IntroductionPhosphorus (P), which plays a vital role in plant growth, is continually added to soil to maximize biomass production, leading to excessive P accumulation and water eutrophication.ResultsIn this study, a pot experiment using a subtropical tobacco-growing soil fertilized with four P levels—no P, low P, medium P, and high P—was conducted and rhizosphere and bulk soils were analyzed.ResultsP addition significantly increased tobacco biomass production (except under low P input) and total soil P and available P content (P<0.05), whereas total nitrogen content decreased in the rhizosphere soils, although this was only significant with medium P application. P fertilization also significantly altered the bacterial communities of rhizosphere soils (P<0.05), but those of bulk soils were unchanged (P>0.05). Moreover, a significant difference was found between rhizosphere soils with low (LR) and high (HR) P inputs (P<0.05). Additionally, compared with rhizosphere soils with no P (CKR), Shannon diversity showed a declining trend, which was significant with LR and HR (P<0.05), whereas an increasing tendency was observed for Chao1 diversity except in LR (P>0.05). Functional prediction revealed that P application significantly decreased the total P and N metabolism of microorganisms in rhizosphere soils (P<0.05).DiscussionCollectively, our results indicate that maintaining sustainable agricultural ecosystems under surplus P conditions requires more attention to be directed toward motivating the potential of soil functional microbes in P cycling, rather than just through continual P input.