Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2015 Mar;197(5):872-81.
doi: 10.1128/JB.02263-14. Epub 2014 Dec 15.

Genome Instability Mediates the Loss of Key Traits by Acinetobacter Baylyi ADP1 During Laboratory Evolution

Affiliations
Free PMC article

Genome Instability Mediates the Loss of Key Traits by Acinetobacter Baylyi ADP1 During Laboratory Evolution

Brian A Renda et al. J Bacteriol. .
Free PMC article

Abstract

Acinetobacter baylyi ADP1 has the potential to be a versatile bacterial host for synthetic biology because it is naturally transformable. To examine the genetic reliability of this desirable trait and to understand the potential stability of other engineered capabilities, we propagated ADP1 for 1,000 generations of growth in rich nutrient broth and analyzed the genetic changes that evolved by whole-genome sequencing. Substantially reduced transformability and increased cellular aggregation evolved during the experiment. New insertions of IS1236 transposable elements and IS1236-mediated deletions led to these phenotypes in most cases and were common overall among the selected mutations. We also observed a 49-kb deletion of a prophage region that removed an integration site, which has been used for genome engineering, from every evolved genome. The comparatively low rates of these three classes of mutations in lineages that were propagated with reduced selection for 7,500 generations indicate that they increase ADP1 fitness under common laboratory growth conditions. Our results suggest that eliminating transposable elements and other genetic failure modes that affect key organismal traits is essential for improving the reliability of metabolic engineering and genome editing in undomesticated microbial hosts, such as Acinetobacter baylyi ADP1.

Figures

FIG 1
FIG 1
A. baylyi ADP1 evolved increased fitness during the 1,000-generation adaptive evolution experiment. Coculture competition assays were conducted between a GFP-tagged variant of the ancestor and the ancestral strain (Anc) or population samples. Error bars represent 95% confidence limits estimated from at least eight replicate assays.
FIG 2
FIG 2
Genome evolution during the 1,000-generation A. baylyi ADP1 adaptive evolution experiment. (A) Types of mutations observed in the sequenced clones from each of the seven independently evolved populations. Lines connecting the columns divide mutations caused by new IS1236 activity from other categories of mutations. Deletions caused by homologous recombination between the two ancestral copies of IS1236 that flank the transposon Tn5613 were not counted as IS mediated. Full details for each mutation are provided in Table S2 in the supplemental material. (B) Schematic of an IS1236 insertion and an IS1236-mediated deletion. A copy-and-paste transposition event results in the duplication of several base pairs (usually three) at the target site flanking the new IS1236 copy. A second, close-by IS1236 insertion can lead to a homologous recombination event between the two IS copies that excises the intervening genomic region. One IS1236 element copy and an adjacent deletion remain after this type of IS-mediated deletion.
FIG 3
FIG 3
ACIAD3148 and clpA are mutated in isolates from many populations in the adaptive evolution experiment. (A) Mutations affecting ACIAD3148 likely evolved in parallel in six populations because it is beneficial to disrupt this gene, encoding a hypothetical protein of unknown function, which is found only in the genus Acinetobacter. The mutations observed include an IS1236 insertion followed by an IS1236-mediated deletion, single-base deletions in a run of seven adenine bases, and in-frame deletions of nine base pairs that could be mediated by a five-base repeat. The resulting amino acid change is shown in parentheses for single-base substitutions. (B) Most mutations affecting the ClpA ATPase and specificity component of the Clp protease disrupt only its C-terminal sequence, but there is also a nonsynonymous base substitution affecting amino acid 562 and a mutation that may affect a putative transcriptional promoter for clpA located inside the clpS gene.
FIG 4
FIG 4
IS1236 drives the evolutionary loss of ADP1 natural transformation. (A) Transformation frequencies for all evolved populations from the adaptive evolution experiment were reduced compared to those of the wild-type ADP1 ancestor (dashed line). However, for the sequenced clones from each population, transformability was significantly reduced only for certain isolates (*, P < 0.05 by one-tailed Welch's t test). Error bars are 95% confidence limits for triplicate assays. (B) In population 3, loss of transformation ability is mediated by the expansion of a noncompetent subpopulation over time as determined by testing individual clones picked from the mixed population. For populations 3 and 8, several 1,000-generation clones were genotyped for the relevant IS1236 insertions in competence genes found in the corresponding sequenced clone. In each case, these IS1236 insertions appear to be responsible for most of the loss of competence in the population, although other, rarer mutations that reduce competence also appear to be present in each case. The measurements shown are for triplicate assays. Error bars are omitted for clarity.
FIG 5
FIG 5
IS1236-mediated loss of genes related to extracellular polysaccharide (EPS) production leads to increased cellular aggregation and reduced bioemulsification activity in evolved A. baylyi ADP1 strains. (A) Sequenced clones have two distinct phenotypes visible by light microscopy: clones with the ancestral phenotype (Anc) occur in clusters of no more than two cells (2a, 4a, 5a, and 7a), and clones with a high-aggregation phenotype typically are found in clusters of tens to hundreds of cells (3a, 6a, and 8a). Scale bars represent 5 μM. (B) The three sequenced clones with the high-aggregation phenotype have loss-of-function mutations in the per and pgi genes related to EPS production. In two cases, these genes are located within large IS1236-mediated deletions. In clone 6a, each gene is inactivated by a separate mutation. (C) Reconstruction of the per and pgi mutations from clone 6a in the ADP1 ancestor shows that either mutation alone confers the high aggregation phenotype and that the double mutant does not exhibit greater aggregation than either single mutant. Scale bars represent 5 μM. (D) Loss of either per or pgi function in strains with the mutations in the 6a clone similarly leads to ADP1 cultures with reduced activity for emulsifying a water-hydrocarbon mixture. Error bars represent standard errors of the means from triplicate assays.
FIG 6
FIG 6
Terminal repeats flanking the large prophage region of A. baylyi ADP1 appear to mediate its rapid loss during the adaptive evolution experiment. The one base pair difference between the two repeats is boldfaced and marked for emphasis. Deletion of this unstable region in all seven sequenced clones results in the loss of three prophage-related genes previously defined as essential for growth in minimal medium (15), ACIAD2139, ACIAD2143, and ACIAD2190, as well as a common site for genomic modification targeted by integration vectors, such as pIM1463 (17).

Similar articles

See all similar articles

Cited by 10 articles

See all "Cited by" articles

Publication types

Substances

LinkOut - more resources

Feedback