FIGURE SUMMARY
Title

Bi-allelic VPS16 variants limit HOPS/CORVET levels and cause a mucopolysaccharidosis-like disease

Authors
Sofou, K., Meier, K., Sanderson, L.E., Kaminski, D., Montoliu-Gaya, L., Samuelsson, E., Blomqvist, M., Agholme, L., Gärtner, J., Mühlhausen, C., Darin, N., Barakat, T.S., Schlotawa, L., van Ham, T., Asin Cayuela, J., Sterky, F.H.
Source
Full text @ EMBO Mol. Med.
Figure 1

Clinical features of the patients

Facial phenotypes of patient A at 3.5 years of age, showing coarse facial features, broad ear lobes, macroglossia, and hypertrichosis.

Radiograph of the hand of patient B at 21 months of age demonstrating rough trabecular structure and thin corticalis as signs of dysostosis multiplex.

Ovoid vertebrae visualized by spine radiographs of both patients.

T2‐weighed brain MRIs of patient A (D1–3) at 28 months of age and patient B at (D4–6) at 16 months of age, showing global brain atrophy as well as periventricular and deep white matter hyperintensities.

Additional MR investigations of patient A at 28 months of age. (E) T1‐weighed coronal images demonstrating thinning of the posterior corpus callosum and reduced volumes of the cerebellar hemispheres and vermis. (F) Magnetic resonance spectroscopy of the white matter showing a decreased N‐acetylaspartate (NAA) peak at 2.02 ppm (arrow). (G) T1‐weighed MRI of the spinal cord showing contrast enhancements around the conus medullaris (arrowheads).

May Grunwald‐Giemsa‐stained bone marrow smear from patient A with myelopoietic cells containing dense granules (arrowheads), suggestive of a lysosomal storage disorder.
Figure 2

Family pedigrees and mutation analysis

Pedigrees of the two families, indicating autosomal recessive inheritance. The genotype of the VPS16 c.2272‐18 position is indicated.?, not available for testing. (A) Family of patient A. The three sisters are healthy, but an older brother had previously died at 11 months from suspected heart failure after showing signs of developmental delay from an age of 3 months of age. (B) Family of patient B. The mother previously had two spontaneous miscarriages but the two sisters are healthy.

Position of the intronic c.2272‐18 variant in the VPS16 gene.

cDNA analysis by Sanger sequencing of leukocyte RNA from patient A and a control sample. Reads beyond the alternatively spliced breakpoint yield double reads in patient samples, corresponding to the predicted mis‐splicing and wild‐type sequences.
Figure 3

The c.2272‐18C>A mutation impairs <italic>VPS16</italic> splicing and leads to reduced levels of VPS16 and HOPS/CORVET components

The c.2272‐18C>A variant generates a new splice‐acceptor sequence 16 base pairs upstream of the normal exon 23 border, resulting in a frameshift and premature stop codon.

Levels of VPS16 transcripts in fibroblasts by qPCR. The probe of the exon 22–24 PCR spans the mis‐spliced intron‐exon border. Data were normalized to levels of POLR2A using the ∆∆Ct method.

Quantifications of VPS16 in fibroblast lysates by immunoblotting. Representative immunoblots (bottom) and summary quantifications (top) of the levels of VPS16 in patient A (n = 4) and patient B (n = 3), normalized to levels of actin (ACTB) and expressed as % of controls.

Schematic illustration of HOPS and CORVET complexes showing their subunit compositions. The Rab5‐binding CORVET subunits VPS3 and VPS8 replace the Rab7‐binding VPS39 and VPS41 of HOPS. Models adapted from Bröcker et al (2012).

Quantifications of HOPS/CORVET subunits in fibroblast lysates (top) and representative immunoblots (bottom), analyzed as in (C). (E) VPS33A, (F) VPS11, (G) VPS18.

Analysis of VPS33A and VPS11 mRNA levels in fibroblasts by qPCR. Data were normalized to levels of POLR2A using the ∆∆Ct method.

Immunoblots (left) and quantifications normalized to control (right) of VPS16, VPS33A, and VPS11 in lysates from fibroblasts transduced with a lentivirus to express VPS16 or control (empty vector).

