Familial Central Hypothyroidism Caused by a Novel IGSF1 Gene Mutation

Abstract

Background:

Congenital hypothyroidism of central origin (CH-C) is a rare disease in which thyroid hormone deficiency is caused by insufficient thyrotropin stimulation of a normal thyroid gland. A recently described syndrome of isolated CH-C and macroorchidism was attributed to loss-of-function mutations of the immunoglobulin superfamily, member 1 gene (IGSF1).

Patients and Methods:

CH-C was diagnosed in three siblings. The TRH, TRHR, and TSHB genes were sequenced followed by whole-exome sequencing in the proband. A mutation identified in IGSF1 was analyzed by direct PCR sequencing in family members. The effects of the mutation were assessed by in vitro studies in HEK293 cells.

Results:

The index case was negative for mutations in TRH, TRHR, and TSHB. Whole-exome sequencing revealed a novel insertion mutation in IGSF1, c.2284_2285insA, p.R762QfsX7, which was confirmed by direct PCR sequencing and was identified in six additional family members. The mutation introduces a frame-shift and premature stop codon in the seventh Ig loop, thereby truncating IGSF1. In vitro studies revealed that the mutated IGSF1–R762QfsX7 migrates as a doublet at ∼28 kDa, which is far smaller than the wild type protein (130–140 kDa). Both bands were endonuclease H sensitive, indicating immature glycosylation and failure of the protein to traffic out of the endoplasmic reticulum to the plasma membrane. Further phenotypic findings in the family included macroorchidism and infertility in the uncle and mild neurological phenotypes in the affected males, such as hypotonia, delayed psychomotor development, clumsy behavior, and attention deficit disorder.

Conclusions:

We identified a novel insertion mutation in the IGSF1 gene and further delineated the phenotype of the IGSF1-deficiency syndrome. Our findings indicate a possible association between an IGSF1 mutation and neurological phenotypes.

Introduction

Congenital hypothyroidism of central origin (CH-C) is a rare disease in which thyroid hormone deficiency is caused by insufficient thyrotropin (TSH) stimulation of a normal thyroid gland. Estimates of CH-C incidence range from about 1:16,000 in the Netherlands (1) to 1:65,000 in the United States (2). This situation can derive from mutations in genes involved in pituitary development, such as POU1F1, PROP1, HESX1, LHX3, LHX4, and SOX3 (3,4). In these cases, additional pituitary hormones are also deficient. Isolated CH-C, in contrast, derives from mutations in genes specific to the hypothalamic–pituitary–thyroid axis, such as the thyrotropin-releasing hormone (TRH) receptor (TRHR) and TSH beta (TSHB). A mutation in the TRHR (Online Mendelian Inheritance in Man [OMIM] 188545) was first reported in 1997 (5), and only one additional case has been reported since (6). Mutations of the TSHB gene (OMIM 188540) are associated with a variable phenotypic spectrum from mild to severe congenital hypothyroidism (CH) (7,8). Targeted deletion of Trh in mice (OMIM 613879) causes hypothyroidism and hyperglycemia; no human TRH mutations have been reported to date (9).

Loss-of-function mutations of the immunoglobulin (Ig) superfamily, member 1 (IGSF1; OMIM 300888) were recently described and linked to a novel syndrome of isolated CH-C and macroorchidism (10). The IGSF1 gene is located on the X chromosome (Xq26.1) and encodes a plasma membrane glycoprotein that is mainly expressed in the pituitary, brain, and testis (11). The IGSF1 protein has 12 Ig-like loops of the C2 (constant region type 2) type, a transmembrane domain and a short intracellular cytosolic tail. Only the C-terminal domain of the protein, containing seven Ig loops, traffics to the plasma membrane (12,13). Many members of the Ig superfamily are cell-adhesion molecules that regulate signaling via cell-surface receptors and are involved in the immune and hematological systems as well as the development of the central nervous system and adult neural cell functions (14). The exact roles of IGSF1 in pituitary hormone secretion and testicular function remain undefined.

The main characteristics of the IGSF1 deficiency syndrome include central hypothyroidism, macroorchidism, delayed pubertal testosterone secretion, increased body mass index (BMI) and variable prolactin and growth hormone (GH) deficiencies (15,16). Sun et al. (10) described 26 patients with 10 distinct IGSF1 mutations (10). Since that first report, an additional five mutations have been described in Japanese patients (17 –19) and in one male patient carrying a Xq26 deletion encompassing the IGSF1 gene (20).

Here, we describe a family with three siblings diagnosed with isolated CH-C. Using whole-exome sequencing (WES), we identified a novel insertion mutation in the IGSF1 gene, c.2284_2285insA, p.R762QfsX7 in the index case, in his two brothers and in four family members. We further delineate the phenotype of this rare condition, expressed the mutant protein in vitro, and correlated genotype and phenotype in the family.

