ hybridization. The student’s t test was used to compare the number of endocrine cells affected with different concentrations (10, 20, 40, 60 μm) of cyclopamine. This experiment was repeated three times.

Bead implantation

Affi-gel blue beads (Bio-Rad, Hercules, CA) were washed with PBS twice for 10 min. The beads were incubated with 1 mg/ml Shh (Curis Inc., Cambridge, MA) or BSA overnight at 4 C. Dechorionated embryos were immobilized by dropping into 1.2% agarose (low melting temperature; Sigma), and each embryo was pipetted up and placed onto the agarose-coated petri dish at 22–24 hpf. Once the agarose was cooled down, the skin behind the eye or at the anterior neural edge was torn gently using a pulled glass needle. Several Shh or BSA-coated beads were pipetted onto the implantation site, and one bead was pushed through the skin with the help of forceps. The bead-implanted embryos were incubated at 28 C until 32 hpf and then fixed using 4% paraformaldehyde. This experiment was repeated three times. A minimum of nine embryos was analyzed from each group.

Fgf signaling inhibition: SU5402 treatments

Embryos were treated with 0.9, 1.8, 3, 6, and 12 μm SU5402 (Calbiochem, San Diego, CA) from a 3 mm stock starting at tailbud stage for the described time intervals at 28.5 C. The low doses for SU5402 treatments were determined based on the intact morphology of the adenohypophysis. Control embryos were treated with dimethylsulfoxide (DMSO). The embryos were incubated in 1ml embryo medium in 12-well plates until 28 or 48 hpf. Another set of embryos was incubated with 20 μm cyclopamine (CyA), and 0.9, 1.8, 3, 6, and 12 μm SU5402 together starting from 10 until 28 hpf. Embryos were fixed at the desired stages and then processed for in situ hybridization. The student’s t test was used to compare the number of endocrine cells affected in different concentrations (0.9, 1.8, 3 μm) of SU5402. This experiment was repeated twice.

Live imaging

POMC-green fluorescent protein (GFP) transgenic embryos were embedded in 1.2% agarose (low melting temperature; Sigma) for live imaging. One to two drops of embryo medium were used to keep the embryos alive. Imaging of POMC-GFP embryos was performed using and Axioplan2 imaging microscope and Axiovision 4.5 software (Zeiss, Thornwood, NY). Time-lapse image stacks were collected every 10 min starting from 23 until 34 hpf. On average, 12 z-stacks were collected at a spacing of approximately 1.5 μm.


A Hh transcriptional response in the adenohypophysis

In the vertebrate spinal cord, Hh dosage helps establish expression domains of transcription factors that then guide dorsal/ventral neuronal cell fates (20). We previously showed that Hh is required for the induction and patterning of the adenohypophysis (8) and that the Hh response gene nkx2.2a is expressed in the anterior region of the placode by 24 hpf, close to the source of Hh in the overlying hypothalamus (8). This expression of nkx2.2a near the source of Hh is similar in the neural tube (20,38) and led us to wonder whether Hh might be regulating the expression of other class I and II Hh-response genes in the pituitary placode, suggestive of a dose sensitive role in anterior/posterior cell differentiation. We therefore analyzed the expression of other Hh-regulated genes in the adenohypophysis. pax7 is a class I Hh-responsive gene expressed dorsally in the spinal cord and repressed by Hh activity (20). We found that Pax7 is expressed in the posterior adenohypophysis in a pattern that is complementary to that of nkx2.2a (Fig. 1A1A).). The pax7 expressing region of the adenohypophysis is distant from the source of Hh, as is the case in the spinal cord (20,38). This expression in the posterior adenohypophysis persists through larval stages and into the adult (Fig. 11,, B and C). Based on expression relative to prl-expressing cells (see Figs. 66 and 88),), this pax7 domain appears to correspond to the PI, whereas the anterior nkx2.2a-expressing region corresponds to the PD. The nkx2.2a and pax7 expression patterns suggest that the adenohypophysis is divided into two subdomains very early in development, and these genes thus provide early markers for PD and PI formation.

Figure 1
A Hh transcriptional response in the adenohypophysis. A, The class II Hh-responsive gene nkx2.2a is expressed in the anterior adenohypophysis (PD) and the class I Hh-responsive transcription factor, Pax7 is expressed in the posterior adenohypophysis (PI) ...
Figure 6
Fgf and Hh independently influence adenohypophysis development. A–H, Lateral views of the adenohypophysis (brackets), anterior left. A, DMSO control showing normal prl (red) expression anteriorly in the PD and pax7 (blue) expression posteriorly ...
Figure 8
Schematic view of adenohypophyseal development relative to sources of Shh and Fgf. A and E, At 20.5 hpf, lim3 is expressed in two distinct bands across the anterior margin of the CNS (E, inset). The Hh-responsive gene nkx2.2a is expressed only in the ...

A broad survey of Hh-responsive gene expression also confirmed that the class I gene pax6a is an early adenohypophyseal marker (39,40). pax6a was expressed throughout the adenohypophysis (Fig. 1E1E),), similar to its broad expression domain in the spinal cord (20,38). Neurally expressed Hh-responsive genes that were not expressed in the adenohypophysis include nkx2.9, nkx6.2, dbx1, dbx2, and pax3 (data not shown).

To verify that Hh regulates the expression of nkx2.2a, pax6a, and pax7 in the adenohypophysis, we next examined expression in the Hh pathway mutants detour (dtr/gli1) and you-too (yot/gli2DR), and in Shh mRNA-injected embryos (29). dtr (gli1) mutant embryos have a reduced Hh response in the spinal cord, whereas yot (gli2 DR) mutants encode a dominant repressor form of Gli2, which lead to a more severe reduction in the Hh response (29,38). In dtr(gli1) mutants, nkx2.2a expression was severely reduced, and pax6a expression was slightly reduced in the adenohypophysis (Fig. 11,, G and H). Interestingly, pax7 expression was slightly expanded into the anterior adenohypophysis in dtr(gli1) mutants (Fig. 11,, F and I) as is seen with reduced Hh signaling in the mammalian spinal cord (41). Pituitary nkx2.2a expression was completely absent in yot(gli2DR) mutants, and an ectopic midline lens formed in the place of the adenohypophysis (Fig. 1J1J)) (30). pax6a expression was relatively normal in yot(gli2DR) mutants that contained ectopic midline lenses (Fig. 1K1K),), consistent with the expression of pax6 in developing lens (42). pax7 expression was largely unaffected in yot(gli2DR) mutants (Fig. 1L1L),), suggesting that the PI is still present (42). To test whether expression of these transcription factors is induced by Hh signaling, we injected mRNA encoding Shh into two cell-staged embryos. In Shh mRNA-injected embryos, nkx2.2a expression was expanded into the posterior adenohypophysis (Fig. 1M1M),), consistent with its activation by Hh in this tissue. Expression of pax6a and pax7 was reduced (Fig. 11,, N and O), consistent with these genes being negatively regulated by Hh, as is the case in the spinal cord (20). Given the anterior/dorsal source of Hh in the overlying hypothalamus, these data suggest that graded Hh signals from the CNS help pattern pituitary precursor cells along the anterior/posterior axis of the placode.

