Polyamine catabolism is rate limited by spermidine/ spermine N 1 -acetyltransferase (SSAT). Although the amino acid sequence of SSAT is known, the substrate binding and catalytic sites are not. The goal of this study was to define the region responsible for acetyl coenzyme A binding. Human SSAT contains a region of 20 amino acids homologous to several microbial antibiotic N-acetyltransferases. The highest homology is represented in the Campylobacter coli streptothricin acetyltransferase sat4 gene, where 16 identical or highly conserved amino acids exist in a 20-residue stretch. The most conserved residues within this region are RGF-GIGS beginning at Arg-101 in the human SSAT. Sitedirected mutations to Arg-101, Gly-104, and Gly-106 resulted in proteins with no measurable activity. The G102D mutation produced a partially active protein with a decreased affinity for acetyl coenzyme A and with a K m >10-fold that of the wild-type protein. Analysis using the PredictProtein program suggests a common structure among the microbial and eukaryotic N-acetyltransferases in the region corresponding to the RGF-GIGS of human SSAT consisting of an ␣-helix usually preceded by a glycine loop. Our data are consistent with the hypothesis that Arg-101 and the proximal glycine loop are necessary for the activity of human SSAT.1 is the rate-limiting step in polyamine catabolism that catalyzes the transfer of the acetyl group from acetyl-CoA to the N 1 position of spermidine or spermine (1-3). The acetylated polyamines are then excreted from the cell in their acetylated forms or are metabolized by the constitutive FAD-requiring polyamine oxidase. SSAT is a key component in the homeostatic system that maintains a narrow range of intracellular polyamines in eukaryotic cells. SSAT is normally induced in response to conditions that lead to rapid increases in the natural polyamines and to stress stimuli (4 -7). The superinduction of SSAT by a series of antitumor polyamine analogues has been implicated in the cell type-specific cytotoxic response of several important human solid tumors (8 -11). Since several of the polyamine analogues are now in or are being considered for clinical trial, a greater understanding of the role of SSAT and its precise mechanism of action is critical for the better design and use of these new antineoplastic agents.Although a kinetic model for the activity of this enzyme has previously been postulated by Della Ragione and Pegg (12) to be an ordered Bi Bi molecular reaction where the natural polyamine is bound first and the acetylated polyamine is released last, little is known about the specific molecular domains required for activity. The molecular cloning of the human SSAT gene, cDNA, and microsequencing of most of the purified human protein have allowed a more detailed study of the residues and domains required for enzyme activity (13-16). The human SSAT gene codes for a 171-amino acid protein with a predicted molecular weight of ϳ20,000 and is thought to be active as a homotrimeric or homotetrameric protein (1...