ZIC3 shapes digit morphogenesis in avian

Abstract

Through the evolutionary diversification of birds, a variety of digit morphologies have evolved as adaptations to diverse environments, with the regression of the anterior digits being a remarkable phenomenon. Transcriptomic analysis of chicken limb development has revealed molecular signatures in the anterior digits, among which Zic family member 3 (Zic3) stands out as a unique highly expressed transcription factor. However, the function of Zic3 in digit development in birds remains unclear. In this study, we investigated the expression pattern of Zic3 across five phylogenetically diverse avian species. Our analysis revealed a consistent Zic3 expression pattern in species from Neoaves and Palaeognathae, except for the ostrich. Ostrich hindlimbs exhibited increased Zic3 expression and an extended expression range from the first toe to the second toe. By locally expanding the expression domain of Zic3 in chicken anterior hindlimb buds we obtained a phenotype similar to that of early ostrich feet, characterized by significant shortening of digits I and II. We further performed mRNA Sequencing (mRNA-Seq) of the Zic3-overexpressed autopods and found that Zic3 inhibited skeletal development through multiple pathways, including the Wnt signaling pathway, ECM-receptor interaction and Focal adhesion. Our results reveal a pivotal role for Zic3 in anterior digits and identify its downstream regulatory mediators in birds.

Introduction

Diversity in avian limb digit morphology has been shaped by the complex interplay of natural selection and developmental constraint (De Bakker et al., 2013). In many Neoaves, the first toe of the hindlimb has a vestigial metatarsal compared to the other toes. However, the fully developed distal part of the toe provides the necessary grip and stability for perching on branches or other supports. In contrast, flightless birds, such as ostriches and emus, have reduced the number of distal anterior toes in their hindlimbs, which is thought to be an adaptive change facilitating energy conservation for fast running. Despite the conserved expression patterns of posterior genes across the various limb morphologies, anterior genes display increased evolutionary variability as observed in species such as chickens, emus, and zebra finches (De Bakker et al., 2013). This heightened variation in anterior genes may underlie the diversification of anterior digits during adaptation of avian species to a range of diverse habitats.

Recent comprehensive transcriptomic analyses of chicken limb development have revealed distinct molecular signatures within the first digit (Wang et al., 2011). We observed that the gene Zic3, which is essential for maintaining the pluripotency of embryonic stem cells, the differentiation of neuroectoderm and mesoderm, and determining the left-right axis of internal organs (Bellchambers and Ware, 2018), is highly expressed in the anterior side of limb buds (Wang et al., 2011). Through chromatin binding and genome-wide transcriptional profiling, Zic3 has been established to be a crucial regulatory transcription factor for embryonic stem cell differentiation and it regulates the activity of a number of other transcription factors (Yang et al., 2019). Loss of Zic3 function eliminates the preaxial polydactyly caused by Gli3 haploinsufficiency in mouse limb buds.

Previous studies have reported higher Zic3 expression levels in the first digit of chicken limbs, compared to other digits, and that its elevated local expression is maintained for an extended period. However, species-specific expression patterns for Zic3 and its role in avian limb development remain to be elucidated. This study aims to investigate the role of Zic3 expression in the regulation of avian digit development, and to offer insight into the molecular mechanisms underlying the diversity of bird digit morphologies.

Materials and methods

Ethics statement

All animal procedures were approved by the Institutional Animal Care and Use Committee of Shenyang Agricultural University (approval ID: 202106038).

Sample collection

This study involved five avian species: the emu (Dromaius novaehollandiae), African ostrich (Struthio camelus), chicken (Gallus gallus), duck (Anas platyrhynchos), and pigeon (Columba livia). Fresh fertile eggs for the experiments were purchased from Fuxin Tianyuan ostriches farm (emu and ostrich), Beijing Merial Vital Laboratory Animal Technology Company (chicken), Haicheng City poultry farm (duck) and Liaoyang County poultry farm (pigeon). Emu and ostrich eggs were incubated in an incubator at 36.5°C with 25 % humidity. Chicken, duck and pigeon eggs were incubated in an incubator at 37.8°C with 60 % humidity.

