Drift and Directional Selection Are the Evolutionary Forces Driving Gene Expression Divergence in Eye and Brain Tissue of Heliconius Butterflies

Abstract

Investigating gene expression evolution over micro- and macroevolutionary timescales will expand our understanding of the role of gene expression in adaptation and speciation. In this study, we characterized the evolutionary forces acting on gene expression levels in eye and brain tissue of five Heliconius butterflies with divergence times of ∼5-12 MYA. We developed and applied Brownian motion (BM) and Ornstein-Uhlenbeck (OU) models to identify genes whose expression levels are evolving through drift, stabilizing selection, or a lineage-specific shift. We found that 81% of the genes evolve under genetic drift. When testing for branch-specific shifts in gene expression, we detected 368 (16%) shift events. Genes showing a shift toward upregulation have significantly lower gene expression variance than those genes showing a shift leading toward downregulation. We hypothesize that directional selection is acting in shifts causing upregulation, since transcription is costly. We further uncovered through simulations that parameter estimation of OU models is biased when using small phylogenies and only becomes reliable with phylogenies having ≥ 50 taxa. Therefore, we developed a new statistical test based on BM to identify highly conserved genes (i.e., evolving under strong stabilizing selection), which comprised 3% of the orthoclusters. In conclusion, we found that drift is the dominant evolutionary force driving gene expression evolution in eye and brain tissue in Heliconius Nevertheless, the higher proportion of genes evolving under directional than under stabilizing selection might reflect species-specific selective pressures on vision and the brain that are necessary to fulfill species-specific requirements.

Keywords: Brownian motion; Ornstein–Uhlenbeck; RevBayes; natural selection; stabilizing selection.

Figure 1

Phylogenetic relationship of the Heliconius species used in this study showing divergence times at each node (Kozak et al. 2015).

Figure 2

Pairwise correlation between five Heliconius species and their respective per gene expression variances. The correlation strength between per gene expression variances was estimated by calculating Pearson’s ρ correlation coefficient.

Figure 3

Testing for a phylogenetic signal in gene expression levels of Heliconius using BM. Significance is shown at model probability > 0.75 (solid red line, Bayes factor > 3, positive support) and model probability > 0.95 (dashed red line, Bayes factor > 20, strong support). (A) Shows the comparison between the two nonphylogenetic models (identical vs. independent species mean). (B) Shows the model probability of the BM model compared with the independent species mean model. (C) Shows the model probability of the BM model compared with the identical species mean model. BM, Brownian motion.

Figure 4

Posterior mean estimates of the rate of gene expression change (σ² in blue) and the 95% threshold computed (red) using Monte Carlo simulations. The genes were sorted by an ascending estimate of σ². Inset: close-up of genes whose σ² is not significantly bigger than zero.

Figure 5

Model probability when testing model suitability when fitting an OU model for the assessment of stabilizing selection. Significance is shown at model probability > 0.75 (solid red, Bayes factor > 3, positive support) and model probability > 0.95 (solid red, Bayes factor > 20, strong support). There are only seven genes with significant support for stabilizing selection through an OU model. OU, Ornstein–Uhlenbeck.

Figure 6

Simulation study for the assessment of parameter estimation bias under an OU model. The relative bias in estimates of the attraction/selection parameter (α) through 1000 simulations under σ values ranging from 0.1 to 10, and α values ranging from 0.01 to 10. Simulations were performed for phylogenies with sizes ranging from 5 to 1000 taxa. OU, Ornstein–Uhlenbeck.

Figure 7

Bar plot showing branch-specific shifts on gene expression levels in Heliconius. Bars in light blue show branch shifts identified by BM models and dark blue bars show branch shifts identified by OU models. BM, Brownian motion; OU, Ornstein–Uhlenbeck.

Figure 8

The gene expression variances for all the genes showing a shift toward an up- and a downregulation are depicted as box plots for each species. Numbers above the box plots show the total number of genes identified with a BM and an OU model. Wilcoxon test: * P < 0.05, ** P < 0.01, and *** P < 0.001. BM, Brownian motion; OU, Ornstein–Uhlenbeck.

Figure 9

Between-species gene expression divergence plotted as a function of divergence time according to the Heliconius phylogeny. Red: σ² from gene expression levels observed in Heliconius. Blue: simulated gene expression divergence under random drift with different values of σ.

Figure 10

Between-species gene expression divergence plotted as a function of divergence time according to the Heliconius phylogeny. Red: σ² from gene expression levels observed in Heliconius. Blue: simulated gene expression divergence under different values of σ. Each panel shows estimates for a different value of α.