Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Functional Characterization of Mutations in the Myosin Vb Gene Associated With Microvillus Inclusion Disease

Szperl, Agata M*; Golachowska, Magdalena R; Bruinenberg, Marcel*; Prekeris, Rytis; Thunnissen, Andy-Mark WH§; Karrenbeld, Arend||; Dijkstra, Gerard; Hoekstra, Dick; Mercer, David#; Ksiazyk, Janusz**; Wijmenga, Cisca*; Wapenaar, Martin C*; Rings, Edmond HHM††; van IJzendoorn, Sven CD

Author Information
Journal of Pediatric Gastroenterology and Nutrition: March 2011 - Volume 52 - Issue 3 - p 307-313
doi: 10.1097/MPG.0b013e3181eea177

Abstract

A structurally, compositionally, and functionally distinct plasma membrane at the apex of the intestinal epithelial cell monolayer provides a selective and protective barrier that regulates the uptake of nutrients from the lumen. The inability of intestinal cells to maintain an apical brush border and the consequence it has on human functioning becomes particularly apparent in patients diagnosed as having microvillus inclusion disease (MVID; Online Mendelian Inheritance in Man 251850).

MVID is a rare autosomal recessive disease presenting with severe intractable diarrhea and malabsorption in neonates (1–5). At the cellular level, variable brush border atrophy with accumulation of lysosomal granules and microvillus inclusions is observed in the apical cytoplasm of MVID enterocytes (1,2,6,7). Periodic acid-Schiff–stained and other apical brush border components (eg, CD10) are typically absent from the cell surface and accumulate in compartments in the apical cytoplasm (8). In contrast to the apical proteins, basolateral proteins display a normal polarized distribution at the surface of MVID enterocytes (8,9), which appear normally arranged in monolayers with distinguishable cell-cell adhesion junctions.

MVID is often described in children born of consanguineous parents and this allowed Müller et al (10) to map the MVID locus to 18q21 using homozygosity mapping in an extended Turkish kindred. Mutation analysis of a positional candidate gene from the region of homozygosity, MYO5B, revealed a homozygous in-frame insertion in the patients with MVID from the Turkish kindred. To date, 25 different nonsense, missense, splice site, or in-frame insertion mutations in the MYO5B gene (Online Mendelian Inheritance in Man 606540) have been identified in 28 patients with MVID from consanguineous and unrelated marriages (10–12). The MYO5B gene encodes myosin Vb, which is an actin filament–based motor protein that interacts with and regulates among others the subcellular spatial distribution of recycling endosomes that express small guanosinetriphosphatase (GTPase) proteins such as Rab11a on their cytoplasmic surface.

In several patients with MVID, MYO5B mutations were found in only 1 allele (heterozygous) or no MYO5B mutation was found (10). Moreover, although knockdown of myosin Vb in human epithelial colorectal adenocarcinoma (Caco-2) cells recapitulates most of the cellular phenotypes of MVID (12), it is not known whether myosin Vb mRNA and protein expression and myosin Vb function are affected in patients with MVID. Such information supports MYO5B gene screening as a diagnostic tool for this difficult-to-recognize rare disease, and allows reliable genetic counseling and prenatal screening. Supporting evidence that MYO5B mutations have consequences for the expression and/or function of the myosin Vb protein in MVID enterocytes as well as mutational analyses of additional patients with MVID are therefore imperative. In the present study, we have used small-intestine biopsies to demonstrate that MVID-associated MYO5B mutations affect the expression and function of the myosin Vb protein in MVID enterocytes. In addition, we have performed mutation analyses of 9 additional patients with MVID of various ethnic backgrounds and report 8 new MYO5B mutations: 3 homozygous and 5 heterozygous mutations that include stop codons/nonsense mutations, missense mutations, splice site mutations, large deletions, and compound heterozygous mutations.