Hh concentration-dependent patterning of the adenohypophysis

To determine whether levels of Hh signaling affect pituitary patterning, we manipulated the doses of Hh signaling using different concentrations of the alkaloid drug Cyclopamine (CyA), (43,44) and assayed the expression of nkx2.2a in the anterior placode and pax7 in the posterior placode at 25 hpf. CyA has been shown to be a potent and specific inhibitor of the Hh receptor-complex molecule Smoothened (45). Zebrafish embryos treated with higher doses of CyA (100 μm) display developmental defects that appear identical with smoothened loss-of-function mutants, further suggesting CyA specifically blocks Hh signaling in the early embryo (8,46,47). Treatment with a low dose of CyA (20 μm), starting at the shield stage, led to loss of nkx2.2a expression in the PD (Fig. 2B2B and Table 11).). This same dose of CyA resulted in no change in the expression of posterior pax7, reduced pax7 expression, or an expansion of pax7 expression into the anterior region (Fig. 2E2E and Table 11).). This result suggests that a high level of Hh signaling is required for the activation of nkx2.2a and the repression of pax7 in the PD. At this dose of CyA, one ectopic lens often formed in the place of the PD, whereas the PI was largely unaffected, as indicated by the expression of pax7, lim3, and pomc posterior to the lens (Fig. 22,, B, E, and H, and Table 11).

Figure 2
Hh signaling levels affect the Hh transcriptional response in the adenohypophysis. A–F, Lateral views of the adenohypophysis (bracketed) of 25-hpf embryos, anterior to the left. A and B, Treatments with 20 μm CyA led to the loss of nkx2.2a ...
Table 1
CyA dose, ectopic midline lenses, and PD/PI formation

Reducing Hh signaling further resulted in the formation of two lenses and the loss of the PI (Fig. 2C2C and Table 11).). The loss of nkx2.2a, pax7, pomc, and lim3 expression (Fig. 22,, C, F, and I) is consistent with previous data, which showed a complete loss of adenohypophyseal cells in the absence of Hh signaling (100 μm CyA or Hh deficient mutants) (Fig. 22,, data not shown) (4,8). These results demonstrate that the anterior and posterior adenohypophysis have different sensitivities to Hh signaling loss. Furthermore, the anterior and posterior regions behave distinctly in the presence of slightly reduced levels of Hh signaling, with only the anterior domain forming an ectopic lens and the posterior domain still expressing posterior specific markers.

Graded Hh signaling and endocrine cell differentiation

To determine how Hh signaling levels affect endocrine cell differentiation and the functional patterning of the pituitary, we used various doses of CyA (10, 20, 40, and 60 μm) and analyzed the expression of pomc, prl, gh, α-gsu, and tsh. These endocrine cell types are spatially organized along the anterior/posterior (A/P) axis of the adenohypophysis of embryos by 48 hpf (4,15). pomc mRNA is expressed in both anteriorly located acth cells of the PD and posterior cells of the PI that later express msh, whereas prl and gh cells are located only in the PD (Fig. 33)) (4). tsh- and α-gsu-expressing cells overlap and are found in both the PD and PI (see Figs. 33 and 88)) (4,15). After a mild reduction of Hh signaling (10 μm CyA), anterior pomc- and gh-expressing cells were significantly reduced or absent, whereas the other cells were not affected (Fig. 33,, B, F, J, N, and Q). A further decrease in Hh signaling levels (20 μm CyA) led to loss of anterior pomc-expressing cells and severe reduction in the number of prl-expressing cells (Fig. 33,, C and Q). The numbers of tsh- and α-gsu-expressing cells were only slightly reduced and posterior pomc cells were not affected (Fig. 33,, C, K, O, and Q). A 40-μm concentration of CyA treatment led to the complete loss of prl cells and posterior pomc cells, a major loss of α-gsu cells, and a smaller, but significant, reduction in tsh cells, and the sizes of the pituitaries were severely reduced (Fig. 33,, D, H, L, P, and Q). Higher doses of CyA completely eliminated the adenohypophysis and no endocrine cell types were detectable (8). Given the spatial arrangement of cells in the placode, these results are consistent with the idea that cells in the PD require the highest levels of Hh signaling to differentiate, whereas cells in the PI can still differentiate normally when Hh levels are moderately reduced. Together these data indicate that Hh signaling helps to functionally pattern the placode, with high levels of Hh being required for endocrine cell differentiation in the PD.

Figure 3
Differential sensitivities of endocrine cell types to changes in Hh signaling levels. A, pomc-expressing cells (blue) are normally seen both in the anterior (PD) and posterior (PI) regions of the adenohypophysis, whereas prl-expressing cells (red) are ...

Ectopic Hh signaling anteriorizes the posterior adenohypophysis

To directly test whether Hh can act at a distance to influence the patterning of the pituitary, we created an ectopic source of Hh by implanting a Shh-coated bead near the posterior region of the developing placode (Fig. 44,, A and B). This exogenous source of Hh led to ectopic expression of nkx2.2a in the posterior adenohypophysis (Fig. 4D4D)) (nine of 10 Shh-bead implanted embryos). nkx2.2a was expressed normally only in the anterior region of the placode when BSA-coated beads were implanted in the same region (Fig. 4C4C,, n = 11). This ectopic, posterior source of Shh thus anteriorized the adenohypophysis.

Figure 4
An ectopic source of Hh can anteriorize the posterior adenohypophysis. A, Lateral view of a 22-hpf embryo implanted with a BSA or Shh-coated affi-gel blue bead (arrow). Inset shows dorsal view of a bead implanted at 20 hpf. B, A Shh-coated bead positioned ...

