Microvillus inclusion disease (MVID) is mainly caused by pathogenic variants in MYO5B gene. MVID may present as a severe congenital enteropathy with intractable watery diarrhea, most often with onset immediately after birth, or present extraintestinal with persistent, low-gamma-glutamyl transferase cholestatic liver disease with increased concentrations of primary bile acids and progressive liver damage. Hence, the latter presentation is clinically indistinguishable from progressive familial intrahepatic cholestasis (PFIC) types I, II, IV, and V (1–3). More than a hundred MYO5B pathogenic variants have been reported (4–7). A complete lack of MYO5B protein causes predominant intestinal disease (MVID), whereas the expression of full-length mutant MYO5B protein with residual function causes predominant cholestatic liver disease (MVID-PFIC), and the expression of mutant MYO5B proteins without residual function causes both intestinal and hepatic dysfunction (MVID-mixed) (6,8).
The diagnosis of MVID is established with a combination of molecular genetic testing, light microscopic detection of vacuoles in the enterocytes, and ultimately, electron microscopic detection of microvillus inclusions. Enterocyte-to-enterocyte variation exists, and the percentage of enterocytes with microvillus inclusions varies greatly between patients and in a single patient. Liver biopsy shows portal and lobular fibrosis and giant cell transformation (5).
In this article, we present 3 cases with MYO5B variants, of which 1 is not previously published, and discuss genotype/phenotype correlations associated with the variants.
This patient was a girl born at gestational age 36 weeks + 4 days, with a birth weight of (BW) 3600 g. She was breastfed. When she was 7 days old, she was hospitalized due to a loss of 17% of her weight, lethargy, and multiple large stools. Blood samples showed hypoglycemia, metabolic acidosis with normal lactate and low bicarbonate (Table 1). Enteral feeding exacerbated the metabolic acidosis and stool output. Metabolic screening of urine was normal and showed no renal loss of bicarbonate. She received total parenteral nutrition and gained weight satisfyingly. Whole genome sequencing was performed at age 14 days showing that she was compound heterozygous for 2 MYO5B variants, (Table 2), of which 1 was novel. These findings were consistent with early onset MVID, which was confirmed by histologic and electron microscopic (EM) examination of small intestinal biopsies. She died at 7 weeks of age.
TABLE 1. -
Initial presentation of the children with MYO5B
|Age at onset, months
|ALT/ALP at admission
(ref. 8–32/17–46) U/L
|GGT (ref. 10–45) U/L
(ref. 5–25/<17) µmol/L
|INR (ref. <1.3)
Child 1 died 7 week of age of septicemia; Child 2 and 3 are siblings.
ALT = alanine aminotransferase; ALP = alkaline phosphatase; GGT = gamma-glutamyl transferase; INR = International Normalized Ratio.
TABLE 2. -
variants in the 3 children
||Amino acid change
This patient was a boy, born at term to first cousin parents. His BW was 3820 g, and he was breastfed. He presented with prolonged, conjugated hyperbilirubinemia at 5 months of age and had no gastrointestinal symptoms. Diagnostic work-up for cholestatic liver disease was inconclusive and he was discharged. At the ages of 15 and 18 months, he had 2 separate fractures in the upper and lower extremities, respectively (battered child was first suspected). Blood samples indicated rickets and he was treated with vitamin D, calcium, and phosphate.
He was admitted to hospital several times in the first years of life with passing diarrhea and vomiting, and a diagnosis of chronic cholestasis was established when he was 2.9 years old. Whole exome sequencing (WES) was performed at 5 years of age and showed a previously reported homozygous variant in MYO5B (Table 2). Intestinal histology and EM were consistent with MVID.
The younger sister of patient 2 was born at term with a BW of 3470 g and was breastfed. During her first year of life, she had episodes with diarrhea and vomiting. At 9 months of age, just 1 month after her older brother’s chronic cholestasis was diagnosed, she was admitted to hospital because of nonfebrile seizure, and she was diagnosed with intracerebral hemorrhage, caused by high prothrombin time with an International Normalized Ratio of 10 (<1.3). The International Normalized Ratio normalized after intravenous vitamin K administration. Blood samples showed mild conjugated hyperbilirubinemia and mild liver transaminase elevation, and a liver biopsy showed progressive familiar intrahepatic cholestasis, with no fibrosis or giant cells. At 3 years of age, WES was performed showing the same homozygous MYO5B variant as in her brother (child 2). Intestinal histology and EM revealed subtle changes that could be compatible with MVID.
The clinical and molecular genetic data are summarized in Tables 1 and 2, and a comparison of the clinical findings in patient 2 and 3 are shown in Table 3 and figure 1.
TABLE 3. -
Comparison of current status of siblings with a homozygous MYO5B
||Case 2 (12-year-old boy)
||Case 3 (10-year-old girl)
|Episodes of diarrhea necessitating hospital admission
|Treatment with fat-soluble vitamin supplements
|Height z score/weight z score
|Bone age, y
|Bone mass density z score
|ALT/ALP (ref. 8–32/17–46) U/L
|Bile salt (ref. <17) µmol/L
|Number of anticholestatic medications
Cholestatic medication: Ursodeoxycholic acid, Rifabutin, Cholestyramine, Naltrexone.
