Understanding the evolutionary transformation of fish fins into tetrapod limbs is a fundamental problem in biology1. The search for antecedents of tetrapod digits in fish has remained controversial because the distal skeletons of limbs and fins differ structurally, developmentally, and histologically2, 3. Moreover, comparisons of fins with limbs have been limited by a relative paucity of data on the cellular and molecular processes underlying the development of the fin skeleton. Here, we provide a functional analysis, using CRISPR/Cas9 and fate mapping, of 5′ hox genes and enhancers in zebrafish that are indispensable for the development of the wrists and digits of tetrapods4, 5. We show that cells marked by the activity of an autopodial hoxa13 enhancer exclusively form elements of the fin fold, including the osteoblasts of the dermal rays. In hox13 knockout fish, we find that a marked reduction and loss of fin rays is associated with an increased number of endochondral distal radials. These discoveries reveal a cellular and genetic connection between the fin rays of fish and the digits of tetrapods and suggest that digits originated via the transition of distal cellular fates.
- Gaining Ground: The Origin and Evolution of Tetrapods. Indiana University Press, Indiana (2012)
- The fin to limb transition: New data, interpretations, and hypotheses from paleontology and developmental biology. Annu. Rev. Earth Planet. Sci. 37, 163–179 (2009)
- The origin of the tetrapod limb: from expeditions to enhancers. Trends Genet. 29, 419–426 (2013) &
- Hoxa-13 and Hoxd-13 play a crucial role in the patterning of the limb autopod. Development 122, 2997–3011 (1996) et al.
- A Hoxa13:Cre mouse strain for conditional gene manipulation in developing limb, hindgut, and urogenital system. Genesis 53, 366–376 (2015) , , &
- The development of the paired fins in the zebrafish (Danio rerio). Mech. Dev. 79, 99–120 (1998) &
- Ever since Owen: Changing perspectives on the early evolution of tetrapods. Annu. Rev. Ecol. Evol. Syst. 39, 571–592 (2008) , &
- The Development of the Teleost Fin and Implications for Our Understanding of Tetrapod Limb Evolution. (Springer US, 1991)
- The apical ectodermal ridge: morphological aspects and signaling pathways. Int. J. Dev. Biol. 52, 857–871 (2008) &
- Vertebrate limb bud development: moving towards integrative analysis of organogenesis. Nat. Rev. Genet. 10, 845–858 (2009) , &
- Mechanism of pectoral fin outgrowth in zebrafish development. Development 139, 2916–2925 (2012) , , , &
- Hox gene expression in teleost fins and the origin of vertebrate digits. Nature 375, 678–681 (1995) , &
- The origin of digits: expression patterns versus regulatory mechanisms. Dev. Cell 18, 526–532 (2010) &
- Cis-regulatory programs in the development and evolution of vertebrate paired appendages. Semin. Cell Dev. Biol. http://dx.doi.org/10.1016/j.semcdb.2016.01.015 (2016) &
- The role of Hox genes during vertebrate limb development. Curr. Opin. Genet. Dev. 17, 359–366 (2007) &
- Biphasic Hoxd gene expression in shark paired fins reveals an ancient origin of the distal limb domain. PLoS One 2, e754 (2007) , &
- An autopodial-like pattern of Hox expression in the fins of a basal actinopterygian fish. Nature 447, 473–476 (2007) , &
- Tri-phasic expression of posterior Hox genes during development of pectoral fins in zebrafish: implications for the evolution of vertebrate paired appendages. Dev. Biol. 322, 220–233 (2008) &
- HoxD expression in the fin-fold compartment of basal gnathostomes and implications for paired appendage evolution. Sci. Rep. 6, 22720 (2016) et al.
- Hoxd13 contribution to the evolution of vertebrate appendages. Dev. Cell 23, 1219–1229 (2012) , , , &
- Deep conservation of wrist and digit enhancers in fish. Proc. Natl Acad. Sci. USA 112, 803–808 (2015) et al.