Patients and Methods

Patients

Blood samples for hormonal and DNA analyses were collected from all participants. Medical history and laboratory data were collected from the family members' computerized medical files. The TRH, TRHR, and TSHB coding regions were sequenced (see below). Because no point mutations were found in these genes, we further performed WES on the index case (patient III-1). After identification of a novel IGSF1 variant, we recruited eight family members and analyzed the identified IGSF1 gene mutation by direct sequencing of PCR amplicons. The study was approved by the Ha'Emek Medical Center Ethics Committee and the National Helsinki Committee for Genetic Studies. Informed consent was obtained from the study participants.

Israeli national neonatal screening for CH

Blood samples are routinely collected by heel puncture 48–72 hours after birth. Neonatal thyroxine (T4) function is measured by PerkinElmer B065-112 AutoDELFIA total T4 (TT4) and B032-312 AutoDELFIA (TSH), both using time-resolved fluoroimmunoassays. The Israeli screening program is based on a TT4 test followed by a confirmatory TSH test. Therefore, when the level of TT4 is below the daily 10th percentile, TSH is measured. TSH values above 20 mIU/L are considered indicative of primary CH. Neonates with abnormal screening results are referred to medical centers.

Sequencing of TRH, TRHR, and TSHB

Genomic DNA was isolated from blood leukocytes using MasterPure™ DNA Purification Kit for Blood Version II from Epicenter (Illumina) according to the manufacturer's instructions. The coding sequences of TSHB (exons 2 and 3), TRH (exons 1 to 3), and TRHR (exons 1 and 2) were amplified by PCR using 20 ng genomic DNA as the template supplemented with 1 U Taq polymerase (Axon Labortechnik), 1 × PCR buffer and 1.5 mM magnesium chloride (Axon Labortechnick), 10 μM dNTPs (Promega Corporation), 1 M betaine (Sigma-Aldrich Inc.), and 0.4 μM each of forward and reverse primers (Supplementary Table S1; Supplementary Data are available online at www.liebertpub.com/thy). PCR products were purified from primers with Exonuclease I (New England Biolabs), treated with shrimp alkaline phosphatase (GE Healthcare), and then sequenced using the Big Dye Terminator v.3.1 kit (Applied Biosystems). The ethanol-precipitated sequencing products were loaded on a 3130 Avant Genetic Analyzer and analyzed with Sequencher software v.5 (Gene Codes,). Sequencing was done in the Johannes Gutenberg University Medical School, Mainz, Germany.

Whole-exome sequencing

Genomic DNA was extracted from peripheral blood mononuclear cells. WES was used to search for candidate variants in the index case (in a recessive model). Genomic DNA libraries were created following standard protocols. Libraries were enriched using a Nextera Exome enrichment kit (62 Mbp; Illumina) and sequenced on a HiSeq 2500 sequencer (2 × 170 bp pair ends run) at an 83 × mean coverage. Reads were aligned to the human genome (hg19; National Center for Biotechnology Information build 37; February 2009). Exome-filtering criteria for both families are summarized in Supplementary Table S2. Sanger sequencing was used to validate all coding and splicing variants in the family that were homo/hemizygous and had a positive logarithm (base 10) of odds score, and were absent or found at very low frequencies in dbSNP, Exome Variant Server (http://evs.gs.washington.edu/EVS/), and 1000 Genomes (see Supplementary Table S2). WES and IGSF1 sequencing were performed in the Galil Genetic Analysis laboratory, Israel.

Sequence analysis of IGSF1

Exon and exon–intron boundaries of IGSF1 were amplified from genomic DNA by PCR using previously reported primers (17). PCR products were sequenced directly with an ABI 373A automated fluorescent sequencer (PE Applied Biosystems). Sanger sequencing was performed for the identified mutation in all subjects. PCR products were sequenced using BigDye Terminator v.3.1. Sequence analysis was performed on an Applied Biosystems 3130xl Genetic Analyzer.

In vitro analysis of mutant IGSF1 protein

Human embryonic kidney (HEK) 293 cells were cultured as described by Sun et al. (10). QuikChange mutagenesis was performed using primers (Forward: 5′-CTC ACA CTG AAA AAA CGC CCC TTC AAG TG-3′; Reverse: 5′-CAC TTG AAG GGG CGT TTT TTC AGT GTG AG-3′) to introduce the 2284_2285insA mutation into the previously described wild type IGSF1-HA expression vector (12). Cells were seeded in six-well plates and transfected with 2 μg of plasmid DNA using 6 μg polyethylenimine. Total protein lysates were prepared 24 hours post-transfection using immunoprecipitation lysis buffer (50 mM Tris pH 7.5, 150 mM sodium chloride, 1 mM ethylenediaminetetraacetic acid, 1% v/v Triton X-100). Peptide-N-Glycosidase F and endonuclease H (EndoH) treatments as well as protein Western blotting were performed as described previously (10).

