1. Introduction
Located underneath and regulated by the hypothalamus, the pituitary (hypophysis) functions as a master endocrine gland, relaying information from the brain to the body, thereby regulating various physiological processes, such as metabolism, growth, the stress response and reproduction. Five main types of hormone-secreting cells are present in the anterior pituitary (AP): corticotrophs that secrete the adrenocorticotrophic hormone (ACTH), which, by inducing glucocorticosteroid production in the adrenal cortex, plays a crucial role in the stress response and carbohydrate metabolism [
1]; thyrotrophs that secrete the thyroid-stimulating hormone (TSH), which stimulates the synthesis and release of the thyroid hormones (THs) triiodothyronine (T3) and thyroxine (T4), thereby affecting development, growth and metabolism [
2]; somatotrophs that secrete the growth hormone (GH), which stimulates growth and affects glucose metabolism [
3]; lactotrophs that secrete prolactin (PRL), which stimulates milk production in mammals and controls osmoregulation in fish [
4]; and gonadotrophs that secrete the gonadotropin follicle-stimulating hormone (FSH) and luteinizing hormone (LH), the key regulators of reproduction [
5].
The fates of these various endocrine cell types are determined by several secreted developmental factors that create an opposing gradient along the forming AP, leading to the particular spatial expressions of a series of transcription factors, which, in turn, control the cells’ identities [
6]. Mutations in such transcription factors affect the AP development, pituitary cell differentiation and pituitary hormone levels and consequently lead to broad systemic effects [
7,
8].
LIM homeobox 4 (LHX4) is a member of the LIM-homeodomain protein family, which bind to DNA through their characteristic helix–turn–helix motive and act as master transcription factors that regulate the gene expression involved in cell differentiation, affecting the body pattern formation during embryonic development, including the development of the endocrine and nervous system structures. LHX4 has gained much attention, as it has been found to be crucial for pituitary development [
6]. In mice,
Lhx4 is expressed in the cerebral cortex, spinal cord, developing hindbrain [
9] and pineal gland [
10], and it plays an important role in the differentiation of ventral motor neurons [
11]. In the developing pituitary gland,
Lhx4 is necessary for the regular differentiation of the pituitary cell types [
7]. Eventually, as the AP develops, the expression of
Lhx4 in the gland declines and is completely absent in differentiated cells. However, its expression has also been documented in cells featuring stem/progenitor cell characteristics among the developed pituitary [
12].
Human patients heterozygous for a mutation in LHX4 protein suffer from combined pituitary hormone deficiency (CPHD), short statures, reproductive and metabolic disorders and abnormalities of the sella turcica [
6,
8,
9,
13]. Unlike human patients, heterozygous
Lhx4-mutant mice display no abnormalities [
8]. However, in homozygous
Lhx4-mutant mice, the pituitary begins to develop but later the AP cells undergo massive apoptosis, leading to AP hypocellularity. Homozygous
Lhx4-mutant mice die shortly after birth due to severe lung defects [
14], limiting research on the role of
Lhx4 in mice.
An alternative animal model to study pituitary development and AP cell determination is the zebrafish. This species is particularly attractive owing to the large number of accessible transparent progeny produced in each cross and their amenability to genetic manipulation and superb real-time imaging. Moreover, the rapid development of the zebrafish AP has been extensively studied. As in mammals, the zebrafish AP originates in the anterior neural ridge (ANR). At 18 h post-fertilization (hpf), the ANR thickens, and the developing AP begins invagination. Throughout the early stages of zebrafish development, the AP migrates posteriorly from the ANR, and it reaches its final position at 60 hpf [
15]. As early as 24 hpf,
prl and
gh mRNAs can already be detected in the AP. The expression of glycoprotein hormone alpha-subunit (
αgsu) starts at 32 hpf, and the expressions of
tshb and
fshb initiate at 42 hpf and 4 days post-fertilization (dpf), respectively [
15];
lhb expression begins only later, at 25 dpf [
16].
