96
[ 2 , 3 ] and the evolution of rifampicin resistance in M. tuberculosis,
caused by mutations in rpoB, that is frequently associated with sec-
ondary mutations in the same gene affecting resistance and/or
relative fitness [ 4 – 7 ].
Experimental evolution of antibiotic resistance can be done in
different ways: in liquid culture or on solid media; with constant or
varying concentrations of antibiotic. The method we describe here is
one we commonly use on our laboratory: evolving a bacterial strain
in liquid culture, by serial passage at successively increasing concen-
trations of the drug, up to a predetermined drug concentration.
The exact trajectory of resistance evolution under selection by,
for example, fluoroquinolones, cannot be accurately predicted a
priori but must be confirmed after the evolution by whole genome
sequencing (WGS) of the selected mutants. In experimental evolu-
tion, cultures or isolates taken after each of the different steps in
the experiment should be stored so that they are available for later
analysis. WGS is the method of choice to provide a complete
genome-wide view into the genetic changes that have occurred.
Typically WGS will be applied initially to strains or cultures from
the starting point and the end point of the experiment. It may also
be necessary to use WGS to analyze some or all of the stored
strains/cultures from the different steps in the experiment so that
the sequence of mutational events can be fully mapped. In this
review we describe in outline how to do WGS to analyze experi-
mental evolution experiments.
It is also necessary to understand the phenotypic changes that
have accompanied the genetic changes revealed by WGS. In the
evolution of antibiotic resistance the most important predicted
change is that susceptibility to the selective drug should be reduced
during the course of the experiment. While the selection pressure
applied (in μg drug per mL culture) predicts a minimal level of
susceptibility at each step in the experiment, the actual trajectory of
susceptibility changes is usually nonlinear and must be measured
for each mutant. The method of choice described in this review is
to measure the minimal inhibitory concentration (MIC) of drug
using a broth dilution protocol.
During the evolution of resistance the mutations that accumu-
late in the selected strains will frequently be the cause of changes in
relative fitness. Methods to measure relative fitness, as a function of
antibiotic concentration under laboratory conditions, have been
recently described [ 8 , 9 ].2 Materials
- Mueller-Hinton II broth (MHII): cation-adjusted (from
Becton Dickinson, cat. no. 212322) for bacterial growth, and
2.1 Serial Transfer
Components
Douglas L. Huseby and Diarmaid Hughes