Abstract
Inherited diarrheal disorders cause serious morbidity resulting in dependence on intensive care and parenteral nutrition. Microvillus inclusion disease (MVID) has been classically described and results from mutations in the gene coding myosin Vb, which is responsible for enterocyte polarization. Newer reports of mutations resulting in truncated syntaxin 3 (STX3) and Munc18-2 (STXBP2) proteins have been elucidated as causative. To date, five cases of STX3 abnormalities resulting in MVID have been described. We report an infant who presented with congenital diarrhea and was determined to have a rare mutation of STX3. This new finding would be beneficial in future functional genotype–phenotype correlation studies.
Keywords: microvillus inclusion disease, congenital diarrheal disease, STX3
Introduction
Congenital diarrheal diseases (CDDs) are a group of inherited enteropathies that cause early-onset symptoms and severe morbidity. 1 CDD presents with varied pathophysiology culminating in chronic malnutrition and dependence on parenteral nutrition. 2 Improved genetic diagnostic tests have identified mutations specific to each disease. 3 Microvillus inclusion disease (MVID) is a secretory diarrhea that is usually noted soon after birth. MYO5B mutations that cause myosin Vb defects have been largely characterized to be a cause of MVID due to abnormal enterocyte polarization. 4 However, in the recent times, syntaxin-3 (coded by STX3) and Munc18-2 (coded by STXBP2) which form parts of a protein complex that controls the membrane fusion of transport vesicles within the apical brush border have been implicated. At least 62 unique MVID-associated MYO5B and 51 STXBP2 mutations have been reported. 5 However, to our knowledge, only five cases of MVID have been reported to result from STX3 mutations. 6 7 8 9 We describe a patient with congenital diarrhea and a challenging clinical course, who was detected to have a homozygous single base pair insertion in exon 9 of the STX3 gene (chr11:59562859:T > TA; Depth: 126x). This case report would expand our understanding of pathogenic STX3 mutations that are associated with CDD.
Case Report
A female infant was referred at 5 weeks of age with chronic diarrhea, dehydration, malnutrition, and metabolic acidosis. She was the first child born to nonconsanguineous parents at 34 weeks' gestation by cesarean section secondary to polyhydramnios and fetal distress. Meconium-stained amniotic fluid was noted intraoperatively. She was started on breast milk on day 1 and attained full enteral feeds by day 5. The treating team noted increasing stool frequency by the end of the first week. She was treated for presumed sepsis with broad-spectrum antibiotics with no improvement. As the purge rate increased, lactose-free formula was used over the next 3 days. The diarrhea persisted and the infant required intravenous fluids. Elemental hydrolyzed formula was administered for 5 to 7 days without improvement. The infant was next started on a carbohydrate-free formula to which fructose was added. Although parenteral nutrition and intravenous fluids could be weaned as stool volumes dropped, no significant weight gain was noted and the infant was referred to us on this diet plan. She was received with a deficit of 300 g from birth weight at 1 month of life, no subcutaneous fat, but looking hungry and alert—the proverbial “skin and bones” appearance. Stools were noted to be semisolid and yellow among normal colored “urine” soaking the diaper. She was started on an expressed breast milk challenge, which she vigorously consumed over the first 24 hours, but which was followed by large watery purges of stools. Initial investigations revealed stool osmolar gap of 190 mOsm/kg ( Table 1 ). These analyses were repeated after bladder catheterization, to rule out urine contamination and were found to be similar.
Table 1. Laboratory results during hospital course.