Graded Fgf signaling and endocrine cell differentiation

We next examined whether Fgf signaling might contribute to cell differentiation along the A/P axis of the early placode. In mammals, Fgf10 is required for pituitary cell survival (26,27), whereas Fgf8 is required for activation of pituitary-specific transcription factors (18,48,49). In zebrafish, fgf3 is expressed posterior to the PI (25), opposite to the source of Shh (8). Previous studies in zebrafish showed that a loss of Fgf function results in a loss of the adenohypophysis (25). Fgf was clearly shown to play a role in survival of placodal cells, and there was also an indication that Fgf might be involved in initial induction of general pituitary fates (25). To test the role of Fgf signaling levels on patterning the placode along the A/P axis, we applied increasing doses (0.9, 1.8, 3 μm) of the Fgf receptor (FgfR) inhibitor SU5402 starting at 10 hpf, before placode formation, and analyzed the expression of endocrine cell types at 48 hpf. SU5402 has been shown to selectively block the kinase activity of FgfR1 (50). Zebrafish embryos treated with higher doses of SU5402 display pituitary defects identical with Fgf3 null mutants (25), suggesting SU5402 specifically blocks Fgf signaling in the zebrafish embryo and that FgfR1 is the main mediator of FGF signaling in the pituitary. To determine whether different doses of SU5402 could be used to modulate Fgf signaling levels, we analyzed expression of the Fgf-regulated gene erm (51) in 0.9, 1.8, and 3 μm SU5402-treated embryos and showed graded loss of erm expression with increasing doses of SU5402, with a complete loss of erm at the highest dose (supplemental figure, published as supplemental data on The Endocrine Society’s Journals Online web site at http://endo.endojournals.org) (52).

A mild reduction in Fgf signaling (0.9 μm SU5402) affected cells in the PI much more severely than those located in the PD. pomc cell numbers were severely reduced in the PI but were unaffected in the PD (Fig. 55,, B and Q). Interestingly, gh cells of the PD were reduced by approximately 50% (Fig. 55,, F and Q). This treatment had a little or no effect on the number of other endocrine cell types (Fig. 55,, B, J, N, and Q). Decreasing Fgf signaling further with 1.8 μm SU5402 reduced posterior pomc cells further and caused a reduction in anterior pomc-, prl-, and α-gsu-expressing cells (Fig. 55,, C, K, and Q). This treatment did not significantly affect gh and tsh cell numbers (Fig. 55,, G, O, and Q). More complete blockage of Fgf with 3 μm SU5402 affected all cell types (Fig. 55,, D, H, L, P, and Q). Still higher doses of SU5402 (6 and 12 μm) completely eliminated all pituitary cell types (data not shown), consistent with previous work showing the requirement for Fgf in pituitary formation and cell survival (25).

Figure 5
Differential sensitivities of endocrine cell types to changes in Fgf signaling levels. A, In DMSO-treated control embryos, pomc (blue) expression overlaps with prl (red) expression in the anterior but not posterior cells. B, Treatment with 0.9 μ ...

Fgf signaling and pituitary induction

To examine how Fgf dosage affects earlier differentiation of the anterior vs. posterior regions of the adenohypophysis, we examined the expression of the early PD and PI markers nkx2.2a and pax7 as well as prl, pomc, and lim3 at 28 hpf. Reducing Fgf signaling slightly (0.9 μm SU5402) led to a mild reduction in pax7 and lim3 expression in the posterior cells but had no effect on prl, nkx2.2a, and pomc expression in anterior cells (Fig. 6B6B and Table 22).). Treatment with 1.8 μm SU5402 led to either loss or reduction of pax7 expression in the posterior adenohypophysis (Fig. 6C6C and Table 22).). Similarly, lim3 expression was reduced in the posterior adenohypophysis, whereas anterior pomc, nkx2.2a, and prl expression appeared unaffected in 1.8 μm SU5402-treated embryos (Fig. 6C6C and Table 22).). Further reduction of Fgf signaling (3 μm SU5402) led to the loss of pax7 and reduced lim3 posteriorly, whereas anterior nkx2.2a, pomc, and prl expression was slightly reduced, indicating that the PD was only mildly affected (Fig. 6D6D and Table 22).). Treatment with higher doses of SU5402 (6 μm) resulted in the complete loss of all pituitary markers including nkx2.2a, anterior pomc, and prl and lim3 (data not shown). These results suggest that high levels of Fgf signaling are required for the induction of the PI, whereas the PD requires lower levels of Fgf signaling to form. This is consistent with the data above showing that melanotropes of the PI are the most sensitive to loss of Fgf signaling, whereas lactotropes and corticotropes of the PD are less sensitive. Interestingly, cell position does not predict sensitivity to altered FGF levels for thyrotropes and somatotropes, suggesting distinct signaling requirements for these cell types or that their precursors may differentiate in a different region of the adenohypophysis (see Fig. 55).

Table 2
CyA, SU5402 dose, and PD/PI formation

To begin to determine whether Fgf signaling might be acting as an early pituitary inducing factor (in addition to its known role as a cell survival factor), we carefully examined lim3 and pax7 expression at the time pax7 first becomes detectable in the posterior placode. At 24–25 hpf, lim3 is expressed in a horseshoe shape at the anterior margin of the CNS, with the lateral regions of expression being thought to contribute to the posterior adenohypophysis (24). Consistent with pax7 being a marker for the posterior region of the adenohypophysis, pax7 is first expressed in these lateral/posterior cells at this age, 25 hpf (Fig. 6K6K).). Because 3 μm SU5402 selectively eliminated the posterior adenohypophysis when embryos were assayed at 48 hpf, we used this concentration to test whether Fgf signaling specifically affects the initial induction of these posterior cells, as assayed by expression of lim3 and pax7. A 3-μm concentration SU5402 eliminated the lateral lim3-expressing cells, whereas medial/anterior lim3-expressing cells were still present (Fig. 6J6J).). Consistently, pax7 was absent in the adenohypophysis in 3 μm SU5402-treated embryos (Fig. 6L6L).). Similar to lim3 and pax7, the expression of the preplacodal gene pitx3 was also absent in the posterior cells with low doses of SU5402 at 28 hpf (data not shown). These results are consistent with the idea that these lateral posterior cells are the source of the posterior adenohypophysis and further suggest that Fgf signaling is responsible for the early induction of the posterior cell fates. Consistent with this early role for Fgf in induction of the posterior adenohypophysis, fgf3 is expressed in bands adjacent to these lateral/posterior lim3-expressing cells [data not shown (25)].

Shh and Fgf signaling act independently in A/P pituitary patterning

Fgf and Hh have been shown to interact during embryonic patterning and can antagonize each other (e.g. in neural precursor differentiation) (53) or synergize (e.g. in patterning the ventral telencephalon) (54). Given the complementary expression of Fgf and Shh on opposite sides of the developing placode and the opposite effects of reduced Fgf and Shh signaling in A/P pituitary patterning, we next wondered whether Hh and Fgf signaling might oppose each other during A/P patterning of the adenohypophysis.

