Glucocorticoid (GC) is thought to have its major effect in perinatal lung via stimulation of surfactant production. We have established the absolute requirement of GC for fetal lung development by analysis of matings of corticotropin-releasing hormone (CRH)-deficient mice, generated by gene targeting in embryonic stem cells. Homozygous mutant mice have impaired GC production due to inadequate ACTH release during stressful and circadian stimuli, and die during the first postnatal day due to inadequate lung development. Histologic studies demonstrate normal lung development until embryonic day 16.5, with a marked divergence in architecture thereafter. Reduced levels of surfactant apoprotein-A,-B, and to a lesser extent -C mRNAs are observed by Northern blot analysis at 16.5- 18.5 days of gestation, though production of dipalmitoyl phophatidylcholine and appearance of lamellar bodies are not significantly different between wildtype and CRH-deficient mice. Alveolar architecture, which should not be affected by surfactant status in the fluid-filled fetal lung, is abnormal in the CRH-deficient fetus. Furthermore, hypercellular mesenchyme in mutant fetuses is associated with persistent cell division, as demonstrated by increased proliferating cell nuclear antigen (PCNA) immunohistochemistry. Clara cell maturation is also delayed in the CRH-deficient fetus as evidenced by Clara cell secretory protein (CC10) immunoreactivity. Abnormalities are reversed by prenatal GC therapy. Thus, GC is essential for morphologic and biochemical pulmonary maturation beyond and possibly independent of its effect on surfactant production.