Abstract
Background
POU1F1 encodes both PIT-1α, which plays pivotal roles in pituitary development and GH, PRL and TSHB expression, and the alternatively spliced isoform PIT-1β, which contains an insertion of 26-amino acids (β-domain) in the transactivation domain of PIT-1α due to the use of an alternative splice acceptor at the end of the first intron. PIT-1β is expressed at much lower levels than PIT-1α and represses endogenous PIT-1α transcriptional activity. Although POU1F1 mutations lead to combined pituitary hormone deficiency (CPHD), no patients with β-domain mutations have been reported.
Results
Here, we report that a three-generation family exhibited different degrees of CPHD, including growth hormone deficiency with intrafamilial variability of prolactin/TSH insufficiency and unexpected prolactinoma occurrence. The CPHD was due to a novel POU1F1 heterozygous variant (c.143-69T>G) in intron 1 of PIT-1α (RefSeq number NM_000306) or as c.152T>G (p.Ile51Ser) in exon 2 of PIT-1β (NM_001122757). Gene splicing experiments showed that this mutation yielded the PIT-1β transcript without other transcripts. The lymphocyte PIT-1β mRNA expression was significantly higher in the patients with the heterozygous mutation than a control. A luciferase reporter assay revealed that the PIT-1β-Ile51Ser mutant repressed PIT-1α and abolished transactivation capacity for the rat prolactin promoter in GH3 pituitary cells.
Conclusions
We describe, for the first time, that the PIT-1β mutation can cause CPHD through a novel genetic mechanism, such as PIT-1β overexpression, and that POU1F1 mutation might be associated with a prolactinoma. Analysis of new patients and long-term follow-up are needed to clarify the characteristics of PIT-1β mutations.
Introduction
POU1F1, also known as pituitary-specific transcription factor 1 (PIT-1), plays pivotal roles in pituitary development and growth hormone(GH), prolactin (PRL) and TSHB gene expression (1). The POU1F1 gene encodes a 291-amino acid (AA) protein, PIT-1 (specifically referred to as PIT-1α, accession number NM_000306). PIT-1α consists of six exons containing three functional domains: an N-terminal transactivation domain (TAD), a POU-specific domain and a C-terminal homeodomain (2, 3). The POU1F1 gene also encodes the splice isoform, PIT-1β (NM_001122757), which includes a 26-AA β-domain insertion in the TAD resulting from a 78-bp in-frame insertion at the end of the first intron of the PIT-1α transcript caused by alternative splicing (4, 5, 6). Researchers have assumed that PIT-1α is the functional isoform for pituitary-specific expression and that PIT-1β may have little physiological significance for two reasons: PIT-1β is expressed at low levels (<5% of total POU1F1 expression) in most species (7), and PIT-1β represses GH, PRL, and TSHβ promoter activity in a pituitary-specific manner (8, 9). The repressive activity of the β-domain toward the PRL promoter has been most extensively studied, and PIT-1β represses both the basal and Ras signaling-stimulated rat PRL (rPRL) promoter in GH4 pituitary cells (8, 9, 10, 11, 12).
Mutations in the POU1F1 gene lead to combined pituitary hormone deficiency (CPHD), which manifests as GH, PRL and TSH deficiency in both mice and humans (2, 13). For POU1F1 mutations, the mode of inheritance is either autosomal recessive with homozygosity or compound heterozygosity for inactivating mutations or autosomal dominant with heterozygosity for dominant-negative mutations. Given that PIT-1β has a dominant-negative effect on PIT-1α, heterozygous mutations that disturb the balance between isoforms in favor of PIT-1β might result in CPHD (14). However, there are to date no reported cases caused by this mechanism, and notably, POU1F1 mutations leading to CPHD all map to regions shared by PIT-1α and PIT-1β (13, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26). For example, Takagi et al. described two brothers with CPHD due to a heterozygous POU1F1 mutation and found that in an in vitro system, an intronic variant in the intron 1 acceptor site caused PIT-1α exon 2 skipping while PIT-1β expression remained intact; however, the lymphocytes of the patients with this mutation do not express PIT-1β (18). In contrast, intact PIT-1α and exon 2-skipped PIT-1 (Δ48-72 PIT-1), which have been reported to have a dominant-negative effect on PIT-1α (25), are expressed. Consequently, the authors concluded that isolated PIT-1α mutations might be sufficient to cause CPHD.
Here, we describe a three-generation family with autosomal dominant inheritance exhibiting different degrees of CPHD due to a novel heterozygous variant identified in the β-subunit domain of the POU1F1 gene, leading to increased expression of PIT-1β. In addition, we report an unexpected phenotype, prolactinoma, observed in the oldest patient carrying this mutation.
Subjects and methods
Subjects
The studied pedigree comprised six individuals (I.2, II.1, II.3, II.4, III.1, III.2, and III.3) from a Japanese family (Fig. 1). All individuals or their parents gave written informed consent. The study was approved by the Institutional Review Board of Asahikawa Medical University.
Family pedigree and POU1F1 screening results. (A) Family pedigree: males, squares; females, circles; M, mutated allele; N, normal allele. The black boxes/circles indicate the affected patients. An oblique line indicates that the corresponding subject is deceased. The arrow indicates the proband. The s.d. of the heights of unaffected adult individuals and affected patients at diagnosis of GHD are also shown. (B) Schemes showing the POU1F1 mutation identified in the affected patients and a chromatogram showing the POU1F1 heterozygous mutation in the proband. Black filled rectangles show exons, and the exon numbers are indicated. The arrow indicates the location of the mutation, which is described as c.143-69T>G for RefSeq number NM_000306 or as c.152T>G (p.Ile51Ser) for NM_001122757 (POU1F1). A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Family pedigree and POU1F1 screening results. (A) Family pedigree: males, squares; females, circles; M, mutated allele; N, normal allele. The black boxes/circles indicate the affected patients. An oblique line indicates that the corresponding subject is deceased. The arrow indicates the proband. The s.d. of the heights of unaffected adult individuals and affected patients at diagnosis of GHD are also shown. (B) Schemes showing the POU1F1 mutation identified in the affected patients and a chromatogram showing the POU1F1 heterozygous mutation in the proband. Black filled rectangles show exons, and the exon numbers are indicated. The arrow indicates the location of the mutation, which is described as c.143-69T>G for RefSeq number NM_000306 or as c.152T>G (p.Ile51Ser) for NM_001122757 (POU1F1). A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Family pedigree and POU1F1 screening results. (A) Family pedigree: males, squares; females, circles; M, mutated allele; N, normal allele. The black boxes/circles indicate the affected patients. An oblique line indicates that the corresponding subject is deceased. The arrow indicates the proband. The s.d. of the heights of unaffected adult individuals and affected patients at diagnosis of GHD are also shown. (B) Schemes showing the POU1F1 mutation identified in the affected patients and a chromatogram showing the POU1F1 heterozygous mutation in the proband. Black filled rectangles show exons, and the exon numbers are indicated. The arrow indicates the location of the mutation, which is described as c.143-69T>G for RefSeq number NM_000306 or as c.152T>G (p.Ile51Ser) for NM_001122757 (POU1F1). A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
POU1F1 genetic testing
All six exons and exon/intron boundaries of the POU1F1 gene (NM_001122757) were analyzed in all the subjects by Sanger sequencing. The primer sequences used are shown in Table 1.