Results

Patient reports

Index case (patient III-1)

The index case was a two-week-old male born to unrelated parents of Ashkenazi Jewish decent. Following a prenatal ultrasound scan that showed a short femur, amniocentesis revealed a normal male karyotype 46, XY. He was born by spontaneous delivery at 38 weeks gestational age with a weight of 2770 g. After birth, bradycardia was observed but cardiac echography demonstrated a normal heart. At the age of 2 weeks, he was referred to our hospital due to prolonged neonatal jaundice (20 mg/dL bilirubin) with elevated liver enzymes (aspartate aminotransferase 104 U/L, alanine aminotransferase 64 U/L, lactic dehydrogenase 965 U/L, alkaline phosphatase 965 U/L). As part of the investigation of the prolonged neonatal jaundice, thyroid functions were checked, revealing a low free T4 (FT4) of 0.47 ng/dL (normal infant range, 0.83–2.0 ng/dL) with a normal TSH of 2.9 mIU/L (normal infant range, 0.5–10 mIU/L) (Table 1). The results of the neonatal screening after the initial diagnosis revealed low TT4 with low TSH (Table 1), consistent with the diagnosis of CH-C. On examination, he was severely hypotonic with severe head lag. Levothyroxine (LT4) therapy was initiated at the age of 3 weeks and the dose was adjusted according to serum FT4 levels (TSH 0.02 mIU/L, FT4 1.72 ng/dL at 3 months). During his first 2 years, the parents discontinued medication for a few months at a time because they deemed it unnecessary. At the age of 2 years, he underwent a technetium-99m scan after L-T4 withdrawal that demonstrated a small gland in normal position. Thyroid function revealed a moderately decreased FT4 (0.76 ng/dL) with normal TSH and triiodothyronine (FT3) concentrations (2.5 mIU/L and 3.7 pg/mL respectively) and a normal thyroglobulin (54 ng/mL; normal range 0–55), confirming the diagnosis of CH-C. A TRH stimulation test showed a basal TSH of 2.5 mIU/L and peak TSH at 13.5 mIU/L, consistent with a normal TRH response (21). An adrenocorticotropic hormone stimulation test (250 μg/m²) revealed a normal cortisol response (peak cortisol >50 μg/dL). Insulin-like growth factor 1 (IGF-1) levels were very low (<25 ng/mL) in repeated measurements; however, he was growing well in the 25th percentile for height. Therefore, GH stimulation tests have not been performed. Milestones for motor and speech development were delayed. He had poor coordination, a tendency to fall, and his speech was heavy and unclear. At his last visit, aged 8.3 years, he was 128 cm (-0.38 SD) in height and 31 kg (1.09 SD) in weight, and his BMI was 18.9 kg/m² (1.43 SD). He had normal cognitive achievements but a clumsy behavior and writing difficulties.

Table 1.

Patient Clinical Data

Patient	III-1	III-3	III-5	II-2	Normal values ^a
Age at presentation	2 weeks	7 days	3 days	27.75 years
Prenatal history	US – short femur	Normal	Polyhydramnion	NA
Presenting symptom	Neonatal jaundice	None	Neonatal jaundice	None
Physical examination	Head lag, hypotonia, large anterior fontanel	Head lag	Head lag	Normal
Identified by neonatal screening	Yes	Yes	No	Yes
Neonatal screening
Total T4 (μg/dL)	2.8	4.74	6.3	–	<10th centile^b
TSH (mIU/L)	<10	7.96	11	—	<20 mIU/L
Laboratory thyroid function
Age	2 weeks	7 days	2 months	27 years
TSH (mIU/L)	2.9	9.77	2.68	1.02	0.4–4.2
Free T4 (ng/dL)	0.47	0.7	0.7	0.93	0.78–1.6
Free T3 (pg/mL)	—	3.2	3.3	2.9	2.3–4.2
Thyroglobulin (ng/mL)	—	—	32.2	ND	0–55
Age at LT4 initiation	3 weeks	11 days	6 weeks	None
Prolactin (ng/mL)	9.66	1.74	7.12	1.01	2.1–17.7
LH (mIU/mL)	<0.07	<0.07	<0.07	NA
Follicle-stimulating hormone (mIU/mL)	2.12	0.63	0.74	NA
IGF-1 (ng/mL)	<25 (63–379)	36.2 (41–207)	23.3 (19–109)	208.7 (93–250)
⁹⁹TC scan	Small gland in normal location with heterogeneous uptake	–	–	–
Others	Hypotonia, psychomotor delay, clumsiness	Psychomotor and speech delay, attention deficit disorder, clumsiness	None	Diabetes mellitus type 1 attention deficit disorder, clumsiness
Age at last visit	8.3 years	4.11 years	10 months	27.75 years
Height (SD)	−0.24	−0.46	0.06	0.55
Weight (SD)	0.68	0.6	−0.59	1.35
Body mass index (SD)	1.43	1.65	−1.05	1.42
Testicular volume (mL)	2	2	2	50

Dashes indicate no value for the patient.