To determine whether Fgf signaling and Shh signaling interact during adenohypophysis induction and A/P patterning, we combined doses of the SU5402 and CyA that separately affect just the PI or PD, respectively. Similar to the data shown in Fig. 22,, 20 μm CyA treatment alone led to the loss of nkx2.2a, anterior pomc and anterior lim3, and slightly reduced prl expression, whereas pax7 expression in the posterior placode was unaffected or expanded to the anterior adenohypophysis (Fig. 6E6E and Table 22).). Because pomc is expressed only anteriorly at 28 hpf (4), these data confirm the selective effect of CyA on anterior cell types. As expected, 0.9 μm SU5402 treatments alone had the complementary effects, causing a slight reduction in pax7 and lim3 expression, whereas nkx2.2a, pomc, and prl expression in the anterior region appeared unaffected (Fig. 6B6B and Table 22).). Treatment with both 20 μm CyA and 0.9 μm SU5402 resulted in the loss of nkx2.2a, anterior pomc, and anterior lim3, whereas prl and pax7 expression were slightly reduced (Fig. 6F6F and Table 22).). Similarly, 20 μm CyA and 1.8 μm SU5402 together eliminated nkx2.2a and pomc expression, reduced lim3 and prl expression, and eliminated pax7 expression (Fig. 6G6G and Table 22).). Treatment of 20 μm CyA together with 3 μm SU5402 led to a severe reduction and/or loss of anterior nkx2.2a, pomc, and prl as well as the loss of lim3 and pax7 expression in the adenohypophysis (Fig. 6H6H and Table 22).). The effects of blocking Fgf and Hh signaling thus seemed to be additive because combining the two inhibitors did not alter the effect seen when each drug was administered separately (Fig. 66,, F–H, and Table 22).). Consistent with independent Fgf and Hh action, ptc1 expression appeared normal in embryos treated with 3 μm SU5402 alone, and erm expression appeared normal in 20 μm CyA-treated embryos (data not shown), suggesting Fgf and Hh do not influence the reciprocal transcriptional response. Together these data show that Fgf and Shh signaling act independently and in a complementary manner to pattern the early adenohypophysis.

Distinct origins of the PD and PI

Based on the distinct signaling requirements for the anterior and posterior regions of the adenohypophysis, we decided to examine early placode formation to determine when these regions of the adenohypophysis first become distinguishable. We took advantage of a newly developed POMC-GFP line (14) to examine early anterior and posterior endocrine cell fate decisions. The POMC-GFP line allows live observations of pomc gene expression in anterior corticotropes and posteriorly positioned melanotropes (14).

It was previously reported that pomc is first expressed around 23–24 hpf in bilateral cells of the pituitary placode (4,14). Consistently, in our time-lapse observations, pomc expression was first seen at the anterior margin of the CNS between 23 and 24 hpf (Fig. 77,, supplemental movie). Surprisingly, careful examination using differential interference contrast optics revealed a morphological border that divides the adenohypophysis into two distinct regions, or lobes, as early as 24 hpf (Fig. 77,, dotted lines). The first pomc-expressing cells appeared in the more anterior of the two lobes (Fig. 7D7D).). By 25 hpf, one pomc-expressing cell had appeared in the posterior lobe, and by 26 hpf bilateral pomc-expressing cells were located in both lobes (Fig. 77,, E and F). Tracking these cells through subsequent hours revealed that the more posterior cells contributed to the posterior population of pomc-expressing cells, whereas the earlier-appearing anterior cells ended up in the anterior region of the adenohypophysis (Fig. 77,, G–L).

Figure 7
Distinct origins for the PD and PI. A and B, Whole-mount, in situ-labeled, 28-hpf embryo showing prl expression (red) in the anterior adenohypophysis and pax7 expression (blue) in the posterior adenohypophysis. The black square is the area enlarged in ...

The adult adenohypophysis is divided into two distinct lobes, the PD and the PI (55,56). To verify that this early A/P organization in the embryonic adenohypophysis corresponds to the adult PD and PI, we analyzed the expression of pomc in larval and adult zebrafish. Similar to the expression at 48 hpf (Fig. 33),), pomc was expressed in two distinct regions at larval and adult stages (Fig. 77,, M and N). These anterior and posterior regions correspond to the PD anteriorly and the PI posteriorly based on published maps of the teleostean pituitary (57,58). As shown above, pax7 expression remains throughout the posterior region of the adenohypophysis, the PI in larval and adult stages (Fig. 11).). These data indicate that the bilobed adult organization of the adenohypophysis (the PD and PI) is established at the earliest stages of adenohypophysis development, starting as early as 23 hpf when the first endocrine cells differentiate. Based on nkx2.2a vs. pax7 expression as well as the appearance of pomc cells, the PD appears to form several hours before the PI. Together, these data indicate the PD and PI have distinct signaling requirements and distinct embryonic origins at the anterior margin, and they may thus represent distinct morphogenetic fields.


Distinct origins for the PI and PD of the adenohypophysis

Fate-mapping studies in both mammals and zebrafish indicate that the adenohypophysis is derived from the midline region of the anterior neural ridge (5,7). Previous work in zebrafish showed that this region contains precursor cells that can become either pituitary or lens, depending on their exposure to Hh signaling at about 15 hpf (5,8). Until now the adenohypophysis has been thought to derive from a single placode that coalesces from this anterior marginal domain. Here we show that there are clear A/P differences in this domain from the earliest stages of placode formation. Given that the adult adenohypophysis consists of two distinct regions arranged along the A/P axis (the PD and PI), we wondered whether the PD and PI might in fact derive from distinct cell populations in the embryo. This study provides several lines of evidence that the adenohypophyseal placode is divided into distinct regions from the beginning of its development and may in fact consist of two distinct morphogenetic fields.

First, we show that the anterior and posterior regions of the adenohypophysis respond distinctly to the loss of Hh and Fgf signaling. A slight reduction in Hh signaling (low concentrations of CyA) caused one ectopic lens to form in place of the PD, whereas the PI appeared unaffected. More severe Hh blockage (higher concentrations of CyA) led to two distinct ectopic midline lenses that formed at the expense of both the PD and PI. Conversely, a slight reduction in Fgf signaling eliminated the PI of the adenohypophysis, whereas the PD appeared largely unaffected. Thus, these two domains clearly respond distinctly to manipulations in developmental signaling systems.

Second, marker analyses suggest that the adenohypophyseal placode is divided into two broad A/P regions from the earliest stages of development and that these regions correspond to the PD and PI in the adult (see schematic in Fig. 88).). nkx2.2a expression was previously shown to be restricted to the anterior region of the placode (8), and here we show that this region corresponds to the PD based on overlap with PD-specific endocrine cells such as anteriorly positioned pomc-expressing corticotropes (Figs. 66 and 77).). We also show that pax7 is expressed in a complementary pattern, in the posterior region of the early placode. This posterior region corresponds to the PI based on overlap with PI-specific endocrine populations, specifically posteriorly positioned pomc-expressing cells that develop into melanotropes (Figs. 33 and 77)) (13,14). pax7 expression continues to be restricted to the PI in larval and adult stages, further suggesting this early division in the placode corresponds to the adult PI.

