Convergent evolution of gene networks by single-gene duplications in higher eukaryotes

Abstract

By combining phylogenetic, proteomic and structural information, we have elucidated the evolutionary driving forces for the gene-regulatory interaction networks of basic helix-loop-helix transcription factors. We infer that recurrent events of single-gene duplication and domain rearrangement repeatedly gave rise to distinct networks with almost identical hub-based topologies, and multiple activators and repressors. We thus provide the first empirical evidence for scale-free protein networks emerging through single-gene duplications, the dominant importance of molecular modularity in the bottom-up construction of complex biological entities, and the convergent evolution of networks.

Figure 1

Two possible patterns of network evolution. We denote hubs as circles, repressors as inverted triangles, factors that have ambivalent or ambiguous function as diamonds and additional dimerization domains as squares. (A) Evolution of a heterodimerization network by single-gene duplication. (i) Initial state of the ancestral network A. (ii) Single-gene duplication of the hub. The duplicated protein has the same dimerization properties as the ancestral hub. (iii) Intrusion of an additional dimerization domain in the emergent hub. Due to this intrusion, the emergent hub has higher affinity for itself (homodimer; indicated by an arrow) than the other members and, thus, is isolated from the ancestral network. (iv) Single-gene duplication of the emergent hub. The two new members can homodimerize as well as heterodimerize with each other. They have higher affinity for each other than members of the ancestral network. (v) Point mutations change the specificity of the newest member such that it can only heterodimerize with the emergent hub; therefore, it behaves as a peripheral member of the emergent network B. (vi) Single-gene duplication of the emergent peripheral member. The new protein has the same dimerization specificities as its parental protein. Both of them heterodimerize with the emergent hub. Members of the emergent network B have a monophyletic origin, and the additional dimerization domains have coevolved with the main dimerization domain. (B) Evolution of a heterodimerization network by large-scale gene duplication. (i) Initial state of the network. (ii) Large-scale gene duplication. Every duplicated gene will have the same dimerization specificities as its ancestral gene, thus forming a complex network. (iii) The duplicated members isolate from the ancestral network. The members of the emergent network are not monophyletic.

Figure 2

bHLH heterodimerization network. (A) Cladogram of the human bHLH domains depicting a summary of the full neighbour joining tree (see supplementary information online), which is in accordance with previous publications (Atchley & Fitch, 1997; Ledent & Vervoort, 2001). (B) Domain architecture of the bHLH class of transcription factors, including the DNA-binding basic region, the HLH dimerization domain and other additional dimerization domains that lie C-terminal to the HLH. These additional dimerization domains are believed to confer dimerization specificity. (C) Topology of the bHLH protein network, based on the protein–protein interactions among members of the bHLH class. The network is compartmentalized according to five phylogenetic groups (A–E), based on both our own analyses and those of others (Atchley & Fitch, 1997; Ledent & Vervoort, 2001). We denote hubs as circles, activators as triangles, repressors as inverted triangles, and factors that have ambivalent or ambiguous function as diamonds. Interactions observed between Drosophila bHLH proteins have also been confirmed for mammalian proteins of the same subfamily. They are assumed to be ancestral and highly conserved, and therefore are denoted with thicker lines.