Sequences of primers for the POU1F1 gene analysis.
| Exon | Primer sequence (5′-3′) | Product size (bp) | |
|---|---|---|---|
| Forward | Reverse | ||
| 1 | TGTATAAAGGGATTTCCTTGCAG | TGGAGGGGTAAAATGAAAGATG | 430 |
| 2 | CGTCAGAGAACTTACCCAAAATG | GGGTGTCACAGAGCAGGAA | 453 |
| 3 | CTGAGACAGGCTAGAGCAGAA | CGTCCACAGTAGAGATGAAGAATG | 461 |
| 4 | AAATAATTCTGATTTGCCAAGAA | TCAAAGAGAAAAGGCGGAAA | 427 |
| 5 | AATGTCTAGCATTTTTATTTTGCAT | GCCTTGGCCTCCCAATTC | 301 |
| 6 | TGCAAGTGTGTTCAGAGTTTTT | AAAGTGGAAAAGTAAAGCTTCTGT | 479 |
Plasmids
The plasmid pET01-Wild, harboring a 926-bp fragment of human POU1F1 spanning intron 1, exon 2 (72 bp in PIT-1α, 150 bp in PIT-1β) and the first 425 bp of intron 2, was provided by Dr Takagi and Prof Hasegawa (18). A plasmid expressing the POU1F1 mutant identified in the current patients was constructed and named pET01-Mutant. The plasmid pcDNA3.1-PIT-1α expressing human PIT-1α and rPRL-Luc, which expresses firefly luciferase under the control of the 0.6-kb rPRL 5′-flanking region, were donated by Prof Okimura (27). pcDNA3.1-PIT-1β was constructed by subcloning full-length human WT PIT-1β cDNA derived from a human pituitary cDNA library (Clontech, Mountain View, USA). pcDNA3.1/V5-His-PTI-1α and pcDNA3.1/V5-His-PIT-1β were also produced, as was pcDNA3.1/V5-His-PIT-1β-Ile51Ser, which expresses mutant PIT-1β with an isoleucine-to-serine substitution at amino acid position 51. DDK-tagged human HRAS expressing pCMV6-Entry-HRAS was purchased from OriGene (Rockville, USA). The plasmid pCMV6-Entry-HRAS-Gly12Val expresses a constitutively active HRAS mutant with a glycine-to-valine substitution at amino acid position 12 (10, 28). pRL-TK, which expresses Renilla luciferase, was purchased from Promega Corp. (Madison, USA).
Cell culture
HeLa human cervical cancer cells were maintained in Dulbecco’s modified Eagle’s medium. GH3 rat pituitary cells were maintained in Ham’s F10 medium with 15% horse serum and 2.5% fetal bovine serum.
Gene splicing experiment
Splicing assays were performed in HeLa cells transfected with pET01-Wild or pET01-Mutant plasmids. cDNA produced from RNA reverse transcription was PCR-amplified with a primer combination from a pET01 vector-specific set and was subsequently sequenced according to the methods used in a previous report (18).
mRNA expression analysis of immortalized patient lymphocytes
Epstein-Barr virus-transformed lymphocytes derived from two patients (II.3 and III.3) and an independent control subject were established. cDNA was PCR-amplified using primers encompassing exons 1 to 4 of POU1F1 and was subsequently sequenced. The PCR products were separated by electrophoresis, stained with ethidium bromide, photographed and quantified by ImageJ software (http://rsbweb.nih.gov/ij/); the values were normalized to those of the loading control GAPDH.
Luciferase reporter assay
As previously shown, recombinant PIT-1α and PIT-1β expression using plasmids leads to different protein levels in transient transfection experiments, and the transactivation potency of these two isoforms must be normalized to their levels of expression (9, 10, 11, 29). A preliminary experiment to assess input DNA levels that would yield similar protein expression levels showed that PIT-1α was expressed at three-fold greater levels than WT PIT-1β or mutant PIT-1β-Ile51Ser (data not shown). Cells were transfected with 0.25 or 0.75 µg of pcDNA3.1/V5-His-PTI-1α or pcDNA3.1/V5-His-PIT-1β (WT or the Ile51Ser mutant) together with rPRL-Luc and pRL-TK in 24-well plates. In some experiments, cells were also transfected with 1 µM control or rat Pou1f1 3′UTR-targeting siRNA (Dharmacon, San Francisco, USA) using Accell siRNA delivery medium (Dharmacon) according to the manufacturer’s protocol. In addition to the abovementioned plasmids, some experiments involved the use of pCMV6-Entry-HRAS-Gly12Val as well. Twenty-four hours after transfection, the cells were lysed and analyzed for luciferase activity using a Dual-Luciferase Reporter Assay System (Promega) and immunoblotting.
Western blotting
Western blotting of whole-cell lysates from transfected GH3 cells was performed. The following primary antibodies were used: anti-PIT-1 (1:1000 dilution; SCB, Santa Cruz, USA), anti-His-tag (1:15000; Proteintech, Manchester, UK), anti-DDK-tag (1:40000; FUJIFILM, Tokyo, Japan), and anti-Actin (1:2000; SCB, Santa Cruz, USA).
Statistical analysis
Data are presented as mean ± s.e.m. Statistical analyses were performed using two-tailed Student’s t-tests for two comparisons or one-way ANOVA with Bonferroni correction for multiple comparisons. Statistical significance was set at P < 0.05.
Results
Family presentation
The proband (II.3), a 44-year-old female, was born at 41 weeks of gestation by cesarean delivery. Her birth history was uncomplicated, with a birth length of 45 cm (−2.70 s.d. and a weight of 3180 g (−0.25 s.d.). At 5 years and 2 months of age, when she was referred to our department for endocrine evaluation, her height was 84 cm (−5.16 s.d.), and her weight was 10.6 kg (−2.9 s.d.) (Fig. 2A). The examination revealed facial dysmorphism with frontal bossing and a depressed and hypoplastic nasal bridge. Investigations revealed GH deficiency (GHD; Table 2). Although total triiodothyronine (T3) (2.26 nmol/L (reference range (RR): 1.23–2.46)) and thyroxine (T4) (92.7 nmol/L (RR: 57.9–148.0)) levels were normal, the TSH response to TRH stimulation was low (Table 2). The proband began GH replacement treatment at 5 years and 7 months of age. As her total T4 decreased to 77.2 nmol/L at 3 months after initiating GH treatment, levothyroxine (L-T4) was started. Her height velocity improved dramatically with these treatments (Fig. 2A). Menarche occurred when she was 14 years old. GH was stopped when she was 15 years and 6 months of age, as she had reached an adult height of 148.7 cm (−1.79 s.d.) and had a normal brain MRI scan. Her thyroid function was re-evaluated when she was 16 years and 11 months old; the TRH stimulation test showed a normal TSH response and a low PRL response (Table 2). She gave birth to her first child (III.1) at the age of 20 and to a second child (III.2) at the age of 22 (Fig. 1A). She did not have galactorrhea during puerperium with either child, and both children were healthy and achieved normal adult heights. At 27 years of age, she was diagnosed with adult GHD (Table 2) and received GH replacement therapy for 2 years, after which she expressed a desire to discontinue GH treatment. At 30 years old, she gave birth to a third child (III.3); she produced very little breast milk for the first month, which was insufficient for breastfeeding. At 32 years of age, she asked to discontinue L-T4 because she was taking little of the medicine at that time. TRH stimulation performed after L-T4 cessation showed a normal TSH response and a relatively low PRL response, though the response was clearly higher than that in her childhood (Table 2). Based on this result, L-T4 supplementation was discontinued. At the last follow-up, she was euthyroid and healthy, although her IGF-1 levels remained low (Table 2).
