The properties and roles of indigenous heavy metal-tolerant microbes have been extensively investigated for plant growth promotion and ecological remediation at contaminated sites, but these studies seldom address the community-level features of the microbes. Three soil samples were collected from the rhizosphere of a Cu-tolerant plant, Artemisia capillaries Thunb., grown on a Cu mine spoil. The MiSeq high-throughput sequencing technique, targeting the V3-V4 region of the bacterial 16S rDNA gene, was used to investigate the bacterial community structure and analyze the links between the bacterial community and soil environmental parameters. The results showed that sampling site Cu3 contained higher concentrations of Cu and Zn than Cu1 and Cu2 did. The Shannon-Wiener diversity index, richness, evenness, ACE index, and Chao1 index of the bacterial communities in Cu3 were all lower than those of Cu1 and Cu2, but coverage by the bacterial communities in Cu3 was higher than that of Cu1. The top 10 dominant phyla of bacteria accounted for 95% of the total relative abundance and 8 out of the top 10 dominant phyla were the same across all three sampling sites; these were Proteobacteria, Acidobacteria, Bacteroidetes, Gemmatimonadetes, Actinobacteria, Verrucomicrobia, Planctomycetes, and Unclassified. The relative abundances of Proteobacteria, Bacteroidetes, and Gemmatimonadetes were higher in Cu1 than in Cu2 and Cu3, but the opposite was true for Acidobacteria, Actinobacteria, Verrucomicrobia, Chloroflexi, and Unclassified phyla, indicating that different bacteria responded differently to Cu stress. The relative abundance of Chloroflexi, which only accounted for 0.54% of the bacterial community in Cu1, was 4.54% and 15.27% in Cu2 and Cu3, respectively. Bacterial communities were clustered into three different groups, according to principal coordinates analysis (PCoA) and redundancy analysis (RDA). The variation in bacterial communities was controlled by 11 principal coordinates, among which the first and second coordinates explained 74.3% and 14.8% of the total variance, respectively. Soil environmental parameters were closely related to the differences in bacterial community and explained 97.5% of the total variance. Total Cu, total P, pH, available P, and organic matter were the significant parameters; altogether, they accounted for 93.9% of the total variance in bacterial community. Total Cu was the most powerful factor, and explained 60.6% of the total variance independently. However, the dominant parameters differed across sampling sites. The RDA bioplots revealed that the bacterial community in Cu3, which showed the highest Cu tolerance, was positively related to total Cu. In contrast, the relatively Cu-sensitive bacterial community in Cu1 was positively correlated with pH and negatively correlated with total Cu. It is of vital importance to study how bacterial communities in the plant rhizosphere change with the environment and screen functional bacteria for plant-microbe remediation of heavy metal contamination.