- An exclusively mesodermal origin of fin mesenchyme demonstrates that zebrafish trunk neural crest does not generate ectomesenchyme. Development 140, 2923–2932 (2013) , , &
- Dermal fin rays and scales derive from mesoderm, not neural crest. Curr. Biol. 23, R336–R337 (2013) , &
- Trunk neural crest origin of caudal fin mesenchyme in the zebrafish Brachydanio rerio. Proc. R. Soc. Lond. B 256, 137–145 (1994) , , , &
- The spotted gar genome illuminates vertebrate evolution and facilitates human-teleost comparisons. Nat. Genet. 48, 427–437 (2016) et al.
- Ubiquitous transgene expression and Cre-based recombination driven by the ubiquitin promoter in zebrafish. Development 138, 169–177 (2011) et al.
- Clustering of tissue-specific sub-TADs accompanies the regulation of HoxA genes in developing limbs. PLoS Genet. 9, e1004018 (2013) et al.
- Loss of fish actinotrichia proteins and the fin-to-limb transition. Nature 466, 234–237 (2010) et al.
- Expression of snail2, a second member of the zebrafish snail family, in cephalic mesendoderm and presumptive neural crest of wild-type and spadetail mutant embryos. Dev. Biol. 172, 86–99 (1995) , &
- Attenuated BMP1 function compromises osteogenesis, leading to bone fragility in humans and zebrafish. Am. J. Hum. Genet. 90, 661–674 (2012) et al.
- Simple and efficient CRISPR/Cas9-mediated targeted mutagenesis in Xenopus tropicalis. Genesis 51, 835–843 (2013) et al.
- Efficient multiplex biallelic zebrafish genome editing using a CRISPR nuclease system. Proc. Natl Acad. Sci. USA 110, 13904–13909 (2013) , &
- ZFIN: Zebrafish Book. 4th ed. (Univ. of Oregon Press, 2000)
Extended data figures and tables
Extended Data Figures
- Extended Data Figure 1: Cre in situ hybridization of lineage tracing fish. (637 KB)
a, Cre is expressed only from 31 hpf to 38 hpf in Dr-CNS65x3–Cre, whereas it is expressed from 38 hpf to 55 hpf in Lo-e16x4–Cre. These temporal expression patterns of Cre indicate that our transgenic lineage tracing labelled the cells which experienced only early or late phase hox. Scale bars are 100 μm. b, Cre expression pattern from 48–120 hpf in independent Lo-e16x4–Cre lines (different founders from a). The fin is outlined by a dashed white line. The expression patterns from different founders were investigated and all expression ceases before 72 hpf. Our in situ results indicate that Lo-e16x4–Cre marks only the cells that experienced late phase hox expression from 38-55 hpf. n = 5 embryos for all stages. Scale bars are 100 μm. c, The expression pattern of and1 and hox13 genes in wild type (10 dpf) and also Cre in Lo-e16x4–Cre line (10 dpf and 3 months, n = 10). Whereas and1 expression can be observed in fin fold (positive control, black arrow), hox13 genes are not expressed at 10 dpf in the wild type. Cre is not expressed at 10 dpf and at 3 months in the fin, indicating that Lo-e16x4–Cre activity is limited to only early embryonic development (38–55 hpf). Three month fins were dissected from the body of Lo-e16x4–Cre lines and subjected to in situ hybridization (n = 3). Scale bars are 500 μm at 10 dpf and 3 months.
