Novel Causative RET Mutation in a Japanese Family with Hirschsprung's Disease: Case Report and Factors Impacting Disease Severity

Abstract

RET gene variances confer susceptibility to Hirschsprung's disease (HSCR) with pathogenetic mutations being identified in half of familial cases. This investigation of familial HSCR was aimed to clarify the relationship between genetic mutations and clinical phenotype using next-generation sequencing. A novel c2313C > G(D771E) RET mutation was identified in all three affected family members. The mutation involved the kinase domain, which is believe to impair RET activity and intestinal function. A second RET mutation, c1465G > A(D489N), was found only in the extensive aganglionosis case. We conclude that the novel c2313C > A(D771E) mutation in RET may be pathogenic for HSCR, while the c1465C > G(D489N) mutation may be related to phenotype severity.

Introduction

Hirschsprung's disease (HSCR) is a congenital disorder characterized by the absence of intestinal ganglion cells that control intestinal tract peristalsis. The length of the aganglionic segment is related to disease severity, with some short segments being associated with no difficulty in daily living while others total intestinal involvement requiring complete parenteral nutrition. Numerous reports have investigated the genetic causes of HSCR, especially following the advent of next-generation sequencing (NGS). ^{1 2} A monoallelic variant in many HSCR-related genes is often sufficient for phenotypic HSCR expression. The cumulative effects of multiple gene mutations on disease phenotype in HSCR have also been reported. ³ Further, only one-third of affected patients have been identified to harbor disease-causing genetic mutations. Among these genes, RET is not only one of the most common genes causing HSCR but also the most extensively one studied regarding the genetics and pathology of HSCR. ^{4 5 6 7}

When there has been recurrence of HSCR in a family, the risk of a subsequent affected child increase 200 folds. ⁸ Moreover, phenotypic differences can exist within the same family. In familial HSCR, pathogenetic RET gene mutations are present in roughly half of cases. ⁸ Further, several studies have speculated on the relationship between genetic mutations and clinical phenotype/prognosis in familial HSCR. ^{4 9 10} Here, we present a genetic investigation of familial HSCR to clarify phenotype/prognosis relationship and to report a novel HSCR-causing RET mutation and a known mutation related to disease severity.

Case Presentation

Patient 1: A 43-Year-Old Japanese Man

He was born after 40 weeks of gestation without pregnancy complications. His birth weight was 3,140 g (25th centile). On day 0, he passed a small amount of meconium and subsequently developed persistent vomiting. By day 7, his abdomen was markedly distended. A contrast enema then showed a transition zone in the sigmoid colon. He was diagnosed as having HSCR (rectosigmoid aganglionosis) and received a pull-through procedure at age 12 months. He recovered without complications and has had no bowel problem since then.

Patient 2: A 5-Year-Old Girl, Daughter of Patient 1

She was born after 39 weeks of gestation without pregnancy complications. Her birth weight was 2,596 g (25th centile). On day 1, she passed a meconium and by day 6, she had developed marked abdominal distension. At 4 months, a rectal biopsy revealed lower-segment HSCR. She subsequently underwent a pull-through procedure at 5 months. Pathological examination of the surgical specimen established the diagnosis of rectosigmoid aganglionosis. She recovered without complications and has experienced no intestinal-related problems since.

Patient 3: A 2-Year-Old Boy, Son of Patient 1

He was born after 38 weeks of gestation following an uncomplicated pregnancy conceived by assisted reproductive technology. His birth weight was 2,960 g (20th centile). Following birth, he failed to pass any stool and had recurrent vomiting. A contrast enema revealed nearly total microcolon. A biopsy of the small intestine done 15-cm distal to the ligament of Treitz contained ganglion cells, whereas a sample at 25-cm distal to the ligament found no ganglions. He was diagnosed with extensive aganglionic HSCR, and subsequently had a jejunostomy located 60-cm distal to the ligament of Treitz. Following this procedure, he can tolerate only a small amount of liquid by mouth receiving most of his nutrition by hyperalimentation infusion.

Methods and Materials

Next-Generation Sequencing Analysis

Sample and DNA Extraction

The family was included in this study after providing informed written consent. DNA was extracted from peripheral blood using a QIAamp DNA Mini Kit (QIAGEN; Hilden, Germany) according to the manufacturer's instructions.

Next-Generation Sequencing Testing

Primer design: primers were designed using the Ion AmpliSeq Designer ( https://www.ampliseq.com ) as panels 1, 2, and 3 to cover an exon region of 29, 20, and 9 genes related to HSCR and its related diseases, respectively ( Table 1 ).

Table 1. List of genes investigated.

