Golachowska, Magdalena R.*; van Dael, Carin M.L.¶; Keuning, Hilda†; Karrenbeld, Arend‡; Hoekstra, Dick*; Gijsbers, Carolien F.M.§; Benninga, Marc A.||; Rings, Edmond H.H.M.†; van IJzendoorn, Sven C.D.*
Microvillus inclusion disease (MVID, Online Mendelian Inheritance in Man no. 251850) is a congenital intestinal malabsorption disorder that presents with intractable secretory diarrhea within a few days (early onset) or weeks (late onset) of life, leading to total parenteral nutrition (TPN) dependency throughout life (1,2). MVID is associated with villous atrophy, intracytoplasmic microvillus inclusions, and intracellular accumulation of the normally brush border–enriched metalloendopeptidase CD10 and periodic acid-Schiff (PAS)–positive material in the subapical cytoplasm of intestinal absorptive cells (3,4). Other brush border proteins including sucrase-isomaltase, alkaline phosphatase, sodium-proton exchanger protein 2 and 3 (NHE-2, NHE-3), cyclic guanosine monophosphate–dependent protein kinase, cystic fibrosis transmembrane conductance regulator, and the sodium-glucose transporter 1 have been reported to be present at reduced levels at the apical brush border membrane and to accumulate in the apical cytoplasm (5). At the ultrastructural level, a variable degree of microvillus atrophy, accumulation of secretory granules, and the presence of microvillus inclusions are typically observed. It has been suggested that the diminished apical surface area, diminished surface expression of apical plasma membrane transporter systems, and consequent abnormally low uptake of nutrients (6) account for most of the clinical symptoms of MVID (7).
Recently, mutations in the MYO5B gene coding for myosin Vb were reported in all but 1 patient with MVID (8–11). MYO5B mutations were shown to correlate with a defective myosin Vb protein expression in MVID enterocytes (11). Myosin Vb is an actin-based motor protein that binds to specific small GTPase rab proteins on recycling endosomes and transports these along actin filaments to the apical plasma membrane. Indeed, the knockdown of myosin Vb or the expression of a dominant-negative myosin Vb motorless tail domain in polarized Madin-Darby canine kidney cells (12), human epithelial colorectal adenocarcinoma Caco-2 cells (10), human sub-bronchial gland Calu-3 cells (13), and rat hepatoma/human fibroblast hybrid WIF-B9 cells (14) inhibits protein recycling to the apical plasma membrane and/or results in the accumulation of resident apical plasma membrane proteins in compartments in the subapical cytosol. It is therefore plausible that impaired apical brush border membrane development and maintenance in MVID enterocytes is caused by defects in the intracellular trafficking of resident apical plasma membrane proteins to the cell surface. This is in agreement with the reported defects in apical recycling endosome organization in MVID enterocytes (11). Myosin Vb has been proposed to play a key role in apical brush border development (8,14,15).
Myosin Vb is ubiquitously expressed in all of the epithelial tissues. It is therefore conceivable that the pronounced structural and functional aberrations as observed in the enterocytes of patients with MVID also occur in other polarized epithelial cells. In support of this proposition, the spectrum of symptoms in this syndrome is suggested not to be exclusive for the small intestine, and microvillus inclusions in the stomach and colon, basolateral plasma membrane inclusions in gallbladder epithelium, and poorly defined microvillus-bearing vesicles in renal tubular epithelial cells have been reported in some patients with MVID (16). Remarkably, multiorgan clinical symptoms are not typically reported in patients with MVID; however, long-term TPN may give rise to hepatic and biliary disease, metabolic bone disease, and renal complications (17). TPN-associated renal complications typically involve a decreased glomerular filtration rate. TPN-associated impairment of renal tubular functions has been reported in some adults but not in children (18). In the present study, we report for the first time 2 MYO5B mutation–carrying patients with MVID that developed proximal tubular renal dysfunction characterized by aminoaciduria, reduced tubular reabsorption of phosphate, glucosuria tubular acidosis, and hypophosphatemic rickets, all characteristics of renal Fanconi syndrome. Renal Fanconi syndrome can be caused by inborn errors of metabolism, secondary to primary Mendelian diseases such as cystinosis, or acquired through exposure to toxic agents. Renal Fanconi syndrome affects the apical reabsorption of various substances and is considered to be a general defect in the function of the proximal tubules. Proximal tubular cells in the kidney contain a microvillus membrane, and loss of apical microvilli has been reported in a child with renal Fanconi syndrome associated with lysinuric protein intolerance (19). The aim of the present study was, therefore, to determine whether MYO5B mutations in these patients correlate with similar apical plasma membrane defects in renal tubular epithelial cells as observed in the intestine.