Third, we show that the PD and PI appear to originate from distinct regions at the anterior margin (Figs. 77 and 88).). lim3 is one of the earliest pituitary-specific genes and is expressed throughout the entire adenohypophysis at 28 and 48 hpf. Careful analysis uncovered two distinct regions of expression of lim3 expression at earlier times that appear to correspond to the PD and PI. At 20.5 hpf, lim3 is expressed in two bands across the midline of the anterior margin, with the more anterior/dorsal band also expressing nkx2.2a (Fig. 88,, data not shown). Four to 8 h later, lim3 is expressed in bilateral rows of cells that extend posteriorly from the midline expression domain to form a horseshoe (Figs. 77 and 88)) (24). These posterior cells appear to subsequently coalesce toward the midline to form a distinct posterior lobe, and as they coalesce, they begin to express the PI marker pax7. Importantly, SU5402 treatments that selectively block pax7 expression and PI formation at 28 and 48 hpf also selectively eliminated the bilateral horseshoe of lim3 expression at 25 hpf, consistent with the idea that these bilateral rows of cells are the source of the PI.

Finally, our live observations of adenohypophysis development allowed us to visualize a distinct border between the PD and PI as early as 25 hpf (Fig. 77).). One possibility is that Shh and Fgf morphogen gradients are converted into a distinct border as has been shown for Fgf signaling during somite formation (59). Alternatively, the PD and PI may in fact represent distinct placodes that coalesce from distinct epithelial populations. In fact, all other cranial sensory placodes in the developing vertebrate (lens, optic, olfactory, and otic placodes) are paired, suggesting this may be a general rule. That the adenohypophysis derives from two placodes may be difficult to observe because of its origin close to the midline and because of the early age at which the placode begins to form. Fine fate-mapping studies will be needed to confirm the distinct origin of the PD and PI, but together our data strongly suggest that the PD and PI originate from distinct regions and have distinct signaling requirements and may thus represent distinct morphogenetic fields.

Graded signaling and pituitary patterning

Both Hh and Fgf are well-known morphogens, with levels of signaling playing an important role in the patterning of different embryonic tissues. Whereas Shh and Fgf signaling have been previously shown to be required for pituitary development in both mammals and fish, there have been no studies investigating how signaling levels affect pituitary induction and patterning. We show that there is a Hh transcriptional response in the adenohypophysis with similarities to the Hh response code in the neural tube (60). A subset of the neuronal Hh transcriptional response is present in this epidermally derived tissue, with expression relative to the source of Hh being conserved (Fig. 11).). The spatial distribution of this Hh transcriptional response suggests a high anterior to low posterior gradient of Hh signaling. Whereas a graded transcriptional response to Fgf signaling is harder to document, previous studies show that fgf3 is expressed closest to the developing PI (25), suggesting a high posterior to low anterior signaling gradient (Fig. 88).

By applying different doses of CyA and/or SU5402, we were able to manipulate Hh and Fgf signaling levels during pituitary induction and patterning. At higher doses, both of these compounds cause defects that appear identical to loss-of-function mutations in their respective signaling pathways, suggesting there is little cross-interference with other signaling systems. By using a series of lower doses of these drugs, we show that graded Hh and Fgf signaling are critical for the induction and patterning of the vertebrate adenohypophysis. More specifically, high levels of Hh are required for the formation of the anterior adenohypophysis, the PD, whereas high levels of Fgf are required for the formation of the posterior adenohypophysis, the PI. Below a certain level of Hh signaling, preplacodal cells behave as lens precursors, and form lenses in a dose-dependent manner. In the absence of all Hh or Fgf signaling the adenohypophysis failed to form consistent with previous studies (8,25).

Furthermore, endocrine cells show differential sensitivities to changes in the levels of Hh and Fgf signaling that roughly correlate with A/P position. Anterior endocrine cell types were more sensitive to reductions in Hh signaling, whereas posterior cells were more sensitive to reductions in Fgf signaling. Given their position relative to the Shh and Fgf3 sources, anterior cells may be exposed to, and require, high levels of Hh signaling to differentiate, whereas posterior cells are exposed to and require high levels of Fgf signaling to differentiate (Fig. 88).). Interestingly, no endocrine cell type was seen to increase in number when either signaling system was disrupted (Figs. 33 and 55),), suggesting complete trans-fating events are not occurring in this system.

gh-producing somatotropes remain an enigma because they appear to show high sensitivity to reduction in both Hh and Fgf signaling. Based on their medial position, it is possible that Gh-expressing cells represent a border population that is highly sensitive to both Hh and Fgf signaling levels. Understanding how morphogen gradients are converted into distinct borders remains a major challenge (61), and this defined system may provide a model for tissue border formation in the vertebrate embryo. Identifying the location of the precursor population that gives rise to these cells may help elucidate why these cells respond uniquely to loss of Hh and Fgf signals.

In the spinal cord, graded Shh signaling establishes an intracellular gradient of Gli activity, with 2- to 3-fold changes in either Shh or Gli activity sufficient to alter fates of progenitor cells (62). The duration of signaling also contributes to morphogen activity of Shh. In fact, genes that are activated at the different doses of Shh in vitro are sequentially activated in vivo (63,64). This same timing of Hh-responsive gene expression is present in the early pituitary placode, with nkx2.2a being activated several hours before pax7, which is expressed further from the source of Shh. Thus, induction of more anterior genes may require extended periods of Shh signaling in addition to higher doses. This same dose/timing relationship has been shown for other developing tissues (65) and is suggested to be a general feature of a morphogen response. The sequential expression of nkx and pax genes is thus consistent with a morphogen response in the placode.

Mechanisms of Shh and Fgf action

There are several possible mechanisms by which graded Hh and Fgf signaling could influence pituitary development, including: 1) induction of the cell fates via regulation of Hh and Fgf-responsive transcription factors, 2) selective regulation of proliferation among unequal precursors, and 3) selective regulation of cell survival. Understanding the cellular mechanisms by which these signaling molecules act is crucial to better understand the potential causes of pituitary adenomas, common benign tumors in humans (66,67).

In the neural tube, initially uncommitted progenitor cells at distinct positions along the dorsoventral axis acquire their fate by exposure to different concentrations of Hh signals (20). These signals interact with intrinsic cellular determinants and specify cell fate by regulating the transcriptional program of the cell. The presence of a Hh transcriptional response in adenohypophyseal precursors suggests that Hh signaling may directly influence pituitary cell fates, similar to the situation in the CNS. Whereas we show that a subset of the Hh transcriptional response code is activated in the pituitary, it remains to be determined exactly how this transcriptional response is translated into positional information and discrete endocrine cell fates.