Gh is fully expressed by somatotrophs at 48 hpf [
15]. The expression of
lhx4 in the adenohypophyseal placode, as well as in the pineal gland and trigeminal ganglion, is detected prior to those of most pituitary hormones [
17,
18]. Thus, utilization of the zebrafish model may further contribute to understanding the role of
LHX4 in AP development and the outcomes of LHX4 deficiency [
19].
Here, we generated and characterized a zebrafish lhx4-knockout (KO) line to investigate the role of lhx4 in pituitary gland development and functioning. Importantly, unlike mice, lhx4-KO fish survive and reach maturity. We discovered that at the embryonic and larval stages, lhx4 mutants produce lower levels of tshb mRNA. At later stages, lhx4 mutants also display decreased gh, tshb, pomca and fshb mRNA levels, undeveloped lhb-producing gonadotrophs and a reduced pomca promoter-driven expression in corticotrophs, along with smaller bodies and reproductive deficiencies, phenotypes that are reminiscent of human CPHD.
3. Discussion
Various dominant LHX4 mutations have been discovered in humans, which has led to a variety of phenotypes in heterozygous carriers, such as a short stature due to GH deficiency, CPHD, abnormalities of the central skull base and cerebellar defects [
8,
26]. A recessive mutation in LHX4 has also been reported. The heterozygous parents were unaffected by the mutation; however, their three homozygous children were born underweight, suffered from poor muscle tone, had severe lung abnormalities, and died within the first week after birth. ACTH, TSH and GH deficiencies were diagnosed in these infants [
26]. Similar to heterozygous humans carrying a recessive LHX4 mutation, heterozygous
Lhx4-mutant mice exhibit no apparent phenotype [
9,
26]. Homozygous
Lhx4-mutant mice successfully develop the Rathke-pouch ectoderm structure but have a severely hypoplastic AP [
27]. They die shortly after birth due to lung defects, hindering studies on the role of
Lhx4 in the pituitary function at later stages [
14].
The
lhx4-KO zebrafish generated in this study (
Figure 1) expressed lower levels of
tshb mRNA at 48 hpf, 7 dpf and 4 months of age (
Figure 2A,B,
Figure 3A,B and
Figure 5A, respectively). Although no differences were found in the expressions of
gh,
pomca and
prl at the early stages (
Figure 2C–E and
Figure 3C–E), the expressions of
gh and
pomca were significantly reduced in the adult
lhx4 mutants (
Figure 5B,C). The development of corticotrophs was also affected by Lhx4 deficiency, as indicated by the reduced
pomca promoter-driven GFP expression in the pituitaries of adult mutants (
Figure 6). Decreased
fshb expression was observed in the adult homozygous
lhx4 mutants (
Figure 5D). Moreover, utilizing a transgenic reporter line in which the expression of RFP is driven by the tilapia
lhb promoter, we have shown that
lhb-expressing gonadotrophs are severely depleted in the pituitaries of adult
lhx4 mutants (
Figure 7B).
Owing to the survival of
lhx4-KO zebrafish and the fact that they reach the adult stage, the phenomenon of the sequential loss of the pituitary hormone-producing cells could be observed. This phenomenon could be explained by the effect of LHX4 deficiency on pituitary precursor cells: In
lhx4-mutant mouse embryos, increased levels of apoptotic pituitary precursor cells have been documented, indicating that
Lhx4 is necessary for the survival of precursor cells and thereby controls the number of differentiated pituitary hormone-secreting cells [
28]. Likewise, it is possible that a population of stem cells within the adult zebrafish AP, as is the case with mouse embryos [
12], is depleted in Lhx4-deffienct fish, resulting in the reduced proliferation of hormone-secreting cells in the AP. Alternatively, the sequential decreased expressions of pituitary hormone-encoding genes may be an indirect outcome of Lhx4 deficiency. The finding that the
tshb expression is reduced in
lhx4 mutants at the embryo and larval stages (
Figure 2A,B and
Figure 3A,B), prior to the observed decrease in other pituitary hormone-encoding genes, implies that at least part of the adult phenotype may be caused by TH deficiency, which is expected to occur only at later life stages, 20 dpf [
22]. Since THs are known to regulate growth, in addition to metabolism and reproduction [
2,
22,
23], the substantially lower expression of
tshb in the homozygous
lhx4 mutants may account for their small size (
Figure 4) and immature gonads (
Figure 7A). Notably, the levels of the mediators of these effects, THs, were not altered by the mutation at 7 dpf (
Figure S3). However, as indicated above, the effect of TSH deficiency on TH signaling is not expected at this stage [
22], and the maternal TH deposited in the yolk, in both WT and
lhx4 mutants, is sufficient for the initiation of the normal development of the nervous system [
21].
