Extremes in chromosome numbers in closely related taxa: A combined approach to studying the underlying karyotype evolution
Final Report Abstract
A long-standing question about plant genomes is how some groups have managed to organize their genes on tremendously varying numbers of chromosomes, sometimes over short evolutionary time spans. This is the overarching question we addressed in this project, using Araceae as our study systems. Specifically, we focused on (i) which karyotype rearrangements have led to chromosome number reduction or increase and (ii) what is the ecological context and time frame of increase or decrease in chromosome number on the focal clades. To test the monophyly and relationships of our focal genera (for which we had cultivated material in Munich), we generated a concatenated DNA sequence nuclear/plastid matrix of some 4500 characters and 86 species (with 98 accessions) representing all geographically and morphologically distinct groups in the tribe Areae. Our first paper from this project presented the phylogeny and formalized some of the nomenclatural changes required by our phylogenetic findings. The new phylogeny matches inflorescence characters and biogeography, with one clade essentially Australian, one essentially SE Asian, and one Mediterranean. We also obtained chromosome counts for key species, which revealed new cases of apparent tetraploidy and octoploidy, with switches to polyploidy appearing three times on the phylogeny. We later found that multiple transitions to polyploidy are very rare in other, similarly species-rich clades of Araceae. To place the observed chromosome number changes in absolute time we applied various molecular-clock models to our DNA matrix of the tribe Areae, which led to the observation of an unusually symmetric tree topology (before and after ultrametrization). This in turn, led to attempts to model diversification in the Areas. At the time when we were carrying out this work, workers were fascinated with so-called lineage-through-time (LTT) plots to study diversification through time. These plots show the number of nodes in a phylogeny in successive time slices, starting from 1 for the root node. Many LTT plots seemingly reveal a slowing in lineage accumulation towards the present, but our simulations revealed this to be a statistical artefact. Specifically, when running models with various combinations of speciation and extinction rates on subsets of our Areae data, we realized that unevenly sampled phylogenies in which the early-branching or especially old (deep) lineages are oversampled relative to young lineages invariable give a down-turning LTT plot. Worse, the statistical tests commonly used to assess the significance of downturns in such plots, namely the γ statistic with the Monte Carlo constant-rates (MCCR) test or with AIC scores in birth-death likelihood analyses, both indicate that such downturns are highly significant. Thirdly, we compiled published chromosome numbers of the Araceae (including our new counts a total of 862 (26% of the family’s species), plotted them on a new DNA phylogeny, and then inferred change in chromosome numbers on a phylogeny. Chromosome number is an unusual trait in neither being continuous nor always having clear spaced 'bins'. Instead, chromosome number changes are thought to change via duplication but also due to the gain or loss of a single chromosome (and very rare other types of changes). Luckily, in 2010, a new approach was published that moved the inference of change in chromosome numbers to maximum likelihood (ML) character state reconstruction. This ML approach showed that (different from previous suggestions) high, not low, chromosome numbers, most likely n = 16 or n = 18, are the ancestral state for the Araceae, with repeated reductions in chromosome number over evolutionary time. Of course, ancient polyploidization events must be harder to "pick up" than recent ones, because of the genomic restructuring that follows polyploidization.
Publications
- 2010. A phylogeny of the Areae implies that Typhonium, Sauromatum, and the Australian species of Typhonium are distinct clades. Taxon 59(2): 439-447
Cusimano, N., M. Barrett, W. L. A. Hetterscheid, and S. S. Renner
- 2010. Slowdowns in diversification rates from real phylogenies may not be real. Systematic Biology 59(4): 458-464
Cusimano, N., and S. S. Renner
- 2012. Maximum likelihood inference implies a high, not a low, ancestral haploid chromosome number in the Araceae, with a critique of the bias introduced by “x”. Annals of Botany 109: 681-692
Cusimano, N., A. Sousa, and S. S. Renner
(See online at https://doi.org/10.1093/aob/mcr302) - 2014. Combining FISH and model-based 36 predictions to understand chromosome evolution in Typhonium (Araceae). Annals of 37 Botany 113(4): 669-680
Sousa, A., N. Cusimano, S. S. Renner
(See online at https://doi.org/10.1093/aob/mct302) - 2014. Ultrametric trees or phylograms for ancestral state 39 reconstruction – does it matter? Taxon 63(4): 721-726
Cusimano, N., and S. S. Renner
(See online at https://doi.org/10.12705/634.14) - . 2015. Interstitial telomere-like repeats in the monocot family 41 Araceae. Botanical Journal of the Linnaean Society 177(1): 15-26
Sousa, A., and S. S. Renner
(See online at https://doi.org/10.1111/boj.12231)