Our studies do not rule out the possibility that Hh signaling also regulates pituitary cell proliferation and survival, as in the neural tube (20). In fact, Hh pathway mutants have significantly smaller pituitaries than wild-type embryos, suggesting a possible mitogenic role for Hh signaling (8). Furthermore, a recent study showed that misregulation of Hh signaling is associated with human adenomas and that CyA acts to induce proliferation of adenoma derived cell line (68), suggesting Hh may negatively regulate pituitary precursor proliferation in adults. Initial analysis of cell proliferation rates in the developing zebrafish pituitary using the antiphosphohistone antibody did not uncover changes in proliferation of adenohypophyseal cells when Hh signaling was blocked (data not shown). However, more sensitive methods such as bromodeoxyuridine (5-bromo-2′-deoxyuridine, BrdU) labeling are needed to detect subtle yet significant changes in cell proliferation. Hh has also been shown to function as a cell survival factor in the ventral neural tube (69,70). Initial analysis of cell death with acridine orange labeling did not show a significant effect of Hh signaling on adenohypophyseal cell survival (data not shown).

Fgf signaling has been clearly shown to play a role in regulating pituitary cell survival, with loss of Fgf signaling, leading to loss of the adenohypophysis and an increase in death of placodal cells (25,27). Fgf signaling has also been shown to regulate cell proliferation in the mammalian pituitary (22,23). These previous studies did not rule out an additional role for Fgf signaling in pituitary induction and the differentiation of endocrine cell fates. Because Fgf plays an important patterning role in the CNS, with graded Fgf signaling from organizing centers in the brain helping to pattern the forebrain, midbrain, and hindbrain along the A/P axis (reviewed in Ref. 71), we examined the effect of graded loss of Fgf signaling on pituitary development. Although a graded transcriptional response code to Fgf signaling is not well established, we showed that the Ets transcription factor erm, a known transcriptional target of Fgf signaling (52), is expressed in the developing adenohypophysis. Furthermore, graded loss of Fgf signaling leads to a graded loss of erm expression (supplemental figure). This suggests that a graded transcriptional response to Fgf signaling could directly affect cell fate decisions. The fact that slight reductions in Fgf signaling led to a selective loss of posterior pituitary fates, including pax7- and pomc-expressing cells in the PI, is consistent with a role in posterior cell fate decisions. Whereas it is possible that posterior cells selectively die in the absence of Fgf signaling, we show that expression of pax7, the earliest marker of the PI, is never initiated when Fgf signaling is blocked, consistent with a role in PI induction as well as survival.

Supplementary Material

[Supplemental Data]


We thank Meng-Chieh Shen for technical assistance, Judy Bennett for fish care, and members of the Karlstrom laboratory for comments on the manuscript. We also thank Drs. Ning-Ai Liu and Shlomo Melmed for sharing the POMC-GFP transgenic line and Curis Inc. for supplying Shh protein. Thanks also go to Will Sillin for help with schematic art work.


This work was supported by National Institutes of Health Grants NS039994 and HD044929.

Current address for J.E.T.: Sunrise Medical Laboratories, Hauppauge, New York 11788.

Disclosure Statement: The authors have nothing to disclose.

First Published Online May 22, 2008

Abbreviations: A/P, Anterior/posterior; CNS, central nervous system; CyA, cyclopamine; DMSO, dimethylsulfoxide; Fgf, fibroblast growth factor; FgfR, Fgf receptor; GFP, green fluorescent protein; Hh, hedgehog; hpf, hours post fertilization; PD, pars distalis; PI, pars intermedia; POMC, pro-opiomelanocortin; Prl, prolactin; Shh, Sonic hedgehog.