Data information: All immunoblots were acquired using fluorescent secondaries and pseudocolored to greyscale. Stars denote unspecific bands. All data were shown as mean ±SEM of 3–4 biological replicates. Statistical comparisons were made by two‐tailed unpaired Student’s t‐tests [panels B, C (right), E (right), F, G and I] or one‐way ANOVA with Dunnett's multiple comparisons tests [panels C (left) and E(left)]. *P < 0.05; **P < 0.01; ***P < 0.001.
Figure 4

Analysis of transferrin uptake and trafficking in fibroblasts

Analysis of the early endosome marker EEA1 in fibroblasts transduced with control or VPS16‐expressing lentivirus by confocal microscopy. (A) Representative confocal micrographs. Scale bar, 20 μm. (B) Quantification of EEA1‐stained puncta number and intensities (ca 80 cells analyzed for each of n = 3 biological replicates).

Analysis of transferrin uptake in fibroblasts transduced with the indicated lentiviruses. Representative confocal micrographs of fibroblasts fed with Alexa488‐conjugated Transferrin for 30 min. Scale bar, 10 μm.

Quantifications of the number and fluorescence intensities of intracellular Transferrin puncta (n = 28–30 fields from three independent experiments). Colored horizontal bars indicate median values and whiskers 5 and 95 percentiles.

Immunoblot for Transferrin receptor (TfR) levels in the indicated fibroblasts under basal and serum‐starved conditions.

Summary quantification of Transferrin receptor levels in non‐starved fibroblasts transduced with the indicated lentiviruses, normalized to actin (ACTB), and expressed as % of controls (n = 6 biological replicates).

Analysis of intracellular Transferrin trafficking. Fibroblasts fed with fluorescently labeled Transferrin were co‐stained with LysoTracker (G) or an antibody against LAMP2 (H). Representative images (left) of fibroblasts transduced with control virus, with magnified areas demonstrating co‐localized puncta (arrowheads). Scale bars 10 μm or 2 μm (inserts). Quantifications (right) of the co‐localization between Transferrin and LysoTracker or LAMP2, respectively, expressed as Pearson correlation coefficients. Colored horizontal bars indicate the median values and whiskers 5 and 95 percentiles (n = 10 optical sections from three independent experiments).

Data information: Bar graphs represent data as mean ±SEM. Statistical comparisons between the indicated groups by one‐way ANOVA with Holm–Sidak’s multiple comparisons tests. *P < 0.05; **P < 0.01; ***P < 0.001.
Figure 5

Analysis of lysosomal distribution and function in fibroblasts

Confocal micrographs of fibroblasts transduced with control or VPS16‐expressing lentivirus and labeled with LysoTracker. Scale bar, 20 μm.

Quantification of LysoTracker‐stained puncta number and intensities (ca 80 cells analyzed for each of n = 3 biological replicates).

Representative immunoblots (bottom) and summary quantifications (top) of the levels of LAMP1 (C) and LAMP2 (D) in fibroblasts, normalized to levels of actin (ACTB) and expressed as % of controls (n = 3 biological replicates).

Analysis of lysosomal Cathepsin D (CtsD) processing by immunoblotting. Immunoblot to detect CtsD isoforms (left) and quantification (right) of processing, calculated as the percentage of CtsD divided by the sum of all isoforms (n = 3 biological replicates).

Data information: Bar graphs represent data as mean ±SEM. Statistical comparisons between the indicated groups by two‐tailed unpaired Student’s t‐tests (panel E) or one‐way ANOVA with Holm–Sidak’s multiple comparisons tests (panel B). *P < 0.05; **P < 0.01.
Figure 6

Analysis of autophagosome formation and turnover in fibroblasts

Analysis of p62/SQSTM1 levels in patient and control fibroblasts transduced with control or VPS16‐expressing lentivirus under basal conditions and after 2 h of serum starvation.

Summary quantification of (A), normalized to actin (ACTB) and expressed as % of controls (n = 3 biological replicates).

Analysis of LC3 isoforms in serum‐starved fibroblasts transduced with control or VPS16‐expressing lentivirus and treated with Bafilomycin A1 (Baf A1), as indicated.

