This study assessed how different
in-situ lysis soil DNA extraction methods influence the DNA yield, quality
and hence the results obtained by
bacterial and fungal automated ribosomal intergenic spacer analysis (ARISA). Of the methods tested in three soils, a modified hexadecyltrimethylammonium bromide-dithiotreitol (CTAB-DTT)-based method produced
3 times more DNA of higher quality than the other methods (260/230 nm ratios=1.64–1.82
and 260/280 nm ratios=1.82–1.89
and extracts were less inhibitory of PCR). DNA extracted by this method also yielded more reproducible ARISA ribotypes (89−119 for bacteria
and 48−88 for fungi;
P<0.05) than DNA extracted by other methods,
and consequently produced more reliable estimates of
bacterial and fungal diversity in all three test soils. The significant correlations observed between the numbers of reproducible ribotypes
and DNA extract 260/230 nm ratios (
r=0.88
and 0.72 for bacteria
and fungi, respectively;
P<0.001) reaffirmed the strong influence of DNA quality on the reliability of microbial diversity indices determined based on PCR-based DNA fingerprinting technique. Results of discriminant function analysis (DFA)
and multivariate analysis of variances (MANOVA) performed on ARISA profiles (number
and relative abundance of ribotype) revealed that the variability associated with DNA extraction methods did not exceed the biological variability among soil types; this supports the conclusion that high-quality DNA underpins DNA fingerprinting techniques.