| Day of life | Therapy/clinical condition | Sodium (mmol/L) | Potassium (mmol/L) | Chloride (mmol/L) | pH | Total protein (g/dL) | Albumin (g/dL) | Cholesterol (mg/dL) | Triglycerides (mg/dL) | Stool pH | Stool reducing substances | Stool osmolar gap | Stool electrolytes (mmol/L) |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Referring hospital 9 |
After 3 d of loose stools were noted, on IV fluids and breast milk feeds | 139 | 5.8 | 96 | 7.29 | 5 | 3.5 | – | 6 | Traces | – | ||
| 13 | Elemental formula + IV fluids | 147 | 6.8 | 110 | 7.18 | – | 5 | – | |||||
| 26 | RCF + Fructose feeds + IV fluids | 132 | 5.2 | – | 7.25 | – | |||||||
| Our unit 38 |
RCF + Fructose feeds + IV fluids | 134 | 3.9 | 98 | 7.19 | 5.1 | 3.6 | 130 | 151 | 7 | + | 190 | 23 Na 26.3 K 22.9 Cl |
| 47 | RCF+ parenteral nutrition | 138 | 4.2 | 107 | 7.28 | – | 7 | – | 178 | 34 Na 22 K 36 Cl |
|||
| 59 | Bowel rest + TPN | 139 | 4.6 | 95 | 7.3 | 6 | 3.8 | – | 6 | – |
Abbreviations: IV, intravenous; RCF, Ross carbohydrate Free; TPN, total parenteral nutrition.
Screens for infection and cultures of blood urine and stool, as well as polymerase chain reaction for enteric pathogens were all negative. Intestinal biopsy was advised and discussed with the family but they declined consent. As the history and findings were suggestive of CDD, a clinical exome was requested after genetic counseling of parents. Carbohydrate-free formula was started in the setting of a stool pH of 7, high stool osmolar gap, and a clinical picture of osmotic diarrhea that worsened with feeds (some improvement noted with fasting and total parenteral nutrition). We hypothesized that an osmotic diarrhea associated with carbohydrate malabsorption (rather than protein or lipid) would best explain most of her findings; however, her history of antenatal polyhydramnios and meconium-stained amniotic fluid suggested the presence of fetal (secretory) diarrhea.
Loose stools continued in spite of trials of carbohydrate-free formula, and hydrolyzed protein formula in tandem, and her general condition deteriorated. Consequently, a period of bowel rest with total parenteral nutrition was attempted. There was partial reduction in diarrhea for a few days (decreased to around 250 mL/kg/d from 430 mL/kg/d on feeds) only to recur again, without meaningful weight gain. Next-generation sequencing reports were obtained on day 84 of life. A homozygous one base pair insertion in exon 9 of the STX3 gene (chr11:59562859:T > TA; depth: 126x) that resulted in the truncation of the protein 19 amino acids downstream to codon 231 (p.Asn231LysfsTerfs19; ENST00000337979.4) was detected. The options of intestinal transplant and lifelong parenteral nutrition were discussed. The infant succumbed in the hospital at 3 months of age to septicemia. Parental testing is planned; the infant's father is held up overseas due to the unprecedented COVID-19 pandemic.
Discussion
Dehydrating diarrhea in the first few days of life is most often severe enough to require parenteral nutrition and prolonged care. Frequent large volume stools may manifest as part of a multisystem disease with varied spectrum of severity. Loose stools in early infancy may result from an allergy to cow milk protein or infections; however, early suspicion of congenital disorders aids in delineating management plans and prognostication. CDD is broadly classified into four groups based on pathological defects in—(1) absorption and transport of nutrients and electrolytes, (2) enterocyte differentiation and polarization, (3) enteroendocrine cell differentiation, and (4) modulation of the intestinal immune response. Differentiating between secretory (diarrhea of fetal onset does not decrease with fasting) and osmotic (onset usually after introduction of specific class of feeds, high stool osmolar gap, abates with fasting and may show response to exclusion diets) types helps in initial management. Some disorders overlap and may worsen with feeding, although the primary disorder is secretory. MVID (OMIM 251850) is one such disorder. It usually presents in the first few days after birth with severe diarrhea, metabolic acidosis, and dehydration as was the case in our infant. However, we did face several diagnostic conundrums—the first one being that of mother having polyhydramnios and meconium-stained amniotic fluid which is only an occasional finding in MVID. 10 11 Albeit transient and inconsistent, the infant did show improvement with fasting. High stool osmolar gap was also a feature which pointed toward osmotic diarrhea. Using a diagnostic algorithm, we clinically ruled out protein and lipid malabsorption (no hypoproteinemic edema, normal stool fat and cholesterol, triglycerides in serum) with detectable reducing substances in stool, thus leading us to consider carbohydrate absorption defects and make the decision to introduce carbohydrate-free formula. Ill-sustained response led us to revise the diagnosis and try other exclusion diets. Complete bowel rest is known to reduce diarrhea moderately in MVID and tufting enteropathy. 11