- Extended Data Figure 2: T7E1 assay of F0 CRISPR/Cas9 adult fish. (208 KB)
PCR products of hoxa13a, hoxa13b or hoxd13a were subjected to a T7E1 assay (Methods) and confirmed by gel electrophoresis. a, The result of the hoxa13a, hoxa13b or hoxd13a T7E1 assay for ten adult fish. ‘M.’ is a 100 bp DNA ladder marker (NEB). In the hoxa13a gel picture, 810 bp (black arrowhead) is the wild-type band as observed in cont. lane (wild type without gRNA injection). All ten fish showed smaller and bottom shifted products (red arrowheads) compared to negative control fish, indicating that all fish have mutations in the target region of hoxa13a. In the hoxa13b gel picture, 1,089 bp is the wild-type band. All ten fish into which hoxa13b gRNAs were injected showed smaller and bottom shifted products compared to negative control fish, indicating that all fish have mutations in the target region of hoxa13b. In the hoxd13a gel picture, 823 bp is the wild-type band. Eight of ten fish showed smaller and bottom shifted products, indicating that 80% of fish have mutations in the target region of hoxd13a. b, The efficiency of CRISPR/Cas9 deletion for hox13 in zebrafish. Almost all adult fish into which gRNAs and Cas9 mRNA were injected have mutations at the target positions. c, The efficiency of germline transmission of CRISPR/Cas9 mutant fish. Identified mutant fish were outcrossed to wild-type fish to obtain embryos and confirmed germline transmission. Obtained embryos were lysed individually at 48 hpf, genotyped by T7E1 assay and sequenced. Because of CRISPR/Cas9 mosaicism, some different mutation patterns, which result in a non-frameshift or frameshift mutation, were observed.
- Extended Data Figure 3: Embryonic phenotypes of hox13 deletion mutants. (640 KB)
a, e, i, m, q, Whole body pictures at 72 hpf. a, Wild type, e, hoxa13a−/− (4 bp del./4 bp del.), i, hoxa13b−/− (4 bp del./14 bp ins.), m, hoxd13a−/− (5 bp ins./17 bp del.), and q, hoxa13a−/−, hoxa13b−/− double homozygous embryo (8 bp del./29 bp del., 14 bp ins./14 bp ins.). The details of mutant sequences are summarized in Extended Data Table 3. Wild-type and single homozygous fish for hoxa13a or hoxa13b were treated by PTU to inhibit pigmentation. The body size and length of mutant embryos are relatively normal at 72 hpf. n = 5 embryos for all genotypes. b, f, j, n, r, Bright field images of pectoral fins. Pectoral fins were detached from the body and photographed (Methods). Hoxa13a−/−, a13b−/− double homozygous embryo shows 30% shorter pectoral fin fold compared to wild type (r, see also Extended Data Fig. 5). n = 5 embryos for all genotypes. c, g, k, o, s, and1 in situ hybridization at 72 hpf. Hox13 mutants show normal expression patterns, which indicates that fin fold development is similar to wild type in these mutants. n = 3 embryos for all genotypes. d, h, l, p, t, shha in situ hybridization at 48 hpf. Hox13 mutants show a normal expression pattern that is related to relatively normal anteroposterior asymmetry of adult fin (Fig. 3, Extended Data Fig. 4 and Supplementary Information). n = 3 embryos for all genotypes. Scale bars are 1 mm (a), 200 μm (b, c) and 100 μm (d).