Whole-mount in Situ Hybridization (WISH)

Embryos were fixed overnight in 4 % paraformaldehyde (PFA). After that, they were dehydrated in a methanol gradient and stored at −80°C. Before hybridization, the embryos were rehydrated in a gradient. Embryos were incubated with 10 μg/ml proteinase K for 10 to 30 min and then washed in PBST and refixed in 0.2 % glutaraldehyde in 4 % PFA. Subsequently, they were placed in preheated hybridization solution containing 1–2 μg/ml digoxin (DIG)-labeled probe and incubated overnight at 65°C. After hybridization, washed embryos were treated with blocking solution and incubated overnight with alkaline phosphatase-coupled anti-DIG antibody. Next, antibodies were removed, and the washed embryos were incubated in NTMT containing NBT-BCIP (Roche). The reaction was stopped when the appropriate color was obtained. Embryos were then washed at least six times overnight in 1 % Triton X-100 solution to reduce background. Finally, the embryos were fixed in 4 % PFA and photographed using a Leica microscope (M165FC). Probes for Zic3 were generated using primers as described previously (Wang et al., 2011). At least three replicates were used for each species, except for emu that had two replicates.

Skeletal staining

In brief, for the staining of cartilage and bone, embryonic limbs were dissected in PBS and the skin was removed. After overnight fixation in 95 % ethanol the samples were transferred to acetone for an overnight incubation, and subsequently stained with alcian blue and alizarin red and stored in 1 % KOH until transparent.

Construction and infection of RCAS-Zic3 in chicken embryo limbs

The chicken Zic3 coding sequence (NCBI Refseq XM_040700079.2) was amplified and cloned into the NotI and CalI sites of RCAS (The RCAS vector was generously provided by Dr. Clifford J. Tabin, Harvard University.). RCAS-Zic3 virus was generated in DF1 cells (CL-0279, Procell Life Science Company, Wuhan, China). Using a microinjection system (Eppendorf, Hamburg, Germany), a virus solution containing fast green dye was injected into limb buds on the anterior side of chicken embryos at Hamburger and Hamilton stage 20 (HH20). Embryos were collected at HH35 and HH45, and the uninfected left wing and leg were used as controls to assess the morphological consequences due to the overexpression of Zic3.

mRNA-Seq and data analysis

We collected autopod samples from the hindlimbs of six chicken embryos: three from embryos infected with RCAS-Zic3 and three from control embryos infected with RCAS, with injections performed at HH10 and collection at HH30. Total RNA from the autopods was extracted using the RNeasy Plus Mini Kit (Qiagen). Purity and concentration of total RNA was assessed with Qubit (Life Technologies, Carlsbad, USA) and Nanodrop (Thermo, Waltham, USA), and RNA integrity was evaluated using the Agilent Bioanalyzer 2100 system (Agilent Technologies, Santa Clara, CA). Samples containing 3 μg of RNA were used as input material for library construction. Libraries were commercially sequenced on the Illumina NovaSeq 6000 platform (Illumina, San Diego, California) by Novogene Biotech (Tianjin, China), yielding approximately 30 million 150 bp paired-end reads per library.

Transcripts Per Million (TPM) was employed to normalize sequencing depth and transcript length for correlation calculations and principal component analysis (PCA) analysis. The dataset was Z-score normalized prior to PCA analysis using the "prcomp" function in R. Differential expression analysis was conducted using the DESeq2 package (version 1.42.0) (Love et al., 2014). Genes with an adjusted P-value < 0.05 and |log₂-FoldChange| ≥ 0.6 were defined as significantly differentially expressed. Gene Ontology (GO) analysis was performed using the Database for Annotation, Visualization, and Integrated Discovery (DAVID, v2024q2 release, https://david.ncifcrf.gov/) with a significance threshold set at P < 0.05. Gene Set Enrichment Analysis (GSEA) was conducted using the official GSEA software (v4.3.3).

Real-time quantitative PCR (RT-qPCR)

Total RNA was extracted from the autopods of hindlimb buds of samples overexpressing Zic3 and RCAS-injected controls using the RNeasy Plus Mini Kit (Qiagen). RNA was then reverse transcribed into cDNA using the PrimeScript™ RT reagent Kit (Takara, Dalian, China). Three independent RT-qPCR experiments were conducted on a real-time PCR detection system (LightCycler® 96, Roche, Switzerland) using TB Green® Premix Ex Taq™ II (Takara, Dalian, China) and specific primers for each gene. The primers used were as follows: chicken-Zic3, forward (F): GCAGCCCATCAAGCAAGAG, reverse (R): GACCAGTTTGTATTTCGCCTTG; ostrich-Zic3, F: CCAGCCCATCAAGCAAGAG, R: CACCAGTTTGTATTTCGCCTTG; Wnt2b, F: GTCTTCGGGAGGGTGATGC, R: GCGGATGCCGTAGTTGATG; Spp1, F: GTGACACCTTTCAACCGT, R: CTCTACGCTCTGATGTTGG; Thbs2, F: GCCAAGACAGAGAAGCAAG, R: TCCATCACCAGCATAACCT. Cav1, F: GGCAACATCTACAAGCCCAATA, R: GCCTTCCAAATCCCATCAA. Data are presented as means ± standard deviations (SD), with P < 0.05 considered to be statistically significant.