Description of Patients and Clinical History
Biopsy material from 2 patients diagnosed with MVID and age-matched non-MVID patients were included in the present study. Patient 1 was a Dutch-Moroccan boy from consanguineous parents who was born at term and admitted at the hospital 3 days after birth because of excessive diarrhea and dehydration. Introduction of oral feeding failed as a result of the progression of diarrhea, and TPN support was started. Jejunal biopsy showed almost total villous atrophy and microvillus inclusions, confirming the diagnosis of MVID. MYO5B gene sequencing revealed a homozygous stop codon in exon 33 (c.4366C>T, p.Gln1456X) (11).
Patient 2 was a white boy from nonconsanguineous parents who was breast-fed from birth on. At the age of 2 months, he was admitted to the hospital because of prolonged icterus. During the course of being admitted, he developed diarrhea, which progressed in severity. Enteral feeding failed as a result of the progression of diarrhea upon introduction, and TPN was started. Duodenal and colon biopsies showed villous atrophy and microvillus inclusions, confirming the diagnosis of MVID. The sequencing of the MYO5B gene revealed compound heterozygous mutations including 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) (11).
Kidney tissue was collected after presentation of renal Fanconi syndrome using a MAGNUM biopsy gun (Bard Biopsy Systems, Tempe, AZ) with a 16-gauge needle (patient 1: 1 year 11 months, patient 2: 2 years 3 months). Control kidney biopsies were taken from noncarcinoma regions of a kidney following nephrectomy because of renal carcinoma. Intestinal samples were taken after removal of the diseased organ during the transplantation procedure (patient 1: aged 5 years 6 months, patient 2: aged 4 years 9 months) from duodenum, jejunum, jejunum/ileum, ileum, and colon. Control small intestinal biopsies were from age-matched patients without MVID with diarrhea of unknown origin with spontaneous recovery. Colon controls were normal proximal colon sections of an age-matched patient diagnosed as having Hirschprung disease. Biopsies were processed for electron and light microscopy using standard methods (11). The present study was reviewed and approved by the University Medical Center Groningen review board. Full informed consent was obtained from the patients’ parents/caregivers.
Kidney and intestinal biopsies of patients 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 during xylol/ethanol washing steps. Epitopes were retrieved by protease K digestion (Sigma-Aldrich, St Louis, MO) or with 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. These included monoclonal mouse antibodies against CD10 (clone 56C6, Monosan, SanBio BV, Uden, the Netherlands), polyclonal antibodies raised against a synthetic peptide derived from the C-terminal hypervariable region of the human Rab11a sequence (Zymed laboratories Inc, South San Francisco, CA), and 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 (60B923; Antagene Inc, San Francisco, CA). Samples were then washed and incubated with secondary horseradish peroxidase–conjugated donkey anti-rabbit or sheep anti-mouse antibodies (GE Healthcare, Giles, UK). Diaminobenzidine was used as a substrate for peroxidase and the nuclei were stained with hematoxyline. Slides were dehydrated with ethanol, dried, and mounted. Immunohistochemical labeling for CD10 and PAS was performed in the automated Ventana BenchMark Ultra system (Ventana Medical Systems Inc, Tucson, AZ) according to standard manufacturer's protocol (11).
Freshly obtained biopsy samples were processed for electron microscopy as described by Szperl et al (11). Briefly, samples were fixed in 2% glutaraldehyde in phosphate buffer, rinsed in 6.8% sucrose in phosphate buffer, and postfixed in a solution of 1% osmium tetroxide in 0.1 mol/L sodium cacodylate buffer containing 11.2% potassium ferrocyanide. Samples were dehydrated with ethanol and processed according to standard procedures upon embedding. Coupes were contrasted with uranyl acetate and lead citrate.