PATIENTS AND METHODS

Description of Patients and Clinical History

Nine patients in whom histological examination of small intestine mucosa confirmed the diagnosis of MVID were included in the present study. Patients 1 to 6 were collected from a larger patient cohort that received a bowel transplant via the Liver/Small Bowel Transplant Program of the University of Nebraska Medical Center. Patient 1 is a 1-year-old Hispanic boy with early-onset MVID of reported nonconsanguineous parents. His brother died of MVID at 21 months of age. Patient 2 is a 3-year-old Hispanic girl with early-onset MVID of consanguineous parents (first cousins). She has 2 sisters with MVID, 1 of which is patient 3. Patient 3 is a 5-year-old Hispanic girl with early-onset MVID of consanguineous parents (first cousins). She has 2 sisters with MVID, 1 of which is patient 2. Patient 4 is a 1-year-old Navajo boy with early-onset MVID of related parents (died of sepsis). Patient 5 is a 12-year-old Navajo girl with early-onset MVID of related parents (died of sepsis with multiorgan system failure); she has 2 healthy sisters. Patient 6 is a 0-year-old white girl with early-onset MVID of reported nonconsanguineous parents (died of sepsis from aspergillis and continuing acute rejection). She has 1 healthy sibling and 2 siblings died of unknown causes. Patient 7 is a 1-year-old Polish white girl with early-onset MVID of reported nonconsanguineous parents. Patient 8 was a 5-year-old Moroccan boy with early-onset MVID of consanguineous parents (first-degree cousins). Patient 9 is a 5-year-old Dutch white boy of unrelated parents who was diagnosed as having late-onset MVID. Unaffected parents and siblings of patients 7, 8, and 9 were also recruited and, after informed consent, saliva samples were collected and genomic DNA was extracted. In addition, available duodenal tissue from patients 8 and 9 and age-matched normal control patients was obtained and processed for immunohistochemistry. Moreover, two 2-year-old Dutch girls (twins) of nonconsanguineous parents who presented with severe secretory diarrhea and nutrient malabsorption directly after birth, but did not display the diagnostic light and electron microscopical hallmarks of MVID, were included in the study (patients 10 and 11). Written consent was obtained for all of the patients. The present study has been reviewed and approved by the University Medical Center Groningen review board.

DNA and RNA Isolation

DNA was isolated from peripheral blood samples using standard laboratory procedures. DNA and RNA from saliva was also collected and isolated (Oragene DNA and Oragene RNA, DNA Genotek Inc, Ottawa, Canada). RNA from in liquid nitrogen snap-frozen biopsy samples was isolated after homogenization using 1-mm glass beads using Trizol (Invitrogen, Carlsbad, CA). Concentration and purity were determined with NanoDrop ND-1000 (Isogen Life Science, De Meern, the Netherlands).

RT-PCR

Real-time polymerase chain reaction (RT-PCR) was performed for the quantification of MYO5B. RNA was isolated from duodenum biopsies of 12 controls and patients 8 and 9. cDNA was generated with a High Capacity cDNA Archive kit (Applied Biosystems, Foster City, CA) using 1 μg total RNA. Primers were designed with Primer Express version 3 (Applied Biosystems); RT_MYO5Bfor: TTGGAAGTGTGGCGATTCAG; RT_MYO5Brev: GCAGTCGGCAGAAGTTGCTT. For β-glucuronidase expression, we used a TaqMan Pre-Developed Assay (Applied Biosystems). Reactions consisted of 1x SYBR Green (or Universal) PCR Mastermix, 1 mmol/L of each primer and 1 μL cDNA. Cycling conditions were 50°C for 2 minutes, 95°C for 10 minutes, and 40 cycles of 95°C for 15 seconds and 60°C for 1 minute. Results were analyzed using SDS version 2.3 (Applied Biosystems).