Clinical features of the patients with the POU1F1 mutation. (A) Growth curve of the proband (subject II.3). GH treatment was given from 5 years of age to 15 years of age. The arrow indicates the time of levothyroxine (L-T4) initiation, 3 months after GH initiation. (B, C and D) Brain MRI of the proband’s mother (subject I.2) at 63 years of age. A 28 × 16 × 23 mm mass lesion in the pituitary gland was detected. The lesion was isointense to gray matter on T1-weighted MRI (B), slightly hyperintense on T2-weighted MRI (C), and almost solidly enhanced on dynamic gadolinium-enhanced imaging (D). (E) Growth curve of the proband’s child (subject III.3). L-T4 and GH were given from 7 days of age and 13 months of age, respectively. A full color version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Clinical features of the patients with the POU1F1 mutation. (A) Growth curve of the proband (subject II.3). GH treatment was given from 5 years of age to 15 years of age. The arrow indicates the time of levothyroxine (L-T4) initiation, 3 months after GH initiation. (B, C and D) Brain MRI of the proband’s mother (subject I.2) at 63 years of age. A 28 × 16 × 23 mm mass lesion in the pituitary gland was detected. The lesion was isointense to gray matter on T1-weighted MRI (B), slightly hyperintense on T2-weighted MRI (C), and almost solidly enhanced on dynamic gadolinium-enhanced imaging (D). (E) Growth curve of the proband’s child (subject III.3). L-T4 and GH were given from 7 days of age and 13 months of age, respectively. A full color version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Clinical features of the patients with the POU1F1 mutation. (A) Growth curve of the proband (subject II.3). GH treatment was given from 5 years of age to 15 years of age. The arrow indicates the time of levothyroxine (L-T4) initiation, 3 months after GH initiation. (B, C and D) Brain MRI of the proband’s mother (subject I.2) at 63 years of age. A 28 × 16 × 23 mm mass lesion in the pituitary gland was detected. The lesion was isointense to gray matter on T1-weighted MRI (B), slightly hyperintense on T2-weighted MRI (C), and almost solidly enhanced on dynamic gadolinium-enhanced imaging (D). (E) Growth curve of the proband’s child (subject III.3). L-T4 and GH were given from 7 days of age and 13 months of age, respectively. A full color version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Clinical characteristics in the affected POU1F1 mutation (c.143-69T>G based on NM_000306, or c.152T>G, p.Ile51Ser based on NM_001122757) carriers.
| Characteristics | RR | II.3 (Proband) | I.2 (Proband’s mother) | III.3 (Proband’s child) | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1981 | 1988 | 1992 | 2003 | 2008 | 2019 | 1981 | 2008 | 2018 | 2008 | 2019 | ||
| Age (years) | 5 | 12 | 16 | 27 | 32 | 43 | 34 | 61 | 71 | 1 | 12 | |
| Height (cm) | 84 | 136.8 | 148.3 | 148.7 | 148.4 | 148.9 | 135 | 133.4 | 133 | 67.2 | 159.8 | |
| Height SDS | −5.16 | −2.39 | −1.86 | −1.79 | NA | NA | −4.02 | NA | NA | −3.36 | +0.57 | |
| Weight (kg) | 10.6 | 28.4 | 46.1 | 43.7 | 48.3 | 49 | NA | 46.8 | 44 | 8.035 | 52.9 | |
| Weight SDS | −2.9 | −1.95 | −0.94 | NA | NA | NA | NA | NA | −1.69 | +0.71 | ||
| BMI | 15.0 | 15.2 | 21.0 | 19.8 | 21.9 | 22.1 | 26.3 | 24.9 | 17.8 | 20.7 | ||
| Hormone replacement | No | GH, L-T4 | L-T4 | L-T4 | No | No | No | No | No (CBG) | L-T4 | GH, L-T4 | |
| IGF-1 (nmol/L)† | NA | 36.0 | 11.6 | 13.6 | 8.6 | 8.4 | NA | 9.5 | 11.4 | 0.5 | 42.4 | |
| IGF-1 SDS | −0.8 | −4.8 | −3.4 | −4.5 | −3.2 | −1.8 | −0.5 | −2.6 | +0.2 | |||
| Glucagon-propranolol test | NA | NA | NA | NA | NA | NA | NA | NA | NA | |||
| RIA/ECLIA 2012‡ | ||||||||||||
| Basal GH (µg/L) | 0.63 | 1.6 | ||||||||||
| Peak GH (µg/L) | c | 2.15 (> 10) | 3.3 (>5) | |||||||||
| ITT | NA | NA | NA | NA | NA | NA | NA | NA | NA | |||
| Basal GH (µg/L) | 0.67 | 0.32 | ||||||||||
| Peak GH (µg/L) | c | 3 (>10) | 2.96 (>5) | |||||||||
| ATT | NA | NA | NA | NA | NA | NA | NA | NA | NA | |||
| Basal GH (µg/L) | 1.05 | 0.29 | ||||||||||
| Peak GH (µg/L) | c | 1.35 (> 10) | 0.92 (>6) | |||||||||
| Levodopa | NA | NA | NA | NA | NA | NA | NA | NA | NA | NA | ||
| Basal GH (µg/L) | 0.67 | |||||||||||
| Peak GH (µg/L) | c | 0.67 (>6) | ||||||||||
| GRH test | NA | NA | NA | NA | NA | NA | NA | NA | NA | |||
| Basal GH (µg/L) | 0.11 | 0.72 | ||||||||||
| Peak GH (µg/L) | c | 1.42 (>6) | 1.06 (>6) | |||||||||
| Total T4 (nmol/L)⁋ | 57.9–148.0 | 92.7 | 135.1 | NA | NA | NA | NA | NA | NA | NA | NA | NA |
| Total T3 (nmol/L)@ | 1.23–2.46 | 2.16 | 2.03 | NA | NA | NA | NA | NA | NA | NA | NA | NA |
| RIA/ECLIA 1994§ | ||||||||||||
| Free T4 (pmol/L) | 11.6–21.9 | NA | 25.6 (8.88–29.1) | NA | 16.7 | 14.9 | 12.4 | NA | 11.2 | 13.1 | 12.5 | 19.0 |
| Free T3 (pmol/L) | 3.53–6.14 | NA | NA | NA | 4.16 | 4.05 | 3.73 | NA | 2.93 | 3.85 | 4.53 | 6.53 |
| Basal TSH (mIU/L) | 0.5–5.0* | 2.2 | 0.003 | 2 | 0.196 | 1.48 | 0.81 | NA | 0.8 | 1.43 | 0.47 | 0.01 |
| Basal PRL (µg/L) | 4.29–13.69 | NA | NA | 0.42 (2.2–22.3) | 3.23 | 2.92 | 4.34 | NA | 616.7 | 0.84 | 6.82 | 2.69 |
| TRH test | NA | NA | NA | NA | NA | |||||||
| Peak TSH (mIU/L) | 4.8 | 10.7 | 8.67 | 13.21 | 1.45 | NA | ||||||
| Peak PRL (µg/L) | NA | 2.64 | 9.71 | 1940 | 16.64 | NA | ||||||
| Basal LH (IU/L) | 1.27–9.57** | 6.6 | 0.484 | NA | 1.12 | 5.39 | NA | NA | 0.76 | NA | 0.51 | 5.06 |
| Basal FSH (IU/L) | 1.18–15.82a | 4 | 7.934 | NA | 2.53 | 3.88 | NA | NA | 9.85 | NA | 0.47 | 4.72 |
| GnRH test | NA | NA | NA | NA | NA | NA | NA | NA | ||||
| Peak LH (IU/L) | 15.3 | 6.2 | 6.44 | |||||||||
| Peak FSH (IU/L) | 24.1 | 26.33 | 1.81 | |||||||||
| Basal ACTH (pmol/L) | 1.63–12.27 | NA | NA | NA | 7.48 | 7.70 | NA | NA | 6.01 | NA | 48.14 | NA |
| Basal cortisol (nmol/L) | 171.1–364.2 | NA | NA | NA | 280.1 | 452.1 | 189.6 | NA | 77.3 | NA | 871.1 | 207.8 |
| CRH test | NA | NA | NA | NA | NA | NA | NA | NA | ||||
| Peak ACTH (pmol/L) | 34.25 | 48.14 | ||||||||||
| Peak cortisol (nmol/L) | 377.0 | 871.1 | ||||||||||
†Assay with RIA; ‡assay with RIA before year 2012, ECLIA after year 2012; ⁋assay with CPBA at year 1981, RIA at RIA; @assay with RIA; §assay with RIA before year 1994, ECLIA after year 1994; *RR for proband (1981–1992): 0.34–3.5; **RR for proband (1981–1992): >7, RR for proband’s child (2008): 0.02–0.15 and for (2019): 1.61–3.63; aRR for proband (1981–1992): <13, RR for proband’s child (2008): 0.38–1.11 and for (2019): 1.21–8.22; cNormal response is given in parentheses.
