Application of high resolution melting assay (HRM) to study temperature-dependent intraspecific competition in a pathogenic bacterium

Abstract

Studies on species' responses to climate change have focused largely on the direct effect of abiotic factors and in particular temperature, neglecting the effects of biotic interactions in determining the outcome of climate change projections. Many microbes rely on strong interference competition; hence the fitness of many pathogenic bacteria could be a function of both their growth properties and intraspecific competition. However, due to technical challenges in distinguishing and tracking individual strains, experimental evidence on intraspecific competition has been limited so far. Here, we developed a robust application of the high-resolution melting (HRM) assay to study head-to-head competition between mixed genotype co-cultures of a waterborne bacterial pathogen of fish, Flavobacterium columnare, at two different temperatures. We found that competition outcome in liquid cultures seemed to be well predicted by growth yield of isolated strains, but was mostly inconsistent with interference competition results measured in inhibition tests on solid agar, especially as no growth inhibition between strain pairs was detected at the higher temperature. These results suggest that, for a given temperature, the factors driving competition outcome differ between liquid and solid environments.

Figure 1

High-resolution DNA melting curve analysis results for a 97-bp amplicon containing one SNP in trpB gene for Flavobacterium columnare. (Right) Normalized melting curves, (Left) derivative plots. The melting curves depict pure genotype A, pure genotype G, and 50/50 ratio of both genotypes. Solid and dashed lines represent replicate measurements.

Figure 2

(Right) Normalized melting curves and (Left) difference plots of reference samples consisting of purified DNA from genotypes G and A pooled in known proportions (13 different proportions of the two genotypes: 100%, 95%, 90%, 80%, 70%, 60%, 50%, 40%, 30%, 20%, 10%, 5% and 0%). Solid and dashed lines represent replicate measurements.

Figure 3

Genotype proportions estimated by HRM runs after 1, 2, 4, 6, 7 and 15 days of competition in liquid medium. For each day, vertical bars show the relative proportions of each competing genotype in five experimental replicates within each competition group (A, B, C and D) and black lines indicates the 90% confidence interval for each proportion estimate based on the triplicate HRM samples. The ordering of experimental replicates in the bar plots are the same from day 1 to day 15 within each group × temperature treatment. The names of the competing genotypes are indicated on the vertical axes. Hatched bars indicate replicates for which HRM data could not be used to estimate genotype proportions. For each day within each competition group, symbols between the 26 °C and 31 °C bar plots indicate the p-values of a Welch’s t-test comparing the genotype proportions for the 26 °C replicates and for the 31 °C replicates (*** < 0.001 < ** < 0.01 < * < 0.05).