Sequencing

The MYO5B coding region and splice sites were PCR amplified and directly sequenced in probands. Their relatives and approximately 50 control individuals (∼100 chromosomes) were screened for the detected mutations. Primers for PCR amplification (Supplemental Digital Content Tables 1 and 2 show primers used for amplification of genomic DNA and cDNA, respectively; http://links.lww.com/MPG/A23 and http://links.lww.com/MPG/A24) were designed using Primer321 on the genomic sequence of MYO5B (NC_000018.8) and its mRNA (NM_001080467). The PCR reaction was performed with 50-ng genomic DNA in 20-μL reaction volume, which included 1x PCR buffer-A (GE Healthcare, Piscataway, NJ), 2.5 mmol/L deoxynucleoside triphosphates, 1 mmol/L primers (Eurogentec, San Diego, CA), 0.5 U Thermus aquaticus polymerase (GE Healthcare). For exons 1, 2, 12, 17, and 18, PCR reaction was performed with 150 ng of genomic DNA in a 25-μL reaction volume with 1x PCR buffer-B made of 0.1 mol/L Tris-HCl (pH 8.8), 0.1 mol/L MgCl2, 0.01 mol/L mercaptoethanol, and 0.05 mol/L ethylenediaminetetraacetic acid/0.1 mol/L (NH4)2SO4. The PCR conditions differed with respect to the annealing temperatures and buffers used. Initial denaturation at 95°C for 5 minutes (4 minutes buffer-B); 40 cycles (33 cycles Buffer-B) of denaturation at 95°C for 30 seconds (1 minute buffer-B), annealing for 30 seconds (1 minute buffer-B), and extension at 72°C for 30 seconds (2 minutes buffer-B); final extension at 72°C for 5 min (7 minutes buffer-B). PCR products were purified (37°C for 15 minutes, 80°C for 15 minutes) with ExoSap-IT (USB, Cleveland, OH) and Sephadex columns. Sequencing reactions were performed using BigDye terminator mix (Applied Biosystems). Sequences were read on a 3730 DNA analyzer and 3130 Genetic analyzer (Applied Biosystems), and we aligned sequencing data with control and reference sequences using ContigExpress software (Invitrogen, Carlsbad, CA).

Deletion Detection

To detect large deletions in MYO5B in patient 7, real-time quantitative PCR was used to determine copy numbers of the exons. Reactions consisted of 1x SYBR Green PCR Mastermix (Applied Biosystems), 1 mmol/L of each primer (Supplemental Digital Content Table 2) and 25 ng of genomic DNA. Cycling conditions were 50°C for 2 minutes, 95°C for 10 minutes, and 40 cycles of denaturation at 95°C for 15 seconds, and annealing for 1 minute.

Immunohistochemistry

Duodenal biopsies of patients 8 and 9 with MVID and age-matched controls were fixed in paraffin and cut in 3-μm thick sections. Slides were dried overnight at 60°C and deparaffinized in xylol-100%-96%-70% ethanol and demiwater. Epitopes were retrieved by protease digestion or in citric acid pH 6.0 (autoclaved; 5 minutes, 120oC). Endogenous peroxidase was deactivated with 3.5% H2O2. Following blocking of nonspecific binding sites in 4% normal goat serum, slides were incubated with primary antibodies, washed, and incubated with appropriate horseradish peroxidase–conjugated secondary antibodies. Diaminobenzidine was used as a substrate for peroxidase. Hematoxylin was used to stain the nuclei. Slides were dehydrated with ethanol, dried, and mounted. Antibodies used were polyclonal antibodies raised against a synthetic peptide derived from the C-terminal hypervariable region of the human Rab11a sequence (Zymed Laboratories Inc); polyclonal antibodies against Rip11/FIP5 (13); polyclonal antibodies raised against a synthetic peptide corresponding to C- or N-terminal residues (amino acids 1093–1112 or 23–41, respectively) of human myosin Vb (Antagene Inc; 60B923) that recognizes a single band of the appropriate molecular mass of ∼214 kDa on Western blot; and horseradish peroxidase–conjugated donkey anti-rabbit, sheep anti-mouse antibodies (GE Healthcare). Immunohistochemistry images show villus cells.

RESULTS

Eight New MYO5B Mutations Associated With Nine Microvillus Inclusion Disease Patients

MYO5B is composed of 40 coding exons, which were separately amplified and subjected to sequence analysis. All 11 patients were included in the mutation analysis by direct sequencing of the entire gene in both forward and reverse directions. Six patients revealed homozygous mutations. Patient 6 revealed 1 heterozygous change, whereas patients 7 and 9 carried compound heterozygous mutations (Table 1). Patients 10 and 11, who presented with nutrient malabsorption and intractable secretory diarrhea after birth but were not diagnosed as having MVID, did not reveal MYO5B mutations.