Results and discussion

To explore the expression patterns of Zic3 across diverse avian species, we examined its expression in five phylogenetically distinct bird species, including representatives from Palaeognathae (emu and ostrich) and Neoaves (chicken, duck, and pigeon). Consistent with previous findings (Wang et al., 2011), Zic3 exhibited highly expression in the metacarpal and metatarsal regions of the first digit in the fore- and hindlimbs of the chicken at HH29. Similar expression patterns were observed in ducks, pigeons and emus. In ostriches, Zic3 expression in the forelimbs aligns with that of other bird species. However, in the hindlimbs of ostriches, Zic3 expression is extended from the first toe to the phalangeal region of the second toe (Fig. 1A). Subsequently, we performed a quantitative analysis of Zic3 expression in digits I–IV (DI–IV) of the hindlimbs of chickens and ostriches. The results revealed that Zic3 is highly expressed only in the first toe of chicken hindlimbs but is also highly expressed in the second toe of the ostrich hindlimbs (Fig. 1B). This conserved and derived expression pattern suggests that Zic3 may play a crucial role in shaping the morphology of anterior digits in birds.

Fig. 1.

Expression of Zic3 in avian limb buds and the effects of Zic3-overexpression in chicken limb buds. A, Zic3 expression in the fore- and hindlimbs of bird embryos. Ostrich, chicken, duck, and pigeon limbs are examined at HH29, while emu hindlimbs are at HH28 and forelimbs at HH32 to adjust for the developmental heterochrony between the fore- and hindlimbs. The phylogenetic tree to the left illustrates the evolutionary relationships among the investigated species. myr, million years. The phylogeny was constructed from TimeTree (http://www.timetree.org). B, quantitative analysis of Zic3 expression in DI–IV of the hindlimbs of chickens and ostriches. C, validation of Zic3 overexpression in chicken limb buds. n = 3, **P < 0.01, ***P < 0.001. D, morphological assessments of chicken embryos injected with RCAS-Zic3 at HH20, with subsequent collection at HH35 and HH45. The model diagram shows the injection locations in green. E, skeletal staining of chicken fore- and hindlimbs injected with RCAS-Zic3 at HH20 and collected at HH35 (controls are shown on the left) and a wild-type (WT) ostrich hindlimb at HH32. Arrows indicate the shortened regions, with the white line represents the metatarsal and the black line represents the phalanx. F, comparison of chicken hindlimbs (RCAS and RCAS-Zic3 injected at HH10 and collected at HH30). RCAS injected and WT limbs are very similar whereas the Zic3-overexpressed limb is distinctly shorter. Purple lines indicate the boundaries used for the dissection to obtain samples for mRNA-Seq. In D–E, all morphological comparisons originate from the right limb infected with RCAS-Zic3, with the left limb serving as the uninfected control from the same individual. I and II represent the first and second digits, respectively, with dorsal views of the limbs presented in A, E, and F. The bar is 1 mm in A, E and F, and 1 cm in D.

To uncover the effect of Zic3 on the development of avian limbs, we injected avian retrovirus carrying RCAS-Zic3 into the anterior distal region of the right fore- and hindlimb buds of chicken embryos at HH20 and surveyed the morphological changes after Zic3-overexpression. RT-qPCR analysis revealed a significant upregulation of Zic3 mRNA in the chicken limb buds infected with RCAS-Zic3 compared to those infected with control RCAS, confirming the successful overexpression of Zic3 in chicken embryos (Fig. 1C). Morphological assessments revealed that the first and second digits of the injected limbs were significantly shortened at HH35 and HH45 compared to the uninjected control digits (Fig. 1D). Skeletal staining further demonstrated that the transgenic chicken hindlimbs at HH35 exhibited a phenotype similar to that of wild-type ostriches, characterized by shortened digits I and II (Fig. 1E). This similarity likely arises from the localized overexpression of Zic3, which mimics the natural expression pattern observed in ostrich hindlimbs. These results highlight the pivotal role of Zic3 in regulating avian limb development, particularly in modulating skeletal growth and maturation.