	Genes	Total coverage
Panel 1	CX3CL1, DHCR7, ECE1, EDN3, EDNRB, GDNF, GFRA1, HOXB5, KIAA1279, L1CAM, NKX2–1, NRG1, NRG3, NRTN, NTF3, NTRK3, PHACTR1, PHACTR2, PHACTR3, PHACTR4, PHOX2B, PROK1, PROKR1, PROKR2, RET, RMRP, SOX10, TCF4, ZEB2	99.1%
Panel 2	ARID1B, ARVCF, ASCL1, COMT, Connexin 26, Connexin 43, DENND3, DNMT3B, GJA1, GJB2, GLI1, GLI2, GLI3, NCLN, NUP98, PSPN, SEMA3A, SEMA3D, TBATA , TUBA1A	97.7%
Panel 3	ACTG2, CALB2, FLNA, MYH11, MYLK, LMOD1, SLC6A20, TLX2 , VCL	93.2%

Library preparation: libraries were prepared using 15 ng of each DNA sample for the Ion AmpliSeq Kit for Chef DL8 (Thermo Fisher Scientific; Waltham, Massachusetts, United States) according to the manufacturer's instructions.

For those regions that could not be covered by the designed panels or those unable to be amplified by the Chef DL8 system, these areas were amplified by single polymerase chain reaction (PCR). Thus was followed by the preparation of a barcoded library using an Ion Xpress Plus Fragment Library Kit (Thermo Fisher Scientific) and IonXpress Barcode Adapters (Thermo Fisher Scientific).

Template preparation and sequencing: the barcode-added library was compiled by emulsion PCR, and then loaded onto Ion 318 Chip V2 BC (Thermo Fisher Scientific) using the Ion PGM Hi-Q View 400 Chef Kit (Thermo Fisher Scientific). This latter step was followed by semiconductor sequencing performed on an Ion PGM System (Thermo Fisher Scientific).

Data analysis: the nucleotide sequence data collected by the Ion PGM System were aligned with the human reference sequence hg19 using Torrent-suite software v5.10 (Thermo Fisher Scientific). Variant call data were extracted using the Torrent-suite Variant Caller plug-in. The obtained VCF file was uploaded into the wANNOVAR ( http://wannovar.wglab.org/ ) pipeline and annotated for genomic variation.

The regions obtaining less than 10-read coverage by NGS and the pathogenic variant candidates extracted from the variations detected by NGS were secondarily analyzed by Sanger's sequencing using the DTCS Quick Start kit (Beckman-Coulter; Brea, California, United States).

Results

The HSCR phenotype was classified according to the extent of aganglionosis. Patient 3 had the extensive aganglionosis phenotype, while his father (patient 1) and sister (patient 2) had the short-segment phenotype. Following their colonic surgeries, patients 1 and 2 had had no subsequent intestinal troubles. However, patient 3 has required continuous hyperalimentation therapy.

NGS and Sanger's sequencing identified a novel RET missense mutation, c2313G > C (D771E) ( Fig. 1A, 1B ) in all three patients. Each was monoallelic for this mutation, and none had any pathogenic mutations in the known or suspected 57 HSCR-related genes evaluated.

Fig. 1.

Results of next-generation sequencing verified by Sanger's sequencing. Ref. seq.: reference sequence, H001: patient 1, H002: patient 2, H003: mother of siblings, H004: patient 3. ( A, B ) RET c.2313C > G mutation and its amino acid replacement D771E. ( C, D ) RET c.1465G > A mutation and its amino acid replacement D489N. ( E ) Pedigree of the studied family with amino acid mutation transmission. The HSCR phenotype is written at the shoulder of each symbol. EA, extensive aganglionosis; HSCR, Hirschsprung's disease; S, short-segment aganglionosis.

Patient 3 also carried a single nucleotide polymorphism (SNP), c1465A > G (D489N), in the RET gene, which was inherited from his healthy mother ( Fig. 1C , 1D ). A pedigree with the gene mutations is presented in Fig. 1E .

In addition to the above studies, we investigated three susceptibility nucleotide foci for HSCR in the RET gene, c135G > A (A45A), c1296G > A (A432A), and c2307T > G (L769L), among the family members. Only one locus (c2307T > G) was common, which was heterozygous among the patients ( Table 2 ). Patients 2 and 3 were also heterozygous for c135G > A (A45A) and c1296G > A (A432A), respectively.

Table 2. Nucleotide analysis of three susceptibility positions in the RET gene in the patients and mother of siblings .

Codon	Nucleotide position	Patient 1	Patient 2	Patient 3	Mother
A45A	c.135	G/G	G/A	G/G	G/G
A432A	c.1296	G/G	G/G	G/A	G/A
L769L	c.2307	T/G	T/G	T/G	G/G

Discussion

We investigated a family with three members affected with HSCR and uncovered a novel disease-causing RET mutation. Two of the three individuals, the father and daughter, had short-segment involvement while the third, the son, had almost total aganglionosis. We also identified a genetic factor that might explain the phenotypic differences seen among the family members.