ATT, arginine tolerance test; CBG, cabergoline; CRH, corticotropin releasing hormone; GH, growth hormone; GnRH, gonadotropin releasing hormone; GRH, growth hormone releasing hormone; ITT, insulin tolerance test; L-T4, levothyroxine; PRL, prolactin; RR, reference range . NA, not assessed.
The proband’s mother (I.2), a 73-year-old female, was noticed to have a short stature (135 cm, −4.02 s.d.) at 34 years of age, the time when the proband (II.2) was examined. The parents and first child of I.2 were of normal stature. I.2 was diagnosed with GHD based on a glucagon–propranolol test (Table 2). Although another endocrine evaluation was not performed, she appeared to have PRL deficiency because she had never had galactorrhea during puerperium with either of her children. Menopause occurred when she was 49 years old. At 63 years old, she underwent an endocrinological screening test showing hyperprolactinemia at 406 µg/L (reference range 4.29–13.69). Medical interviews revealed that she had been lactating for 10 years without taking any medication that causes hyperprolactinemia. A brain MRI scan showed a 28 × 16 × 23 mm mass in the pituitary gland (Fig. 2B, C and D). The lesion was isointense to gray matter on T1-weighted MRI and slightly hyperintense on T2-weighted MRI. The dynamic gadolinium-enhanced MRI revealed that most of the tumor was solidly enhanced, indicating a macroadenoma. Although her free T3 (2.93 pmol/L (reference range 3.54–6.16)) and free T4 (11.20 pmol/L (reference range 11.58–21.88)) levels were low, the TSH response to TRH was normal (Table 2). Other endocrine functions were normal despite GHD (Table 2). Therefore, she was clinically diagnosed with prolactinoma, and the cabergoline treatment was started without performing biopsy or surgery. As she exhibited no clinical findings suggestive of hypothyroidism, her thyroid function was followed up without L-T4 treatment. Her PRL levels and thyroid function were normalized at 3 months after the cabergoline treatment (Table 2). Therefore, her mild TSH deficiency was potentially related to macroprolactinoma.
The proband’s third child (III.3), a 13-year-old boy, was born at 37 weeks of gestation by cesarean delivery. His birth length and weight were 46.3 cm (−0.84 s.d.) and 2710 g (−0.56 s.d.), respectively. At 1 day of age, a screening test for blood glucose showed hypoglycemia at 36 mg/dL. He had required glucose infusion for 6 days. At 7 days of age, endocrine screening revealed central hypothyroidism (TSH: 2.51 mIU/L (RR: 1.21–3.00); free T3: 313 pmol/L (RR: 3.73–7.48); free T4: 8.88 pmol/L (RR: 18.8–44.0)). Other pituitary hormone levels were normal for his age (IGF-1: 2.75 nmol/L (RR: 1.43–19.5); cortisol: 436.1 nmol/L (RR: 171.1–364.2); LH: 1.43 IU/L (RR: 0.2–3.0); FSH: 0.72 IU/L (RR: 0.2–2.1)). The L-T4 treatment was started. Despite normal neuromuscular development, his growth velocity gradually decreased from 4 months of age (Fig. 2E). At 13 months, he was admitted to our department for endocrine examination because of his short stature (body length: 67.2 cm (−3.36 s.d.), and investigations revealed GHD (Table 2). His TSH response to the TRH stimulation test was low, though the PRL level pre- or post-TRH loading was normal. Moreover, adrenal function was normal, and gonadotropin levels were appropriate for his age (Table 2). A brain MRI scan showed a normal pituitary grand. His responsiveness to GH treatment was very good with catch-up growth (Fig. 2E). Spontaneous puberty occurred at age 10 years and 2 months with a testicular volume of 4 mL.
The affected individuals carried a novel mutation at the β-domain in POU1F1
POU1F1 analysis, which was performed because the phenotype was compatible with POU1F1 mutation, revealed the presence of a heterozygous variant in all three affected family members. The variant is described as c.143-69T>G based on NM_000306 or c.152T>G (p.Ile51Ser) based on NM_001122757. In other words, this variant is located in intron 1 of PIT-1α or exon 2 of PIT-1β. This variant was not found in the unaffected first-degree relatives tested (Fig. 1), and the variant is not present in Genome Aggregation Database (gnomAD) or TogoVar, a comprehensive Japanese genetic variation database.
The novel POU1F1 mutation yielded PIT-1β, but not PIT-1α, in vitro using a minigene construct
We tested the effect of the novel mutation on in vitro splicing efficiency using a minigene construct (Fig. 3A). HeLa cells transfected with pET01-Mutant exclusively expressed mutant PIT-1β (c.152T>G, p.Ile51Ser) mRNA without activation of a cryptic donor splice site. In contrast, the WT pET01-Wild vector produced PIT-1α as well as a small amount of PIT-1β (Fig. 3B).
The c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation yields only the PIT-1β transcript. (A) Representations of the pET01-Wild and pET01-Mutant minigene constructs. A genomic fragment containing human POU1F1 spanning intron 1 (IVS1), exon 2, and the first 425 bp of intron 2 (IVS2) of the wild or mutant type was inserted into the exon-trapping vector using the XhoI and NotI restriction sites. Gray box, extended exon 2 specific for the PIT-1β isoform; white boxes, pET01 vector exons; horizontal arrows, vector-derived primers; vertical arrows, position of the c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation. (B) In vitro splicing assay. HeLa cells were transiently transfected with pET01-Wild or Mutant minigene constructs or empty pET01 vector. Subsequent RT-PCR analysis was performed with the primers indicated above. The pET01-Wild construct yielded two bands of 190 and 268 bp indicating inclusion of PIT-1α exon 2, which was dominantly expressed, and PIT-1β exon 2, respectively, while only a 268 bp band was generated by the pET01-Mutant, which was confirmed by sequencing. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
The c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation yields only the PIT-1β transcript. (A) Representations of the pET01-Wild and pET01-Mutant minigene constructs. A genomic fragment containing human POU1F1 spanning intron 1 (IVS1), exon 2, and the first 425 bp of intron 2 (IVS2) of the wild or mutant type was inserted into the exon-trapping vector using the XhoI and NotI restriction sites. Gray box, extended exon 2 specific for the PIT-1β isoform; white boxes, pET01 vector exons; horizontal arrows, vector-derived primers; vertical arrows, position of the c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation. (B) In vitro splicing assay. HeLa cells were transiently transfected with pET01-Wild or Mutant minigene constructs or empty pET01 vector. Subsequent RT-PCR analysis was performed with the primers indicated above. The pET01-Wild construct yielded two bands of 190 and 268 bp indicating inclusion of PIT-1α exon 2, which was dominantly expressed, and PIT-1β exon 2, respectively, while only a 268 bp band was generated by the pET01-Mutant, which was confirmed by sequencing. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
The c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation yields only the PIT-1β transcript. (A) Representations of the pET01-Wild and pET01-Mutant minigene constructs. A genomic fragment containing human POU1F1 spanning intron 1 (IVS1), exon 2, and the first 425 bp of intron 2 (IVS2) of the wild or mutant type was inserted into the exon-trapping vector using the XhoI and NotI restriction sites. Gray box, extended exon 2 specific for the PIT-1β isoform; white boxes, pET01 vector exons; horizontal arrows, vector-derived primers; vertical arrows, position of the c.143-69T>G (in NM_000306) or c.152T>G (p.Ile51Ser) POU1F1 mutation. (B) In vitro splicing assay. HeLa cells were transiently transfected with pET01-Wild or Mutant minigene constructs or empty pET01 vector. Subsequent RT-PCR analysis was performed with the primers indicated above. The pET01-Wild construct yielded two bands of 190 and 268 bp indicating inclusion of PIT-1α exon 2, which was dominantly expressed, and PIT-1β exon 2, respectively, while only a 268 bp band was generated by the pET01-Mutant, which was confirmed by sequencing. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
PIT-1β mRNA expression was increased in patient lymphocytes
RT-PCR with a forward primer targeting exon 1 and a reverse primer targeting exon 4 yielded two products in both the patients and a healthy control (Fig. 4A, top panel). Sequencing of these mRNA products showed the short product to be PIT-1α and the long product to be PIT-1β. The patients harbored a c.152T>G (p.Ile51Ser) heterozygous mutation in the region encoding PIT-1β (Fig. 4B), and expression levels of PIT-1β were approximately two-fold higher in the patients than in the healthy control (Fig. 4A, bottom panel). No other bands were detected in the patient or control, indicating that the c.152T>G mutation does not activate a cryptic splice donor site or cause PIT-1α exon 2 skipping.
