MTMR5 domain topology, pathogenic variants, and spatiotemporal expression pattern in zebrafish. (A) Human MTMR5 contains six functional domains: an N-terminal tripartite DENN domain [upstream DENN (a.a. 1–86), DENN (a.a. 129–311) and downstream DENN (a.a. 364–433)], an SBF2 domain (a.a. 543–765), a GRAM domain (a.a. 883–969), a catalytically inactive myotubularin-phosphatase domain (a.a. 1108–1533), a coiled-coil domain (CC, a.a. 1651–1665) and a C-terminal PH domain (a.a. 1763–1868). Of note, the domain topology of the MTMR5 protein is highly conserved between human and zebrafish. Pathogenic variants are found throughout the gene (for details, see Supplementary Table 1). Note: autism SD (spectrum disorders). (B) RT-PCR using total cDNA shows temporal expression pattern of zebrafish mtmr5 from 6 h post-fertilization (hpf) to 5 dpf (days post-fertilization). Note: ppib is a housekeeping control. (C) Whole-mount in situ hybridization using DIG-conjugated RNA probes in 1 dpf and 3 dpf embryos. The sense probe (negative or −ve control) gives no staining at 1 dpf or 3 dpf. At 1 dpf, from the lateral view, the anti-sense probe shows ubiquitous staining in the brain, including the forebrain (FB), midbrain-hindbrain boundary (MHB), and the hindbrain; the dorsal view at 1 dpf also revealed staining in fine structures such as the retina and the hypothalamus (inset scale bar 250 μm). Similar brain-specific staining can be observed at 3 dpf for the anti-sense probe, in the forebrain (FB), midbrain (MB) and hindbrain (HB). Staining is also visible in the jaw region at 3 dpf (lateral view, arrow). Scale bars: 1 mm. Domain topology of MTMR5 proteins was mapped using SMART¹⁸ and DeepCoil.¹⁹ Domain illustrations were generated using IBS 2.0.²⁰ Uncropped gels of Fig. 1B are included in Supplementary Fig. 5A.

Generation of a full mtmr5 gene deletion/knockout zebrafish. (A) Schematic diagram of the workflow. Single-guide RNAs (sgRNA) and Cas9 (Clustered regularly interspaced short palindromic repeats or CRISPR associated protein 9) mRNA were co-injected into 1-cell stage zebrafish embryos (wild-type or WT; AB strain). The full mtmr5 gene deletion was first identified by PCR at F0 and further confirmed at F1 by PCR and Sanger sequencing. (B) Sanger sequencing revealed an 86 kilobase (kb) deletion at the mtmr5 gene, starting 40 base pairs (bp) after the translation start site to 3 bp after the translational stop site (STOP). (C) Genotyping was achieved using a 3-primer PCR method, showing the WT band at 300 bp (one flanking and one inside mtmr5), the mutant band at 500 bp (primers flanking mtmr5) and the heterozygous (Het) bands at both 300 and 500 bp. (D) Reverse transcription (RT)-PCR using specific primers against the deleted mtmr5 region shows a ∼100 bp band in the 7 dpf (days post-fertilization) WT cDNAs, and the lack of amplification in the mutant cDNAs. Housekeeping control (ppib) shows bands in both WT (wild-type) and mtmr5-knockout (KO) samples, and H₂O (water) control shows no bands. Note: four biological replicates per condition, each lane represents n = 15–20 zebrafish heads (7 dpf). Uncropped gels of Fig. 2C and D are included in Supplementary Fig. 5B and C, respectively.

Loss of mtmr5 does not affect motor functions. (A) Embryos generated from a mtmr5-heterozygous (+/−) in-cross (inx) were subjected to Optovin treatment at 3 days post-fertilization (dpf) followed by the quantification of the total distance travelled (mm) by each embryo via ZebraBox. There was no apparent difference in swim performance between wild-type (WT) n = 36, heterozygous (Het) n = 77, and knockouts (KO) n = 23. (B and C) Embryos resulting from a heterozygous in-cross were subjected to a free swim assay (no Optovin treatment) at (B) 6 dpf (WT n = 25, Het n = 47, KO n = 15) and (C) 14 dpf (WT n = 35, Het n = 30, KO n = 15). There were no observed differences in swim performance between the three genotypes at both developmental stages. (D and E) Embryos resulting from a homozygous knockout female crossed with a heterozygous male were subjected to swim assays at (D) 3 dpf (with Optovin treatment; Sib (Het) n = 13, KO n = 16), and at (E) 6 dpf (free swimming assay) (Sib (Het) n = 17, KO n = 22), and again, at both developmental time points, there were no differences in swimming performance between the genotypes. (F) A representative image of the swim trace tracking used by the ZebraBox program to quantify the total distance travelled (mm) for each embryo per well. All statistical analyses include at least three independent experiments. Each dot on the graphs represents one zebrafish, at least n = 4 zebrafish were used per group per experiment. Quantitative data are mean±SEM, normalized to the average of WT and presented as ratios. One-way ANOVA analysis was performed for all heterozygous in-cross experiments and an unpaired two-tailed Student’s t-test was used for the homozygous female × heterozygous male experiment; ns = not significant, P > 0.05.

