We evaluated the effect of depth on microbial communities from mesotrophic lake sediments as measured by two DNA extraction methods. DNA yields and the number of ribotypes differed significantly by depth. The mean yields at the depth interval of 0À8 cm from the two extraction methods were both higher than those within the depth range of 10À44 cm. Significant differences were also observed between the two methods in DNA yields from the deeper (10À44 cm) samples. An analysis of denaturing gradient gel electrophoresis (DGGE) profiles demonstrated that the choice of the DNA extraction method has a profound effect on the bacterial community profiles generated, which was reflected in the number of bands or ribotypes detected from samples from all depths. Higher DNA yield did not lead to a higher diversity of phylotypes. In almost all cases, extraction method 2 (M2, soil DNA test kit) resulted in a greater diversity of phylotypes compared to method 1 (M1, liquid nitrogen/cetyl trimethylammonium bromide extraction). Moreover, notable changes in bacterial diversity were detected in the uppermost layers (0À5 cm). The results of canonical correspondence analysis illustrated that the DNA extraction method did not affect the evaluation of the relationship between bacterial community structure and environmental variables in lake sediments at multiple depths, and the differences in community structure in the two extraction methods were both related to the same environmental variables (pH, total phosphorus concentration) and depth. Vertically, the community structure of sediment layers formed several separate clusters along the depth gradient. This study expands our understanding of the depth-related microbial community structure in lake sediments.