TABLE 1
TABLE 1:
Summary of MYO5B mutations associated with MVID as reported in the present study

Patient 1 carries a homozygous nonconservative missense mutation in exon 8 (c.946G>A, p.Gly316Arg), which replaces a small aliphatic glycine (conserved in myosin Va and Vc; Supplemental Digital Content Figure 1, available at http://links.lww.com/MPG/A25) with a large and charged arginine in the protein's conserved head domain region. In patients 2 and 3, we found a shared homozygous deletion in exon 19 (c.2330_del G; Supplemental Digital Content Figure 2, available at http://links.lww.com/MPG/A26). This mutation disturbs the reading frame and leads to a premature stop codon (p.Gly777AsnfsX6; Supplemental Digital Content Figure 3, available at http://links.lww.com/MPG/A27) in the first calmodulin-binding IQ1 motif of myosin Vb. Any resultant protein will therefore not be able to dimerize and function as a processive motor protein, and lacks the entire cargo-binding tail domain. Patients 4 and 5 are homozygous for a nonconservative missense mutation in exon 16 (c.1979C>T, p.Pro660Leu). This mutation was recently described (11) in 7 Navajo patients with MVID. In patient 6, we found 1 heterozygous mutation in exon 19, which results in a premature stop codon (c.2246C>T, p.Arg749X) in the head domain of myosin Vb (p.Arg749 is conserved in myosin Va and Vc; Supplemental Digital Content Figure 4, available at http://links.lww.com/MPG/A28). Resultant protein will not be able to dimerize and function as a processive motor protein and lack the entire cargo-binding tail domain.

For patients 7, 8, and 9, we also obtained DNA samples from unaffected siblings and/or parents. Patient 7 reveals a compound heterozygous mutation, which includes a paternal allele with a nonconservative asparagine-to-serine (c.1367A>G, p.Asn456Ser) substitution in exon 11 of the head domain (Fig. 1A, B, D) (conserved in myosin Va and Vc; Supplemental Digital Content Figure 5, available at http://links.lww.com/MPG/A29), together with a missense variant p.Met1688Val (c.5062A>G) in exon 37 (p.Met1688 is substituted in MYO5A and MYO5C; Supplemental Digital Content Figure 6, available at http://links.lww.com/MPG/A30). p.Met1688Val represents an infrequent polymorphism because it was found in Polish and Dutch controls with allele frequencies of 5.8% (6/104) and 1.7% (2/116), respectively. When searching for a maternally transmitted mutation in patient 7, a Mendelian inconsistency in the inheritance of the exon 11 variant c.1367A>G was observed: the mother appeared to be homozygous A/A, whereas the patients were homozygous G/G (Fig. 1A). This could point toward a maternal transmission of a deletion. Using RT-PCR to determine the copy number of the MYO5B gene, we found that the maternal allele in patient 7 contained a deletion involving exons 2 to 12 of MYO5B (Fig. 1C), rendering any protein formed incapable of binding actin and function as a motor protein. Sequencing of MYO5B in patient 8 revealed a homozygous stop codon in exon 33 (c.4366C>T, p.Gln1456X) (Fig. 1A). This removes the terminal Rab11a-binding sites (1799–1814) (Fig. 1B), which only functions in unison with the more proximal Rab11a-binding site (1400–1415). Sequencing of MYO5B in patient 9 showed that this patient is a compound heterozygote carrying a de novo nonconservative substitution mutation in exon 12 (c.1540T>C, p.Cys514Arg), and a maternally derived mutation in intron 33 (c.4460-1G>C) that destroys the canonical splice acceptor site (Fig. 1A, B). Intron 33 harbors 3 clusters of potent candidate cryptic splice acceptor sites (Supplemental Digital Content Figure 7, available at http://links.lww.com/MPG/A31). PCR on the patient's intestinal cDNA with primers for intron 33 and exon 35 demonstrated retention of >100 bp of intron 33 immediately upstream of exon 34. This “extended exon 34” contains 9 stop codons, at least 1 in each of the 3 reading frames (Supplemental Digital Content Figure 8, available at http://links.lww.com/MPG/A32). The p.Cys514 residue forms part of the helix-turn-helix motif in the motor domain that is associated with actin binding (14–17) (Fig. 1E). All of the mutations identified in the present study are listed in Table 1. The position of all of the mutated residues in the crystal structure of the myosin Vb head domain are depicted in Fig. 1E.