  • Zhu X, Gleiberman AS, Rosenfeld MG 2007 Molecular physiology of pituitary development: signaling and transcriptional networks. Physiol Rev 87:933–963 [PubMed]
  • Schlosser G 2006 Induction and specification of cranial placodes. Dev Biol 294:303–351 [PubMed]
  • Treier M, Rosenfeld MG 1996 The hypothalamic-pituitary axis: co-development of two organs. Curr Opin Cell Biol 8:833–843 [PubMed]
  • Herzog W, Zeng X, Lele Z, Sonntag C, Ting JW, Chang CY, Hammerschmidt M 2003 Adenohypophysis formation in the zebrafish and its dependence on sonic hedgehog. Dev Biol 254:36–49 [PubMed]
  • Dutta S, Dietrich JE, Aspock G, Burdine RD, Schier A, Westerfield M, Varga ZM 2005 pitx3 defines an equivalence domain for lens and anterior pituitary placode. Development 132:1579–1590 [PubMed]
  • Osumi-Yamashita N, Ninomiya Y, Doi H, Eto K 1994 The contribution of both forebrain and midbrain crest cells to the mesenchyme in the frontonasal mass of mouse embryos. Dev Biol 164:409–419 [PubMed]
  • Toro S, Varga ZM 2007 Equivalent progenitor cells in the zebrafish anterior preplacodal field give rise to adenohypophysis, lens, and olfactory placodes. Semin Cell Dev Biol 18:534–542 [PubMed]
  • Sbrogna JL, Barresi MJ, Karlstrom RO 2003 Multiple roles for Hedgehog signaling in zebrafish pituitary development. Dev Biol 254:19–35 [PubMed]
  • So WK, Kwok HF, Ge W 2005 Zebrafish gonadotropins and their receptors: II. Cloning and characterization of zebrafish follicle-stimulating hormone and luteinizing hormone subunits—their spatial-temporal expression patterns and receptor specificity. Biol Reprod 72:1382–1396 [PubMed]
  • Liem KF, Bemis WE, Walker Jr WF, Grande L 2001 Functional anatomy of the vertebrates. 3rd ed. New York: Harcourt College Publishers
  • Lamonerie T, Tremblay JJ, Lanctot C, Therrien M, Gauthier Y, Drouin J 1996 Ptx1, a bicoid-related homeo box transcription factor involved in transcription of the pro-opiomelanocortin gene. Genes Dev 10:1284–1295 [PubMed]
  • Wilson DB, Wyatt DP 1986 Ultrastructural immunocytochemistry of somatotrophs and mammotrophs in embryos of the dwarf mutant mouse. Anat Rec 215:282–287 [PubMed]
  • Treier M, Gleiberman AS, O'Connell SM, Szeto DP, McMahon JA, McMahon AP, Rosenfeld MG 1998 Multistep signaling requirements for pituitary organogenesis in vivo. Genes Dev 12:1691–1704 [PMC free article] [PubMed]
  • Liu NA, Huang H, Yang Z, Herzog W, Hammerschmidt M, Lin S, Melmed S 2003 Pituitary corticotroph ontogeny and regulation in transgenic zebrafish. Mol Endocrinol 17:959–966 [PubMed]
  • Nica G, Herzog W, Sonntag C, Hammerschmidt M 2004 Zebrafish pit1 mutants lack three pituitary cell types and develop severe dwarfism. Mol Endocrinol 18:1196–1209 [PubMed]
  • Bertagna X 1994 Proopiomelanocortin-derived peptides. Endocrinol Metab Clin North Am 23:467–485 [PubMed]
  • Liu NA, Liu Q, Wawrowsky K, Yang Z, Lin S, Melmed S 2006 Prolactin receptor signaling mediates the osmotic response of embryonic zebrafish lactotrophs. Mol Endocrinol 20:871–880 [PubMed]
  • Treier M, O'Connell S, Gleiberman A, Price J, Szeto DP, Burgess R, Chuang PT, McMahon AP, Rosenfeld MG 2001 Hedgehog signaling is required for pituitary gland development. Development 128:377–386 [PubMed]
  • Ingham PW, McMahon AP 2001 Hedgehog signaling in animal development: paradigms and principles. Genes Dev 15:3059–3087 [PubMed]
  • Briscoe J, Pierani A, Jessell TM, Ericson J 2000 A homeodomain protein code specifies progenitor cell identity and neuronal fate in the ventral neural tube. Cell 101:435–445 [PubMed]
  • Lupo G, Harris WA, Lewis KE 2006 Mechanisms of ventral patterning in the vertebrate nervous system. Nat Rev Neurosci 7:103–114 [PubMed]
  • Ericson J, Norlin S, Jessell TM, Edlund T 1998 Integrated FGF and BMP signaling controls the progression of progenitor cell differentiation and the emergence of pattern in the embryonic anterior pituitary. Development 125:1005–1015 [PubMed]
  • Norlin S, Nordstrom U, Edlund T 2000 Fibroblast growth factor signaling is required for the proliferation and patterning of progenitor cells in the developing anterior pituitary. Mech Dev 96:175–182 [PubMed]
  • Pogoda HM, Hammerschmidt M 2007 Molecular genetics of pituitary development in zebrafish. Semin Cell Dev Biol 18:543–558 [PubMed]
  • Herzog W, Sonntag C, von der Hardt S, Roehl HH, Varga ZM, Hammerschmidt M 2004 Fgf3 signaling from the ventral diencephalon is required for early specification and subsequent survival of the zebrafish adenohypophysis. Development 131:3681–3692 [PubMed]
  • Revest JM, Suniara RK, Kerr K, Owen JJ, Dickson C 2001 Development of the thymus requires signaling through the fibroblast growth factor receptor R2-IIIb. J Immunol 167:1954–1961 [PubMed]
  • Ohuchi H, Hori Y, Yamasaki M, Harada H, Sekine K, Kato S, Itoh N 2000 FGF10 acts as a major ligand for FGF receptor 2 IIIb in mouse multi-organ development. Biochem Biophys Res Commun 277:643–649 [PubMed]
  • Kimmel CB, Ballard WW, Kimmel SR, Ullmann B, Schilling TF 1995 Stages of embryonic development of the zebrafish. 203:253–310 [PubMed]
  • Karlstrom RO, Tyurina OV, Kawakami A, Nishioka N, Talbot WS, Sasaki H, Schier AF 2003 Genetic analysis of zebrafish gli1 and gli2 reveals divergent requirements for gli genes in vertebrate development. Development 130:1549–1564 [PubMed]
  • Karlstrom RO, Talbot WS, Schier AF 1999 Comparative synteny cloning of zebrafish you-too: mutations in the Hedgehog target gli2 affect ventral forebrain patterning. Genes Dev 13:388–393 [PMC free article] [PubMed]
  • Glasgow E, Karavanov AA, Dawid IB 1997 Neuronal and neuroendocrine expression of lim3, a LIM class homeobox gene, is altered in mutant zebrafish with axial signaling defects. Dev Biol 192:405–419 [PubMed]
  • Barth KA, Wilson SW 1995 Expression of zebrafish nk2.2 is influenced by sonic hedgehog vertebrate hedgehog-1 and demarcates a zone of neuronal differentiation in the embryonic forebrain. Development 121:1755–1768 [PubMed]
  • Krauss S, Johansen T, Korzh V, Moens U, Ericson JU, Fjose A 1991 Zebrafish pax[zf-a]: a paired box-containing gene expressed in the neural tube. EMBO J 10:3609–3619 [PMC free article] [PubMed]
  • Seo HC, Saetre BO, Havik B, Ellingsen S, Fjose A 1998 The zebrafish Pax3 and Pax7 homologues are highly conserved, encode multiple isoforms and show dynamic segment-like expression in the developing brain. Mech Dev 70:49–63 [PubMed]
  • Jensen AM, Walker C, Westerfield M 2001 Mosaic eyes: a zebrafish gene required in pigmented epithelium for apical localization of retinal cell division and lamination. Development 128:95–105 [PubMed]
  • Barresi MJ, Hutson LD, Chien CB, Karlstrom RO 2005 Hedgehog regulated Slit expression determines commissure and glial cell position in the zebrafish forebrain. Development 132:3643–3656 [PubMed]