β2-Microglobulin, creatinin, and carnitin were measured using standard techniques. Kidney function tests were performed based on creatinin clearance measured in 24-hour urine collections, with iothalamate and para-iaminohippurate for renal plasma flow, and in patient 1 with inulin. Amino acid profiles in plasma and urine were measured by cation exchange column chromatography coupled to postcolumn ninhydrine derivatization and spectrophotometric detection on a Biochrome 20 amino acid analyzer (Biochrome, Cambridge, UK) according to the manufacturer's protocols. Amino acid fractional resorptions were calculated and normalized to creatinine.
Renal Fanconi Syndrome in 2 Patients With MVID
Two Dutch patients diagnosed as having MVID, showing the typical diagnostic features such as intestinal brush border atrophy and microvillus inclusions at the ultrastructural level (Fig. 1) and carrying MYO5B mutations (see Methods), developed renal Fanconi syndrome.
Patient 1 was screened for renal tubular dysfunction during evaluation for intestinal transplantation. Analysis of fractional excretion of sodium, tubular reabsorption of phosphate, β2-microglobulin clearance, and generalized aminoaciduria indicated renal Fanconi syndrome (Fig. 2, Table 1). No disturbance in the glomerular function was measured (Table 1). The clinical picture and management resembled that of patient 2 (see below). At the age of 5 years, patient 1 received a combined small intestine and colon transplant, and renal Fanconi syndrome spontaneously resolved after enteral feeding was fully restored.
Patient 2 was hospitalized for the evaluation of growth failure during which excessive urinary losses of phosphate were observed without rapid catch-up of weight gain. Further examination showed tubular resorption of phosphate of 41%, hypercalciuria (Ca/creatinin ratio 5:6), proteinuria (1.7 g/L), generalized aminoaciduria (Fig. 2, Table 1), and severe rickets (Fig. 2), which are characteristics of renal Fanconi syndrome. No disturbances in the glomerular function were measured (Table 1). Phosphorus in the parenteral nutrition was stepwise increased up to a maximum of 21 mmol/L, so as not to risk crystallization. It was decided to add oral phosphate (NaKP 1 mL = 1 mmol phosphate) as well. With the addition of oral phosphate, serum phosphate gradually increased. The parenteral and oral supplementation of phosphate resulted in a decrease of alkaline phophatase, a normalization of the dual-energy x-ray absorptiometry scan results for bone density, and resolution of the signs of rachitis on x-ray. Also, growth was established; however, laboratory results indicated the persistence of renal Fanconi syndrome. The latter resolved gradually after the patient received a multiorgan (small intestine, large intestine, pancreas, and liver) transplant at the age of 5 years, and enteral feeding was fully restored.
Laboratory analyses including calcium, phosphate, pH, bicarbonate, alkaline phosphatase, vitamin D, parathyroid hormone, and dual-energy x-ray absorptiometry scan results are depicted in Tables 2 to 4 and Figure 3.
Apical Brush Border Membrane and Recycling Endosome Organization Are Unaffected in Kidney Epithelial Cells
When evaluating overall renal tubular histology, dilated tubules were observed in patient 2 when compared with those of age-matched non-MVID control patients and patient 1 (Fig. 4A). PAS staining of kidney biopsies from age-matched non-MVID control patients showed a clear staining of the apical brush border membrane (Fig. 2A, black arrows), which appeared fuzzy because of the microvilli-rich and therefore extensively folded brush border membrane. In addition, a sharp outline of the tubular structure reflecting the basement membrane is observed (Fig. 4A, arrowheads). A similar staining pattern is observed in the renal tubules of both patients with MVID (P1 and P2). No intracellular PAS staining is detected in the proximal tubular epithelial cells of either patient with MVID. At the ultrastructural level, proximal tubular epithelial cells of both patients with MVID revealed a well-developed apical brush border (Fig. 4B1 and 4B2, black arrows). No microvillus atrophy or presence of microvillus inclusions was observed. It should be noted that many electron-dense deposits of unknown origin (Fig. 4B3; arrowhead) were observed in patient 1, a feature also frequently observed in MVID enterocytes. Also, the mitochondria in the tubular cells of patient 1 appeared swollen (Fig. 4B1; white arrow). These data demonstrate that the apical brush border is not visibly affected in patients with MVID.
