Root system architecture (RSA) contributes to nitrogen (N) uptake and utilization in maize. In this study, a germplasm enhancement of maize double haploid population of 226 lines genotyped with 61,634 SNPs was used to investigate the genetic basis of RSA under two N levels using a genome-wide association study (GWAS). GLM + PCA, FarmCPU, and MLM models were utilized to balance false positives and false negatives. In total, 33 and 51 significant SNP-trait associations were detected under high and low N conditions, respectively. Under high N, SNP S9_2483543 was detected by all models. Linkage disequilibrium (LD) regions of some SNPs overlapped with the intervals of QTL for RSA and N response that were detected in previous studies. In particular, several known genes, Rtcs, Rtcl, Rtcl, and Ms44, were located in the LD regions of S1_9992325, S9_151726472, S9_154381179, and S4_197073985, respectively. Among the candidate genes identified by this study, GRMZM2G139811, GRMZM2G314898, GRMZM2G054050, GRMZM2G173682, GRMZM2G470914, GRMZM2G462325, GRMZM2G416184, and GRMZM2G064302 were involved in seedling, seed, and root system development or N metabolism in Arabidopsis or rice. The markers identified in this study can be used for marker-assisted selection of RSA traits to improve nitrogen use efficiency in maize breeding, and the candidate genes will contribute to further understanding of the genetic basis of RSA under diverse N conditions.