The extremely sick state of the infant, parental hesitation, along with newer recommendations in favor of molecular studies over invasive procedures led us to defer an intestinal biopsy. 1 12 Pathology from intestinal biopsy is known to show shortened microvilli and villous atrophy with an increased number of secretory granules and membrane-bound periodic acid Schiff's stain positive inclusions within intestinal mucosal cells in MVID. 4 12 13
Genetic studies have enabled identification of candidate genes in each class of CDD. Most disorders are found to be monogenic in origin. Myosin Vb is an actin-based molecular motor protein that regulates endosome/transport vesicle trafficking to the apical brush border. MYO5B defects may block the apical traffic of intracellular vacuoles causing aggregation of these intracellular inclusions in MVID. Genes regulating STX3 and STXBP2 have been identified and functionally related to MVID, thereby improving diagnostic spectrum. STX3 (a member of the Qa-SNARE protein family) and STXBP2 are part of a protein complex that controls fusion of transport vesicles with the apical brush border membranes. There is evidence that MYO5B , STX3 , and STXBP2 genes are functionally linked in MVID patients. 5 There is very close similarity of phenotypes reported in patients carrying MYO5B, STX3, or STXBP2 mutations and the defects in apical membrane trafficking in epithelial cells suggest a common disease mechanism in MVID.
In contrast to classical MVID due to MYO5B defects where apical brush border alone is affected, multiple microvilli at the basolateral membranes have been noted in the other variants. Disease due to STX3 mutations have been reported to be clinically less severe unlike in our infant. 1 6 It would be appropriate to mention here that these children were probably dependent on parenteral nutrition making the illness severe enough to warrant intensive life supports. The early availability of next-generation sequencing for diagnosis in our case helped us provide counseling and pragmatic information transfer to the parents about expected quality of life.
To the best of our knowledge, there are five reported cases of STX3 mutations causing CDD due to MVID. Three unique STX3 disease relevant mutations were reported (OMIM no. 600876). 1 6 7 8 9 12 These include a homozygous nonsense mutation (c.739C > T, p.Arg247*, Ensembl COSM193004) in exon 9 and a homozygous frame-shifting 2-bp insertion (c.372_373dup, p.Arg125Leufs*7) in exon 6 of STX3. Both mutations were predicted to cause cellular STX3 protein depletion and truncation. 5 Others are (c.424C > T; p.Arg142) and (c.363 366delinsGA; p.Val122fs). It is notable that all the variants associated with MVID phenotype are nonsense variations, where as a missense variation reported by Chograni et al in 2015 (c.122A-G transition in exon 3 resulting in a glu41-to-gly [E41G] substitution at a highly conserved residue within the syntaxin N-terminal domain, and classified as a variant of uncertain significance) was identified in three siblings with congenital cataract and developmental delay. The mutation was present in heterozygosity in their unaffected parents, and was not found in 50 ethnically matched controls; variants detected in the 3 other candidate genes did not segregate with disease. 7
We report a fifth variant (c.690_691insA, p.Asn231LysfsTerfs19; ENST00000337979.4). Our infant had a homozygous one base pair insertion in exon 9 of the STX3 gene (chr11:59562859:T > TA; depth: 126x) detected by next-generation sequencing. This introduces a stop codon in the central SNARE domain resulting in a truncated protein lacking part of the SNARE motif and the entire C-terminal transmembrane domain and extracellular part of the protein. The in silico predictions based on Variant Effect Predictor (SIFT version 5.2.2; PolyPhen 2.2.2) and MutationTaster2 were damaging.
We admit that an intestinal biopsy revealing pathognomonic findings might have added value; however, the invasive nature of the test was not acceptable to the family. Parental testing is planned. Although the couple were nonconsanguineous, we would like to point out here that the Indian society has traditionally been highly endogamous . Tamil Nadu (a state from where the couple hail), while relatively industrialized, has a caste endogamy rate of 97%. 14 It also interesting to note the words of David Reich here, “….. the truth is that India is composed of a large number of small populations.”
Conclusion
This rare mutation on causal gene STX3 could add to the registry on MVID genetics. Functional assessment of the gene mutation for genotype–phenotype correlation in model systems is the way forward.