We next analyzed the distribution of the apical recycling endosomal system. The apical recycling endosome marker Rab11a localizes predominantly in the subapical area of the cells of control samples (Fig. 4C, control, black arrow), similar as observed in the epithelial cells of the control intestine (11); however, whereas in the enterocytes of patients with MVID Rab11a labeling is dispersed or cannot be detected (11), Rab11a displays a normal subapical localization in the proximal tubular epithelial cells of both patients with MVID, indistinguishable from control patients (Fig. 4C, arrows). These data reveal that in contrast to the strikingly impaired organization of the apical brush border and apical recycling endosomes distribution in intestinal epithelial cells, no obvious abnormalities are seen in the tubular epithelial cells of the kidney.
Apical Brush Border Membrane and Recycling Endosome Organization in Intestinal Epithelial Cells
The absence of morphological apical plasma membrane defects in proximal tubular epithelial cells in patients with MVID may suggest that the apical plasma membrane defects, as observed in the enterocytes of these patients, are the result of intestine-specific factors. To obtain further support for this suggestion, we examined the intestinal epithelial cells from all of the segments of the MVID intestine. For this, samples from explanted MVID intestines were taken from the duodenum, jejunum, ileum, and colon of both patients, fixed in paraffin, and prepared for PAS, CD10, myosin Vb, and Rab11a labeling. In the control intestine, a typically abundant PAS (Fig. 5) and CD10 (Fig. 6) staining was observed at the intestinal brush border membrane. Significant CD10 staining was also observed in the supranuclear area, possibly reflecting newly synthesized CD10 passing through the Golgi apparatus (Fig. 6). Myosin Vb (Fig. 7) and Rab11a (Fig. 8) displayed a typical subapical staining pattern, consistent with the location of apical recycling endosomes in the control samples. In striking contrast, in both patients with MVID, the PAS-positive material accumulated in the apical cytoplasm in enterocytes, and this was observed in all of the segments of the small intestine and colon (Fig. 5). Of interest, residual PAS staining at the brush border was observed in the enterocytes of patient 2, which presents late-onset MVID. In addition, patients with MVID show no (patient 1; early onset) or severely reduced (patient 2; late onset) labeling of CD10 at the brush border and extensive intracellular subapical deposits (Fig. 6) were observed. These intracellular deposits likely reflect microvillus inclusions observed with electron microscopy. The intensity of CD10 labeling decreases in the most distal part of the small intestine and is absent in the colon in both patients with MVID and in the control (Fig. 6). The staining for myosin Vb (Fig. 7) and Rab11a (Fig. 8) was below detection level in both patients with MVID in all parts of the intestine and colon tested. These data reveal that defects in apical plasma membrane apical recycling endosome organization appear along the entire horizontal axis of the MVID intestine.
We describe 2 patients with MVID, carrying MYO5B mutations, who have developed renal Fanconi syndrome with no disturbances in glomerular function. Renal Fanconi syndrome is characterized by aminoaciduria, glucosuria tubular acidosis, reduced tubular reabsorption of phosphate, and resultant hypophosphatemic rickets. Cystinosis, the most common cause of renal Fanconi syndrome in children, was not observed in these patients.
Mutations in the MYO5B gene have been associated with MVID, and loss of expression or function of the myosin Vb protein has been proposed to impair apical brush border membrane formation and cell polarity (8,14). At the cellular level, MVID and renal Fanconi syndrome have been attributed to defects in apical protein trafficking, in particular the recycling of brush border proteins via endosomes in the apical cytoplasm in kidney epithelial (20) and intestinal epithelial cells (6,11), resulting in (re)absorption defects. Apical recycling also plays an important role in apical surface development (reviewed in (15)). Apical microvillus atrophy is a prominent feature of MVID, and loss of the apical brush border has been reported in a child with renal Fanconi syndrome associated with lysinuric protein intolerance (19). Kidney biopsies taken from these patients with MVID allowed us to compare for the first time the organization of the apical plasma membrane and apical recycling endosomes in kidney versus intestinal epithelial cells.
