3-Guanidinopropionic acid (1, PNU-10483) has been demonstrated to both improve insulin sensitivity and to promote weight loss selectively from adipose tissue in animal models of non-insulin-dependent diabetes mellitus (NIDDM). However, 1 has also been shown to be a substrate for both the creatine transporter and creatine kinase, leading to marked accumulation in muscle tissue as the corresponding N-phosphate 4. In an effort to identify novel entities that maintain antidiabetic potency without susceptibility to creatine-like metabolism, an analogue program was undertaken to explore the effects of various structural modifications, including homologation, simple substitution, single atom mutations, and bioisosteric replacements for the guanidine and carboxylic acid. Overall, the scope of activity encompassed by the set of new analogues proved to be exceedingly narrow. Notable exceptions demonstrating equivalent or improved antidiabetic activity included the alpha-amino derivative 29, aminopyridine 47, isothiourea 67, and aminoguanidine 69. On the basis of its superior therapeutic ratio, aminoguanidine 69 was selected for preclinical development and became the foundation for a second phase of analogue work. Furthermore, in vitro studies demonstrated that 69 is markedly less susceptible to phosphorylation by creatine kinase than the lead 1, suggesting that it should have less potential for accumulation in muscle tissue than 1.