Analysis of POU1F1 mRNA expression by semiquantitative RT-PCR in Epstein-Barr virus-transformed lymphocytes derived from patients and independent control subjects. (A) cDNA produced by RT was PCR-amplified with primers encompassing exons 1 to 4. The gray filled rectangle in exon 2 of Pit-1β indicates the β-domain. RT-PCR generated two products in the control and patients. Sequencing revealed that the long and short products corresponded to PIT-1β and PIT-1α transcripts, respectively (top panel). The bands were integrated by ImageJ software analysis (bottom panel). The values were normalized to those of the loading control GAPDH. The bars and error bars indicate the mean ± s.e.m. of the fold changes in PIT-1α or PIT-1β expression compared to control PIT-1α expression. *P < 0.05; **P < 0.01 compared to PIT-1α mRNA expression in the control. The results were derived from three different experiments, and representative images are shown. (B) Chromatogram showing PIT-1β mRNA as determined by sequencing of the long bands of RT-PCR products from patients and a control. The arrow indicates the heterozygous c.152T>G (p.Ile51Ser) mutation based on NM_001122757 for POU1F1. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Analysis of POU1F1 mRNA expression by semiquantitative RT-PCR in Epstein-Barr virus-transformed lymphocytes derived from patients and independent control subjects. (A) cDNA produced by RT was PCR-amplified with primers encompassing exons 1 to 4. The gray filled rectangle in exon 2 of Pit-1β indicates the β-domain. RT-PCR generated two products in the control and patients. Sequencing revealed that the long and short products corresponded to PIT-1β and PIT-1α transcripts, respectively (top panel). The bands were integrated by ImageJ software analysis (bottom panel). The values were normalized to those of the loading control GAPDH. The bars and error bars indicate the mean ± s.e.m. of the fold changes in PIT-1α or PIT-1β expression compared to control PIT-1α expression. *P < 0.05; **P < 0.01 compared to PIT-1α mRNA expression in the control. The results were derived from three different experiments, and representative images are shown. (B) Chromatogram showing PIT-1β mRNA as determined by sequencing of the long bands of RT-PCR products from patients and a control. The arrow indicates the heterozygous c.152T>G (p.Ile51Ser) mutation based on NM_001122757 for POU1F1. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Analysis of POU1F1 mRNA expression by semiquantitative RT-PCR in Epstein-Barr virus-transformed lymphocytes derived from patients and independent control subjects. (A) cDNA produced by RT was PCR-amplified with primers encompassing exons 1 to 4. The gray filled rectangle in exon 2 of Pit-1β indicates the β-domain. RT-PCR generated two products in the control and patients. Sequencing revealed that the long and short products corresponded to PIT-1β and PIT-1α transcripts, respectively (top panel). The bands were integrated by ImageJ software analysis (bottom panel). The values were normalized to those of the loading control GAPDH. The bars and error bars indicate the mean ± s.e.m. of the fold changes in PIT-1α or PIT-1β expression compared to control PIT-1α expression. *P < 0.05; **P < 0.01 compared to PIT-1α mRNA expression in the control. The results were derived from three different experiments, and representative images are shown. (B) Chromatogram showing PIT-1β mRNA as determined by sequencing of the long bands of RT-PCR products from patients and a control. The arrow indicates the heterozygous c.152T>G (p.Ile51Ser) mutation based on NM_001122757 for POU1F1. A full colour version of this figure is available at https://doi.org/10.1530/EJE-20-1313.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
The PIT-1β-Ile51Ser mutant repressed endogenous Pit-1, abolished transactivation capacity and impaired synergistic interaction with Ras at the rPRL promoter in GH3 cells
The functional impact of PIT-1β-Ile51Ser on transcriptional activity was assessed using a luciferase reporter assay. As depicted in Fig. 5A, baseline rPRL reporter gene activity was significantly elevated, reflecting endogenous Pou1f1 expression in GH3 cells. Although PIT-1α transfection enhanced rPRL promoter activity, PIT-1β transfection significantly suppressed; PIT-1β-Ile51Ser transfection also significantly suppressed the rPRL promoter activity to the same degree as WT PIT-1β. Next, to explore whether PIT-1β-Ile51Ser exhibits transcriptional activity at the rPRL promoter in GH3 cells, we knocked down endogenous Pou1f1 using a 3′UTR siRNA construct and subsequently rescued it by exogenously expressing the open reading frame of PIT-1α and WT or mutant PIT-1β. As shown in Fig. 5B, Pou1f1 knockdown did not induce rPRL promoter activity; as expected, forced PIT-1α expression reversed the reduction in rPRL promoter activity. However, WT PIT-1β and PIT-1β-Ile51Ser displayed little transcriptional activity at the rPRL promoter. Finally, we examined the effect of PIT-1β-Ile51Ser on Ras signaling-stimulated rPRL promoter activity. Cotransfection of PIT-1α with a constitutively active HRAS mutant resulted in synergistic enhancement of rPRL transactivation, whereas WT PIT-1β and PIT-1β-Ile51Ser not only failed to enhance the Ras response but also inhibited activation of the rPRL promoter to the same extent (Fig. 5C).