Quantification of LC3‐II, normalized to levels of LC3‐I, actin and expressed as % of controls (n = 6 biological replicates).

Autophagic flux, calculated as the relative increase in LC3‐II ratios upon addition of Baf A1 (data from D; n = 6 biological replicates).

Analysis of autophagosome numbers in fibroblasts transfected to express mRFP‐GFP‐LC3. (F) Representative confocal micrographs (optical sections) of fibroblasts imaged live under basal or serum‐starved conditions. Scale bar, 20 μm. (G and H) Quantification of autophagosomes, defined as green + red puncta (G) and autophago‐lysosomes, calculated as the number of red (but not green) puncta divided by the total number of puncta. Colored horizontal bars denote median values and whiskers 5 and 95 percentiles (n = 50–113 cells from a total of three independent experiments; for exact values see Appendix Table S1). (I) Cell of patient A transfected with mRFP‐GFP‐LC3, labeled with LysoTracker, and imaged live by confocal microscopy under serum‐starved conditions. Insert shows red+blue autolysosomes (arrowheads) and a green+red autophagosome devoid of LysoTracker signal (arrow). Scale bars 10 μm or 2 μm (inserts).

Data information: Bar graphs represent data as mean ±SEM. Statistical comparisons between the indicated groups by one‐way ANOVA with Holm–Sidak’s multiple comparisons tests. *P < 0.05; ***P < 0.001.
Figure 7

Zebrafish embryos deficient in <italic>vps16</italic> demonstrate abnormalities in LROs and myelination

Schematic representation of the HOPS complex, indicating the high amino acid identity between human and zebrafish for each subunit.

Comparison of control and vps16 embryos at 5 dpf. Compared with control, vps16 embryos have reduced pigmentation in the skin and retinal pigment epithelium and do not develop a swim bladder (arrows). Magnified views show altered melanocyte morphology on the head and pale, translucent eyes.

Myelination, as reported using the Tg(mbp:gfp‐caax) transgenic, is significantly reduced in vps16 crispants. The yellow box indicates the region assessed in (D). Scale bar 100 μm. Bar graphs represent data as mean ±SEM. Statistical comparisons by Mann–Whitney U‐tests (n = 6). **P < 0.01.
Figure 8

LysoTracker and GFP‐Lc3 signal are enhanced in <italic>vps16</italic> embryos, suggesting glial dysfunction

LysoTracker‐stained acidic compartments in zebrafish larvae brains at 3 and 5 dpf. Total signal was increased across the brains of vps16 crispants at 5 dpf, and abnormal, large puncta observed in the optic tectum from 3 dpf (arrows).

LysoTracker signal colocalizes with the astroglial reporter slc1a2b:citrine and does not appear to overlap with the pan‐neuronal reporter elavl3:GCaMP5G.

The puncta indicated in A exist within cells expressing the mpeg1:EGFP transgene, revealing microglia with a grossly altered cellular morphology.

Microglia at 3 dpf contain an increased number of acidic compartments as compared with controls.

Global GFP‐Lc3 signal is increased in unstimulated vps16 embryos (left panel), with clusters of puncta present particularly along the midline of the optic tectum (right panel, magnified view) and anterior hindbrain.

Magnified view of the anterior hindbrain showing numerous GFP‐Lc3 puncta in the vps16 embryo that do not colocalize with LysoTracker.

Data information: Scale bars 100 μm in (A, F left), 30 μm in (C), 20 μm in (F right), and 10 μm in (D, G). Bar graphs represent data as mean ±SEM. Statistical comparisons by Mann–Whitney U‐tests. **P < 0.01; ***P < 0.001. n = 6 each for 3dpf, 3 (ctrl) and 12 (vps16) for 5 dpf in (B). n = 28 (ctrl) and 14 (vps16) in (E). All confocal images representative of a minimum of 6 embryos each.
Figure EV1

Additional clinical features of patient A

Photographs at 3.5 years of age. (A) Barrel‐shaped chest with pectus carinatum. (B) Pes cavus with hammertoe deformity of first toes bilaterally.

