Soil harbors remarkably stabilize bacterial communities at the phylum level. However, no two soils have exactly the same structure of bacterial phyla. The structure of microbial community is strongly influenced by the type of land-use through changes in soil attributes. Using high-throughput pyrosequencing and quantitative polymerase chain reaction techniques, soil microbial community structures were investigated along a land-use gradient of 100- and 27-year farmlands, a 33-year Pinus forest, a 28-year poplar forest, and a 21-year shrubland, as well as a native desert from which all cultivated systems were converted. The results revealed that the dominant phylotypes in the native soil comprised primarily of Alphaproteobacteria, Actinobacteria, Bacteroidetes, and Firmicutes, accounting for > 71.4% of the total bacterial 16S rRNA sequence reads. Changes in land-use led to a significant decrease in these dominant phylotypes down to 33.4%. In contrast, the phylotypes with low abundance, such as Acidobacteria, Chloroflexi, Nitrospira, and Gammaproteobacteria, increased sharply from 4.5%-5.9% in the native soil to 20.9%-30.2% of the total 16S rRNA gene sequences in the cultivated soils except for the soil from the shrubland. These contrasting changes in the major taxa appear to be correlated with the changes in soil attributes. For instance, bacterial and archaeal amoA genes were found to be 960- and 3800-fold more abundant in the soil from the 100-year farmland than the native soil. The changes in numerically less dominant nitrifying phylotypes are consistent with soil inorganic nitrogen dynamics. Quantification of the 16S rRNA genes demonstrated that bacteria and archaea were about two to three orders of magnitude more abundant in the cultivated soil than in the native soil. Hence, land-use type affects the soil bacterial community structure, which has profound consequences on ecosystem function.