An additional explanation for the reduction in the pituitary cell types in adult Lhx4-deficient zebrafish could be reduced innervation or decreased blood flow to the gland. This assumption is based on the fact that
LHX4 mutations in humans lead to a reduction in the size of the pituitary stalk [
29]. Furthermore,
lhx4 is also expressed outside the AP [
8,
11,
17,
30] and (
Figure S1), the phenotypes described here may be induced by pathways that are unrelated to the pituitary function. For example, the reduced locomotor activity of the
lhx4-mutant larvae (
Figure S2) could have stemmed from poor muscle tone, as is the case in human patients [
26]. Thus, the poor mobilization of the
lhx4 mutants may have led to a failure in the competition over food when raised with their WT siblings, which may account for their malnutrition and reduced body size [
31], a possibility that warrants further inquiry.
The reproductive impairments of the homozygous
lhx4-mutant females could also be explained by malnutrition and a reduced body size, although our findings of lower
fshb mRNA levels and depleted
lhb-expressing gonadotrophs in adult homozygous
lhx4 mutants would be a more reasonable explanation [
25,
32]. Thus, we conclude that gonadotropin deficiency, or the combination of gonadotropin and Tshb deficiencies, is the source for the female infertility of homozygous
lhx4-mutant females. As opposed to the females, the
lhx4-mutant males were fertile, indicating that the reduced pituitary hormone levels have a stronger effect on ovarian development in comparison with testicular development. This implies that the combined hormonal profile required for ovarian development and vitellogenesis, absent in the
lhx4 mutant, is different from that required for testicular development. An alternative explanation could be a direct effect of Lhx4 on gonadal development: In mice, an RNAseq analysis of developing gonads revealed a significantly higher expression of
Lhx4 mRNA in the developing ovary compared to that in the developing testes [
33], possibly explaining the sex-specific effect of LHX4 deficiency in zebrafish. This could be an interesting avenue to investigate despite the apparent differences between the development of mammalian and fish reproductive systems.
In summary, our lhx4-KO model exhibits phenotypes that resemble those observed in human patients carrying a LHX4 mutation, such as CPHD, impaired growth and fertility abnormalities. Unlike other models, lhx4-mutant zebrafish survive the larval stage and reach adulthood. We found that although lhx4 is mainly expressed in the developing AP, it exerts its function throughout all the zebrafish life stages. Hence, the characterized lhx4-mutant line constitutes a valuable model to further investigate the consequence of lhx4 mutation on pituitary development, pituitary functioning and beyond. Further research is required to define the cellular, temporal and spatial expression of lhx4 in the zebrafish AP, and to understand how this expression pattern affects the various AP cell types and the physiological processes they regulate.
4. Materials and Methods
4.1. Fish and Embryos
Zebrafish (Danio rerio) were grown and maintained in a recirculating-water system at 28 °C under 12:12 h LD cycles and fed twice a day. The fish were naturally mated in an appropriate tank, and the embryos were collected and kept in a Petri dish with embryo water containing methylene blue (0.3 ppm) in an incubator at 28 °C under 12:12 h LD cycles. On the 7th day, larvae were transferred to 10 L tanks in the recirculating-water system. Once the fish reached adulthood, they were genotyped and transferred accordingly into 3 L tanks.