Adult normal values.

The Israeli screening program is total T4-based; when the level of total T4 is below the daily 10th percentile, TSH is measured.

Normal FT4 levels for infant aged 1–30 days, 0.83–2.0 ng/dL.

Normal TSH levels for infant aged 1–30 days, 0.5–10 mIU/L.

NA, not available; ND, not done; T3, triiodothyronine; T4, thyroxine; TSH, thyrotropin; LH, Luteinizing hormone; ⁹⁹TC, technetium-99m; IGF, insulin-like growth factor; SD, standard deviation.

Patient III-3 (brother of the proband)

Patient III-3 was an infant male born following a normal pregnancy, weighing 3160 g at term. Thyroid function tests were performed at the age of 7 days due to his brother's CH and revealed low FT4 with normal TSH and FT3 levels (Table 1). The results of the neonatal screening that were reported after his first visit revealed a borderline low TT4 with a normal TSH (Table 1). On examination, he had severe head lag without any other findings. LT4 medication was initiated at the age of 11 days. He was hypotonic and exhibited delayed motor and speech development. Clumsy behavior, speech problems, and attention deficit disorder (ADD) were diagnosed at the age of 4 years. Although IGF-1 levels were low, GH stimulating tests were not undertaken since he was growing well. At his last visit at age 4.11 years, his height was 104 cm (-0.35 SD), weight 19 kg (0.61 SD), and BMI 17.57 kg/m² (1.54 SD).

Patient III-5 (brother of the proband)

Patient III-5 was a 3-day-old male born following a pregnancy with polyhydramnios, weighing 4030 g at 42 weeks gestational age. Neonatal screening had shown normal TT4 and TSH levels (Table 1), and thyroid function at the age of 2 days was also within the reference range for his age (TSH = 15.7 mIU/L, FT4 = 0.84 ng/dL). Repeated thyroid function tests at the age of 7 days were also within the reference range (TSH = 3.61 mIU/L, FT4 = 0.99 ng/dL, and FT3 = 3.3 pg/mL). On examination, he had severe head lag and neonatal jaundice without other signs of hypothyroidism. At the age of 6 weeks, repeated thyroid function tests revealed a borderline low FT4 with normal serum levels of TSH and FT3 (Table 1); he was diagnosed with CH-C and LT4 therapy was initiated. His IGF-1 level was low, but he had normal growth. He had normal milestones but a clumsy motor behavior. At his last visit at age 10 months, his height was 72.5 cm (0.06 SD) and his weight 8.5 kg (-0.6 SD).

Patient II-2 (uncle of the index case)

A 27-year-old male was evaluated at our clinic due to his nephew's familial CH. After birth, he had been diagnosed by the national neonatal screening with CH (data not available); however, he never received supplemental LT4 therapy. In an evaluation of infertility, a sperm analysis revealed pathological morphology (98%; normal <96%) and decreased sperm mobility (25%; normal >40%). Twin pregnancy was attained after intracervical insemination. After the birth of his sons, at the age of 27 years, he was diagnosed with diabetes mellitus type 1 (glutamic decarboxylase antibodies positive) and insulin therapy was required. His height was 178 cm (0.5 SD), weight 78 kg (-0.04 SD), and BMI 25 kg/m². Enlarged testes of 50 mL (>25 mL by Prader orchidometer) were found. Thyroid function was within the reference range but FT4 was in the lower normal range without taking any medication; his serum prolactin was low (Table 1). He had been diagnosed with ADD in childhood but attained normal intellectual achievements. He reported clumsy motor behavior.

Family history

The mother and father were both healthy and had normal thyroid function tests (Fig. 1). Parental height was 183 cm and maternal height 164 cm. In addition to the three affected sons, they had two daughters with normal psychomotor development and normal thyroid function. The maternal grandmother had had a toxic multinodular goiter and had been treated with radioactive iodine, resulting in acquired hypothyroidism. Another uncle (II-1) was diagnosed with hypothyroidism at the age of three months, but he was never treated. He had significant psychomotor delay at infancy and as an adult he had clumsy motor behavior, as well as speech problems. He declined to participate in the study.

FIG. 1.

Pedigree of the Israeli family with the novel IGSF1 gene mutation. The age (in years) indicates the age of each subject at the time of analysis. In patients III-1, III-3, and III-5, the data represent thyroid function prior to levothyroxine (LT4) therapy; in the other family members, the data were collected at the time of the study. Patient I-1 was on LT4 therapy. The shading of symbols indicates either a heterozygous or hemizygous status for the p.R762QfsX7 insertion mutation. Normal free T4 levels for infants aged 1–30 days is 0.83–2.0 ng/dL. Normal thyrotropin (TSH) levels for infants aged 1–30 days is 0.5–10 mIU/L.

