Evolvability, observed

The most successful E. coli strains in a long-term evolution experiment were those that showed the greatest potential for adaptation

By Jef Akst

Natural selection picks the most well adapted organisms to survive and reproduce. But what if the most beneficial mutations in the short term meant less room for adaptation in the future?

New research suggests that the capacity of a species to further adapt to its environment, or evolvability, can be just as important, or more so, than the adaptations it's already acquired. The results, published this week in Science, give an empirical foundation to a theory that, in addition to beneficial mutations that confer immediate fitness advantages, long-term evolvability may be important for determining a species' success.

"[The idea of] selection for evolvability has been in the air for a long time, but this is one of the first real systematic and explicit demonstrations of this actually happening," said evolutionary biologist and population geneticist Michael Desai of Harvard University, who was not involved in the study.

Researchers at Michigan State University and the University of Houston in Texas took advantage of a long-term evolution experiment on Escherichia coli that's been running for more than 50,000 generations. Characterizing archived strains from 500, 1000, and 1500 generations, the team identified two beneficial mutations that arose in some strains prior to 500 generations and eventually spread through the entire population. The researchers dubbed the strains that carried these mutations at 500 generations the eventual winners (EWs) and those lacking the mutations the eventual losers (ELs).

The first surprise came when the team compared the fitness of four strains -- two EWs and two ELs -- and found that while all four strains had significantly higher fitness than the ancestral strain, the ELs appeared more fit than the EWs. Comparing the four strains directly confirmed the result: The two EW strains were at a significant disadvantage to the ELs. If these strains had not accumulated any more mutations, the researchers estimated the EWs would have gone extinct in just 350 additional generations.

But it wasn't just evolutionary luck that allowed the EWs to prevail over the ELs. When the researchers "replayed" evolution by culturing 10 replicate populations for each of the four strains for 883 generations, the EW populations almost always came out on top, acquiring more beneficial mutations that allowed them to overcome their fitness disadvantage to the ELs. The results suggested that the EWs, while initially at a disadvantage, prevailed in the long-term because they were more likely to acquire more beneficial mutations. In other words, the EWs had greater evolvability.

"It's exciting to see some evidence that this sort of thing, which everyone has been speculating about," Desai said. "The question is how common are these sort of interactions."

The large population sizes used in this study reduced competition between strains, allowing the less fit EWs to stick around long enough to accumulate the mutations that would eventually give them a fitness advantage. And the laboratory conditions in which the bacteria were held were unnaturally constant. Resource availability and environmental variation could affect the likelihood that evolution will select for strains on the basis of their evolvability.

"Whether natural populations can really select on those differences in evolvability will depend on a lot of things that we don't understand," said Desai. "But the point is that such differences in evolvability exist."

The researchers didn't find any changes in overall mutation rate among the strains, but proposed that the differences in evolvability may be due to interactions between the mutations that existed at the start of the replay experiments and those that later evolved. A mutation called topA1 in the EL strains, for example, which affects an enzyme involved in winding and unwinding DNA, appeared to dampen the fitness advantage of a subsequent mutation in the spoT gene, which encodes a global regulator of gene expression. While the spoT mutation conferred a significant fitness advantage to the EW strains that lacked the topA1 mutation, it had no significant effect on the EL strain, possibly due to the changes in DNA supercoiling that resulted from the EL mutation in topA.

"It's a pretty exciting result," said evolutionary biologist Paul Turner of Yale University, who did not participate in the study. "Evolutionary biology is not generally known as a predictive science," he said, noting the exception that scientists have used historical data on the success of flu virus lineages to predict which strains to target with the vaccine for the following year. "Maybe a study like this helps us do this in a more generalized way."

R.J. Woods et al., "Second-order selection for evolvability in a large Escherichia coli population," Science, 331: 1433-6, 2011.

Initially complex, always heterogeneous

Darren J. Burgess

The genetic complexity and heterogeneity of cancer is becoming increasingly appreciated through genomic and histological analyses. Two recent studies add further weight to this concept, revealing that even the subpopulation of leukaemia-initiating cells in individual patient samples can have surprising genetic heterogeneity.

By analysing individual leukaemic cells in each sample population, researchers found that the leukaemia initiating cell subpopulation (defined by cell surface markers or by serial xenotransplantation in mice)maintained a genetic heterogeneity that was similar to the population of leukaemia cells in the sample. This suggests that the linear clonal succession model of cancer evolution, in which cancers progress through single-cell clone bottlenecks, might be an oversimplification. The genetic heterogeneity and the branching evolutionary trajectories evident in the samples evoke a remarkably Darwinian perspective of the evolution of leukaemia-initiating cells.