- Extended Data Figure 4: Phenotype of adult hox13 mutant fish. (499 KB)
a, c, e, g, i, k, m, Whole body morphology of hox13 deletion mutants were photographed at 4 months old; hoxa13a−/− (8 bp del./29 bp del.), hoxa13b−/− (4 bp del./14 bp ins.), hoxd13a−/− (5 bp ins./10 bp ins.), hoxa13a−/−, hoxa13b−/− double homozygous fish (8 bp del./29 bp del., 14bp ins./14 bp ins.) and triple knockout (k, m, mosaic for hoxa13a hoxa13b and hoxd13a) fish (Methods). n = 3 fish for wild type, single and double mutants and n = 5 fish for triple mosaic mutants (same specimens were used as in Fig. 3). The details of mutant sequences are summarized in Extended Data Table 3. Each homozygous mutant fish shows normal morphology at 4 months old except for slightly short pectoral fin rays of hoxa13a−/− or a13b−/− single mutants. Hoxa13a−/−, hoxa13b−/− double homozygous fish shows a severe reduction of fin rays in pectoral, pelvic, dorsal and anal fins compared with wild type. The triple knockout (mosaic for hoxa13a, hoxa13b and hoxd13a) fish also showed a reduction in fin rays. Scale bar is 5 mm. Owing to the size of the adult fish, three different pictures for anterior, centre and posterior of the body were merged to make whole-body pictures. b, d, f, h, j, l, n, Bone staining pictures of mutant fish. The endochondral bones of pectoral fins are shown. Whereas single homozygous fish show relatively normal proximal radials (b, d, f, h and Fig. 3), double homozygous mutants show fused third and fourth proximal radials (j). One triple knockout (mosaic for hoxa13a, hoxa13b and hoxd13a, 0, 25, 50%) fish had fused third and fourth proximal radials (i), but another triple knockout (0, 0, 0%) had more broken down proximal radials (n). n = 3 fish for wild type, single and double mutants and n = 5 fish for triple mosaic mutants (same specimens were used as in Fig. 3). The scale bar is 500 μm. o, p, Examples of counting distal radials in wild-type and hoxa13a−/−, hoxa13b−/− double homozygous fish. First distal radials are not shown in CT segmentation because of a fusion with first fin ray. q, The number variation of distal radials in mutant fish. Multiple fins were investigated in wild type (25 fish/50 fins), hoxa13a−/− (4 bp del./4 bp del., 3 fish/6 fins), hoxa13b−/− (4 bp del./14 bp ins., 3 fish/6 fins), hoxd13a−/− (5 bp ins./17 bp del., 3 fish/6 fins), hoxa13a−/−, hoxa13b−/− double homozygous (8 bp del./29 bp del., 14 bp ins./14 bp ins., 3 fish/6 fins) and triple knockout (mosaic for hoxa13a, hoxa13b and hoxd13a) fish (five fish/10 fins). The number of distal radials increased to 10 and 13 in double and triple mutants, respectively. The difference in distal radial number between wild-type and double homozygous or wild-type and triple knockout fish (mosaic for hoxa13a, hoxa13b and hoxd13a) is statistically significant (P = 0.0014 or P = 0.00001, respectively, t-test comparing the means, two-tailed distribution).
- Extended Data Figure 5: Analysis of embryonic fin fold and endochondral disk in hoxa13a−/−, hoxa13b−/− embryos. (436 KB)
a, A bright field image of wild-type and hoxa13a−/−, hoxa13b−/− pectoral fins at 72 hpf. Pectoral fins were detached from the body and photographed (Methods). Scale bar is 150 μm. b, The difference in fin fold length between wild-type and hoxa13a−/−, hoxa13b−/− embryos. The length of the fin fold was measured in wild-type (n = 8) and hoxa13a−/−, hoxa13b−/− double homozygous (n = 5) embryos at 72 hpf and 96 hpf (Methods). The length of the fin folds was decreased to about 70% of wild type in double homozygous embryos (72 hpf; P = 0.006, 96 hpf; P = 0.004, t-test comparing the means, one-tailed distribution, see Source Data). The error bars indicate s.e.m. c, d, Images of DAPI staining of wild-type (c) and hoxa13a−/−, hoxa13b−/− mutant (d) pectoral fins captured by confocal microscopy. White circles indicate nuclei in the endochondral disks. Scale bar is 200 μm. e, The average number of cells in the endochondral disk of wild-type and hoxa13a−/−, hoxa13b−/− mutant fins (see Methods and Source Data). The difference is statistically significant (P = 0.041 by Student’s t-test, one-tailed distribution). The error bars indicate s.e.m.
Extended Data Tables
- Supplementary Figure (5.4 MB)
This file contains a 3D pdf of wild-type and double mutant fin.