Theories
Genome size and complexity variation has been a long-term debate during the last decades.
In multi-cellular eukaryotes, genome expansion is a consequence of noncoding DNA proliferation [1]. Several theories have emerged to explain variation in genome size and complexity. Among them, the most generally accepted are the bulk-DNA hypothesis, followed by the selfish –DNA hypothesis [2]. However, theses hypotheses explain only partially divergent patterns observed in eukaryotes.
Mutational burden hypothesis (MBH), which is mostly based on population genetics principles, is a unifying concept that attempts to reconcile different points of view. This hypothesis implies that “…noncoding element are generally deleterious but proliferate nonadaptively when small effective population reduce the effectiveness of selection relative to genetic drift”. In other words[3], the genome is constantly under two nonadaptative forces: random genetic drift and mutation pressure.
What was expected?
If the MBH is correct, a genome under high mutation rate would be reduce in term of size and complexity.
A glimpse of plant mitochondrial genomes: what make them special
Mitochondrial genomes exhibited a broad range of diversity in term of genome structure and diversity among eukaryotes [4]. The plant mitochondrial genome contain usually more than 90% of non coding DNA with usually low point mutation rate whereas animal mitochondrial seems refractory to such expansion of noncoding DNA [5]. The authors select the genus Silene, which include members with high mitochondrial mutations rate, while other members within the same genus have maintained their low rates.
Findings and Interpretations
A massive expansion of genome associated to massive acceleration of mutation rates at DNA level was clearly established in S. noctiflora and S. conica, as compared to S. vulgaris and S. latifolia (Figs 1,2 and 3). However, during our round table discussion, it was unclear how the branch length of the tree presented in the figure 1 was computed. As no branch values were shown, was it done based on pre-computed data?
Theses observations were neither correlated to gene nor intron content. Usually genome growth is largely dependent on intronic and intergenic sequences. Intronic sequences did not shown significant variation among Silene species (shown in Table 1). As expected, this massive genome expansion was mostly due to intergenic sequences, which constitute 99% of the total genome size. These intergenic sequences in S. conica and S. noctifloralack detectable homology when compared to other genomes. A possible explanation may be that high mutation rates may have exerted such pressure that made them significantly diverge from their counterparts in other Silene.
A striking feature in S. conica and S. noctiflora, was the large number of imperfect repeats observed, which were linked to the presence of large number of small circular-mapping chromosomes. It is also worthwhile to see that these chromosomes shared only short repeats with other parts of the genome. At the opposite of what was found in S. vulgarisand S. latifolia, fast-evolving genomes in S. conica and S. noctiflora had a reduced recombination rate (figure 6). The underlying idea is that high mutational rate may favor changes in the repeats that make them less efficient for recombination. However this argument has to be considered with caution, as recombination may also favor formation of novel sequences or chimeras, which may potentially contribute to genome instability instead of maintenance. It is still unclear whether this impaired recombination activity in fast evolving genome may be responsible for the expansion, but at least it would partially agree with the MBH.
The authors investigated, if the biparental inherence and heteroplasmy may play a role in genome expansion and finally claim that there is no significant impact, even if the supporting data was not shown, their logic behind was quite forward. ii) The same conclusion was draw from intraspecific nucleotide polymorphism.
Discussion
Although the exact origin of expanded intergenic regions is still unclear, the authors discussed several potential answers:
1. “Intergenic content may derive from nuclear genome”
This is unlikely as no significant homology with nuclear data could be readily identified.
2. “Intergenic content may be due to selfish element proliferation”
The selfish DNA proliferation does not explain at all this genome expansion as no drastic change in terms of identifiable repeated elements was identified between fast evolving genomes and their counter parts.
3. “Increase in intergenic content may be due to impairment of DNA repair mechanism coupled to high mutation rate?”