The PIT-1β-Ile51Ser mutant inhibits the activity of the rPRL promoter in GH3 cells. (A) PIT-1α enhanced rPRL promoter activity, whereas PIT-1β-Ile51Ser inhibited rPRL promoter activity, similar to WT PIT-1β. GH3 cells were transfected with an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. The results of Western blotting for His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (B) PIT-1β-Ile51Ser and WT PIT-1β seldom had rPRL promoter activity. The rPRL promoter activity was reduced by knockdown of rat Pou1f1. The forced PIT-1α expression reversed the reduction in rPRL promoter activity due to Pou1f1 knockdown, whereas PIT-1β-Ile51Ser and WT PIT-1β expression did not. GH3 cells were transfected with control or Pou1f1 siRNA, which targeted the 3′UTR of Pou1f1, in the presence of an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. *P < 0.05; **P < 0.01 compared to the condition with Pou1f1 siRNA and rPRL-Luc transfection. ##P < 0.01 compared to the condition with Pou1f1 siRNA, rPRL-Luc, and pcNDA3.1/V5-His-PIT-1α transfection. The results of Western blotting for endogenous rat Pit-1 and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (C) Reduced rPRL transactivation ability and impaired synergy with HRAS of PIT-1β-Ile51Ser. To evaluate the effect of PIT-1β-Ile51Ser on the synergistic interactions of HRAS on PRL promoters, pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser was transfected with a constitutively active HRAS construct, rPRL-Luc and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. ##P < 0.01 compared to the condition with HRAS and rPRL-Luc transfection. The results of Western blotting for HRAS and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. The results were derived from three different experiments, and representative images are shown.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
The PIT-1β-Ile51Ser mutant inhibits the activity of the rPRL promoter in GH3 cells. (A) PIT-1α enhanced rPRL promoter activity, whereas PIT-1β-Ile51Ser inhibited rPRL promoter activity, similar to WT PIT-1β. GH3 cells were transfected with an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. The results of Western blotting for His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (B) PIT-1β-Ile51Ser and WT PIT-1β seldom had rPRL promoter activity. The rPRL promoter activity was reduced by knockdown of rat Pou1f1. The forced PIT-1α expression reversed the reduction in rPRL promoter activity due to Pou1f1 knockdown, whereas PIT-1β-Ile51Ser and WT PIT-1β expression did not. GH3 cells were transfected with control or Pou1f1 siRNA, which targeted the 3′UTR of Pou1f1, in the presence of an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. *P < 0.05; **P < 0.01 compared to the condition with Pou1f1 siRNA and rPRL-Luc transfection. ##P < 0.01 compared to the condition with Pou1f1 siRNA, rPRL-Luc, and pcNDA3.1/V5-His-PIT-1α transfection. The results of Western blotting for endogenous rat Pit-1 and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (C) Reduced rPRL transactivation ability and impaired synergy with HRAS of PIT-1β-Ile51Ser. To evaluate the effect of PIT-1β-Ile51Ser on the synergistic interactions of HRAS on PRL promoters, pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser was transfected with a constitutively active HRAS construct, rPRL-Luc and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. ##P < 0.01 compared to the condition with HRAS and rPRL-Luc transfection. The results of Western blotting for HRAS and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. The results were derived from three different experiments, and representative images are shown.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
The PIT-1β-Ile51Ser mutant inhibits the activity of the rPRL promoter in GH3 cells. (A) PIT-1α enhanced rPRL promoter activity, whereas PIT-1β-Ile51Ser inhibited rPRL promoter activity, similar to WT PIT-1β. GH3 cells were transfected with an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. The results of Western blotting for His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (B) PIT-1β-Ile51Ser and WT PIT-1β seldom had rPRL promoter activity. The rPRL promoter activity was reduced by knockdown of rat Pou1f1. The forced PIT-1α expression reversed the reduction in rPRL promoter activity due to Pou1f1 knockdown, whereas PIT-1β-Ile51Ser and WT PIT-1β expression did not. GH3 cells were transfected with control or Pou1f1 siRNA, which targeted the 3′UTR of Pou1f1, in the presence of an rPRL-Luc-expressing plasmid together with pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. *P < 0.05; **P < 0.01 compared to the condition with Pou1f1 siRNA and rPRL-Luc transfection. ##P < 0.01 compared to the condition with Pou1f1 siRNA, rPRL-Luc, and pcNDA3.1/V5-His-PIT-1α transfection. The results of Western blotting for endogenous rat Pit-1 and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. (C) Reduced rPRL transactivation ability and impaired synergy with HRAS of PIT-1β-Ile51Ser. To evaluate the effect of PIT-1β-Ile51Ser on the synergistic interactions of HRAS on PRL promoters, pcDNA3.1/V5-His-PIT-1α, PIT-1β, or PIT-1β-Ile51Ser was transfected with a constitutively active HRAS construct, rPRL-Luc and pRL-TK. The bars and error bars represent the mean ± s.e.m. values of firefly luciferase activity normalized to Renilla luciferase activity. **P < 0.01 compared to the condition with control vector and rPRL-Luc transfection. ##P < 0.01 compared to the condition with HRAS and rPRL-Luc transfection. The results of Western blotting for HRAS and His-tagged PIT-1α or β (wild or mutant) in the samples used for reporter assays are shown in the bottom panels. The results were derived from three different experiments, and representative images are shown.
Citation: European Journal of Endocrinology 185, 1; 10.1530/EJE-20-1313
Discussion
To the best of our knowledge, we report the first example of a PIT-1β mutation linked to CPHD, of which GHD is a common feature together with intrafamilial variability of PRL or TSH insufficiency. Moreover, we report an unexpected clinical finding of prolactin-producing adenoma in a mutation carrier who must have previously had PRL deficiency.
From a molecular perspective, we identified the mechanism by which the p.Ile51Ser mutation of PIT-1β results in overproduction of PIT-1β rather than PIT-1α (Figs 3 and 4). Compared with PIT-1α, mutant PIT-1β exhibited severely reduced transcriptional activity and inhibited the transcriptional activity of endogenous Pou1f1 in GH3 pituitary cells (Fig. 5). However, the basal and Ras signaling-stimulated PRL promoter activity of the p.Ile51Ser mutant was the same as that of WT PIT-1β. Hence, this is the first example of imbalance between POU1F1 isoforms in favor of PIT-1β production resulting in deleterious effects on pituitary development and hormone expression, even though our functional assays were performed in heterospecific systems and under non-physiological conditions. Overall, the mechanisms resulting in elevated expression of the p.Ile51Ser PIT-1β mutant in the present family members and reduced expression of WT PIT-1β in humans remain to be clarified. Indeed, although we examined whether the p.Ile51Ser mutation significantly disturbs the balance between exonic splicing enhancers or silencers in silico using the Human Splicing Finder system (https://www.genomnis.com/access-hsf) (30), we did not identify a known alteration, contrary to our experimental results. Therefore, clinicians and molecular biologists should be aware that in silico splicing prediction algorithms do not always provide reliable conclusions. The vast majority of hexamer sequences that function as enhancers or silencers for alternative exons are reported to be located in alternative exons (31). Given that p.Ile51Ser is located in an alternative exon (4, 5, 6), we speculate that mutation at this site might enhance transcription of mutant PIT-1β. Obviously, analysis of more patients is required to confirm that the PIT-1β mutation results in congenital hypopituitarism.
From a clinical perspective, PRL and TSH deficiencies are not always observed in patients with previously reported POU1F1 mutations, as in the present family with PIT-1β mutation. For example, two unique families without PRL deficiency have been reported (17, 32). In one, patients with a Pro76Leu heterozygous mutation exhibited isolated GHD due to selective binding of the mutant protein to the GH promoter (17). In the second, two siblings carried a homozygous Lys230Glu mutation; one had only GHD, whereas the other had combined deficiency of GH, PRL and TSH (32). In addition, inter- and intrafamilial variability with respect to onset time and extent have been reported for TSH deficiency due to POU1F1 mutation (33, 34). This may be because thyrotropes are derived from two different populations: one being POU1F1-independent and another being POU1F1-dependent (35). Moreover, TSH deficiency is a difficult condition to diagnose in clinical practice because of suboptimal detection of clinical and biochemical parameters (36, 37). The diagnosis of TSH deficiency is generally based on the combined findings of low free T4 and low or normal TSH (36, 37). Although a TRH stimulation test may help support the diagnosis of TSH deficiency in unclear situations (36), it is not always accurate; in fact, some patients diagnosed with central hypothyroidism respond normally (38). GH treatment causes a reduction in serum T4 due to either enhanced peripheral conversion of T4 to T3 or development of central hypothyroidism (37). In the proband in our study (II.3), TSH showed a low response to TRH, and T4 decreased after GH treatment, though it did remain in the normal range. L-T4 was started due to possible TSH deficiency; however, as the patient had normal thyroid function thereafter, she may not have had TSH deficiency. Thus, POU1F1 gene mutation should be considered in differential diagnosis for familial CPHD with GHD irrespective of the concurrence of PRL or TSH deficiency.