To investigate the impact of Zic3 overexpression on autopod development, we performed mRNA-Seq analysis on autopods from chicken embryos injected with RCAS-Zic3 or RCAS at HH10 and collected at HH30 (Fig. 1F). PCA of the transcriptomes revealed distinct clustering between Zic3-overexpression and control samples (Fig. 2A), with 445 differentially expressed genes (DEGs) identified (Fig. 2B). Many downregulated DEGs are associated with skeletal development, including key genes in the Bmp signaling pathway such as Bmp8a, Bmp3, and Mstn (Fig. 2B). The Bmp signaling pathway is crucial for the development of digit segmentation patterns (Grall et al., 2024). However, no significant changes were observed in the Hoxd-related genes, which may be due to Zic3 overexpression affecting digit length but not digit identity. Gene Ontology-Biological Process (GO-BP) enrichment analysis of DEGs highlighted the significant impact of Zic3 overexpression on skeletal development (Fig. 2C), corroborating its role as a transcriptional regulator of cell (Yang et al., 2019). Further pathway analysis using GSEA based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database indicated that the Wnt signaling pathway, ECM-receptor interaction, and Focal adhesion were inhibited in Zic3-overexpressing limbs (Fig. 2D). Heatmaps displaying the core genes of these pathways are shown (Fig. 2E). RT-qPCR validation of selected DEGs, including Wnt2b, Spp1, Thbs2, and Cav1, confirmed the mRNA-Seq results (Fig. 2F). In previous studies, ECM-receptor interaction and Focal adhesion were shown to play crucial roles in skeletal development (Brachvogel et al., 2013; Chen et al., 2023), and the Wnt signaling pathway is essential for osteoblast differentiation and bone metabolism (Vlashi et al., 2023). Zic3 has been shown to suppress Wnt/β-catenin signaling in the control of development of the notochord and organizer in Xenopus (Fujimi et al., 2012). Therefore, we concluded from our transcriptomic that Zic3 delays skeletal development in the autopods of chicken embryos by inhibiting the Wnt signaling pathway, ECM-receptor interaction and Focal adhesion.

Fig. 2.

mRNA-Seq analysis of chicken autopods with Zic3 overexpression. A, PCA analysis of transcriptomes derived from RCAS-Zic3 and control groups (n = 3). B, volcano plot showing DEGs between the RCAS-Zic3 and control groups. C, Top 15 terms identified in the GO-BP analysis of the RCAS-Zic3 DEGs, with six GO terms associated with skeletal development highlighted in red. D, GSEA analysis of genes in the Wnt signaling pathway, ECM-receptor interaction and Focal adhesion between RCAS-Zic3 and control. E, heatmap displaying the differential expression of core genes associated with the Wnt signaling pathway, ECM-receptor interaction, and focal adhesion. Genes with significant differences following injection of RCAS-Zic3 and control groups are highlighted. F, RT-qPCR analysis showing the expression levels of Wnt2b, Spp1, Thbs2 and Cav1 genes in the RCAS-Zic3 and control groups (n = 3, **P < 0.01, ***P < 0.001).

At the single-cell resolution during three stages of chicken hindlimb development, the synergistic actions of Fgf, Bmp, and Wnt signal pathway are suggested to regulate the growth of autopods and the elongation of digits (Feregrino et al., 2019). In this study, overexpression of Zic3 led to changes in some genes of the Fgf and Bmp signaling pathways, but only the Wnt signaling pathway was significantly suppressed. On one hand, this may indicate that in chicken embryos, Zic3 primarily modulates the expression of other signaling pathway genes by regulating the Wnt signaling pathway, thereby affecting limb morphology. On the other hand, the fact that overexpression of Zic3 in chicken embryo limbs only altered the length of the digits, without changing the identity of digits. Therefore, future studies should employ more precise sampling methods for mRNA-Seq with additional direct experiments to validate these findings.

Differential expression of Zic3 in avian limb buds, leading to morphological differences, provides an excellent example of developmental mechanisms shaping the distal digits of tetrapod. Whether this mechanism is universally applicable to other amniotes with reduced or lost digits, such as odd- and even-toed ungulates requires further investigation. In summary, our studies using overexpression of Zic3 in chicken embryos together with mRNA-Seq analysis of these tissues provides insight into mechanisms, and downstream regulatory networks, driving the diversity of avian limb morphology. This not only provides a new perspective on the mechanisms used by Zic3 to alter digital development but also identifies potential molecular targets for poultry science that could be used to improve skeletal characteristics and prevent disease.

Data accessibility

The mRNA-Seq data generated in this study has been deposited into the Gene Expression Omnibus (GEO) with accession number GSE 281985.

Author contributions

Z.W., S.Z., S.L. and S.B. designed the study. S.L. and S.B. analyzed data for the work. Y.T., X.X and J.Y. acquired the data; S.S. and J.Z. verified the data. S.L., S.B. and Z.W. wrote the manuscript. Z.W., S.Z. and D.M.I. reviewed it critically.

Declaration of competing interest

The authors declare that they have no competing interests.