Defective adult oligodendrocyte and Schwann cell development, pigment pattern, and craniofacial morphology in puma mutant zebrafish having an alpha tubulin mutation

Abstract

The processes of myelination remain incompletely understood but are of profound biomedical importance owing to the several dysmyelinating and demyelinating disorders known in humans. Here, we analyze the zebrafish puma mutant, isolated originally for pigment pattern defects limited to the adult stage. We show that puma mutants also have late-arising defects in Schwann cells of the peripheral nervous system, locomotor abnormalities, and sex-biased defects in adult craniofacial morphology. Using methods of positional cloning, we identify a critical genetic interval harboring two alpha tubulin loci, and we identify a chemically induced missense mutation in one of these, tubulin alpha 8-like 3a (tuba8l3a). We demonstrate tuba8l3a expression in the central nervous system (CNS), leading us to search for defects in the development of oligodendrocytes, the myelinating cells of the CNS. We find gross reductions in CNS myelin and oligodendrocyte numbers in adult puma mutants, and these deficits are apparent already during the larval-to-adult transformation. By contrast, analyses of embryos and early larvae reveal a normal complement of oligodendrocytes that nevertheless fail to localize normal amounts of myelin basic protein (mbp) mRNA in cellular processes, and fail to organize these processes as in the wild-type. This study identifies the puma mutant as a valuable model for studying microtubule-dependent events of myelination, as well as strategies for remyelination in the adult.

Fig. 1

puma mutant adult zebrafish exhibit defects in pigment pattern, craniofacial morphology, Schwann cell development, and locomotor behavior. (A,A′) Adult stripes and craniofacial morphology are disrupted in puma mutants. (B,B′) Clearing and staining to reveal bone (red) and cartilage (blue) shows thinner dysmorphic bones of the skull in puma mutants (arrowheads) and ventral displacement of the jaw (arrow). (C,C′) External anatomy of head for wild-type and a severely affected puma mutant. (D–F) Development of Schwann cell defects. (D,D′) Early larvae exhibit mbp+ Schwann cells covering the lateral line nerve (arrow) in both wild-type and puma mutants (here, 4.0 mm standardized standard length, 4.0 SSL). (E,E′) During later post-embryonic development, mbp+ cells are more sparsely arranged in puma mutants (here, 6.3 SSL). (F,F′) mbp+ cells are nearly absent in puma mutants, with only rare residual cells found along the lateral line (arrow in F′); mbp+ cells found along dorsoventrally oriented nerves in wild-type (arrowheads in F) are completely lacking in puma (8.0 SSL). (G) Quantitative analysis of craniofacial defects in puma mutants. Shown are the proportions of individuals placed into different categories for severity of craniofacial defect, divided by sex and genotype. Numbers of individuals examined are given above the bars. Contingency table analysis shows that homozygous puma mutants were more likely than wild-type to exhibit craniofacial defects and such defects were more prevalent among males than females (genotype: χ²=93.0, P<0.0001; sex: χ²=43.7, P<0.0001; n=171). Scale bars: in A, 4 mm for A, A′. In B, 1 mm for B,B′; in D, 60 μm for D, D′; in E, 60 μm for E, E′; in F, 60 μm for F, F′.

Fig. 2

Identification of an alpha tubulin lesion in the puma mutant. (A) Critical genetic and physical interval for the puma mutant locus. Genetic markers and their associated genetic distances are shown above the horizontal scale; PAC end sequences in this region are shown below the horizontal scale. (B) Electropherograms for tuba8l3a showing single nucleotide polymorphism between genetic backgrounds and lesion present in puma mutant (nucleotides highlighted with pink). Top to bottom: wild-type wik sequence; wild-type SJD with silent G→T transversion; homozygous puma mutant illustrating SNP shared with SJD and novel T→A transversion, resulting in a non-conservative N→K substitution; puma/+ map-cross individual heterozygous for both polymorphic sites. (C) Schematic of the tuba8l3a cDNA indicating unique and shared polymorphisms between strains. Only the puma haplotype exhibits the T747A substition. Red, beta/alpha domain interface. (D) N249K substitution mapped onto structural model of an αβ-tubulin dimer (Nogales et al., 1998b) (MMDB ID:8900, PDB ID: 1TUB).