Interestingly, a comparison of multiple leukaemia samples from individual patients during disease progression and post-treatment relapse revealed that shifts can occur in the dominance of the subclones, although the subclonal diversity is generally maintained.

It will be interesting to see whether these diverse subclones of cancerinitiating
cells equally fulfil all the biological properties of cancer stem cells. Another key question is whether the extent of genetic complexity seen for ALL-initiating cells will be mirrored in other cancers, given that acute leukaemias have a greater proportion of cancer-initiating cells compared with some solid tumours.

Finally, this genetic heterogeneity suggests that the therapeutic targeting of cancer-initiating cells will be a considerable challenge. However, the ability to isolate and study rare subclones in xenografts opens a pathway to developing therapies that would target all subclones.

See more:
Anderson, K. et al. Genetic variegation of clonal architecture and propagating cells in leukaemia. Nature 469,356–361 (2011).

Notta, F. et al. Evolution of human BCR–ABL1 lymphoblastic leukaemia-initiating cells. Nature 469, 362–367 (2011).

The dark side of induced pluripotency

MARTIN F. PERA

Induced pluripotent stem cells (iPSCs) are generated through the reprogramming of differentiated adult cells and can be coaxed to develop into a wide range of cell types. They therefore have far-reaching potential for use in research and in regenerative medicine. But the ultimate value of these cells as disease models or as sources for transplantation therapy will depend on the fidelity of their reprogramming to the pluripotent state, and on their maintenance of a normal genetic and epigenetic (involving aspects other than DNA sequence) status. Recent surveys show that the reprogramming process and subsequent culture of iPSCs in vitro can induce genetic and epigenetic abnormalities in these cells. The studies raise concerns over the implications of such aberrations for future applications of iPSCs.

It has long been known that, during cultivation in vitro, human embryonic stem cells (ESCs) can become aneuploid; that is, they acquire an abnormal number of chromosomes. The new papers have applied various state-of-the-art genomic technologies to assess in detail the occurrence and frequency of genetic and epigenetic defects in both human iPSCs and ESCs.

Hussein et al. studied copy number variation (CNV) across the genome during iPSC generation, whereas Gore and colleagues looked for point mutations in iPSCs using genome-wide sequencing of protein-coding regions. Lister et al. examined DNA methylation — an epigenetic mark — across the genomes of ESCs and iPSCs at the single-base level. These studies, along with other investigations into changes in chromosome numbers and CNV in the two kinds of stem cell, lead to the conclusion that reprogramming and subsequent expansion of iPSCs in culture can lead to the accumulation of diverse abnormalities at the chromosomal, subchromosomal and single-base levels. Specifically, three common themes, regarding the genetic and epigenetic stability of ESCs and iPSCs, emerge.

First, by several measures, iPSCs display more genetic and epigenetic abnormalities than do ESCs or fibroblasts — the cells from which they originated. Chromosomal abnormalities appear early during the culturing of iPSCs5, a phenomenon not generally observed in ESCs. Also, the frequency of mutations in iPSCs is estimated to be ten times higher than in fibroblasts. And there are greater numbers of novel CNVs (CNVs not found in the cell of origin or in human genomes of comparable background) in iPSCs than in ESCs. Similarly, the epigenome of iPSCs features incomplete reprogramming (with cells retaining epigenetic marks of the cell of origin), aberrant methylation of CG dinucleotides, and abnormalities in non-CG methylation — an epigenetic feature seen only in pluripotent cells.

The research groups report clues to the potential function of the genetic lesions that arise in ESCs and iPSCs. For example, regions prone to amplification, deletion or point mutation seem to be enriched in genes involved in cell-cycle regulation and cancer. Although the changes observed do not strongly implicate any particular gene functionally as a target for change during the amplification of iPSCs or during their adaptation to culture conditions, the frequent association of the affected genes with cancer gives cause for concern.

With regard to evaluating the safety of ESCs and iPSCs, a key issue is the biological significance of the changes that these studies report. Clearly, aneuploid cell lines would not be used in therapy (although they might be useful for research into the basis of genetic disorders associated with anomalies in chromosome number or other genetic abnormalities). Cell lines bearing mutations of established functional consequence in oncogenes or tumour suppressors, or in genes associated with Mendelian disorders (those usually due to a single gene), could equally not be used therapeutically.

Read more in Nature