Lyubetsky VA et al. (2023), Wide-scale identification of novel/eliminated g...

XB-ART-60207

Biol Direct 2023 Aug 11;181:45. doi: 10.1186/s13062-023-00405-6.

Show Gene links Show Anatomy links

Wide-scale identification of novel/eliminated genes responsible for evolutionary transformations.

Lyubetsky VA , Rubanov LI , Tereshina MB , Ivanova AS , Araslanova KR , Uroshlev LA , Goremykina GI , Yang JR , Kanovei VG , Zverkov OA , Shitikov AD , Korotkova DD , Zaraisky AG .

???displayArticle.abstract???
BACKGROUND: It is generally accepted that most evolutionary transformations at the phenotype level are associated either with rearrangements of genomic regulatory elements, which control the activity of gene networks, or with changes in the amino acid contents of proteins. Recently, evidence has accumulated that significant evolutionary transformations could also be associated with the loss/emergence of whole genes. The targeted identification of such genes is a challenging problem for both bioinformatics and evo-devo research. RESULTS: To solve this problem we propose the WINEGRET method, named after the first letters of the title. Its main idea is to search for genes that satisfy two requirements: first, the desired genes were lost/emerged at the same evolutionary stage at which the phenotypic trait of interest was lost/emerged, and second, the expression of these genes changes significantly during the development of the trait of interest in the model organism. To verify the first requirement, we do not use existing databases of orthologs, but rely purely on gene homology and local synteny by using some novel quickly computable conditions. Genes satisfying the second requirement are found by deep RNA sequencing. As a proof of principle, we used our method to find genes absent in extant amniotes (reptiles, birds, mammals) but present in anamniotes (fish and amphibians), in which these genes are involved in the regeneration of large body appendages. As a result, 57 genes were identified. For three of them, c-c motif chemokine 4, eotaxin-like, and a previously unknown gene called here sod4, essential roles for tail regeneration were demonstrated. Noteworthy, we established that the latter gene belongs to a novel family of Cu/Zn-superoxide dismutases lost by amniotes, SOD4. CONCLUSIONS: We present a method for targeted identification of genes whose loss/emergence in evolution could be associated with the loss/emergence of a phenotypic trait of interest. In a proof-of-principle study, we identified genes absent in amniotes that participate in body appendage regeneration in anamniotes. Our method provides a wide range of opportunities for studying the relationship between the loss/emergence of phenotypic traits and the loss/emergence of specific genes in evolution.

???displayArticle.pubmedLink??? 37568147
???displayArticle.pmcLink??? PMC10416458
???displayArticle.link??? Biol Direct

???attribute.lit??? ???displayArticles.show???