4.2. Generation of lhx4-Mutant Zebrafish and Genotyping
The CRISPR-Cas9 system was used to establish the lhx4-KO zebrafish line, registered in the Zebrafish Information Network (ZFIN) database as lhx4tlv12. Oligos 5′-taggagtgccactgcaacgtaa-3′ and 5′-aaacttacgttgcagtggcact-3′ were designed to target a sequence at the end of exon 1 (5′-GGAGTGCCACTGCAACgtaa-3′), which contains the BtsI-v2 restriction site (underlined). The oligos were ligated into a pT7-gRNA zebrafish-optimized vector (plasmid #46759, Addgene, Watertown, MA, USA), followed by linearization with BamHI (R3136, New England Biolabs, Ipswich, MA, USA), and the synthesis of gRNA was performed with a MAXIscript T7 Transcription Kit (AM1312, Invitrogen, Waltham, MA, USA). An injection mix was prepared by mixing gRNA (60 ng μL−1) and TrueCut Cas9 Protein V2 (1 μg μL−1; A36498, Invitrogen), followed by 5 min incubation in 37 °C prior to co-injection into one-cell-stage WT zebrafish embryos. Injected embryos (F0 generation) were raised to maturity and crossed with WT fish to identify carriers of an indel within the lhx4 gene at the F1 generation.
For genotyping, whole larvae or fin samples from mature fish were lysed in lysis buffer [10 mM Tris (pH 8), 2 mM EDTA (pH 8), 0.2% Triton X-100 and 0.1 mg mL−1 protein kinase]. Fixated post-whole-mount ISH and immunostained samples were lysed using the Extract-N-Amp™ FFPE Tissue PCR kit (XNAT2-1KT, Sigma, St. Louis, MO, USA), according to the manufacturer’s protocol. Lysis was performed overnight at 52 °C, followed by 10 min inactivation at 95 °C. The isolated genomic DNA served as the template to amplify a 490 bp fragment of the lhx4 gene using forward 5′-atgaaaatgatgcaaagtgcg-3′ and reverse 5′-tgcccagctatgcgatctaac-3′ primers. Identification of the lhx4-mutant allele was based on the incomplete digestion of the PCR product by BtsI-v2 (R0667, New England Biolabs, Ipswich, MA, USA), in contrast to the full digestion of the WT allele into two fragments (of 76 bp and 414 bp). Sequence analysis of the selected F1 founder genomic DNA indicated a 5 bp deletion at the end of lhx4 exon 1, and the lhx4tlv12 line was propagated by further crossings to produce homozygous mutants and WT siblings at future generations.
4.3. Reverse Transcription PCR
RNA was purified from brain samples dissected from adult homozygous lhx4 mutants and WT siblings using the RNeasy Lipid Tissue Mini kit (74804, Qiagen, Hilden, Germany). An amount of 1 µg of the purified RNA served as the template for the cDNA synthesis using the qScript cDNA Synthesis Kit (95047, Quantabio, Beverly, MA, USA). PCR was performed on cDNA templates using forward primer 5′-atgaaaatgatgcaaagtgcg-3′ targeting the beginning of exon 1, and reverse primer 5′-cgaaacgcttgaagaagtcc-3′ spanning the exon 2–3 junction, yielding a 265 bp product.
4.4. Whole-Mount In Situ Hybridization
An 815 bp fragment of the zebrafish lhx4 (RefSeq NM_001122973.1) coding sequence (CDS) was amplified using forward 5′-ggacttcttcaagcgtttcg-3′ and reverse 5′-tcagagcttgacccacactg-3′ primers and cloned into pGEM-T Easy (A1360, Promega, Madison, WI, USA). In addition, plasmids containing CDS fragments of zebrafish gh1 (RefSeq NM_001020492.2), tshba (RefSeq NM_181494.2), prl (RefSeq NM_181437.3) and pomca (RefSeq NM_181438.3) were kindly provided by The Hammerschmidt Plasmid Stocks (Spemann Labs, Freiburg, Germany). Plasmids were linearized, and digoxigenin (DIG)-labeled anti-sense riboprobes were synthesized using the Dig RNA Labeling Kit (SP6/T7; 11175025910, Roche, Basel, Switzerland), according to the manufacturer’s instructions.