ALT = alanine aminotransferase; ALP = alkaline phosphatase; INR = International Normalized Ratio.
Here we present 3 patients with MYO5B variants and different clinical manifestations, ranging from isolated intestinal disease to intestinal disease combined with cholestatic liver disease, predominant cholestatic liver disease clinically similar to low-gamma-glutamyl transferase PFIC, seizures, and fractures.
The homozygous variant found in the 2 siblings (p.Glu82Lys) has previously been reported homozygous in 1 patient with MVID presenting at age 6 months (7) and later development of PFIC, thus a mixed phenotype. It was also reported in a pair of siblings with onset of MVID within the first months and first year of life, respectively, which later resolved, and the patients have no sign of liver affection (10). The 2 siblings reported here belong to 2 different phenotype categories, that is, PFIC for patient 2 who had no clinical symptoms of MVID, and the mixed phenotype for patient 3 (Tables 1 and 3). Thus, for this specific variant, it is not possible to predict phenotype from genotype, because it has now been found homozygous in both the patients with PFIC, MVID, and MVID-PFIC. To our knowledge, this is the only MYO5B variant that may be associated with all 3 phenotypes.
Patient 1 had severe MVID with early death. She harbored a novel nonsense variant (leading to p.[Tyr559*]) and a missense variant (p.Arg219Cys) that has been reported, but without any phenotype information. Another amino acid change at the same position (p.Arg219His) was reported in a patient with MVID who harbored another missense variant on the other allele (5). The combination of a nonsense variant and (p.Arg219Cys) in a patient with a severe phenotype as found here could indicate that p.Arg219Cys is not associated with any residual function.
Possible genotype–phenotype correlation in MYO5B-associated disease has been addressed in a few studies (5,8). In the study by Wang et al (8) of 130 patients with MYO5B variants, they found that the severity of intestinal manifestation was positively correlated to the number of null variants. Children with cholestasis carried at least 1 non-null variant. Comparing MVID and combined patients to FIC patients, the latter were more likely to carry missense/in-frame variants affecting the nonmotor regions of MYO5B. In line with these results, Aldrian et al (5) studied 114 cases and found that biallelic truncating variants were associated with a phenotype of MVID or mixed MVID/PFIC, but not isolated PFIC, and concluded that MVID results from the lack of MYO5B in enterocytes causing disrupted enterocyte localization. Solitary primary cholestatic liver disease may result from the expression of mutant MYO5B proteins that cause aberrant protein–protein interactions in hepatocytes, while nonfunctional MYO5B protein causes both intestinal and hepatic disease. However, as the siblings we report here illustrate, care should be taken when drawing firm conclusions on predicted phenotype based solely on the identified variant(s), because the variant they carry is associated with all 3 phenotypes, indicating that other factors modify the phenotype.
We think the use of whole genome sequencing/WES is a valuable diagnostic tool for cholestatic liver disease, and we suggest that genetic testing is included early during diagnostic investigations of children with gastrointestinal and cholestatic presentation.
Informed patient consent was obtained from all patients for publication of the case details.
1. Schlegel C, Weis VG, Knowles BC, et al. Apical membrane alterations in non-intestinal organs in microvillus inclusion disease. Dig Dis Sci. 2018;63:356–365.
2. Gonzales E, Taylor SA, Davit-Spraul A, et al. MYO5B
mutations cause cholestasis
with normal serum gamma-glutamyl transferase activity in children without microvillous inclusion disease. Hepatology. 2017;65:164–173.
3. Amirneni S, Haep N, Gad MA, et al. Molecular overview of progressive familial intrahepatic cholestasis
. World J Gastroenterol. 2020;26:7470–7484.
4. Girard M, Lacaille F, Verkarre V, et al. MYO5B
and bile salt export pump contribute to cholestatic liver disorder in microvillous inclusion disease. Hepatology. 2014;60:301–310.
5. Aldrian D, Vogel GF, Frey TK, et al. Congenital diarrhea and cholestatic liver disease: phenotypic spectrum associated with MYO5B
mutations. J Clin. Med. 2021;10:481.
6. van der Velde KJ, Dhekne HS, Swertz MA, et al. An overview and online registry of microvillus inclusion disease patients and their MYO5B
mutations. Hum Mutat. 2013;34:1597–1605.
7. Cockar I, Foskett P, Strautnieks S, et al. Mutations in Myosin 5B in Children With Early-onset Cholestasis
. J Pediatr Gastroenterol Nutr. 2020;71:184–188.
8. Wang LI, Qui YL, Xu HM, et al. MYO5B
-associated diseases: novel liver-related variants and genotype-phenotype correlation. Liver Int. 2022;42:402–411.
9. Vogel GF, Janecke AR, Krainer IM, et al. Abnormal Rab11-Rab8-vesicles cluster in enterocytes of patients with microvillus inclusion disease. Traffic. 2017;18:453–464.
10. Nathavitharana KA, Green NJ, Raafat F, et al. Siblings with microvillus inclusion disease. Arch Dis Child. 1994;71:71–73.