This web page was produced as an assignment for an undergraduate course at Davidson College.


In Situ Evolutionary Rate Measurements Show Ecological Success of Recently Emerged Bacterial Hybrids

Vincent J. Benef and Jillian F. Banfield

Paper Review by Nisha Crouser


There is very limited information available on the rate at which microorganisms evolve. Denef and Banfield analyzed biofilms for over 9 years to investigate the processes and determine the rates of bacterial evolution in their own environment. Previous experiments had developed evolutionary rate estimates based on laboratory measurements; this was one of the first experiments that looked at natural populations of bacteria.

Denef et al. chose to collect biofilms from a well-defined acid mine drainage system which provided a relatively homogenous microbial community because of the metal-rich, low pH environment. They found that the chemolithoautotrophic Leptospirillum group II dominates most biofilms in this system. Previous data indicate that large-scale recombination events occurred resulting in six hybrid genotypes (types I-VI), the researchers wanted to better understand when these events occurred. Over a 9-year period they collected microbial biofilms from six different locations at intervals of 3 weeks to no more than a year.

They chose to look specifically at location C75 because based on proteomics (the study of proteins) inferences the Leptospirillum group II type III genotype dominated the site. From the 13 samples taken from the C75 site the researchers were able to come up with a substitution rate of 1.4*10-9 substitutions per nucleotide per generation. This number was calculated by tracking the genome change at this location from August 2006-December 2010. The rate they determined is similar to the rate of 1.3*10-9 substitutions per nucleotide per generation that was predicted for the type III genome using the universal mutations-per-genome rate proposed by Drake. To validate that this substitution rate was consistent throughout the biofilms at the C75 location, they sampled three different locations approximately 1 m apart in June 2008. They found low levels of variation both within populations and across space, indicating that nucleotide changes at this site could be used to determine substitution rates.

The researchers then compared genomic data from across the various sites to types I and VI and found that the six other genotypes had mosaic blocks tens to hundreds of kilobases in length of type I and VI within their genome. From this information they were able to compare the six genotypes and determine when various recombination events occurred. From this information they constructed a phylogenetic tree for all the types of Leptospirillum in the acid mine drainage system. They were unable to determine why these hybrid genotypes arose but realized that the mine is subject to human and seasonal perturbations, which could have led to hybrid proliferation. They also concluded that hybrids were the result of selective pressures based on several lines of evidence.

Overall the researchers were able to successfully analyze the accumulation of genome change in free-living bacterial populations. Their data is the first real insight into how microbial systems evolve in response to both natural and anthropogenic change.


Overall I felt the article was very informative, but at times difficult to follow. The methods mentioned in the figure legends were not always clearly stated or explained in the text. Also the figure legends often referred to other supplemental figures, making it difficult for the figures to stand on their own. Figure 2 was especially difficult to understand because the genome map was very crowded and it was hard to discern between the individual genotypes. Also the text provided little evidence for how the phylogenetic tree was constructed. Overall, this figure is very busy and has too much going on all at once. The authors may have tried to cram in a lot of information to save space, but in doing so a lot of the data is lost in the overwhelming presentation.

Also since the researchers did several different studies and comparisons on samples from the various different sites, it was difficult to follow exactly what the researchers did and how they were able to generate their results according to the text. They also never explained what read recruitment was and I was unable to find a definition for this method in any literature sources. This method was used throughout the paper and should have been explained at some point.

Despite some issues in presentation of their findings, the accomplishments of Benet et al are very impressive. The methods used by the researchers can now be applied to other environmental systems to better understand the evolution of various bacteria. Clearly, the acid mine drainage system represents a unique, isolated environment so the results from this study cannot be directly applied to other situations; but the fact that the researchers were able to identify when and where recombination events occurred and the emergence of new types of Leptospirillum is remarkable. Future research will definitely want to understand how other microbial communities evolve over time and the rate at which they do so. This information could eventually have implications for a multitude of fields, including medicine.


