Podobnik et al., 2020 - Evolution of the potassium channel gene Kcnj13 underlies colour pattern diversification in Danio fish. Nature communications   11:6230 Full text @ Nat. Commun.

Fig. 1 Colour patterns in <italic>Danio</italic> fish and interspecific hybrids.

a Colour pattern of zebrafish, D. rerio. bD. rerio Meox1 (choker) mutants, which lack a horizontal myoseptum. cD. aesculapii. dD. tinwini. e Hybrid between D. rerio and D. aesculapii and f hybrid between D. rerio and D. tinwini. gD. choprae. hD. margaritatus. i Hybrid between D. rerio and D. choprae, j and hybrid between D. rerio and D. margaritatus. k Hybrid between D. aesculapii and D. choprae. l Hybrid between D. tinwini and D. margaritatus. mD. erythromicron. n Hybrid between D. rerio and D. erythromicron. o Hybrid between D. aesculapii and D. erythromicron. p Hybrid between D. aesculapii and D. margaritatus. qD. dangila. r Hybrid between D. rerio and D. dangila. All pictures are representative for the corresponding species or hybrids; for the variability of hybrid patterns see Supplementary Fig 1. Please note that not all panels are shown to the same scale; the sizes of the fish are ~18 mm (D. margaritatus and D. erythromicron), 24 mm (D. tinwini), 30 mm (D. choprae), 35 mm (D. rerio, D. aesculapii) and 75 mm (D. dangila).

Fig. 2 Development of colour patterns in <italic>D. rerio</italic> and <italic>D. aesculapii</italic>.

aD. rerio fish at stage PR, iridophores (arrowhead) emerge along the horizontal myoseptum (asterisk) to form the first light stripe. bD aesculapii fish at stage PR. cD. rerio at stage SP, the first light stripe is flanked dorsally and ventrally by emerging dark stripes. dD. aesculapii at stage SP, iridophores emerge in a scattered fashion. eD. rerio at stage J++, light stripes are covered by compact xanthophores. fD. aesculapii at stage J++, melanophores and xanthophores broadly intermix. gD. rerio at stage JA, the stripes are fully formed. hD. aesculapii at stage JA, melanophores and xanthophores sort out loosely into vertical bars of low contrast; no dense iridophores are visible between the dark bars. ad Incident light illumination to highlight iridophores, eh bright field illumination to visualise xanthophores and melanophores. All pictures are representative for the corresponding species and stages (n ≥ 3). Staging of animals according to Parichy et al.83. PB (pectoral fin bud, 7.2 mm SL). SP (squamation posterior, 9.5 mm SL). J++ (juvenile posterior, 16 mm SL). JA (juvenile-adult, >16 mm SL). Scale bars correspond to 250 μm.

Fig. 3 Mutant phenotypes in <italic>D. rerio</italic> and <italic>D. aesculapii</italic> of genes required for individual pigment cell types.

In D. rerio loss of one type of pigment cell type, a melanophores in Mitfa (nacre) mutants, c xanthophores in Csf1ra (pfeffer) mutants, or, e iridophores in Mpv17 (transparent) mutants, still permits rudimentary aggregation of dense iridophores (a) or melanophores (b, c). In D. aesculapii, loss of melanophores, b, in Mitfa mutants (n > 100) or loss of xanthophores, d, in Csf1ra mutants (n > 50), abrogate any residual pattern formation. However, vertical melanophore bars still form in Mpv17 mutants (n = 8), f despite the absence of iridophores. All images show representative examples of the corresponding genotypes. Scale bars correspond to 1 mm.

Fig. 4 Mutant phenotypes in <italic>D. rerio</italic>, <italic>D. aesculapii</italic> and their hybrids of genes required for heterotopic interactions.

In D. rerio mutations in, a, Cx39.4 (luchs), b, Cx41.8 (leopard), and c, Igsf11 (seurat) lead to spotted patterns, whereas, d, mutations in Kcnj13 (obelix) result in fewer and wider stripes. In D. aesculapii, eh, mutations in the orthologous genes lead to the complete loss of any pattern. In D. rerio dominant alleles of Kcnj13, i, cause broader stripes and irregularities when heterozygous. Double mutants, j, Cx39.4 k.o.; Kcnj13 k.o., loose almost all melanophores and pattern. Interspecific hybrids between D. rerio and D. aesculapii, which are both mutant, k, for Cx41.8, or, l, for Kcnj13, show patterns of spots or wider stripes similar to the corresponding D. rerio mutants (b, d; n = 15). All images show representative examples of the corresponding genotypes. Scale bars correspond to 1 mm.

Fig. 5 A reciprocal hemizygosity test to identify <italic>Kcnj13</italic> evolution.

Two hybrids between D. rerio and D. aesculapii, which are hemizygous for a Kcnj13 loss-of-function mutation. a stripes are interrupted in hybrids carrying the mutant allele from D. rerio (n > 60, nick in the blue line representing the zebrafish genome). b hybrids carrying the mutant allele from D. aesculapii (n = 6, nick in the magenta line, representing the D. aesculapii genome) are indistinguishable from wild-type hybrids (Fig. 1e).

Fig. 6 One-way complementation tests suggest repeated <italic>Kcnj13</italic> evolution.

On the left the phylogenetic tree depicts the relationship between the Danio species; the asterisk denotes a node with lower bootstrap support. The left column shows the patterns of the different species. In the middle column patterns of wild-type hybrids with D. rerio are shown (see also Supplementary Fig. 1). In the right column patterns of hybrids that carry a mutant Kcnj13 allele from D. rerio are shown. Pattern defects are obvious in three cases: hybrids with D. aesculapii (n > 60, magenta), with D. tinwini (n = 12, yellow) and D. choprae (n = 40, cyan). In the other six cases the patterns in hemizygous hybrids do not differ from the striped patterns of wild-type hybrids (D. rerio Kcnj13 k.o./D. kyathit, n = 32; D. rerio Kcnj13 k.o./D. nigrofasciatus, n = 16, D. rerio Kcnj13 k.o./D. albolineatus, n = 4; D. rerio Kcnj13 k.o./D. erythromicron, n = 38; D. rerio Kcnj13 k.o./D. margaritatus, n = 12; and D. rerio Kcnj13 k.o./D. dangila, n = 16). All pictures show representative examples of the corresponding species/hybrids/genotypes; for variability of the hybrid patterns see also Supplementary Fig. 3. Scale bars correspond to 1 mm.

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 @ Nat. Commun.