FIGURE 1
FIGURE 1:
Identification and mutation analysis of MYO5B in patients with microvillus inclusion disease (MVID) and parents/siblings. (A) Pedigrees in 3 families with patients 7, 8, and 9 with MVID (designated P7, P8, and P9). The c.1540T>C substitution in patient 9 is a de novo mutation on the paternal chromosome (indicated with asterisk). Haplotype analysis in patient 7 and her family was indicative of a deletion in MYO5B on the maternal chromosome. The MYO5B variants tested were (top to bottom): c.1367A>G (exon 11), c.3276+11 (rs2276176, intron 24), c.4222-73 (rs490648, intron 31), c.4315+5 (rs488890 intron 32), c.5062A>G (exon 37), c.5313+72(rs621101, intron 38). (B) MYO5B mutations identified in patients with MVID. (C) Deletion mapping results of patient 7. The maternally derived deletion, spanning exons 2 to 12, was determined by quantitative polymerase chain reaction using DNA from the proband's brother (2 copies for each exon) to normalize the signal. (D) Ribbon diagram of the nucleotide-free structure of the motor domain of chicken myosin V (rigor-like/strong actin-binding state, Protein Data Bank code: 1OE9), indicating the locations of mutated residues (figure prepared with PyMOL [DeLano Scientific LLC]).

Resequencing mutation-containing exons revealed that none of the identified mutations were detected in 50 ethnically matched controls or have been reported as known variants (in HapMap, dbSNP, and the 1000 genome database), unless stated otherwise.

MYO5B Mutations Affect the Expression and Function of the Myosin Vb Protein in MVID Enterocytes

We analyzed the expression levels of myosin Vb mRNA from the biopsies of patients 8 and 9 by RT-PCR and compared these with control patients. In patient 8, myosin Vb mRNA expression was reduced by 50% when compared with 14 non-MVID control patients (Fig. 2A), which is in agreement with the identified nonsense mutation p.Gln1456X, which is predicted to result in nonsense-mediated RNA decay (18). In patient 9, myosin Vb mRNA levels were comparable with controls (Fig. 2A).

FIGURE 2
FIGURE 2:
Real-time polymerase chain reaction detection of myosin Vb mRNA and immunohistochemical labeling of myosin Vb. (A) RNA was extracted from small intestinal biopsies of microvillus inclusion disease (MVID) patients 8 and 9 (designated P8 and P9) as described in Patients and Methods, and relative myosin Vb mRNA levels were determined with real-time polymerase chain reaction. (B) Small intestinal biopsies of patients 8 and 9 with MVID and age-matched controls were labeled with antibodies against human myosin Vb. Negative (nonimmune first antibody) control staining is shown. The accumulation of myosin Vb in the apical cytoplasm in control enterocytes (arrows) is lost in MVID enterocytes.

We also analyzed the cellular expression pattern of myosin Vb protein in duodenal biopsies of patients 8 and 9 and age-matched controls. The myosin Vb protein is present in the villus enterocytes and mainly concentrated at their apical aspect below the brush border of control enterocytes (Fig. 2B, arrow). In contrast, no or little specific myosin Vb signal was detected in the enterocytes of patients 8 and 9, respectively (Fig. 2B). The MYO5B mutations did not involve residues that were used in the synthetic peptides to generate these antibodies, and the antibodies should recognize the mutant protein if present. The lack of clear myosin Vb signal in MVID enterocytes may reflect the absence of the protein (in accordance with the reduced myosin Vb mRNA levels in patient 8; as given above), and/or may reflect a dispersion of remaining myosin Vb protein throughout the cells, rendering myosin Vb below the detection limit.

