In aging men, the prostate gland becomes hyperproliferative and displays a propensity toward carcinoma. Although this hyperproliferative process has been proposed to represent an inappropriate reactivation of an embryonic differentiation program, the regulatory genes responsible for normal prostate development and function are largely undefined. Here we show that the murineNkx3.1 homeobox gene is the earliest known marker of prostate epithelium during embryogenesis and is subsequently expressed at all stages of prostate differentiation in vivo as well as in tissue recombinants. A null mutation for Nkx3.1 obtained by targeted gene disruption results in defects in prostate ductal morphogenesis and secretory protein production. Notably, Nkx3.1 mutant mice display prostatic epithelial hyperplasia and dysplasia that increases in severity with age. This epithelial hyperplasia and dysplasia also occurs in heterozygous mice, indicating haploinsufficiency for this phenotype. Because human NKX3.1 is known to map to a prostate cancer hot spot, we propose that NKX3.1 is a prostate-specific tumor suppressor gene and that loss of a single allele may predispose to prostate carcinogenesis. The Nkx3.1 mutant mice provide a unique animal model for examining the relationship between normal prostate differentiation and early stages of prostate carcinogenesis.