Blood and plasma iron status at indicated ages. The patient received oral iron supplementation. Values outside reference range are colored red. MCV, mean corpuscular volume; MCH, mean corpuscular hemoglobin; TIBC, total iron‐binding capacity.
Figure EV2

Whole‐genome sequencing and mutation prediction

Assembly of whole‐genome sequencing reads, identifying the VPS16 c.2272‐18 variant (in patient A).

Predicted impact by the c.2272‐18C>A variant on splicing, as predicted by algorithms used by Alamut Visual (Interactive Biosoftware). Green bars indicate predicted splice‐acceptor sites.
Figure EV3

Analysis of HOPS/CORVET subunits in fibroblasts (related to Fig 3)

Effect of proteasome inhibition on VPS16 and VPS33A. (A) Immunoblots for VPS16 and VPS33A and quantifications of protein levels in lysates of fibroblasts treated with MG132 for 30 or 120 min, respectively. VPS16 band densities were quantified in an extended area (indicated by red boxes) to capture the presence of a possible truncated protein with a molecular weight of 87 kDa (wild‐type VPS16 is 95 kDa). Stars denote unspecific bands. (B) Quantifications of (A) normalized to control conditions and represented as mean ± SEM for VPS16 (n = 3 biological replicates) or mean for VPS33A (n = 2 biological replicates).

Rescue of cellular phenotype by VPSN52K. (C) Representative immunoblots of fibroblast lysates using antibodies against VPS16, VPS33A, and actin. Stars denote unspecific bands. (D) Levels of VPS16 and VPS33A quantified in patient cells transduced with control or VPS16N52K‐expressing lentiviruses and normalized to levels of actin (n = 3). Data represented as mean ± SEM; ***P < 0.001 by unpaired Student’s t‐tests. (E) Cross‐species sequence alignment of VPS16 residues surrounding the asparagine in position 52; (bottom row) asterisks, colons, and periods indicate residues that are fully, strongly, or weakly conserved, respectively.
Figure EV4

Analysis of endolysosomal compartments (related to Figs 4 and 5)

Quantification of cellular uptake of fluorescently labeled transferrin in fibroblast lysates, normalized to total protein levels (n = 3 biological replicates). Data represented as mean ±SEM. Statistical comparisons by unpaired two‐tailed Student’s t‐test (n = 3; P = 0.0023). **P < 0.01.

Immunoblot (left) of RAB11B in the indicated fibroblasts, under basal and serum‐starved conditions, and (right) summary quantifications normalized to levels of actin (ACTB) and expressed as % of controls. Bar graphs represent data as mean ±SEM (n = 3 biological replicates).

Representative confocal micrographs of fibroblasts stained for LAMP2 and LysoTracker. Scale bar 10 μm.

Quantification of the co‐localization between LysoTracker and LAMP2, expressed as Pearson correlation coefficients. Colored horizontal bars indicate the median values and whiskers 5 and 95 percentiles (n = 10 optical sections from three independent experiments).

Quantification of the number and intensities of LAMP2‐stained puncta. Data represented as mean ± SEM (n = 6).
Figure EV5

Sequence validation of the <italic>vps16</italic> crispant zebrafish embryos

Amino acid sequence alignment of the human and zebrafish VPS16 proteins, colored according to homology and with the region of the gRNA target site indicated.

Representative sequence trace for a zebrafish vps16 embryo as compared to a control, showing significant decomposition. The gRNA target sequence is highlighted by the shaded box.

Representative indel distribution for a sequenced vps16 embryo.
Acknowledgments
This image is the copyrighted work of the attributed author or publisher, and ZFIN has permission only to display this image to its users. Additional permissions should be obtained from the applicable author or publisher of the image. Full text @ EMBO Mol. Med.
Your Input Welcome
We welcome your input and comments. Please use this form to recommend updates to the information in ZFIN. We appreciate as much detail as possible and references as appropriate. We will review your comments promptly.
Please check the highlighted fields and try again.
Thank you for submitting comments. Your input has been emailed to ZFIN curators who may contact you if additional information is required.
Oops. Something went wrong. Please try again later.