Diversification dynamics of rhynchostomatian ciliates: the impact of seven intrinsic traits on speciation and extinction in a microbial group

Abstract

Ciliates are a suitable microbial model to investigate trait-dependent diversification because of their comparatively complex morphology and high diversity. We examined the impact of seven intrinsic traits on speciation, extinction, and net-diversification of rhynchostomatians, a group of comparatively large, predatory ciliates with proboscis carrying a dorsal brush (sensoric structure) and toxicysts (organelles used to kill the prey). Bayesian estimates under the binary-state speciation and extinction model indicate that two types of extrusomes and two-rowed dorsal brush raise diversification through decreasing extinction. On the other hand, the higher number of contractile vacuoles and their dorsal location likely increase diversification via elevating speciation rate. Particular nuclear characteristics, however, do not significantly differ in their diversification rates and hence lineages with various macronuclear patterns and number of micronuclei have similar probabilities to generate new species. Likelihood-based quantitative state diversification analyses suggest that rhynchostomatians conform to Cope's rule in that their diversity linearly grows with increasing body length and relative length of the proboscis. Comparison with other litostomatean ciliates indicates that rhynchostomatians are not among the cladogenically most successful lineages and their survival over several hundred million years could be associated with their comparatively large and complex bodies that reduce the risk of extinction.

Figure 1

Maximum clade credibility species tree and corresponding cloudogram of the subclass Rhynchostomatia. The cloudogram is shown in green colour and consists of 7,500 superimposed post burn-in trees from the posterior distribution of the BEAST coalescent analysis. Branch colours designate tree partitions inferred to evolve at three significantly different rates by MEDUSA, with maximum-likelihood estimates of net-diversification rate (r) and extinction fraction (ε) for each tree partition. The tracheliid lineage diversified at a net-diversification rate of 4.88 (violet colour), while the dileptid clade at a net-diversification rate of 22.95 with an extinction fraction of 0.85 (red colour). However, there was a statistically significant shift in net-diversification rate to 14.91 within dileptids at the node uniting members with predominantly scattered macronuclear nodules (blue colour). The scale bar indicates number of substitutions. Only posterior probabilities > 0.95 are shown.

Figure 2

Maximum-parsimony character state reconstruction of five binary traits. The maximum clade credibility tree shown in Fig. 1 served as a scaffold for reconstruction of character state evolution. Squares at tips of branches show the character state of each taxon. For further details, see Supplementary Table S2. Explanations: CG, cortical granules; CV, contractile vacuoles; DB, dorsal brush; EX, extrusomes; P, proboscis.

Figure 3

Posterior densities of differences in net-diversification rates between lineages with two-rowed and multi-rowed dorsal brush, between lineages with two types and one type of extrusomes, between lineages with dorsally and dorsally as well as ventrally located contractile vacuoles, and between lineages with more then two and two vacuoles. The x-axis represents the difference in net-diversification rates (∆r) between lineages with particular character states (∆r = r ₁ − r ₀; subscripts indicate different character states as specified in Supplementary Table S8). The ∆r distribution was obtained from MCMC simulations on a sample of 100 trees from the posterior distribution of the BEAST coalescent analysis. The y-axis represents posterior densities of differences in net-diversification rates, as calculated on the basis of the MCMC simulations. Arrows denote median of the difference in net-diversification rates (∆r). For interquartile ranges and further details, see Table 1. Ninety-five per cent intervals of highest posterior density are indicated below densities.

Figure 4

Maximum-parsimony character state reconstruction of micro- and macronuclear traits as well as posterior densities of speciation rates for four macronuclear patterns in Bayesian MuSSE analyses. The maximum clade credibility tree shown in Fig. 1 served as a scaffold for both reconstruction analyses. Squares at tips of branches show the character state of each taxon. The x-axis represents distribution of speciation rates for mononodulate macronucleus (yellow colour), binodulate macronucleus (blue colour), moniliform macronuclear strand (green colour), and many scattered macronuclear nodules (grey colour). Distributions of speciation rates were obtained from MCMC simulations on the maximum clade credibility tree in the R-package diversitree. The y-axis represents posterior densities of speciation rates calculated on the basis of the MCMC simulations with exponential prior and 10,000 samples. Ninety-five per cent intervals of highest posterior density are indicated below densities.