  • Ekker SC, Ungar AR, Greenstein P, von Kessler DP, Porter JA, Moon RT, Beachy PA 1995 Patterning activities of vertebrate hedgehog proteins in the developing eye and brain. Curr Biol 5:944–955 [PubMed]
  • Guner B, Karlstrom RO 2007 Cloning of zebrafish nkx6.2 and a comprehensive analysis of the conserved transcriptional response to Hedgehog/Gli signaling in the zebrafish neural tube. MOD, Gene Expr Patterns 7:596–605 [PMC free article] [PubMed]
  • Bentley CA, Zidehsarai MP, Grindley JC, Parlow AF, Barth-Hall S, Roberts VJ 1999 Pax6 is implicated in murine pituitary endocrine function. Endocrine 10:171–177 [PubMed]
  • Kioussi C, O'Connell S, St. Onge L, Treier M, Gleiberman AS, Gruss P, Rosenfeld MG 1999 Pax6 is essential for establishing ventral-dorsal cell boundaries in pituitary gland development. Proc Natl Acad Sci USA 96:14378–14382 [PMC free article] [PubMed]
  • Persson M, Stamataki D, te Welscher P, Andersson E, Bose J, Ruther U, Ericson J, Briscoe J 2002 Dorsal-ventral patterning of the spinal cord requires Gli3 transcriptional repressor activity. Genes Dev 16:2865–2878 [PMC free article] [PubMed]
  • Amirthalingam K, Lorens JB, Saetre BO, Salaneck E, Fjose A 1995 Embryonic expression and DNA-binding properties of zebrafish pax-6. Biochem Biophys Res Commun 215:122–128 [PubMed]
  • Feng X, Adiarte EG, Devoto SH 2006 Hedgehog acts directly on the zebrafish dermomyotome to promote myogenic differentiation. Dev Biol 300:736–746 [PubMed]
  • Incardona JP, Gaffield W, Kapur RP, Roelink H 1998 The teratogenic Veratrum alkaloid cyclopamine inhibits sonic hedgehog signal transduction. Development 125:3553–3562 [PubMed]
  • Chen JK, Taipale J, Cooper MK, Beachy PA 2002 Inhibition of Hedgehog signaling by direct binding of cyclopamine to Smoothened. Genes Dev 16:2743–2748 [PMC free article] [PubMed]
  • Chen W, Burgess S, Hopkins N 2001 Analysis of the zebrafish smoothened mutant reveals conserved and divergent functions of hedgehog activity. Development 128:2385–2396 [PubMed]
  • Barresi MJ, D'Angelo JA, Hernandez LP, Devoto SH 2001 Distinct mechanisms regulate slow-muscle development. Curr Biol 11:1432–1438 [PubMed]
  • Scully KM, Rosenfeld MG 2002 Pituitary development: regulatory codes in mammalian organogenesis. Science 295:2231–2235 [PubMed]
  • Dasen JS, Rosenfeld MG 2001 Signaling and transcriptional mechanisms in pituitary development. Annu Rev Neurosci 24:327–355 [PubMed]
  • Mohammadi M, McMahon G, Sun L, Tang C, Hirth P, Yeh BK, Hubbard SR, Schlessinger J 1997 Structures of the tyrosine kinase domain of fibroblast growth factor receptor in complex with inhibitors. Science 276: 955–960 [PubMed]
  • Raible F, Brand M 2001 Tight transcriptional control of the ETS domain factors Erm and Pea3 by Fgf signaling during early zebrafish development. Mech Dev 107:105–117 [PubMed]
  • Roehl H, Nusslein-Volhard C 2001 Zebrafish pea3 and erm are general targets of FGF8 signaling. Curr Biol 11:503–507 [PubMed]
  • Fogarty MP, Emmenegger BA, Grasfeder LL, Oliver TG, Wechsler-Reya RJ 2007 Fibroblast growth factor blocks Sonic hedgehog signaling in neuronal precursors and tumor cells. Proc Natl Acad Sci USA 104:2973–2978 [PMC free article] [PubMed]
  • Bertrand N, Dahmane N 2006 Sonic hedgehog signaling in forebrain development and its interactions with pathways that modify its effects. Trends Cell Biol 16:597–605 [PubMed]
  • Chapman SC, Sawitzke AL, Campbell DS, Schoenwolf GC 2005 A three-dimensional atlas of pituitary gland development in the zebrafish. J Comp Neurol 487:428–440 [PubMed]
  • Zhu Y, Stiller JW, Shaner MP, Baldini A, Scemama JL, Capehart AA 2004 Cloning of somatolactin α and β cDNAs in zebrafish and phylogenetic analysis of two distinct somatolactin subtypes in fish. J Endocrinol 182:509–518 [PubMed]
  • Yan HY, Thomas P 1991 Histochemical and immunocytochemical identification of the pituitary cell types in three sciaenid fishes: Atlantic croaker (Micropogonias undulatus), spotted seatrout (Cynoscion nebulosus), and red drum (Sciaenops ocellatus). Gen Comp Endocrinol 84:389–400 [PubMed]
  • Toubeau G, Poilve A, Baras E, Nonclercq D, De Moor S, Beckers JF, Dessy-Doize C, Heuson-Stiennon JA 1991 Immunocytochemical study of cell type distribution in the pituitary of Barbus barbus (Teleostei, Cyprinidae). Gen Comp Endocrinol 83:35–47 [PubMed]
  • Delfini MC, Dubrulle J, Malapert P, Chal J, Pourquie O 2005 Control of the segmentation process by graded MAPK/ERK activation in the chick embryo. Proc Natl Acad Sci USA 102:11343–11348 [PMC free article] [PubMed]
  • Briscoe J, Sussel L, Serup P, Hartigan-O'Connor D, Jessell TM, Rubenstein JL, Ericson J 1999 Homeobox gene Nkx2.2 and specification of neuronal identity by graded Sonic hedgehog signalling. Nature 398:622–627 [PubMed]
  • Wolpert L 2003 Cell boundaries: knowing who to mix with and what to shout or whisper. Development 130:4497–4500 [PubMed]
  • Stamataki D, Ulloa F, Tsoni SV, Mynett A, Briscoe J 2005 A gradient of Gli activity mediates graded Sonic Hedgehog signaling in the neural tube. Genes Dev 19:626–641 [PMC free article] [PubMed]
  • Freeman M 2000 Feedback control of intercellular signalling in development. Nature 408:313–319 [PubMed]
  • Jeong J, McMahon AP 2005 Growth and pattern of the mammalian neural tube are governed by partially overlapping feedback activities of the hedgehog antagonists patched 1 and Hhip1. Development 132:143–154 [PubMed]
  • Pages F, Kerridge S 2000 Morphogen gradients. A question of time or concentration? Trends Genet 16:40–44 [PubMed]
  • Shimon I, Melmed S 1998 Management of pituitary tumors. Ann Intern Med 129:472–483 [PubMed]
  • Asa SL, Ezzat S 2002 The pathogenesis of pituitary tumours. Nat Rev Cancer 2:836–849 [PubMed]
  • Vila G, Theodoropoulou M, Stalla J, Tonn JC, Losa M, Renner U, Stalla GK, Paez-Pereda M 2005 Expression and function of sonic hedgehog pathway components in pituitary adenomas: evidence for a direct role in hormone secretion and cell proliferation. J Clin Endocrinol Metab 90:6687–6694 [PubMed]
  • Charrier JB, Lapointe F, Le Douarin NM, Teillet MA 2001 Anti-apoptotic role of Sonic hedgehog protein at the early stages of nervous system organogenesis. Development 128:4011–4020 [PubMed]
  • Thibert C, Teillet MA, Lapointe F, Mazelin L, Le Douarin NM, Mehlen P 2003 Inhibition of neuroepithelial patched-induced apoptosis by sonic hedgehog. Science 301:843–846 [PubMed]
  • Sanchez-Camacho C, Rodriguez J, Ruiz JM, Trousse F, Bovolenta P 2005 Morphogens as growth cone signalling molecules. Brain Res Brain Res Rev 49:242–252 [PubMed]

Articles from Endocrinology are provided here courtesy of The Endocrine Society

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...