As opposed to the pronounced intestinal epithelial brush border atrophy in the studied patients with MVID, we found that MYO5B mutations did not affect apical brush border morphology and the PAS staining pattern in renal tubular epithelial cells. In addition, in contrast to the pronounced dispersal of the apical recycling endosome system marked by Rab11a in the intestinal epithelial cells, the spatial organization of the apical recycling endosome system was unaffected in renal tubular epithelial cells of patients with MVID. Analysis of the brush border morphology and apical recycling endosome distribution along the intestinal horizontal axis (ie, from duodenum to colon) revealed that the entire intestine is affected in patients with MVID.
Temporal kidney dysfunction has been reported in patients with MVID (21,22). Long-term TPN has been reported to cause renal complications including glomerular changes, which were partially related to infections and nephrotoxic medications or to the dehydration (18,23,24); however, tubular disturbances, as measured by β2-microglobinuria, have not been associated with TPN (17,23), and disturbances in glomerular function were not observed in the patients with MVID (Table 1). It is unclear whether hypophosphatemic rickets and renal insufficiency are a result of complications from stool losses in MVID as proposed by Kagitani et al (22). The observations that the renal Fanconi syndrome declined in both patients following a bowel transplant and that the apical brush border and apical recycling endosome organization appear unaffected in the proximal tubular epithelial cells makes it unlikely that the renal Fanconi syndrome is a direct consequence of the MYO5B mutations. We can, however, not formally exclude that MYO5B mutations render the proximal tubular cells of some patients with MVID more susceptible for changes in electrolyte or fluid homeostasis, subsequently leading to failure of tubular transporters.
An important finding in the present study is that MYO5B mutations exert divergent effects on the apical membrane system and structural polarity of kidney and intestinal epithelial cells. The typical apical brush border defects presented in MVID are therefore likely to be triggered by intestine-specific factors. GTPase rab proteins and molecular motors such as myosin Vb are involved in many membrane-recycling pathways, and different protein complexes will carry different cargo (13,14,25–27). The key difference between renal and intestinal situations may thus be the cargo that is carried. We propose that the intestine-specific factors may be intrinsic (eg, gene expression/cell turnover rate) and/or external (eg, the gut-lumen environment). Membrane recycling can be influenced by metabolic and signaling pathways (28,29), which may differ between cell types. It is conceivable that these factors preclude the intestinal epithelial cells from adjusting or mending the effects of MYO5B mutations on the apical membrane system and structural polarity.
Brush border atrophy in the intestine is believed to diminish the absorptive surface and expression of proteins involved in digestion and absorption at the apical plasma membrane and is, in this way, responsible for the malabsorption in MVID and consequent lifelong dependency on TPN. With intestine transplantation as the present standard treatment for MVID, which is suboptimal at best, the identification of intestine-specific factors that trigger brush border defects in MVID enterocytes may provide new targets and open new avenues for the development of alternative therapeutic strategies to combat this devastating disease.
The authors thank the patients, their parents, and the transplantation teams of the University Medical Center, Groningen. We thank the Dutch Digestive Diseases Foundation (MLDS) for supporting the national collaboration for patients with intestinal failure. We thank Dirk-Jan Reijngoud for expert laboratory analyses and Julius Baller for expert technical assistance. We thank Marcory van Dijk for help with the evaluation of kidney biopsies.