Genetic results

The index case was negative for mutations in the TRH, TRHR, and TSHB genes. Using WES, we identified a total of 25,100 variants, which were filtered according to various criteria (Supplementary Table S2). From the eight remaining variants, only the IGSF1 mutation segregated with the abnormal phenotype in the family. Sanger sequencing confirmed the presence of the mutation. A novel mutation, c.2284_2285insA, p.R762QfsX7, was identified in IGSF1 in the proband (Fig. 2). IGSF1 is located on the X chromosome and the mutation was hemizygous in the three affected siblings and the maternal uncle and heterozygous in the mother, one sister, and the maternal grandmother (Fig. 1).

FIG. 2.

Sequence of the IGSF1 gene in (A) a wild type (WT) male and (B) the heterozygous mother of the proband. (C) Hemizygous mutated immunoglobulin (Ig) superfamily, member 1 gene (IGSF1) in the index case. The c.2284_2285insA, p.R762QfsX7 insertion mutation introduces a frame shift.

We next assessed the pathogenicity of the identified mutation by expressing the p.R762QfsX7 mutant in heterologous HEK293 cells. The wild type IGSF1 migrates as a doublet on an sodium dodecyl sulfate–polyacrylamide gel (Fig. 3A, lane 2). As previously reported (10), the higher molecular weight band reflects the mature (EndoH-resistant, lane 3) glycoform of the protein that traffics to the plasma membrane, whereas the lower molecular weight band represents the immature (EndoH-sensitive) glycoform that is retained in the endoplasmic reticulum (ER) (10). The identified novel mutation introduces a frame shift and a premature stop codon in the seventh Ig loop (the second Ig loop of the C-terminal domain), truncating the IGSF1 protein. When examined by Western blot, IGSF1–p.R762QfsX7 migrated as a low abundance doublet at ∼28 kDa (Fig. 3B, lane 2), which is far smaller than the wild type protein at ∼130–140 kDa. In contrast to the wild type, both bands of the IGSF1–p.R762QfsX7 doublet were EndoH sensitive (Fig. 3B, lane 3), indicating immature glycosylation and failure of the protein to traffic out of the ER to the plasma membrane. Similar results have been reported with other truncation mutants (10). The generated protein is small and unstable, with impaired transport out of the ER.

FIG. 3.

Protein analysis of wild type (A) and mutant IGSF1 (B). HEK293 cells were transfected with pcDNA3.0 or with expression vectors for HA-tagged WT or mutant (Arg762ThrfsX7) forms of IGSF1. Protein lysates were deglycosylated with EndoH or PNGaseF, resolved by sodium dodecyl sulfate polyacrylamide gel electrophoresis, and immunoblotted with an HA antibody (A) for the lysates containing the wild type protein, and an IGSF1-C-terminal domain antibody (B) for the lysates containing the mutant protein (as the C-terminal HA tags are lost as a result of the frame-shift mutation).

Discussion

In three siblings with CH-C and four other family members, we identified a novel insertion mutation, c.2284_2285insA, p.R762QfsX7, which introduces a frame shift and thus a premature stop codon in the seventh Ig loop of IGSF1. Sun et al. (10) identified 10 distinct mutations in the IGSF1 gene in males from 11 unrelated European families presenting with central hypothyroidism and testicular enlargement. An additional five novel IGSF1 mutations were described in Japanese patients presenting with either CH-C or failure to thrive (17 –19). In all, 15 different mutations have been reported to date in IGSF1, including 6 nonsense, 5 missense, 1 insertion, and 3 deletion mutations. The main initially described phenotypic characteristics of patients with IGSF1 deficiency include central hypothyroidism and macroorchidism, but further clinical characterization of these patients also revealed hypoprolactinemia, variable GH deficiency, delayed pubertal testosterone secretion with normal timing of testicular development and increased body weight (16). Since IGSF1 is located on the X chromosome, the affected patients are males, but a subset of female carriers also exhibit mild central hypothyroidism (16). In the family reported here, all affected males had CH-C but only three were treated with L-T4. Two had hypoprolactinemia (prolactin levels below the lower normal range) and one had macroorchidism. In addition, male carriers of the mutation exhibited hypotonia, delayed psychomotor development, motor and speech clumsiness, and ADD. None of the affected females had central hypothyroidism, but the maternal grandmother had had a toxic multinodular goiter.

The exact frequency of IGSF1 mutations is not known, but a frequency of approximately 1:100,000 has been predicted (15). Adult male Igsf1-deficient mice exhibit reduced pituitary and serum TSH and reduced serum FT3 and/or FT4 concentrations, likely due to impaired TRH signaling (10). The clinical phenotype of IGSF1 deficiency in humans with respect to thyroid function is similar to that of mice lacking Igsf1.

