We have studied male sexual differentiation of null mutant mice (-/-) for the thyroid-specific enhancer-binding protein (T/ebp or Nkx2.1) gene, a homeodomain transcription factor that plays a role in organogenesis of the thyroid, lung, ventral forebrain, and pituitary gland. Because the T/ebp/Nkx2.1 (-/-) mice do not develop the pituitary gland, their sexual differentiation, if any, must occur in the absence of action of gonadotropins and other pituitary hormones. The (-/-) mice survive only until birth (embryonic d 19-19.5 of pregnancy), and when their external and internal genitals were inspected at embryonic d 18.5, they were indistinguishable from the (+/-) and (+/+) control mice. The testis weights of (-/-) mice were 20% lower than in (+/+) and (+/-) mice. The testosterone content of the (-/-) testes (13.5 +/- 2.4 pg/gonad, mean +/- SEM, n = 11) was dramatically reduced, compared with (+/-) (165 +/- 22.5 pg, n = 14) and (+/+) (234 +/- 37.3 pg, n = 10) littermates. Light microscopy revealed no difference in seminiferous tubules, interstitial tissue, or relative proportions of the two-cell compartments between the (-/-) and (+/+) testes. However, electron microscopy confirmed that Leydig cells in the (-/-) testes were much smaller, with smaller mitochondria and proportion of smooth endoplasmic reticulum than found in the controls, which was in support of the low androgen content of the knockout testes. In conclusion, this study on T/ebp/Nkx2.1 knockout mice, devoid of the pituitary gland, demonstrates that pituitary hormone secretion is not needed for stimulation of sufficient fetal testicular androgen synthesis to induce male sexual differentiation. The endogenous testosterone level in the null mutant testes is 5-10% of the control level, which suggests that there is a considerable safety margin in the amount of testosterone that is needed for the male fetal masculinization.