Loss of mtmr5 results in microcephaly and an overall body size reduction from as early as 10 dpf. (A, top) A representative image of a WT 7 dpf zebrafish with lines delineating the measured gross anatomy. This system is used to measure all age groups (7 dpf, 10 dpf, 14 dpf and 2 months). (A, middle and bottom) At 7 dpf, there was no observed difference in (Ai) body length, (Aii) brain height or (Aiii) brain length between the siblings and KO. WT n = 20, heterozygous (Het) = 36, and KO n = 17. (B) Microcephaly (i.e. reduced absolute brain size) starts to appear in the KO at 10 dpf, when (Bi) there was a significant reduction in body length, (Bii) brain height and (Biii) brain length in the KO. WT n = 16, Het n = 40, KO n = 16. (C and D) Similarly, at 14 dpf (C) and at 2 months (D), there was a significant decrease in body length (Ci and Di), brain height (Cii and Dii) and brain length (Ciii and Diii) in KO compared to WT siblings. 14 dpf: WT n = 27, Het n = 45, KO n = 18. 2 months: WT n = 5, Het n = 8, KO n = 10. All statistical analyses include at least three independent experiments. Each dot on the graphs represents one zebrafish, at least n = 5 zebrafish were used per group per experiment. Quantitative data are mean±SEM, normalized to the average of WT and presented as ratios. One-way ANOVA was done for all measurements. *P < 0.05; **P < 0.01; ****P < 0.0001; ns, not significant. Scale bars: 500 μm (7, 10 and 14 dpf images) or 20 mm (adult images).

Loss of mtmr5 does not result in excessive cell death in the nervous system and does not cause changes in brain chamber size. (A and B) Representative average Z-projections depicting laterally orientated 2 days post-fertilization (dpf) (A) and 7 dpf (B) zebrafish embryos stained with acridine orange (AO) to visualize and quantify apoptotic cells in live zebrafish brains. No differences in AO intensity were detected between the sibling and the knockout (KO) (A—2 dpf: Sib n = 32, KO n = 9; B—7 dpf: WT n = 7, Het n = 9, KO n = 18). (C) Representative images from fixed and DAPI-stained 7 dpf zebrafish embryos to visualize and quantify brain chamber size. There were no qualitative or quantitative differences in both forebrain and midbrain size between the siblings and KOs at 7 dpf (Sib n = 9, KO n = 5). (D and E) Representative images of AO-stained (D) anterior lateral regions of 14 dpf zebrafish, encompassing the central nervous system, show no visual differences between the siblings and KOs, and (E) posterior-lateral regions of 14 dpf zebrafish, encompassing the peripheral nervous system, show no visual differences between the siblings and KOs. All statistical analyses include at least three independent experiments. Each dot on the graphs represents one zebrafish, at least n = 3 zebrafish were used per group per experiment. Quantitative data are mean±SEM, normalized to the average of WT and presented as ratios. Unpaired two-tailed Student’s t-test was used. ns, not significant. Scale bars: 200 μm (A), 100 μm (B and C) or 500 μm (D).

Loss of mtmr5 leads to less defined axonal organization and dysregulated axon myelination. (A and B) Seven days post-fertilization (dpf) zebrafish embryos were fixed and probed with an acetylated tubulin antibody to label mature axons. (A and B) Representative images of Z-projections (standard deviation via Fiji ImageJ) of acetylated tubulin staining. Axons appear less organized in the mtmr5-knockout (KO) compared to the sibling. (Ai and Bi) Two-dimensional skeletonized images generated using Binary function in Fiji ImageJ to allow for quantification of axonal organizational parameters. (C-Cii) The number of total axonal branches (C) and axon end-points (Ci) significantly increased (by ∼25%) in the KO compared to the siblings (Sib), while the average axon branch length (Cii) in KOs significantly decreased (by ∼10%). All quantification was normalized to the sibling average of the respective parameter. All statistical analyses include at least three independent experiments. Each dot on the graphs represents one zebrafish. Quantitative data are mean±SEM, normalized to the average of WT and presented as ratios. Unpaired two-tailed Student’s t-test was used. *P < 0.05; **P < 0.01, ns, not significant. Scale bars (A and B): 20 μm. Sib n = 11, KO n = 10. (D) Transmission electron microscopy was used to visualize axons in the posterior lateral line at 14 dpf. WT axons were in close contact with their myelin sheaths. This tight and organized wrapping (D, box) can be observed in more detail in the zoomed-in inset (Di). SC = Schwann cell. (E) In the 14 dpf KO, the myelin is still present; yet, there is an apparent blank space between the axon and the myelin (arrows), showing the detachment or loose myelin in the KOs (E, yellow box; Ei). Scale bars: (D and E) 2 μm or (Di and Ei) 0.5 μm.

Bulk RNA sequencing of 7 dpf mtmr5-knockout versus wild-type brain-enriched samples. (A) Volcano plot displaying the differentially expressed genes in 7 days-post-fertilization (dpf) mtmr5-knockout (KO) zebrafish heads compared to wild-type (WT) siblings. A significance threshold of adjusted P-value (padj) < 0.01 and absolute log₂ fold-change >1 was used. Upregulated genes are in the top right segment and downregulated genes are in top left segment. Number of significant genes in each category are indicated (upregulated, n = 1693; downregulated, n = 347). Each dot represents one zebrafish gene. See Supplementary Table 2 for the full list of genes, padj values, and log₂fold-changes. (B) Gene ontology (GO) terms significantly enriched in the up- and down-regulated genes (padj < 0.01), sorted by the number of genes associated with the respective GO term (gene count). All enriched GO terms (padj < 0.05) are included in Supplementary Table 3. (C) Normalized read counts for the three CMT4B-associated MTMR genes: mtmr5, mtmr2 and mtmr13. ****padj < 0.00001, ns > 0.05. (D) Log₂ fold-change values for all mtmr genes with significant values indicated (padj < 0.01). mtmr5 and mtmr13 gene names are used in place of sbf1 and sbf2, respectively.