1. Cutz E, Rhoads JM, Drumm B, et al. Microvillus inclusion disease: an inherited defect of brush-border assembly and differentiation. N Engl J Med
2. Davidson GP, Cutz E, Hamilton JR, et al. Familial enteropathy: a syndrome of protracted diarrhea from birth, failure to thrive, and hypoplastic villus atrophy. Gastroenterology
3. Groisman GM, Amar M, Livne E. CD10: a valuable tool for the light microscopic diagnosis of microvillous inclusion disease (familial microvillous atrophy). Am J Surg Pathol
4. Ruemmele FM, Jan D, Lacaille F, et al. New perspectives for children with microvillous inclusion disease: early small bowel transplantation. Transplantation
5. Phillips AD, Schmitz J. Familial microvillous atrophy: a clinicopathological survey of 23 cases. J Pediatr Gastroenterol Nutr
6. Ameen NA, Salas PJ. Microvillus inclusion disease: a genetic defect affecting apical membrane protein traffic in intestinal epithelium. Traffic
7. Bijlsma PB, Kiliaan AJ, Scholten G, et al. Increased paracellular macromolecular transport and subnormal glucose uptake in duodenal biopsies of patients with microvillus inclusion disease. Comparisons to other chronic diarrhea patients and to nondiarrhea patients. Ann N Y Acad Sci
8. Müller T, Hess MW, Schiefermeier N, et al. MYO5B mutations cause microvillus inclusion disease and disrupt epithelial cell polarity. Nat Gen
9. Erickson RP, Larson-Thome K, Valenzuela RK, et al. Navajo microvillous inclusion disease is due to a mutation in MYO5B. Am J Med Genet A
10. 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
11. Szperl AM, Golachowska MR, Bruinenberg M, et al. Functional characterization of mutations in the myosin Vb
gene associated with microvillus inclusion disease. J Pediatr Gastroenterol Nutr
12. Lapierre LA, Goldenring JR. Interactions of myosin Vb with Rab11 family members and cargoes traversing the plasma membrane recycling system. Methods Enzymol
13. 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
14. Wakabayashi Y, Dutt P, Lippincott-Schwartz J, et al. Rab11a and myosin Vb are required for bile canalicular formation in WIF-B9 cells. Proc Natl Acad Sci U S A
15. Golachowska MR, Hoekstra D, van IJzendoorn SC. Recycling endosomes in apical plasma membrane domain formation and epithelial cell polarity. Trends Cell Biol
16. Cutz E, Sherman PM, Davidson GP. Enteropathies associated with protracted diarrhea of infancy: clinicopathological features, cellular and molecular mechanisms. Pediatr Pathol Lab Med
17. Buchman AL. Complications of long-term home total parenteral nutrition: their identification, prevention and treatment. Dig Dis Sci
18. Moukarzel AA, Ament ME, Buchman A, et al. Renal function of children receiving long-term parenteral nutrition. J Pediatr
19. Benninga MA, Lilien M, de Koning TJ, et al. Renal fanconi syndrome with ultrastructural defects in lysinuric protein intolerance. J Inherit Metab Dis
20. Marshansky V, Ausiello DA, Brown D. Physiological importance of endosomal acidification: potential role in proximal tubulopathies. Curr Opin Nephrol Hypertens
21. Gambarara M, Diamanti A, Ferretti F, et al. Intractable diarrhea of infancy with congenital intestinal mucosa abnormalities: outcome of four cases. Transplant Proc
22. Kagitani K, Yamamoto T, Miki K, et al. Hypophosphatemic rickets accompanying congenital microvillous atrophy. J Bone Miner Res
23. Buchman AL, Moukarzel A, Ament ME, et al. Serious renal impairment is associated with long-term parenteral nutrition. JPEN J Parenter Enteral Nutr
24. Lauverjat M, Hadj AA, Vanhems P, et al. Chronic dehydration may impair renal function in patients with chronic intestinal failure on long-term parenteral nutrition. Clin Nutr
25. van IJzendoorn SC, Mostov KE, Hoekstra D. Role of rab proteins in epithelial membrane traffic. Int Rev Cytol
26. Roland JT, Bryant DM, Datta A, et al. Rab GTPase-Myo5B complexes control membrane recycling and epithelial polarization. Proc Natl Acad Sci U S A
27. Gardner LA, Hajjhussein H, Frederick-Dyer KC, et al. Rab11a and its binding partners regulate the recycling of the β1-adrenergic receptor. Cell Signal
28. van der Wouden JM, Maier O, van IJzendoorn SC, et al. Membrane dynamics and the regulation of epithelial cell polarity. Int Rev Cytol
29. van IJzendoorn SC, van der Wouden JM, Liebisch G, et al. Polarized membrane traffic and cell polarity development is dependent on dihydroceramide synthase-regulated sphinganine turnover. Mol Biol Cell