The embryos/larvae were fixed at 24 and 48 hpf and 7 dpf, and whole-mount ISH was performed as previously described [
34], with the following modification: the 24 and 27 hpf sampled embryos were not treated with proteinase K. Images were acquired (see
Section 4.11), and the staining signal was quantified using ImageJ software 2.1.0. (National Institute of Health, Bethesda, MD, USA). The staining signal, presented as the integrated (optical) density, was computed by multiplying the area (pixels) by the mean intensity value. After image quantification, each embryo/larva was genotyped (see ‘Generation of
lhx4-Mutant Zebrafish and Genotyping’). Statistical differences between genotypes were determined by Mann–Whitney test.
4.5. Whole-Mount Immunostaining
The immunostaining protocol was carried out as published in [
35], with minor adjustments. In short, 7 dpf fixed larvae were incubated in 10% H
2O
2 at room temperature for 4 h and then washed 3 times with PBT solution (0.25% Triton X-100 in PBS). The samples were blocked for 2 h at room temperature with 4% blocking solution (containing donkey serum), followed by overnight incubation at 4 °C with primary antibodies: anti-mouse triiodothyronine (T3) (1:100; ME-124, sc-57481, Santa Cruz Biotechnology, Dallas, TX, USA) or anti-rabbit thyroxine (T4) (1:100; 8658501, MP bio, Irvine, CA, USA).
After 6 washes with PBT, larvae were incubated for 4 h with secondary antibodies: donkey anti-mouse Alexa Fluor 488 (1:500; 715-545-150, Jackson ImmunoResearch, West Grove, PA, USA) or donkey anti-rabbit Cy3 (1:500; 711-165-152, Jackson ImmunoResearch, West Grove, PA, USA) for T3 and T4, respectively. Afterwards, the fluorescent signal was captured (see
Section 4.11) and quantified as described for the whole-mount ISH signal (see
Section 4.4). Subsequently, larvae were genotyped as described in ‘Generation of
lhx4 mutant zebrafish and genotyping’. Statistical differences between groups were evaluated by Mann–Whitney test.
4.6. Histology
Homozygous
lhx4-mutant females (N = 2) and their WT siblings (N = 6) at the age of 4 months were fixed in 4% PFA. After decalcification and paraffin embedding, longitude sections (4 µm) were prepared and stained with H&E by Gavish Research Services. Ovary slide images were acquired (see
Section 4.11).
4.7. Larval Locomotor Activity Assay
Progeny of heterozygous lhx4-mutant intercross were raised in an incubator under 12:12 h LD cycles. At 9 dpf, the larvae were individually placed in wells of a 24-well plate in the observation chamber of the DanioVision tracking system (Noldus Information Technology, Wageningen, The Netherlands). The activity of each larva was tracked for 4 h under constant light and analyzed by Ethovision 15.0 software (Noldus Information Technology, Wageningen, The Netherlands) for the total activity (logcm) and top speed (cm s−1). Following activity monitoring, larvae were lysed and genotyped as described (see ‘Generation of lhx4-Mutant Zebrafish and Genotyping’). Statistical differences between genotypes were determined by t-test with Benjamini–Hochberg correction for multiple comparisons to maintain a false discovery rate of 0.05.
4.8. Body Size Measurement
To quantify the body size, adult homozygous
lhx4 mutants and their siblings at the age of 5 months were anesthetized with 0.16 mg mL
−1 tricaine (A-5040, Sigma), laterally placed on a Petri dish plate and photographed (see
Section 4.11). The body size was evaluated as the distance from the head to the tail-base, using an in-house custom RStudio version 2023.09.1+494 script. Statistical differences between genotypes were determined by Mann–Whitney test.