	Fig. 1. Declining capability to regenerate large body appendages, i.e., limbs and tail, in the line of vertebrates from fish and amphibians to placental mammals. Shown are groups of vertebrates that emerged sequentially in evolution. Some extant species from these groups were collected into the lower, middle, and upper species sets to search for genes absent in reptiles, birds, and mammals but involved in tadpole tail regeneration in Xenopus tropicalis, which was chosen as the reference species R (see main text for details)
	Fig. 2. Scheme of the WINEGRET method using the example of a search for genes absent in mammals, birds, and reptiles but present in fish and amphibians and involved in the tail regeneration of Xenopus tropicalis frog tadpoles
	Fig. 3. The principles of orthologs search in Task I. A Forward check of a homolog X′ in species A of a given gene X in species R that involves, e.g., two witnesses: Y and Z in R and Y′ and Z′ in A. The dashed arrow represents the α-homology condition, and the solid arrow represents the β-homology (see for the explanation of α- and β-homology Step 2 in section A of Methods). The brackets indicate the chosen neighborhoods of the genes X and X′, in which the witnesses are sought. The witnesses may be arbitrarily positioned and directed within the neighborhood. B Backward check for the existence of an alternative U that is more similar to candidate X′ than X that involves, e.g., two witnesses. The dashed arrow represents the α-homology condition, and the solid arrow represents the β-homology. The brackets indicate the chosen neighborhoods of the genes X and X′, in which the witnesses are sought. The witnesses may be arbitrarily positioned and directed within the neighborhood
	Fig. 4. Checking whether gene X from R has an ortholog X′ in upper species A through a fish genome F that involves, e.g., two witnesses. Here, R is a fixed reference species; gene X has the ortholog X in fish F confirmed by two witness pairs, Y–Y* and Z–Z. The gene X in turn has the ortholog X*′ in upper species A confirmed by two witness pairs, V–V′ and T–T′. The dashed arrow represents the α-homology condition, and the solid arrow represents the β-homology (see for the determination of α- and β-homology Step 2 in section A of Methods). The brackets indicate the chosen neighborhoods with radii r, in which the witnesses are sought. The witnesses may be arbitrarily positioned and directed within the neighborhood
	Fig. 5. Unrooted phylogenetic tree of SOD1, SOD2, SOD3, and SOD4 family proteins in the following jawed vertebrates: shark (Callorhinchus milii), bony fish (Danio rerio), frog (Xenopus tropicalis), lizard (Gekko japonicus), chicken (Gallus gallus) and human (Homo sapiens). Protein alignment and phylogenetic tree were built using MAFFT v7.511 [27] and IQ-TREE v2.2.0 [25, 44] tools for proteins under the following accessions: AFM87136.1 for Sod1Shark, NP_571369.1 for Sod1Fish, NP_001016252.1 for Sod1Frog, XP_015271096.1 for Sod1Lizard, NP_990395.2 for Sod1Chick, CAG46542.1 for SOD1Human, NP_001279581.1 for Sod2Shark, NP_956270.1 for Sod2Fish, NP_001005694.1 for Sod2Frog, XP_015268482.1 Sod2Lizard, NP_989542.2 for Sod2Chick, NP_001019636.1 for SOD2Human, AFM90279.1 for Sod3Shark, XP_001332758.1 for Sod3Fish, NP_001106630.1 for Sod3Frog, XP_015272245.1 for Sod3Lizard, XP_040525137.1 for Sod3Chick, CAG46651.1 for SOD3Human, XP_042193029.1 for Sod4Shark, XP_001343650.5 for Sod4Fish, and XP_017953150.2 for Sod4Frog. The branch support values were calculated using the ultrafast bootstrap approximation [43] with 1000 replicates
	Fig. 6. Testing selected gene expression in Xenopus laevis tadpole tail regeneration. A Schema of experiments. The level of tail amputation is shown by red dotted line. B Results of qRT-PCR analysis of the expression of three genes selected from the final list in cells of tip of the amputated tail stump
	Fig. 7. The effect of vivo-MO-induced knockdown of c-c motif chemokine 4, eotaxin-like, and sod4 on the regeneration of the amputated Xenopus laevis tadpole tail. A Schema of experiments. Tail stumps were successively injected at 0, 1 and 2 dpa with either control vivo-MO or with vivo-MO to mRNA of each of the aforementioned three genes. The level of amputation is shown by red line. B–E Typical samples of tails of Xenopus laevis tadpoles injected with the indicated vivo-MO at 4th day after amputation. The level of amputation is shown by red line. F Percent distribution of three types of the regenerating tails of Xenopus laevis tadpoles (normal regeneration, partial and no regeneration) among the 4 dpa tails injected with the control vivo-MO and vivo-MOs to c-c motif chemokine 4, eotaxin-like, and sod4 mRNA. Common scale bar for B–E of 300 microns is shown in B
	Fig. 8. The length of the list-4 of genes depending on \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document}λ value. For \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda$$\end{document}λ from 0.02 to 0.42, the number of genes in the list decreases from 107 to 68. When \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$\lambda = 0$$\end{document}λ=0, this number is 109

References [+] :

Altenhoff, Inferring Orthology and Paralogy. 2019, Pubmed