CH-C is only detected by TT4-based neonatal screening and is missed by TSH-based screening. Since the Israeli national newborn screening program for CH is based on a TT4 test followed by a confirmatory TSH test, these cases are identified by the screening program, whereas in TSH-based screening they are missed and the diagnosis is made later in infancy or childhood following evaluation of prolonged neonatal jaundice, failure to thrive, short stature, or delayed psychomotor development. Furthermore, some patients are diagnosed with CH-C only after familial genetic analysis in the fifth to eighth decade with no clinical symptoms of hypothyroidism (10,16). Interestingly, while patients III-1 and III-3 were identified by the national screening program, in the third sibling (III-5) the TT4 screening levels were within the 10th percentile of the normal range. Only at the age of six weeks did laboratory FT4 levels decrease below the reference range, indicating that the hypothyroidism developed over time. These findings demonstrate variation in the extent of hypothyroidism within a family. Moreover, despite the fact that he was never treated with supplemental LT4, the affected uncle (patient II-2) had a normal FT4, albeit in the lower normal range, at the age of 27 years, suggesting that thyroid hormone secretion was maintained over the years. This finding indicates that some individuals do not require supplemental therapy for their entire lives, and therefore, discontinuation of hormonal therapy to assess reversibility of hypothyroidism is recommended. Similarly, in the first report of IGSF1 mutations in humans by Sun et al. (10), four patients aged over 50 years (out of 26 affected patients) were not treated with LT4.

IGSF1 has 12 Ig-like C2-type loops, a transmembrane domain, and a short intracellular cytosolic tail. The protein is cleaved into N- and C-terminal domains and only the latter traffics to the plasma membrane (12,13). Using in vitro studies in HEK293 cells, we show that the identified p.R762QfsX7 mutation causes a premature truncation of the IGSF1 C-terminal domain, which does not acquire complex carbohydrates. This indicates that the protein does not traffic from the ER through the Golgi to the plasma membrane. Similar results have been reported with other IGSF1 truncation mutations (10,17,18). Indeed, all of the intragenic mutations identified to date, apart from one (19), have been located in the C-terminal domain, indicating its important role in IGSF1 function.

IGSF1 is expressed in rodent pituitary thyrotrophs, lactotrophs, and somatotrophs (10,11). Consistent with this pattern of expression, the phenotype of IGSF1 deficiency in humans includes central hypothyroidism and variable prolactin and GH deficiencies (16). In our family, all affected males had normal growth patterns and the adult male was within the normal range for adult height, indicating a lesser prominence of GH deficiency in this syndrome. Hypoprolactinemia has been reported in about 70% of male patients (10,16) and in two of our patients, but its clinical significance in males is still undefined. Mice lacking Igsf1 are fertile and have normal testicular volumes (10). IGSF1-deficient males exhibit macroorchidism but are fertile and have normal gonadotropin and testosterone levels (10,16). Pubertal development has been reported to be disharmonious, with testicular growth starting at a normal age, but testosterone secretion, pubertal spurt and the development of pubic hair being delayed (16). It is not clear when testicular enlargement develops; however, enlarged testes have been observed at as early as 11 years of age. In the family described here, all three young (prepubertal) siblings had normal testicular volumes, whereas patient II-2, who was 27 years old, exhibited macroorchidism, abnormal sperm motility and increased pathological morphology. Since only one adult male exhibited infertility in this family, other etiologies for abnormal spermatogenesis cannot be excluded.

It was initially hypothesized that IGSF1 acts as an inhibin co-receptor, and the consequence of loss of inhibin action was expected to result in enhanced follicle-stimulating hormone (FSH) secretion; however, further studies challenged this hypothesis (22). The association between hypothyroidism and macroorchidism has been reported in the past in males with prolonged untreated hypothyroidism (23 –25). It has been hypothesized that macroorchidism results from increased FSH stimulated by hypothalamic TRH, which in turn stimulates the proliferation of Sertoli cells (which determines testis size) (25). Interestingly, the increase in basal and pulsatile 24 hour FSH in IGSF1 deficient males compared with controls might at least partially explain the macroorchidism (26). Furthermore, T₃ itself has direct effects on Sertoli cell proliferation (27,28). Our finding is supported by a study showing that IGSF1 is expressed in elongating spermatids in rats (11), and by the fact that the other members of the Ig superfamily are expressed at the membrane of germ cells and spermatozoa, where they play roles in cell adhesion and sperm–egg interactions (11). In other reports of adult males with macroorchidism, sperm analysis was not reported.

Our findings of hypotonia, delayed motor development, ADD, and clumsiness (without cognitive impairment) in the IGSF1-deficient males may point to a role for IGSF1 in neural cell functions. Indeed, the neural cell adhesion molecules in the Ig superfamily have previously been implicated in the development of the central nervous system, facilitating neural cell migration, axon guidance, and synapse formation (14). In adult life, they maintain synaptic connections, cell–cell contacts, neuron–glial interactions, and synaptic plasticity (14). Dysfunction of these cells has been linked to psychiatric conditions as well as to Alzheimer's disease and autism (14). We have shown an association between mutations in IGSF1 and subtle neurological symptoms in affected males. It might be assumed that this subset of neurological symptoms is due to delayed initiation of LT4 therapy; however, the fact that therapy was initiated early in all three siblings (between two and six weeks), argues against this possibility. Indeed, it was shown recently, using neuropsychological assessment, that adult male patients with IGSF1 deficiency exhibit mild deficits in attentional control, whereas other cognitive tests were unaffected (29). However, it is possible that the minor neurological signs in the reported affected males are attributed to other reasons in the family reported here.