Since the population size and environmental conditions are important for MBH, at first glance it seems that the paper did not describe sufficiently the factors. For example, we discussed that the S. vulgaris and S. latifolia are known to be invasive, whereas their fast evolving S. conica and S. noctifloraare not invasive. A partial answer of this question is provide by Lynch [2], who wrote “that forces driving the evolution of genomic architecture are unlikely to be a direct consequences of organisms difference in lifestyle”. Since the genetic drift is important for MBH, the accumulation of intergenic sequences may be due also to a deficiency in removal mechanisms due to small population size.
Aspects not covered by the paper
The epigenome status and a potential link with genome expansion were not investigated at all in the paper. To which extent these factors affect variation in genome size and complexity remains an open question.
My take home message
The authors started with very interesting observations (i.e. massive genome expansion in two Silene species) and compared to the predictions from MBH. The finding described in the paper does not support prediction derived from the MBH. Despite significant effort from the authors, there is still no clear answer about the exact origin of overwhelming intergenic DNA in genome expansion neither the driving forces behind it.
Sloan, D., Alverson, A., Chuckalovcak, J., Wu, M., McCauley, D., Palmer, J., & Taylor, D. (2012). Rapid Evolution of Enormous, Multichromosomal Genomes in Flowering Plant Mitochondria with Exceptionally High Mutation Rates PLoS Biology, 10 (1) DOI: 10.1371/journal.pbio.1001241
Other References
1. Lynch M, Conery JS (2003) The origins of genome complexity. Science 302: 1401-1404.
2. Lynch M (2006) Streamlining and simplification of microbial genome architecture. Annu Rev Microbiol 60: 327-349.
3. Lynch M, Koskella B, Schaack S (2006) Mutation pressure and the evolution of organelle genomic architecture. Science 311: 1727-1730.
4. Lang BF, Gray MW, Burger G (1999) Mitochondrial genome evolution and the origin of eukaryotes. Annu Rev Genet 33: 351-397.
5. Wolfe KH, Li WH, Sharp PM (1987) Rates of nucleotide substitution vary greatly among plant mitochondrial, chloroplast, and nuclear DNAs. Proc Natl Acad Sci U S A 84: 9054-9058.
posted by MRR for Ousmane H. Cissé
Mitochondrial genomes are odd though: there are ~20 copies per mitochondrion, and 1000-2000 mitochondria per cell.One can envisage a situation where some mitochondria have a mutation inactivating gene A, while others have a mutation inactivating gene B: now you have to maintain both of these populations within the cell. Depending on the dynamics of mutation rate, mitochondrial genome replication, recombination and gene conversion, these (initially identical) copies could subsequently diverge from each other.Rinse and repeat – with a high mutation rate – and you end up with a large population of mitochondrial chromosomes, each retaining only a small number of functional genes, with the remainder of each circle degenerated into random rubbish.Is that really “genome expansion” in the same way you’d talk about changes in nuclear genome size? I guess it’s related, in that one mechanism for nuclear genome expansion is by duplication of a genome segment followed by degeneration or neofunctionalisation of the duplicated genes – but in the mitochondrial case, the “duplication” is inherent in the fact that you started with thousands of copies per cell!There may be an interesting analogy to be drawn with the proliferation of repeats on Y chromosomes as they degenerate.
Indeed, assembly of such template containing thousand of mitochondrial genomes may be challenging. Repeat regions are particularly difficult to assemble, and there is a risk to collapse repeat regions. However, such “errors” would decrease the assembly size. The presence of multiples copies of mitochondrial genomes can theoretically detect by an extremely high coverage compared to the nuclear genome. But, because the nuclear genome is not yet available, it was not possible to perform such comparison. Evidences of mitochondrial genome expansion provided by the authors in Silene conica and Silene noctiflora are quite convincing, as compared to other members within the genus. I agree that many of the mitochondrial chromosomes lacks intact genes, which may indicates that genes undergoes duplication followed by degeneration – as you suggested -. Nevertheless, the genome expansion is not correlated with the gene content. As shown in table 1, the gene repertoire in fast evolving species, do not exhibit significant degeneration. Here, the vast majority of the genome is composed by noncoding regions without significant contribution of a particular type of repeat.