Fig. 3

Orthology of tuba8l3a relative to other other alpha tubulin genes. (A) Bayesian phylogeny with clade credibility values for each node. (B). Parsimony phylogeny with bootstrap support for each node. Zebrafish tuba8l3a (puma) is boxed; protein colors correspond to synteny comparison in Fig. 4. See text for details. Species identifiers: Cf, Canis familiaris, Dm, Drosophila melanogaster; Dr, Danio rerio; Gg, Gallus gallus; Hs, Homo sapiens; Mm, Mus musculus; Ol, Oryzias latipes (medaka); Sc, Saccharomyces cerevisiae; Tn Tetraodon nigroviridis (pufferfish). Protein identifiers (Genbank) or transcript identifiers (Ensembl): Cf Tuba1, XM_858678; Cf Tuba2L, XM_848175; Dm CG7794, NP_650674.1; Dm TA67C, NP_524009.2; Dm TA84B, NP_476772.1; Dm TA84D, NP_524264.1; Dm TA85E, NP_524297.1; Dr tuba, NP_001098596.1; Dr tuba1, NP_919369.1; Dr tuba2, NP_998195.1; Dr tuba4, NP_956089.1; Dr tuba6, XP_001334783.1; Dr tuba7L, NP_001002230.1; Dr tuba8, NP_997937.1; Dr tuba8l2, NP_956985.1; Dr tuba8l3a (puma), ; Dr tuba8l3b, ; Dr tuba8l4, NP_956479.1; Dr WU:fb37a10, XP_688383.1; Gg Tuba3/7, XP_422851.1; Gg Tuba8, NP_990775.1; Hs TUBA, NP_006073.2; Hs TUBA1A, NP_006000.2; Hs TUBA3C, NP_005992.1; Hs TUBA3E, NP_997195.1; Hs TUBA3L, NP_079079.1; Hs TUBA4A, NP_005991.1; Hs TUBA6, NP_116093.1; Hs TUBA8, NP_061816.1; Hs TUBB1, NP_110400.1; Hs TUBG1, NP_001061.2; Mm Tuba1, NP_035783.1; Mm Tuba1b, NP_035784.1; Mm Tuba1c, NP_033474.1; Mm Tuba3L, NP_001029051.2; Mm Tuba4, NP_033473.1; Mm Tuba8, NP_059075.1; Ol *20148, ENSORLT00000020148; Ol *20157, ENSORLT00000020157; Ol *20159, ENSORLT00000020159; Sc TUB1, NP_013625; Tn *19770, ENSTNIT00000019770; Tn *19771, ENSTNIT00000019771; Tn* 08261, ENSTNIT00000008261.

Fig. 4

Synteny analyses reveal tuba8l3a orthology. Shown are loci inferred to be orthologues by comparison of conserved synteny across vertebrate genomes based on Ensembl genome assemblies. tuba8l3a is orthologous to the human pseudogene ψTUBA4B and dog Tuba2L. Regions are not to scale. For Tetraodon and Oryzias, unnamed tubulins are indicated by transcript identifiers where the preceding “*” indicates “ENSTNIT000000” and “ENSORLT000000”, respectively.

Fig. 5

tuba8l3a is not spatially restricted in the embryo but is expressed most prominently in the central nervous system during post-embryonic development. Shown is staining with antisense and sense probes (diluted to equal concentrations) targeted to the 5′ untranslated region of tuba8l3a, (A,A′) Widespread expression of tuba8l3a at 24 hpf. (B,B′) Lateral views of larvae (7.2 SSL) showing tuba8l3a transcript in the brain (arrow) and cranial ganglia (arrowhead). (C,C′) Dorsal views of the same individuals, illustrating tuba8l3a mRNA in the brain (arrow). Larvae in B–C are homozygous nacre mutants that lack otherwise obscuring melanophores due to an autonomously acting mutation in the mitfa transcription factor. Expression in wild-type larvae was indistinguishable from nacre mutants (not shown). (D,D′) tuba8l3a transcript is detectable in the inner nuclear layer of the retina (arrow). (E,E′, F,F′) More posteriorly, tuba8l3a is expressed in the periventricular grey zone of optic tectum. (G,G′,H) tuba8l3a staining (arrow) in the cranial ganglia. Larvae in D–H are 8–9 SSL.

Fig. 6

Deficient central nervous system myelination in puma mutant adults and juveniles. (A,A′) Decreased opacity of adult brain from puma as compared to wild-type; this difference is especially notable at the lateral edges and posteriorly (arrowheads). Anterior to the left. Tel, telencephalon; TeO, tegmentum opticum; CCe, corpus cerebelli; MO, medulla oblongata; MS, medulla spinalis. (B,B′) Black Gold II staining for myelin in transverse sections of wild-type and puma mutant adults. Myelin is stained blue–black whereas other tissue is purple-red. Reduced staining in puma is evident to some degree within the optic tectum (TeO) but is more pronounced in ventral–medial regions (arrow), internal to the periventricular gray zone (PGZ). Radially oriented black gold staining (arrowhead) also is largely absent from the PGZ itself of the puma mutant. (C,C′) Antiserum against zebrafish mbp (red) stains numerous punctate foci in the ventral–medial brain of wild-type but far fewer in the corresponding region of puma mutants. Green, background tissue fluorescence. Yellow, fluorescence of blood vessels in green and red channels. (D,D′–G′G) In situ hybridization for mbp and plp in juveniles and adults (13 SSL, 24 SSL). (D,D′) In juveniles, staining for mbp is pronounced in a band (arrowhead) near the outer surface of the optic tectum (TeO) in wild-type (D) but is not apparent in puma (D′). (E,E′) In adults, abundant mbp transcripts are found radially in the PGZ (arrowhead) and at the outer surface of the optic tectum (arrow) in wild-type but are found at lower levels in puma. (F,F′) In juveniles, plp staining is detected in individual cell bodies both interiorly (arrowhead) and near the tectum surface (arrow) of wild-type but there are far fewer of these cells in puma mutants. (G,G′) In adults, numerous plp-expressing cell bodies are observed in wild-type, whereas in puma mutants there are fewer cells and those present are less densely stained. Scale bars: in B, 100 μm; in C, 20 μm; in D, 100 μm for D–G.

