AurA (Aurora-A) is a ubiquitous protein kinase regulating entry into mitosis and shown to promote transformation upon overexpression. In order to gain information on the structural features determining its substrate specificity, we assayed human recombinant AurA on a variety of phosphoacceptor peptide substrates including a series of properly modified derivatives of the Kemptide (ALRRASLGAA). The data presented here show that AurA is a basophilic Ser/Thr protein kinase recognizing the consensus R/K/N-R-X-S/T-B, where B denotes any hydrophobic residue with the exception of Pro. We show that the presence of a Pro at position n+1 fully abrogates phosphorylation of the peptide substrate. Although the consensus for AurA is reminiscent of that of PKA (protein kinase A), it significantly differs from the latter for a much more stringent dependence on the hydrophobic residue at n+1 and for its tolerance of residues other than Arg at position n-3. Based on the finding that the peptide ALKRASLGAA is not a substrate of PKA while still providing a sensitive assay of AurA activity, we suggest that this peptide may be used for differential screening of the two kinases. We have further validated the AurA consensus by generating a peptide (APSSRRTT288LCGT) that comprises the main AurA autophosphorylation site and by showing that AurA phosphorylated this peptide exclusively at one site fulfilling its consensus (Thr288). Moreover, we show that AurA could autophosphorylate at Thr288 through an intermolecular mechanism of reaction and that, in vivo, PKA was not involved with Thr288 phosphorylation. The evidence obtained in the present study provides a rational tool for predicting AurA sites in potential substrates of physiological significance.