Unidirectionally chiral inversion of N(G)-nitro-D-arginine (D-NNA) to its L-enantiomer (L-NNA) occurred in rats, and it was blocked markedly (ca. 80%) by renal vascular ligation, and entirely (100%) by the D-amino acid oxidase (DAO) inhibitor sodium benzoate, suggesting that renal DAO is essential for the inversion. However, the doses of sodium benzoate administrated were extremely high (e.g., 400 mg/kg) due to its low potency. It is thus possible that sodium benzoate-mediated blockade of D-NNA inversion might be due to its nonspecific (or non-DAO-related) effects. In addition, after D-NNA was incubated with the pure enzyme of DAO in vitro without tissue homogenates, L-NNA was not produced, even though D-NNA was disposed. We propose that this occurred because D-NNA was first converted to its corresponding alpha-keto acid by DAO and then to L-NNA by transaminase(s); however, there was no direct evidence for this process. The goal of this study is to further elucidate the process of D-NNA chiral inversion both in vivo and in in vitro tissue homogenates by comparing mutant ddY/DAO(-/-) mice that lack DAO activity entirely compared to normal ddY/DAO(+/+) mice and Swiss mice. Furthermore, the ability to produce L-NNA from D-NNA-corresponding alpha-keto acids (N(G)-nitroguanidino-2-oxopentanoic acid) produced by porcine kidney-derived DAO (pkDAO) was also studied in the DAO inhibitor-pretreated rats. We found that D-NNA chiral inversion occurred in Swiss mice and ddY/DAO(+/+) mice both in vivo and in in vitro kidney homogenates, but not in ddY/DAO(-/-) mice, correlated to their DAO activities. The alpha-keto acid (N(G)-nitro-guanidino-2-oxopentanoic acid) from D-NNA was able to produce L-NNA, and subsequent vasoconstriction and pressor responses. These results indicate that the role of renal DAO is indispensible but insufficient for chiral inversion of D-NNA and other neutral and polar D-amino acids, and unidentified aminotransferase(s) are involved in a subsequent mechanism for the process of chiral inversion.