Fig. 7

Reduced myelination in puma mutant brains revealed by transmission electron microscopy. (A–C) Representative regions of wild-type at low magnification (5000×, A and B) and higher magnification (60,000×, C) show numerous well-myelinated axon tracts. (D–F) Corresponding regions and magnifications in puma reveal far fewer myelinated axon tracts (arrowheads in D and E) though layering of individual myelin sheaths that are present is indistinguishable from wild-type (arrowheads in F; arrow, mitochondrion). Scale bars: in A, 5 μm, for A,B,D,E; in C, 500 nm for C, F.

Fig. 8

Defects in oligodendrocyte number and patterning during the larval-to-adult transformation of puma mutants. (A,A′,B,B′) Wild-type zebrafish larvae exhibit many more plp+ oligodendrocytes in the brain than do puma mutants both during the middle larval period (A,A′; ∼6.5 SSL) and the late larval period (B,B′ ; ∼8.0 SSL). (C,C′,D,D′,E,E′) mbp expression also differs between wild-type and puma mutants during middle and later larval development (C,C′ and D,D′, respectively); a higher magnification view of different larvae is shown in E, E′. Note especially the absence of most myelinated fibers in the puma mutant and the concentration of mbp mRNA in cell bodies. (F,F′,G,G′) The early oligodendrocyte marker olig1 does not differ in expression between genotypes at middle or later larval stages, nor do several additional markers of particular cell lineages or activities (H–K; see text for details). Scale bars: in A, 100 μm, for A,A′,B,B′,C,C′,D,D′,F,F′. In E, 40 μm for E,E′; in G, 100 μm for G,G′; in H, 100 μm for H,H′; in J, 100 μm for J,J′; in K, 100 μm for K,K′.

Fig. 9

Normal oligodendrocyte early differentiation but defective morphogenesis in puma mutant early larvae and rescue of puma mutant phenotype by morpholino knockdown. All panels are of 5 dpf larvae with anteriors to the left. (A,A′,B,B′) sox10 is expressed similarly in wild-type and puma mutants both in the brain (A,A′) and in the spinal cord (B,B′). Arrowheads in A, dispersed sox10+ cells within the brain. Arrow in A, more superficial sox10 staining within cranial ganglia. Arrowhead in B, sox10+ cells in the ventral spinal cord. (D,D′,E,E′) plp and mag expression are similar between wild-type and puma mutants as well, with mRNA staining localizing principally to cell bodies. (F,F′) mbp staining, however, differs between genotypes, with wild-type mbp+ transcript detectable principally in oligodendrocyte processes, rather than in cell bodies, whereas in puma mutants, transcript is present both within processes and in cell bodies. (G,G′) Higher magnification view of different embryos, showing in wild-type compact and organized mbp staining in myelinating processes, but in puma mutants diminished staining in processes and greater staining in cell bodies (arrowhead). Although tuba8l3a expression in embryos was consistent with that of a maternally derived transcript, neither mbp mis-patterning at 5 dpf nor other phenotypes segregated as would be predicted for a maternal effect mutation (data not shown). (H) Morpholino knockdown of tuba8l3a results in moderate perturbations to mbp staining in wild-type larvae. (H′) In contrast, knockdown of tuba8l3a in puma mutants restores a wild-type pattern of mbp staining. (I) Partial disruption of mbp staining in another wild-type morphant. (I′) An uninject puma mutant larva from the same clutch as the individual in H′, stained for the same duration. Morphino dosages in H,H′,I: 0.5 ng tuba8l3a-tb1; similar results were observed with tuba8l3a-tb2 (data not shown). Scale bars: in A, 60 μm for A,A′,B,B′; in D, 100 μm for D,D′,E,E′,F,F′; in G, 40 μm for G,G′,H,H′,I,I′.

Fig. 10

Defects in oligodendrocyte patterning and myelination in puma mutant early larvae. (A–C) sox10:GFP transgene reveals oligodendrocyte cell bodies and processes, that are more disorganized in puma mutants than in wild-type at 5 dpf. Cell bodies are deliberately overexposed (shown here in white) to reveal fainter processes. (A,A′) Low magnification showing the orderly array of myelinating processes in wild-type (arrow), and more disorganized processes in puma mutants. (B,B′) Higher magnification of different individuals. (C,C′) Relatively long processes are observed in both wild-type and puma mutants (arrowheads). Scale bars: in A, 40 μm for A,A′; in 40 μm for B,B′; in C, 40 μm for C, C″.