In conclusion, we identified a novel mutation in IGSF1 and further delineate the phenotype of the IGSF1-deficiency syndrome. The data presented here suggest an association between IGSF1 mutations and neurological phenotypes.

Footnotes

Acknowledgments

We thank Camille Vainstein for professional language editing and the families for participating in and providing samples for this study. This work was supported by a grant from the Rashut Lepituach HaGalil and the Negev (Y.T.-R.) and by operating grants from the Canadian Institutes of Health Research (MOP-133557) and Natural Sciences and Engineering Research Council of Canada (2015-05178) to D.J.B. M.-O.T. was supported by fellowships from Réseau Québécois en reproduction and Natural Sciences and Engineering Research Council of Canada.

Author Disclosure Statement

The authors have nothing to disclose.

References

Lanting

, van Tijn

, Loeber

, Vulsma

, de Vijlder

, Verkerk

. 2005. Clinical effectiveness and cost-effectiveness of the use of the thyroxine/thyroxine-binding globulin ratio to detect congenital hypothyroidism of thyroidal and central origin in a neonatal screening program. Pediatrics, 116:168–173.

LaFranchi

. 2010. Newborn screening strategies for congenital hypothyroidism: an update. J Inherit Metab Dis, 33:S225–S233.

Nettore

, Fenzi

, Macchia

. 2000. Genetic defects in thyroid hormone supply. In: De Groot

, Beck-Peccoz

, Chrousos

, Dungan

, Grossman

, Hershman

, Koch

, McLachlan

, New

, Rebar

, Singer

, Vinik

, Weickert

(eds) Endotext [Internet]. MD Text.com, Inc., South Dartmouth, MA. Available at: www.ncbi.nlm.nih.gov/books/NBK279158/ (accessed December 16, 2014 ).

Schoenmakers

, Alatzoglou

, Chatterjee

, Dattani

. 2015. Recent advances in central congenital hypothyroidism. J Endocrinol, 227:R51–71.

Collu

, Tang

, Castagné

, Lagacé

, Masson

, Huot

, Deal

, Delvin

, Faccenda

, Eidne

, Van Vliet

. 1997. A novel mechanism for isolated central hypothyroidism: inactivating mutations in the thyrotropin-releasing hormone receptor gene. J Clin Endocrinol Metab, 82:1561–1565.

Bonomi

, Busnelli

, Beck-Peccoz

, Costanzo

, Antonica

, Dolci

, Pilotta

, Buzi

, Persani

. 2009. A family with complete resistance to thyrotropin-releasing hormone. N Engl J Med, 360:731–734.

Partsch

, Riepe

, Krone

, Sippell

, Pohlenz

. 2006. Initially elevated TSH and congenital central hypothyroidism due to a homozygous mutation of the TSH beta subunit gene: case report and review of the literature. Exp Clin Endocrinol Diabetes, 114:227–234.

Miyai

. 2007. Congenital thyrotropin deficiency-from discovery to molecular biology, postgenome, and preventive medicine. Endocr J, 54:191–203.

Yamada

, Saga

, Shibusawa

, Hirato

, Murakami

, Iwasaki

, Hashimoto

, Satoh

, Wakabayashi

, Taketo

, Mori

. 1997. Tertiary hypothyroidism and hyperglycemia in mice with targeted disruption of the thyrotropin-releasing hormone gene. Proc Natl Acad Sci USA, 94:10862–10867.

10.

Sun

, Bak

, Schoenmakers

, van Trotsenburg

, Oostdijk

, Voshol

, Cambridge

, White

, le Tissier

, Gharavy

, Martinez-Barbera

, Stokvis-Brantsma

, Vulsma

, Kempers

, Persani

, Campi

, Bonomi

, Beck-Peccoz

, Zhu

, Davis

, Hokken-Koelega

, Del Blanco

, Rangasami

, Ruivenkamp

, Laros

, Kriek

, Kant

, Bosch

, Biermasz

, Appelman-Dijkstra

, Corssmit

, Hovens

, Pereira

, den Dunnen

, Wade

, Breuning

, Hennekam

, Chatterjee

, Dattani

, Wit

, Bernard

. 2012. Loss-of-function mutations in IGSF1 cause an X-linked syndrome of central hypothyroidism and testicular enlargement. Nat Genet, 44:1375–1381.

11.

