Genetic basis for soma is present in undifferentiated volvocine green algae

Abstract

Somatic cellular differentiation plays a critical role in the transition from unicellular to multicellular life, but the evolution of its genetic basis remains poorly understood. By definition, somatic cells do not reproduce to pass on genes and so constitute an extreme form of altruistic behaviour. The volvocine green algae provide an excellent model system to study the evolution of multicellularity and somatic differentiation. In Volvox carteri, somatic cell differentiation is controlled by the regA gene, which is part of a tandem duplication of genes known as the reg cluster. Although previous work found the reg cluster in divergent Volvox species, its origin and distribution in the broader group of volvocine algae has not been known. Here, we show that the reg cluster is present in many species without somatic cells and determine that the genetic basis for soma arose before the phenotype at the origin of the family Volvocaceae approximately 200 million years ago. We hypothesize that the ancestral function was involved in regulating reproduction in response to stress and that this function was later co-opted to produce soma. Determining that the reg cluster was co-opted to control somatic cell development provides insight into how cellular differentiation, and with it greater levels of complexity and individuality, evolves.

Keywords: VARL domain; gene duplication; germ-soma differentiation; major transitions; multicellularity; volvocine algae.

Figure 1

Species phylogeny and micrographs of exemplar species of volvocine algae. A. Bayesian species tree, consistent with previously published species trees. Color of species without (Chlamydomonas reinhardtii, red; Gonium pectorale, orange) and with the reg cluster (undifferentiated Pandorina morum, Platydorina caudata, Yamagishiella unicocca, Eudorina elegans UTEX 1212, green; soma differentiated Pleodorina californica, blue) correspond to other figures, Volvox (germ and soma differentiated) species for which the reg cluster has been previously sequenced are shown in black species in grey are not included in this analysis. Note that numbers following E. elegans species refer to UTEX strain numbers. Inferred origin of the reg cluster is denoted. See Figure 5 for maximum likelihood and Bayesian support values. B. C. reinhardtii (scale bar, 10 μm); C. G. pectorale (10 μm); D. Pla. caudata (25μm); E. V. ferrisii (50 μm); F. Pan. morum (10 μm); G. Y. unicocca (20 μm); H. E. elegans UTEX 1212 (10 μm); I. V. carteri f. nagariensis (50 μm); J. Ple. californica (25 μm).

Figure 2

Gene synteny near the reg cluster and closely related regA-like genes (bold). Synteny of C. reinhardtii (red), G. pectorale (orange), Pan. morum (green), Pla. caudata (green), Y. unicocca (green), E. elegans UTEX 1212 (green), Ple. californica (blue), and V. carteri (black) is shown. All available data from Pan. morum, Pla. caudata, E. elegans UTEX 1212, and Ple. californica are shown, while representative genomic regions from C. reinhardtii, G. pectorale, Y. unicocca, and V. carteri are shown. Scaffold or chromosome numbers are indicated for C. reinhardtii, G. pectorale, Y. unicocca, and V. carteri. Putative reg cluster and RLS1/rlsD orthologs are connected with black lines, syntenic genes are connected with gray lines.

Figure 3

Maximum likelihood VARL domain tree. Color of species without (C. reinhardtii, red; G. pectorale, orange) and with the reg cluster (Pan. morum, Pla. caudata, Y. unicocca, and E. elegans UTEX 1212, green; Ple. californica, blue; V. carteri, V. ferrisii, and V. gigas, black) correspond to other figures. Nodes with 80% or higher bootstrap support values are labeled with support values; unlabeled nodes have less than 80% bootstrap support.

Figure 4

Protein similarity plots for all reg cluster and RLS1/rlsD proteins based on syntenic position (rlsA, regA, rlsB, rlsO, rlsC, and rlsD). Regions showing high similarity are highlighted with gray boxes. The two peaks in the shaded VARL region represent the N-terminal extension and core VARL domain separated by the less conserved linker region (Duncan et al., 2007).

Figure 5

Ancestral character state reconstruction of somatic differentiation. Branch color refers to undifferentiated (gray) or somatic differentiation (black) inferred by maximum likelihood methods using the equal transition rates model. Dashed branches indicate an ambiguous maximum likelihood reconstruction. Large font numbers at selected nodes indicate Bayes Factors using the equal rates model; negative, support for undifferentiated; positive, support for somatic differentiated. Bayes Factors are interpreted following Kass and Raftery (1995): 0-2 weak evidence, 2-6 positive evidence, 6-10 strong evidence, >10 very strong evidence. Small font numbers along branches indicate Bayesian posterior probabilities (left of slash) and maximum likelihood bootstrap values (right of slash). Unlabeled nodes are supported with 1.00 PP and 100% bootstrap values. Bayes Factors and support values are colored consistent with the reconstructed state at that node.

Figure 6

Conceptual schematic of hypothesized VARL gene expression patterns in unicellular, undifferentiated, and differentiated species. A. Temporal expression of RLS1 (dashed) in C. reinhardtii in response to environmental change. B. The reg cluster (shown as four genes, though some species have five reg cluster genes) maintains expression in response to environmental change following its origin from the duplication of rlsD (dashed). C. The reg cluster is developmentally co-opted to control cell division throughout the life cycle. D. The reg cluster is developmentally co-opted to control the AP axis. Panels B, C, and D are not mutually exclusive as different hypotheses may apply to different undifferentiated species and a single hypothesis may not uniformly apply to all genes within the reg cluster of a given species. E. The reg cluster is co-opted to regulate somatic differentiation in V. carteri. For all panels, darker shading represents higher gene expression.