Application of Whole Exome Sequencing in Congenital Secretory Diarrhea Diagnosis

What Is Known Congenital diarrheal disorders are rare. Congenital sodium diarrhea is a rare variant of this rare disorder. What Is New We found a novel homozygous change (D405G) in the gene known to cause secretory sodium diarrhea (SLC93A). This case report underscores the importance of whole exome sequencing as early diagnostic tool. We have also discussed post-discharge course for this patient adding to natural history of the disease. Congenital diarrheal disorders (CDDs) are a rare, complex group of enteropathies that typically present early in infancy and can be a diagnostic challenge. These disorders generally have similar presentation but have locus heterogeneity and are often inherited in an autosomal recessive fashion (1). Congenital sodium diarrhea (CSD), among the rarest of the CDDs, was first described in 1985 determined to be caused by a biallelic loss-of-function mutation of SLC93A encoding NHE3(2). Clinical symptoms usually occur prenatally with polyhydramnios and dilation of bowel loops. Shortly after birth, there is profuse watery diarrhea with excessive fecal sodium loss and metabolic acidosis. CASE We encountered a term newborn female weighing 3740 g born to consanguineous South Asian parents with no significant family history. Prenatal labs, initial level II ultrasound, and fetal echocardiogram were normal. Repeat ultrasound at 14 weeks of gestation revealed an estimated fetal weight >90th percentile, polyhydramnios, and multiple dilated loops of bowel. Upon delivery, marked abdominal distention was noted. Abdominal ultrasound showed uniform dilatation of fluid-filled small and large bowel down to the rectum without signs of mechanical obstruction (Supplemental figure, Supplemental Digital Content, https://links.lww.com/MPG/B555). Within 24 hours of life, 620 mL of fluid was passed per rectum. Initial laboratory tests revealed normal serum sodium and stool electrolytes except for elevated fecal sodium of 107 meq/L. Fluid losses were replaced and sodium citrate supplementation was provided to maintain normal hydration and manage metabolic acidosis. Considering the diagnosis of CDD, an upper endoscopy and sigmoidoscopy were performed revealing normal gross, histopathologic, and electron microscopic findings. SNP microarray analysis did not detect abnormal copy number variation. Long contiguous stretches of homozygosity (LCSH) were identified consistent with reported consanguinity of the parents. On careful analysis none of the known genes causing secretory diarrhea fell within the regions of homozygosity. However, recognizing the limitation of SNP array and that no potential CDD gene was located in the LCSH, whole exome sequencing was performed to identify an underlying genetic cause of secretory diarrhea in this infant. In the interim, after short-term total parenteral nutrition, oral feeds were tolerated with adequate growth. Before discharge, a gastrostomy tube was placed for fluid and electrolyte replacement. At 16 months of age, infrequent diarrhea was reported and she continues to require sodium replacement and supplementation with citric acid. Growth and development remain normal. METHODS AND RESULTS Whole exome sequencing was performed at the Center for Applied Genomics at the Hospital for Sick Children, Toronto, on the patient, parents, and 2 healthy siblings revealing 148,943 variants, 739 of which were predicted to be rare, coding, and damaging. Autosomal recessive inheritance modeling was carried out using VarSeq software (Golden Helix) identifying 4 homozygous recessive, 28 compound heterozygous, and no high-quality de novo variants. One of the homozygous recessive variants is in a gene known to cause secretory sodium diarrhea, SLC9A3 encoding NHE3. In this proband, a novel homozygous change (D405G NHE3) was identified, whereas both parents and unaffected siblings were heterozygous carriers. This variant is not present in public databases and is predicted to be deleterious by multiple algorithms (3). In addition, D405G NHE3 is conserved across many species and is within the transmembrane region close to several previously reported pathogenic variants (Fig. 1). Interestingly, the region of homozygosity of SLC9A3 was not identified in an LCSH region or on SNP array.FIGURE 1: A, A predicted transmembrane domain architecture of SLC9A3 highlighting the D405G variant along with the location of several other SLC9A3 variants reported in congenital diarrhea cases. B, The D405G variant (NP_004165.2:p.Asp405Gly) maps to the 11th transmembrane domain of SLC9A3 and is conserved across species.DISCUSSION Given the potential morbidity and mortality associated with CDDs, much work has been done to determine the pathophysiology and molecular basis for these diseases. In recent years, a newer classification system is being used based on the mechanism of diarrhea: defects in digestion and absorption of nutrients and electrolytes; defects in enterocyte structure; defects in enteroendocrine cell differentiation; and defects in intestinal immune–related homeostasis (4). Evaluation CDDs typically starts by characterizing the diarrhea as osmotic, secretory, or inflammatory. Osmotic diarrhea is the result of unabsorbed nutrients reaching the colon thus driving fluid into the lumen and responds to fasting. Secretory diarrhea involves water flux toward the lumen due to abnormalities in ion transport and does not respond to fasting. Inflammatory diarrhea, caused by immune-mediated mucosal damage, may partially or not at all respond to fasting (5). Our patient was found to have CSD which typically presents with profuse secretory diarrhea after birth due to impaired sodium-proton exchange, manifesting as increased fecal sodium and metabolic acidosis. Hyponatremia is common but was not seen in our patient. The molecular etiology of CSD now includes 3 identified genes in its classic form; SPINT2, GUCY2C, and SLC9A3 (NHE3). GUCY2C encodes guanylate cyclase C, and gain of function mutation of guanylate cyclase C results in elevated levels of cyclic guanosine monophosphate causing a signaling cascade, which activates protein kinases and the cystic fibrosis transmembrane conductance regulator. This results in excessive chloride secretion in the lumen, which creates an osmotic pressure gradient; leading to increased secretion of sodium ions and water in the intestinal lumen (6). SLC9A3 encodes the sodium/proton antiporter 3 (NHE3) located in the apical membrane of enterocytes and is responsible for the majority of intestinal sodium absorption and acid-base homeostasis. Biallelic mutants in SLC9A3 have been identified in 18 patients from 16 unrelated families with CSD (7). Our patient, with a novel NHE3 D405G variant, has done clinically well, needing only enteral supplementation for sodium losses appears to confer a milder disease form. At 16 months of age, having proven the ability to maintain acid-base homeostasis and steady growth, we expect to transition enteral to oral supplementation. In recent studies, the association between loss of NHE3 function as a predisposing factor toward the development of inflammatory bowel disease, is something she will need to be monitored closely for (8). Whole exome sequencing with careful functional studies including in silico, cellular, and animal studies have provided a power tool for the discovery of new phenotypes and genetic mutations associated with these disorders and have enabled advances in diagnostic, therapeutic, and prognostic indications for patients and their families. Here, with our patient we hope to add to that pool of knowledge.