Myosin Vb regulates the subcellular positioning of recycling endosomes by binding to small GTPase Rab proteins such as Rab11a at the cytosolic surface of endosomes and attaching these endosomes to and moving them along actin filaments (19–24). Alterations in the typical spatial organization of recycling endosomes in MVID enterocytes can therefore be used as a read-out for altered myosin Vb function. To address this, the expression and distribution of recycling endosome-associated proteins Rab11a (19–24) and the Rab11a effector protein FIP5 (Rip11) (13) was investigated. The recycling endosome-associated proteins Rab11a and FIP5 (also known as Rip11) accumulate just below the enterocyte brush border close to the apical membrane in control duodenal tissue, similar to myosin Vb (Fig. 3A, arrows). In contrast, in MVID enterocytes, Rab11a and FIP5 did not accumulate in the apical region and instead, no specific staining pattern (compare to negative antibody control) could be observed (Fig. 3A). Sequence analysis revealed single nucleotide polymorphisms but no functional mutations in the coding regions of the genes RAB11A and RAB11FIP5 in patients 8 and 9 (summarized in Fig. 3B). Early endosomal antigen 1, a marker of early sorting endosomes and typically excluded from myosin Vb-positive recycling endosomes, and the late endosome- and lysosome-associated protein LAMP-1 displayed comparable staining patterns in controls and MVID patient 9 (Fig. 4, arrows). The distribution of the Golgi complex was also apparently unaffected in MVID enterocytes, although it appeared somewhat more concentrated in the supranuclear region (Fig. 4).

FIGURE 3
FIGURE 3:
Immunohistochemical labeling and variant analysis of Rab11a and FIP5(/Rip11). (A) The distribution of Rab11a and FIP5(/Rip11) in control and microvillus inclusion disease enterocytes of patients 8 and 9 (designated P8 and P9) is shown. Negative (nonimmune first antibody) staining is shown (antibody control). Note that the accumulation of Rab11a in the apical cytoplasm of control enterocytes (arrows) is lost in microvillus inclusion disease enterocytes. (B) Sequence variants in the genes RAB11A and RAB11FIP5.
FIGURE 4
FIGURE 4:
Immunohistochemical labeling of early endosome-, late endosome/lyosome-, and golgi-associated proteins. The subcellular distributions of giantin, EEA1, and LAMP-1 in control and microvillus inclusion disease enterocytes (patient 9 [P9]) are shown. Negative (nonimmune first antibody) control staining is shown (antibody control).

DISCUSSION

We have analyzed the sequence of MYO5B in 9 patients with MVID and identified 8 new mutations (∼25% of all reported mutations) including a large deletion, a single nucleotide deletion, 2 missense mutations, and 1 nonsense mutation. We also report 2 additional compound heterozygous MYO5B mutations. In 2 patients, we found a homozygous missense mutation that has been described previously (11). Our study adds 8 mutations to the 25 earlier reported by Müller et al (24 mutations/21 patients) (10,12) and Erickson et al (1 mutation shared by 7 Navajo patients) (11), yielding a total of 33 distinct MYO5B mutations in 37 patients with MVID who have been identified to date. With our data, providing 25% of all reported MYO5B mutations and patients, we make a first analysis of the current MYO5B mutation spectrum. Of the 33 thus far published MYO5B mutations, 24 are localized in the N-terminal head domain that includes actin-binding and adenosine triphosphate catalytic sites, 2 in calmodulin-binding IQ motifs that form the light chain–binding lever arm domain, 1 in a potential coiled-coil region that mediates the association of the heavy chain into dimers, and 6 are localized in the cargo-binding globular tail domain. All of the MYO5B mutations are distinct from those reported in MYO5A and other nonconventional myosins. Furthermore, the reported heterozygous mutations are exclusively found in white patients (Poland, Ireland, France, and the United States), and include at least 1 nonsense mutation or large deletion. Interestingly, all but 1 missense mutation clusters in the myosin Vb head domain, whereas the nonsense, splice site, and deletions/insertions are found randomly in the motor, lever arm, and tail domain. Although some of the mutations are predicted to result in nonsense-mediated RNA decay (eg, the homozygous p.Gln1456X mutation in patient 8 and the c.4460-1G>C mutation in patient 9), which is supported by the observed reduction in myosin Vb mRNA levels in patient 8, other MYO5B mutations involve residues that are important for the function of the myosin Vb protein. Indeed, the N456 residue mutated in patient 7, for instance, is part of a set of conserved motifs shared in all myosins that participate in coupling changes in the adenosine triphosphatase active site (P-loop and switch I) to conformational changes in the actin-binding and force-generating domains, and proposed to have a pivotal role in motor function as mediator of allosteric communication (14–17).