A characteristic clinical feature of the present patients with the PIT-1β mutation was fluctuating PRL secretion. In the proband (II.3), levels of PRL increased from childhood to adulthood (Table 2); there was clinical evidence of scant lactation after her third delivery, whereas lactation did not occur after her two previous deliveries. On the other hand, the basal PRL levels of the proband’s child (III.3) decreased from infancy to childhood. In the proband’s mother (I.2), prolactinoma developed, even though she had never exhibited galactorrhea during puerperium with either of her children. To date, there have been no reports of prolactinoma in hypopituitarism with a genetic etiology, including POU1F1 mutation. In general, it is unclear how the PIT-1β mutation is associated with the development of prolactinoma. A report examining PIT-1α and PIT-1β in pituitary adenomas, including prolactinoma, found that the isoforms are expressed in both functional and nonfunctional pituitary adenomas (39). Therefore, it has been suggested that the major role of the POU1F1 gene in pituitary adenoma may not be related to hormone production. Additionally, some POU1F1-negative PRL-secreting adenomas have been reported, suggesting a possible independent mechanism of transcription in prolactinoma (40). Hence, even if a genetic abnormality of a transcription factor that is important for pituitary development in patients with CPHD is present, long-term follow-up considering the possibility of adenoma development is necessary.
In conclusion, we show for the first time that PIT-1β mutation can cause CPHD. The evidence suggests that this mutation induces pathological conditions by increasing mutant PIT-1β alternative splicing, which has low transcriptional activity and inhibits endogenous PIT-1α. A previously unreported phenotype of POU1F1 mutations, prolactinoma, was observed in the oldest family member. Studies of new patients and long-term follow-up data on POU1F1 mutations are needed to determine whether the occurrence of prolactinoma is specific for the PIT-1β mutation.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this study.
Funding
The study was generously funded by JSPS KAKENHI Grant Number JP16H06608, the KOUEKIZAIDANNHOUJIN SUHARAKINENNZAIDANN, the Japanese Society for Pediatric Endocrinology Future Development Grant supported by Novo Nordisk Pharma Ltd and the Foundation for Growth Science in Japan.
Acknowledgements
The authors thank the present family for their cooperation with this publication, Drs Yasuhiko Okimura from Department of Nutrition and Food Science, Kobe Women’s University Graduate School of Life Sciences, Tomonobu Hasegawa and Masaki Takagi from Department of Pediatrics, Keio University School of Medicine for providing us with their plasmids, Mses. Nami Iguchi and Risa Taniguchi for technical support in molecular analysis, and the members for 'the next generation research meeting of east Japan pediatric endocrinology' group for a fruitful discussion.
References
- 1↑
Andersen B, Pearse 2nd RV, Jenne K, Sornson M, Lin SC, Bartke A & Rosenfeld MG The Ames dwarf gene is required for Pit-1 gene activation. Developmental Biology 1995 172 495–503. (https://doi.org/10.1006/dbio.1995.8040)
- 2↑
Andersen B & Rosenfeld MG POU domain factors in the neuroendocrine system: lessons from developmental biology provide insights into human disease. Endocrine Reviews 2001 22 2–35. (https://doi.org/10.1210/edrv.22.1.0421)
- 3↑
Ingraham HA, Flynn SE, Voss JW, Albert VR, Kapiloff MS, Wilson L & Rosenfeld MG The POU-specific domain of Pit-1 is essential for sequence-specific, high affinity DNA binding and DNA-dependent Pit-1-Pit-1 interactions. Cell 1990 61 1021–1033. (https://doi.org/10.1016/0092-8674(9090067-o)
- 4↑
Delhase M, Vila V, Hooghe-Peters EL & Castrillo JL A novel pituitary transcription factor is produced by alternative splicing of the human GHF-1/PIT-1 gene. Gene 1995 155 273–275. (https://doi.org/10.1016/0378-1119(9400757-j)
- 5↑
Morris AE, Kloss B, McChesney RE, Bancroft C & Chasin LA An alternatively spliced Pit-1 isoform altered in its ability to trans-activate. Nucleic Acids Research 1992 20 1355–1361. (https://doi.org/10.1093/nar/20.6.1355)
- 6↑
Theill LE, Hattori K, Lazzaro D, Castrillo JL & Karin M Differential splicing of the GHF1 primary transcript gives rise to two functionally distinct homeodomain proteins. EMBO Journal 1992 11 2261–2269. (https://doi.org/10.1002/j.1460-2075.1992.tb05285.x)
- 7↑
Wallis M Evolution of the POU1F1 transcription factor in mammals: rapid change of the alternatively-spliced beta-domain. General and Comparative Endocrinology 2018 260 100–106. (https://doi.org/10.1016/j.ygcen.2018.01.005)
- 8↑
Sporici RA, Hodskins JS, Locasto DM, Meszaros LB, Ferry AL, Weidner AM, Rinehart CA, Bailey JC, Mains IM & Diamond SE Repression of the prolactin promoter: a functional consequence of the heterodimerization between Pit-1 and Pit-1 beta. Journal of Molecular Endocrinology 2005 35 317–331. (https://doi.org/10.1677/jme.1.01678)
- 9↑
Jonsen MD, Duval DL & Gutierrez-Hartmann A The 26-amino acid beta-motif of the Pit-1beta transcription factor is a dominant and independent repressor domain. Molecular Endocrinology 2009 23 1371–1384. (https://doi.org/10.1210/me.2008-0137)
- 10↑
Diamond SE & Gutierrez-Hartmann A A 26-amino acid insertion domain defines a functional transcription switch motif in Pit-1beta. Journal of Biological Chemistry 1996 271 28925–28932. (https://doi.org/10.1074/jbc.271.46.28925)
- 11↑
Diamond SE & Gutierrez-Hartmann A The Pit-1beta domain dictates active repression and alteration of histone acetylation of the proximal prolactin promoter. Journal of Biological Chemistry 2000 275 30977–30986. (https://doi.org/10.1074/jbc.M006048200)
- 12↑
Sánchez-Pacheco A, Peña P, Palomino T, Güell A, Castrillo JL & Aranda A The transcription factor GHF-1, but not the splice variant GHF-2, cooperates with thyroid hormone and retinoic acid receptors to stimulate rat growth hormone gene expression. FEBS Letters 1998 422 103–107. (https://doi.org/10.1016/s0014-5793(9701609-8)
- 13↑
Giordano M Genetic causes of isolated and combined pituitary hormone deficiency. Best Practice and Research: Clinical Endocrinology and Metabolism 2016 30 679–691. (https://doi.org/10.1016/j.beem.2016.09.005)
- 14↑
Pfäffle R & Klammt J Pituitary transcription factors in the aetiology of combined pituitary hormone deficiency. Best Practice and Research: Clinical Endocrinology and Metabolism 2011 25 43–60. (https://doi.org/10.1016/j.beem.2010.10.014)
- 15↑
Baş F, Uyguner ZO, Darendeliler F, Aycan Z, Çetinkaya E, Berberoğlu M, Şiklar Z, Öcal G, Darcan Ş & Gökşen D Molecular analysis of PROP1, POU1F1, LHX3, and HESX1 in Turkish patients with combined pituitary hormone deficiency: a multicenter study. Endocrine 2015 49 479–491. (https://doi.org/10.1007/s12020-014-0498-1)