4.9. Quantitative Real-Time RT-PCR Analysis
Pituitary glands were dissected from homozygous lhx4 mutants (N = 10) and their WT siblings (N = 13), and RNA was extracted using the RNeasy Micro Kit (74004, Qiagen, Hilden, Germany), according to the manufacturer’s instructions. cDNA was synthesized with the qScript cDNA Synthesis Kit (95047, QuantaBio, Beverly, MA, USA). qRT-PCR was carried out using the following primer sets: tshba (RefSeq NM_181494.2): forward 5′-cccccactgactacaccatctac-3′ and reverse 5′-gcatcccctctgaacaataaaacgag-3′ primers yielding a 149 bp product; gh1 (RefSeq NM_001020492.2): forward 5′-gctgcttcgtatctctttccgcc-3′ and reverse 5′-ggctgtccatcgagacatccc-3′ primers yielding a 174 bp product; pomca (RefSeq NM_181438.3): forward 5′-cgagcaaacgcaaagacaac-3′ and reverse 5′-gccaagcaggacacaacatc-3′ primers yielding a 121 bp product; fshb (RefSeq NM_205624.1): forward 5′-ggactatgctggacaatggatcg-3′ and reverse 5′-tcagagccacggggtac-3′ primers yielding a 154 bp product; lhb (RefSeq NM_205622.2): forward 5′-acggtatcggtggaaaaagagg-3′ and reverse 5′-tacgtgcacactgtctggtg-3′ primers yielding a 134 bp product. The reference gene used for calculating the relative expression was actb2 (RefSeq NM_181601.5), using forward 5′-ccccaaacccaagttcagcc-3′ and reverse 5′-acccacgatggatgggaaga-3′ primers that yielded a 128 bp product.
The qRT-PCR was performed using PerfeCTa SYBR green FastMix (95074-250-2, QuantaBio, Beverly, MA, USA) in a QuantStudio 1 instrument (Thermo Fisher Scientific, Waltham, MA, USA) and analyzed by QuantStudioTM Design & Analysis Software v1.5.1. The qRT-PCR amplification protocol consisted of 20 s of initial denaturation at 95 °C, followed by 40 cycles of 1 s denaturation at 95 °C, annealing and extension at 60 °C for 20 s and a final melting-curve stage. The reactions were performed in triplicates and the relative gene expression was calculated by the comparative-threshold-cycle method (2−∆∆Ct). The WT expression was set to 1, and the gene expression of the lhx4 mutant compared to that of the WT was calculated. Statistical differences in gene expression between genotypes were determined by Mann–Whitney test.
4.10. Transgenic Reporter Lines
Tg(−1.
0pomca:GFP)
zf44 [
24] and Tg(Oni.
lhb:TagRFP,myl7:TagRFP) [
5] reporter lines were utilized; the latter also expresses RFP in heart cells for the detection of positive transgenic larvae, as
lhb expression initiates only at a later stage. For accurate fluorescence-level comparisons, only reporter fish harboring a single transgenic insertion of GFP/RFP were used. The transgenic reporter lines and homozygous
lhx4 mutants were crossed, yielding heterozygous
lhx4 mutants. GFP/RFP-positive progeny were raised to adulthood and crossed with heterozygous
lhx4 mutants to produce homozygous
lhx4 mutants and WT siblings with a single transgenic allele.
When reaching maturity, fish were sacrificed, the pituitary was exposed by removing the jaws and the fluorescence was documented (see
Section 4.11). Subsequently, fish were genotyped as previously described (see ‘Generation of
lhx4-Mutant Zebrafish and Genotyping’). The mean intensity and area (pixels) of the GFP fluorescence were computed using ImageJ software 2.1.0. (National Institute of Health, Bethesda, MD, USA) and multiplied to produce the integrated density, and differences between genotypes were analyzed by
t-test.
4.11. Imaging
Images were taken with an SZX16 Research Stereo Microscope (Olympus, Waltham, MA, USA) equipped with a camera (DP74) and cellSens Entry 2.1 software, using an Oblique high-contrast cartridge (SZX2-COBH). The X-Cite Xylis Broad Spectrum LED Illumination System (Excelitas technologies, Waltham, MA, USA) was used for fluorescence excitation.
The immunostaining signal was captured with an AX10 fluorescence microscope (Zeiss, Oberkochen, Germany) equipped with a camera and Zen 2.3 lite software.