The RET mutation D771E was present in exon 13 and involved the intracellular kinase domain, the most highly evolutionarily conserved area of the gene. This mutation was caused by the nucleotide change c2313C > G, which produced an amino acid change from aspartic acid to glutamic acid. Another mutation involving the same amino acid has also been described, D771N, caused by the nucleotide change c2311(G > A), which involved a change from aspartic acid to asparagine. ^{7 11 12} Those patients who have possessed D771N have had total-colonic aganglionosis. Functional consideration has revealed that this residue situated in the αC helix helps with proper positioning of the helix, formation of a salt bridge between D771 and R912 in the activation loop, and stabilization of the two lobes of RET kinase. ⁷ Hyndman et al also noted that the mutant, RET D771N, had a consistently reduced ability to phosphorylate STAT3 as compared with the wild type. ⁷ These authors further concluded that the mutation could impair the proliferation and survival of neural crest cells or path finding of early neuroblasts both of which are required for the development of enteric neurogenesis. Failure of either of these processes can give rise to the HSCR phenotype. ⁷ Thus, we speculate that the RET residue D771 mutation decreased the function of the RET protein and adversely impacted cell function, which caused improper neuroblast activity culminating in HSCR. Moreover, the difference in the clinical phenotype between D771N (total-colonic aganglionosis) and D771E (short-segment aganglionosis) could be modulated by the replaced amino acid to affect the function of the mutant RET protein. Aspartic acid and glutamic acid share a common characteristic in that they both have an electric charge, whereas asparagine does not. Lubin et al described that the presence or absence of a charge did affect protein stability and function using in EST3 yeast gene, ¹³ which might explain the phenotypic difference between the D771N and D771E mutations.

Kim et al reported that the RET mutation D489N caused by c1465(G > A) resulted in HSCR in patients with the short-segment phenotype. ¹⁴ D489N mutation, which is found primarily in the Asian population, involves the extracellular domain of the RET protein. ^{10 15 16 17} In addition, Ishii et al reported this mutation in a patient with total-colonic aganglionosis who was homozygous for both D489N and L769L and for a heterozygous mutation of V778D. ¹⁰ In the same report, four other HSCR patients were cited with a heterozygous D489N mutation in the RET gene without other pathogenic RET mutations. Each had the rectosigmoid phenotype. ¹⁰ Ishii et al also stated that this mutation affected RET processing in the endoplasmic reticulum and disturbed RET expression on the cell surface. These authors also created cells carrying triple mutations (D489N, L769L, and V778D) in RET and demonstrated a significant decrease in proliferation in comparison to wild-type and single-mutant RET. ¹⁰ In our familial cohort, both the mother and son carried the D489N mutation. The son had a more severe form of HSCR than did his father or sister, while the mother was apparently healthy. This observation indicates that D489N, itself in monoallelic form does not cause HSCR. Thus, we postulate that the RET D489N mutation imparts a negative effect on RET function and the clinical HSCR phenotype only when coupled with another pathogenic RET mutation.

Lastly, many reports have discussed the clinical impact of three synonymous SNPs, c135G > A (A45A), c1296G > A (A432A), and c2307T > G (L769L), on the HSCR phenotype. ^{18 19 20} Garcia-Barceló et al reported that among these SNPs, the most common HSCR-associated genotype was homozygosity for allele A of c135G > A (A45A) and allele G of c2307T > G (L769L). ²⁰ This notion appears not to apply in our cases since all patients were heterozygous for c2307T > G (L769L), while patients 2 and 3 were heterozygous for c135G > A (A45A) and c1296G > A (A432A), respectively. Garcia-Barceló et al also investigated the impact of the haplotype of these three loci and concluded that allele A of c1296G > A was associated with the more severe manifestation of HSCR. ²⁰ Our patient 3 with the most severe phenotype had this haplotype. Although we cannot compare the impact of D489N and the c1296G > A haplotype on disease severity, our findings suggested that the c1296G > A haplotype had some influence on the severity of HSCR.

Conclusion

In conclusion, we uncovered a novel RET D771E mutation in a family with three affected members who had variable ganglionic involvement. We postulate that this represents a phenotype-enhancement effect of the RET D489N mutation. More genetic analyses of HSCR cases, especially in the Asian population, will be required to confirm our findings.

Funding Statement

Funding This study was financially supported by JSPS KAKENHI, under grant nos.: JP16H00645, JP17H00639, and JP19H00445.