Acknowledgment
We thank immensely, Dr. G.K. Madhavilatha, senior scientist and genetic counselor, MedGenome Laboratories, for her significant academic inputs.
Funding Statement
Funding None.
Footnotes
Conflict of Interest None declared.
References
- 1.PediCODE Consortium . Thiagarajah J R, Kamin D S, Acra S. Advances in evaluation of chronic diarrhea in infants. Gastroenterology. 2018;154(08):2045–2.059E9. doi: 10.1053/j.gastro.2018.03.067. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Duggan C P, Jaksic T. Pediatric intestinal failure. N Engl J Med. 2017;377(07):666–675. doi: 10.1056/NEJMra1602650. [DOI] [PubMed] [Google Scholar]
- 3.Berni Canani R, Terrin G, Cardillo G, Tomaiuolo R, Castaldo G. Congenital diarrheal disorders: improved understanding of gene defects is leading to advances in intestinal physiology and clinical management. J Pediatr Gastroenterol Nutr. 2010;50(04):360–366. doi: 10.1097/MPG.0b013e3181d135ef. [DOI] [PubMed] [Google Scholar]
- 4.Ruemmele F M, Müller T, Schiefermeier N. Loss-of-function of MYO5B is the main cause of microvillus inclusion disease: 15 novel mutations and a CaCo-2 RNAi cell model. Hum Mutat. 2010;31(05):544–551. doi: 10.1002/humu.21224. [DOI] [PubMed] [Google Scholar]
- 5.Dhekne H S, Pylypenko O, Overeem A W. MYO5B, STX3, and STXBP2 mutations reveal a common disease mechanism that unifies a subset of congenital diarrheal disorders: a mutation update. Hum Mutat. 2018;39(03):333–344. doi: 10.1002/humu.23386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Wiegerinck C L, Janecke A R, Schneeberger K. Loss of syntaxin 3 causes variant microvillus inclusion disease. Gastroenterology. 2014;147(01):65–6.8E11. doi: 10.1053/j.gastro.2014.04.002. [DOI] [PubMed] [Google Scholar]
- 7.Chograni M, Alkuraya F S, Ourteni I, Maazoul F, Lariani I, Chaabouni H B. Autosomal recessive congenital cataract, intellectual disability phenotype linked to STX3 in a consanguineous Tunisian family. Clin Genet. 2015;88(03):283–287. doi: 10.1111/cge.12489. [DOI] [PubMed] [Google Scholar]
- 8.Alsaleem B MR, Ahmed A BM, Fageeh M A. Microvillus inclusion disease variant in an infant with intractable diarrhea. Case Rep Gastroenterol. 2017;11(03):647–651. doi: 10.1159/000479624. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Julia J, Shui V, Mittal N, Heim-Hall J, Blanco C L. Microvillus inclusion disease, a diagnosis to consider when abnormal stools and neurological impairments run together due to a rare syntaxin 3 gene mutation. J Neonatal Perinatal Med. 2019;12(03):313–319. doi: 10.3233/NPM-1852. [DOI] [PubMed] [Google Scholar]
- 10.Vogel G F, Hess M W, Pfaller K, Huber L A, Janecke A R, Müller T. Towards understanding microvillus inclusion disease. Mol Cell Pediatr. 2016;3(01):3. doi: 10.1186/s40348-016-0031-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Ruemmele F M, Schmitz J, Goulet O. Microvillous inclusion disease (microvillous atrophy) Orphanet J Rare Dis. 2006;1:22. doi: 10.1186/1750-1172-1-22. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Terrin G, Tomaiuolo R, Passariello A. Congenital diarrheal disorders: an updated diagnostic approach. Int J Mol Sci. 2012;13(04):4168–4185. doi: 10.3390/ijms13044168. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Iyngkaran N, Abdin Z, Davis K. Acquired carbohydrate intolerance and cow milk protein-sensitive enteropathy in young infants. J Pediatr. 1979;95(03):373–378. doi: 10.1016/s0022-3476(79)80509-0. [DOI] [PubMed] [Google Scholar]
- 14.Intercaste marriages in India: has it really changed over time?Accessed July 19, 2020 at:https://epc2010.princeton.edu/papers/100157
