Microbial transformation of crop residue is the key process of soil organic matter (SOM) formation and mineralization, which determines soil fertility and affects global climate change. However, utilization dynamics of residue-derived carbon (residue C) by various microbial communities is still not well understood, especially under different residue quality and soil fertility conditions over a long-term scale (i.e., >1 year). In this study, a 500-day in-situ field experiment was conducted using 13C-labeled maize (Zea mays L.) root and shoot (composed of both stem and leaf) to examine the role of microbial community composition on the C processing. Specifically, the mineralization of residue C and incorporation of residue C into microbial biomass in low fertility (LF) and high fertility (HF) soils were investigated. The abundance of 13C in soil samples and microbial phospholipid fatty acids (PLFAs) were measured after 60, 90, 150 and 500 days since the residues added. The results showed that the mineralization rate of residue C was significantly higher in the LF than that in the HF soil for the first 150 days, and the shoot-derived C was more susceptible to degradation than root-derived C, but the final mineralization rates (~78%) were not significantly different among treatments on the day 500. Soil fertility significantly affected the relative composition of different microbial groups and distribution of residue C in microbial communities, but residue type did not do so. Furthermore, residue C contributed more to PLFA-C pool in the LF than HF treatments, and the proportion of root C in PLFA-C pool was higher than that of shoot C, indicating easier immobilization of root C by soil microbial anabolism. Accordingly, soil fertility and residue quality could both regulate the kinetics of the microbial immobilization of crop residue C, but overall the available residual quantity of applied (plant-derived) C to enhance or maintain soil C pool did not depend on them in a long term in the agricultural Alfisols.