We demonstrate that a nonsense MYO5B mutation correlates with reduced myosin Vb mRNA expression, and that MYO5B mutations correlate with an aberrant cellular expression pattern of the myosin Vb protein. A main function of myosin Vb is to regulate the subcellular distribution and positioning of recycling endosomes. It does so by interacting with small GTPase Rab proteins such as Rab11a at the cytosolic surface of recycling endosomes and coupling these endosomes to and positioning them along actin filaments. We demonstrate that the typical and myosin Vb–controlled accumulation of Rab11a- and FIP5-positive recycling endosomes in the apical cytoplasm of the cells is abolished in MVID enterocytes. These data are indicative of an altered myosin Vb function in MVID enterocytes. It should be noted that our conclusions are based on 2 cases of this rare disease, and that future experiments are necessary to further consolidate these. Further in-depth analysis of MYO5B mutations and their molecular and cell biological consequences are warranted to expand our understanding of how they are related to MVID pathogenesis.

It is encouraging that all of the patients with MVID (except for one (10)) who have been screened thus far carry mutations in their MYO5B gene, and the discovery of additional compound heterozygous mutations by Ruemmele et al (12) and us (the present study) significantly strengthens the correlation between MYO5B and MVID. This correlation is further supported by a recent study (12) in which knockdown of myosin Vb in human epithelial colorectal adenocarcinoma (Caco-2) cells recapitulates most of the cellular phenotypes of MVID, and by our observation that MYO5B mutations were not found in 2 patients who presented with secretory diarrhea and malabsorption after birth but were not diagnosed as having MVID. A firm association of MYO5B mutations with MVID is a major advance in the diagnosis of this rare but fatal disease, in which variable phenotypes are seen among patients. It will also facilitate reliable genetic counseling and prenatal screening. Because total parenteral nutrition and bowel transplants are, at best, not permanent solutions for treating this devastating disease, the continuing identification of MYO5B mutations will pave the way for the development of alternative therapeutic strategies.

Acknowledgments

We thank the patients, their parents, and siblings, and the transplantation teams of the UMC Groningen and the Children's Memorial Health Institute in Warsaw. We thank Carolien Gijsbers (Juliana Children's Hospital, The Hague, the Netherlands) and Marc Benninga (Academic Medical Center Amsterdam, the Netherlands) for the shared care of the 2 Dutch patients with MVID, and the Dutch Digestive Diseases Foundation (MLDS) for supporting the national collaboration for patients with intestinal failure. We thank Julius Baller and Mathieu Platteel for expert technical assistance, and Hilda Keuning for analysis of histological findings. Control DNA samples were provided by Yvonne Vos (Dutch and Moroccan) and Anna Rybak (Polish). Hanna Romanowska collected DNA samples from the Polish patient and her family. We are indebted to Henkjan Verkade for critically reading the manuscript.