Figure 1. Permission pending

A.This figure shows all of the samples and the locations from which they were taken within the acid mine drainage system. Each pie chart shows the date when the sample was collected and the genotype proportion of each sample based on read recruitment. At the C75 site twelve biofilms contained only the type III genotype (>99% of cells) while the remaining one contained 84% type III and 16% type I genotype. The figure also notes the direction of water flow through the acid mine drainage system, with the C75 location near the beginning of the mine therefore not interacting much with the other samples. The question the researchers were trying to answer was: what are the genotypic make-ups of the samples taken from different locations within the mine? As shown by the various colors corresponding to different Leptospirillum types, the genotype varied depending on the location from which samples were taken. Clearly, type III dominated the C75 location meanwhile only type I was found at the 5-way point.

B. This graph shows the acid mine drainage flow rate (L/min) measured at the mine entrance from April 2006- December 2010. As seen in the graph the flow rate changes both seasonally and yearly. Both February 2009 and 2010 show relatively high flow rates while 2007 and 2008 show lower rates. Also depicted in this graph is the community composition at the C75 location at various time points. These values were obtained using fluorescence in situ hybridization. The different bars show the percentage of each of the four possible populations: Archaea (Arc), Leptospirillum group III (L3), Leptospirillum group II (L2). The asterisks represent the Leptospirillum group II type III strain transitions revealed by SNP analysis, but this these transitions are never explained in the text detracting from the value of the figure. Overall this figure attempts to show that community population and flow rate are constantly changing over time, but either archaea or L2 (types II-VI) seem to dominate these populations throughout the various time points.

C. This graph simply shows the flow rate (L/min) of the acid mine drainage from 2001 to 2011. The shaded region shows the portion that is illustrated in Fig. 1B. Overall, the graph shows that flow rate changes over time.

Figure 2. Permission pending

A & B. In this figure the researchers wanted to display the results they had found when they compared the six different Leptospirillum genotypes. Using read recruitment to types VI and I, they found that each of these different types had blocks of tens to hundreds of kilobases in length that were a mosaic of type I and type VI. They also found that the transition point between the recombinant blocks was the same in each of the genotypes. The figure illustrates this. The outer circle of part A shows the peptides unique to type I (red) and type VI (blue) reference genomes at each protein locus for the C75 type III population. The inner circles show the prevalence of type III, IV, IVa, V, VI recombinants. The important point being made is shown in figure B. This magnified portion shows the transition points between the recombinant blocks. A transition point is where the blue switches to red or vice versa. The fact that all of these transition points occur at the same location in each sample indicates that each major recombination event must have occurred in a single cell; suggesting that all of these types evolved from a common ancestor in which the recombination first took place.

C. When the researchers found that recombination most likely occurred in a single cell, they wanted to look at the mutations that were specific to each of the six genotypes. They identified these fixed mutations by comparing the six genotypes to the type III genome using read recruitment (which is not fully explained in the article). From this information they were able to create the phylogenetic tree shown in the figure. Since read recruitment was only done relative to the type-III genome, positions within type I-like regions not shared by type IV, V, or VI genotypes were designated as gaps for tree construction.

The phylogenetic tree displays bootstrap values next to the branch points to show the certainty associated with the branching (higher values indicate greater certainty). The red dotted arrows indicate the points where recombination events occurred and the circles with red markers show the locations of recombination on the genome. On the far right of the tree, the researchers mapped the phylogenetic location of all of the samples they collected in the acid mine drainage system. The timeline underneath the tree shows the calculated time between recombination events and historical events that occurred at the mine.

Figure 3. Permission pending

This figure compares the group II high-frequency variants to the C75 June 2006 type III genotype based on read recruitment. Dots in the figure indicate variant nucleotides relative to the last common ancestor of types III to VI; these dots are color-coated based on the type of substitution that took place. The reference genome (outer circle) is color-coated so that type I portions are shown in red, type VI in blue, and deletions shown in gray. Recombination events in the inner circles also are shown in red and blue, while insertions/deletions are shaded gray. This figure is another way of showing that all recombination events must have occurred in a single cell since the recombinants line up in all of the variants. It also shows that genotype III is the common ancestor from which all the other genotypes (besides genotypes I and VI) evolved from.


Denef, J.Vincent and Banfield, Jillian F. 2012. In Situ Evolutionary Rate Measurements Show Ecological Success of Recently Emerged Bacterial Hybrids. SCIENCE 336:462-467.



Genomics Page
Biology Home Page

Email Questions or Comments.

© Copyright 2013 Department of Biology, Davidson College, Davidson, NC 28035