Joustra

, Meijer

, Heinen

, Mol

, Laghmani el

, Sengers

, Carreno

, van Trotsenburg

, Biermasz

, Bernard

, Wit

, Oostdijk

, van Pelt

, Hamer

, Wagenaar

. 2015. Spatial and temporal expression of immunoglobulin superfamily member 1 (IGSF1) in the rat. J Endocrinol, 226:181–191.

12.

Mazzarella

, Pengue

, Jones

, Schlessinger

. 1998. Cloning and expression of an immunoglobulin superfamily gene (IGSF1) in Xq25. Genomics, 48:157–162.

13.

Robakis

, Bak

, Lin

, Bernard

, Scheiffele

. 2008. An internal signal sequence directs intramembrane proteolysis of a cellular immunoglobulin domain protein. J Biol Chem, 26:36369–36376.

14.

Katidou

, Vidaki

, Strigini

, Karagogeos

. 2008. The immunoglobulin superfamily of neuronal cell adhesion molecules: lessons from animal models and correlation with human disease. Biotechnol J, 3:1564–1580.

15.

Joustra

, van Trotsenburg

, Sun

, Losekoot

, Bernard

, Biermasz

, Oostdijk

, Wit

. 2013. IGSF1 deficiency syndrome: a newly uncovered endocrinopathy. Rare Dis, 1:e24883.

16.

Joustra

, Schoenmakers

, Persani

, Campi

, Bonomi

, Radetti

, Beck-Peccoz

, Zhu

, Davis

, Sun

, Corssmit

, Appelman-Dijkstra

, Heinen

, Pereira

, Varewijck

, Janssen

, Endert

, Hennekam

, Lombardi

, Mannens

, Bak

, Bernard

, Breuning

, Chatterjee

, Dattani

, Oostdijk

, Biermasz

, Wit

, van Trotsenburg

. 2013. The IGSF1 deficiency syndrome: characteristics of male and female patients. J Clin Endocrinol Metab, 98:4942–4952.

17.

Nakamura

, Bak

, Silander

, Lam

, Hotsubo

, Yorifuji

, Ishizu

, Bernard

, Tajima

. 2013. Three novel IGSF1 mutations in four Japanese patients with X-linked congenital central hypothyroidism. J Clin Endocrinol Metab, 98:E1682–E1691.

18.

Tajima

, Nakamura

, Ishizu

. 2013. A novel mutation of IGSF1 in a Japanese patient of congenital central hypothyroidism without macroorchidism. Endocr J, 60:245–249.

19.

Asakura

, Abe

, Muroya

, Hanakawa

, Oto

, Narumi

, Hasegawa

, Adachi

. 2015. Combined growth hormone and thyroid-stimulating hormone deficiency in a Japanese patient with a novel frameshift mutation in IGSF1. Horm Res Paediatr, 84:349–354.

20.

Reynaert

, Braat

, de Zegher

. 2015. Congenital nystagmus and central hypothyroidism. Int J Pediatr Endocrinol, 1:7.

21.

van Tijn

, de Vijlder

, Vulsma

. 2008. Role of the thyrotropin-releasing hormone stimulation test in diagnosis of congenital central hypothyroidism in infants. J Clin Endocrinol Metab, 93:410–419.

22.

Bernard

, Burns

, Haupt

, Matzuk

, Woodruff

. 2003. Normal reproductive function in InhBP/p120-deficient mice. Mol Cell Biol, 23:4882–4891.

23.

Castro-Magaña

, Angulo

, Cañas

, Sharp

, Fuentes

. 1998. Hypothalamic-pituitary gonadal axis in boys with primary hypothyroidism and macroorchidism. J Pediatr, 112:397–402.

24.

Jannini

, Ulisse

, D'Armiento

. 1995. Macroorchidism in juvenile hypothyroidism. J Clin Endocrinol Metab, 80:2543–2544.

25.

Bruder

, Samuels

, Bremner

, Ridgway

, Wierman

. 1995. Hypothyroidism-induced macroorchidism: use of a gonadotropin-releasing hormone agonist to understand its mechanism and augment adult stature. J Clin Endocrinol Metab, 80:11–16.

26.

Joustra

, Roelfsema

, Endert

, Ballieux

, van Trotsenburg

, Fliers

, Corssmit

, Bernard

, Oostdijk

, Wit

, Pereira

, Biermasz

. 2016. Pituitary Hormone Secretion Profiles in IGSF1 Deficiency Syndrome. Neuroendocrinology, 103:408–416.

27.

Holsberger

, Cooke

. 2005. Understanding the role of thyroid hormone in Sertoli cell development: a mechanistic hypothesis. Cell Tissue Res, 322:133–140.

28.

Maran

. 2003. Thyroid hormones: their role in testicular steroidogenesis. Arch Androl, 49:375–388.

29.

Joustra

, Andela

, Oostdijk

, van Trotsenburg

, Fliers

, Wit

, Pereira

, Middelkoop

, Biermasz

. 2016. Mild deficits in attentional control in patients with the IGSF1 deficiency syndrome. Clin Endocrinol, 84:896–903.