REFERENCES

1. Phillips AD, Jenkins P, Raafat F, et al. Congenital microvillous atrophy: specific diagnostic features. Arch Dis Child 1985; 60:135–140.
2. Phillips AD, Schmitz J. Familial microvillous atrophy: a clinicopathological survey of 23 cases. J Pediatr Gastroenterol Nutr 1992; 14:380–396.
3. Cutz E, Rhoads JM, Drumm B, et al. Microvillus inclusion disease: an inherited defect of brush-border assembly and differentiation. N Engl J Med 1989; 320:646–651.
4. Sherman PM, Mitchell DJ, Cutz E. Neonatal enteropathies: defining the causes of protracted diarrhea of infancy. J Pediatr Gastroenterol Nutr 2004; 38:16–26.
5. Goulet O, Ruemmele F, Lacaille F, et al. Irreversible intestinal failure. J Pediatr Gastroenterol Nutr 2004; 38:250–269.
6. Ruemmele FM, Schmitz J, Goulet O. Microvillous inclusion disease (microvillous atrophy). Orphanet J Rare Dis 2006; 1:22.
7. Iancu TC, Mahajnah M, Manov I, et al. Microvillous inclusion disease: ultrastructural variability. Ultrastruct Pathol 2007; 31:173–188.
8. Ameen NA, Salas PJ. Microvillus inclusion disease: a genetic defect affecting apical membrane protein traffic in intestinal epithelium. Traffic 2000; 1:76–83.
9. Michail S, Collins JF, Xu H, et al. Abnormal expression of brush-border membrane transporters in the duodenal mucosa of two patients with microvillus inclusion disease. J Pediatr Gastroenterol Nutr 1998; 27:536–542.
10. Müller T, Hess MW, Schiefermeier N, et al. MYO5B mutations cause microvillus inclusion disease and disrupt epithelial cell polarity. Nat Genet 2008; 40:1163–1165.
11. Erickson RP, Larson-Thomé K, Valenzuela RK, et al. Navajo microvillous inclusion disease is due to a mutation in MYO5B. Am J Med Genet Part A 2008; 146A:3117–3119.
12. Ruemmele FM, Müller T, Schiefermeier N, et al. 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; 31:544–551.
13. Prekeris R, Klumperman J, Scheller RH. Rab11/Rip11 protein complex regulates apical membrane trafficking via recycling endosomes. Mol Cell 2000; 6:1437–1448.
14. Holmes KC, Schroder RR, Sweeney HL, et al. The structure of the rigor complex and its implications for the power stroke. Philos Trans R Soc Lond B Biol Sci 2004; 359:1819–1828.
15. Coureux PD, Sweeney HL, Houdusse A. Three myosin V structures delineate essential features of chemo-mechanical transduction. EMBO J 2004; 23:4527–4537.
16. Tang S, Liao JC, Dunn AR, et al. Predicting allosteric communication in myosin via a pathway of conserved residues. J Mol Biol 2007; 373:1361–1373.
17. Cecchini M, Houdusse A, Karplus M. Allosteric communication in myosin V: from small conformational changes to large directed movements. PLoS Comput Biol 2008; 4:e1000129.
18. Isken O, Maquat LE. The multiple lives of NMD factors: balancing roles in gene and genome regulation. Nat Rev Genet 2000; 9:699–712.
19. Lapierre LA, Kumar R, Hales CM, et al. Myosin vb is associated with plasma membrane recycling systems. Mol Biol Cell 2001; 12:1843–1857.
20. Swiatecka-Urban A, Talebian L, Kanno E, et al. Myosin Vb is required for trafficking of the cystic fibrosis transmembrane conductance regulator in Rab11a-specific apical recycling endosomes in polarized human airway epithelial cells. J Biol Chem 2007; 282:23725–23736.
21. Nedvetsky PI, Stefan E, Frische S, et al. A Role of myosin Vb and Rab11-FIP2 in the aquaporin-2 shuttle. Traffic 2007; 8:110–123.
22. Roland JT, Kenworthy AK, Peranen J, et al. Myosin Vb interacts with Rab8a on a tubular network containing EHD1 and EHD3. Mol Biol Cell 2007; 18:2828–2837.
23. Hoekstra D, Tyteca D, van IJzendoorn SC. The subapical compartment: a traffic center in membrane polarity development. J Cell Sci 2004; 117:2183–2192.
24. van IJzendoorn SC. Recycling endosomes. J Cell Sci 2006; 19:1679–1681.
Keywords:

apical recycling endosome; brush border; microvillus inclusion disease; myosin Vb

Supplemental Digital Content

Copyright 2011 by ESPGHAN and NASPGHAN