- 16↑
Birla S, Khadgawat R, Jyotsna VP, Jain V, Garg MK, Bhalla AS & Sharma A Identification of novel PROP1 and POU1F1 mutations in patients with combined pituitary hormone deficiency. Hormone and Metabolic Research 2016 48 822–827. (https://doi.org/10.1055/s-0042-117112)
- 17↑
Sobrier ML, Tsai YC, Pérez C, Leheup B, Bouceba T, Duquesnoy P, Copin B, Sizova D, Penzo A & Stanger BZ Functional characterization of a human POU1F1 mutation associated with isolated growth hormone deficiency: a novel etiology for IGHD. Human Molecular Genetics 2016 25 472–483. (https://doi.org/10.1093/hmg/ddv486)
- 18↑
Takagi M, Kamasaki H, Yagi H, Fukuzawa R, Narumi S & Hasegawa T A novel heterozygous intronic mutation in POU1F1 is associated with combined pituitary hormone deficiency. Endocrine Journal 2017 64 229–234. (https://doi.org/10.1507/endocrj.EJ16-0361)
- 19↑
Gat-Yablonski G, Frumkin-Ben David R, Bar M, Potievsky O, Phillip M & Lazar L Homozygous microdeletion of the POU1F1, CHMP2B, and VGLL3 genes in chromosome 3-a novel syndrome. American Journal of Medical Genetics: Part A 2011 155A 2242–2246. (https://doi.org/10.1002/ajmg.a.34136)
- 20↑
Pfaffle RW, DiMattia GE, Parks JS, Brown MR, Wit JM, Jansen M, Van der Nat H, Van den Brande JL, Rosenfeld MG & Ingraham HA Mutation of the POU-specific domain of Pit-1 and hypopituitarism without pituitary hypoplasia. Science 1992 257 1118–1121. (https://doi.org/10.1126/science.257.5073.1118)
- 21↑
Carlomagno Y, Salerno M, Vivenza D, Capalbo D, Godi M, Mellone S, Tiradani L, Corneli G, Momigliano-Richiardi P & Bona G A novel recessive splicing mutation in the POU1F1 gene causing combined pituitary hormone deficiency. Journal of Endocrinological Investigation 2009 32 653–658. (https://doi.org/10.1007/BF03345736)
- 22↑
Tenenbaum-Rakover Y, Sobrier ML & Amselem S A novel POU1F1 mutation (p.Thr168IlefsX7) associated with an early and severe form of combined pituitary hormone deficiency: Functional analysis and follow-up from infancy to adulthood. Clinical Endocrinology 2011 75 214–219. (https://doi.org/10.1111/j.1365-2265.2011.04028.x)
- 23↑
Gat-Yablonski G, Klar A, Hirsch D, Eliakim A, Lazar L, Hurvitz H & Phillip M Three novel mutations in POU1F1 in Israeli patients with combined pituitary hormone deficiency. Journal of Pediatric Endocrinology and Metabolism 2005 18 385–393. (https://doi.org/10.1515/jpem.2005.18.4.385)
- 24↑
Hendriks-Stegeman BI, Augustijn KD, Bakker B, Holthuizen P, van der Vliet PC & Jansen M Combined pituitary hormone deficiency caused by compound heterozygosity for two novel mutations in the POU domain of the Pit1/POU1F1 gene. Journal of Clinical Endocrinology and Metabolism 2001 86 1545–1550. (https://doi.org/10.1210/jcem.86.4.7371)
- 25↑
Inoue H, Mukai T, Sakamoto Y, Kimura C, Kangawa N, Itakura M, Ogata T, Ito Y, Fujieda KJapan Growth Genome Consortium. Identification of a novel mutation in the exon 2 splice donor site of the POU1F1/PIT-1 gene in Japanese identical twins with mild combined pituitary hormone deficiency. Clinical Endocrinology 2012 76 78–87. (https://doi.org/10.1111/j.1365-2265.2011.04165.x)
- 26↑
Turton JP, Strom M, Langham S, Dattani MT & Le Tissier P Two novel mutations in the POU1F1 gene generate null alleles through different mechanisms leading to combined pituitary hormone deficiency. Clinical Endocrinology 2012 76 387–393. (https://doi.org/10.1111/j.1365-2265.2011.04236.x)
- 27↑
Kishimoto M, Okimura Y, Yagita K, Iguchi G, Fumoto M, Iida K, Kaji H, Okamura H & Chihara K Novel function of the transactivation domain of a pituitary-specific transcription factor, Pit-1. Journal of Biological Chemistry 2002 277 45141–45148. (https://doi.org/10.1074/jbc.M202991200)
- 28↑
Shih C & Weinberg RA Isolation of a transforming sequence from a human bladder carcinoma cell line. Cell 1982 29 161–169. (https://doi.org/10.1016/0092-8674(8290100-3)
- 29↑
Konzak KE & Moore DD Functional isoforms of Pit-1 generated by alternative messenger RNA splicing. Molecular Endocrinology 1992 6 241–247. (https://doi.org/10.1210/mend.6.2.1569967)
- 30↑
Desmet FO, Hamroun D, Lalande M, Collod-Béroud G, Claustres M & Béroud C Human Splicing Finder: an online bioinformatics tool to predict splicing signals. Nucleic Acids Research 2009 37 e67. (https://doi.org/10.1093/nar/gkp215)
- 31↑
Rosenberg AB, Patwardhan RP, Shendure J & Seelig G Learning the sequence determinants of alternative splicing from millions of random sequences. Cell 2015 163 698–711. (https://doi.org/10.1016/j.cell.2015.09.054)
- 32↑
Gat-Yablonski G, Lazar L, Pertzelan A & Phillip M A novel mutation in PIT-1: phenotypic variability in familial combined pituitary hormone deficiencies. Journal of Pediatric Endocrinology and Metabolism 2002 15 325–330. (https://doi.org/10.1515/jpem.2002.15.3.325)
- 33↑
Salemi S, Besson A, Eble A, Gallati S, Pfaffle RW & Mullis PE New N-terminal located mutation (Q4ter) within the POU1F1-gene (PIT-1) causes recessive combined pituitary hormone deficiency and variable phenotype. Growth Hormone and IGF Research 2003 13 264–268. (https://doi.org/10.1016/s1096-6374(0300015-7)
- 34↑
Turton JP, Reynaud R, Mehta A, Torpiano J, Saveanu A, Woods KS, Tiulpakov A, Zdravkovic V, Hamilton J & Attard-Montalto S Novel mutations within the POU1F1 gene associated with variable combined pituitary hormone deficiency. Journal of Clinical Endocrinology and Metabolism 2005 90 4762–4770. (https://doi.org/10.1210/jc.2005-0570)
- 35↑
de Moraes DC, Vaisman M, Conceição FL & Ortiga-Carvalho TM Pituitary development: a complex, temporal regulated process dependent on specific transcriptional factors. Journal of Endocrinology 2012 215 239–245. (https://doi.org/10.1530/JOE-12-0229)
- 36↑
Persani L, Brabant G, Dattani M, Bonomi M, Feldt-Rasmussen U, Fliers E, Gruters A, Maiter D, Schoenmakers N & van Trotsenburg ASP 2018 European Thyroid Association (ETA) guidelines on the diagnosis and management of central hypothyroidism. European Thyroid Journal 2018 7 225–237. (https://doi.org/10.1159/000491388)
- 37↑
Beck-Peccoz P, Rodari G, Giavoli C & Lania A Central hypothyroidism - a neglected thyroid disorder. Nature Reviews: Endocrinology 2017 13 588–598. (https://doi.org/10.1038/nrendo.2017.47)
- 38↑
Hartoft-Nielsen ML, Lange M, Rasmussen AK, Scherer S, Zimmermann-Belsing T & Feldt-Rasmussen U Thyrotropin-releasing hormone stimulation test in patients with pituitary pathology. Hormone Research 2004 61 53–57. (https://doi.org/10.1159/000075239)
- 39↑
Hamada K, Nishi T, Kuratsu J & Ushio Y Expression and alternative splicing of Pit-1 messenger ribonucleic acid in pituitary adenomas. Neurosurgery 1996 38 362–366. (https://doi.org/10.1097/00006123-199602000-00026)
- 40↑
Sanno N, Teramoto A, Matsuno A, Takekoshi S, Itoh J & Osamura RY Expression of Pit-1 and estrogen receptor messenger RNA in prolactin-producing pituitary adenomas. Modern Pathology 1996 9 526–533.
