ABSTRACT: Akt1 plays a protective role in the postnatal C57BL6 mouse testis following lactational exposure to the neonatal goitrogen, propylthiouracil (PTU). To elucidate the transcriptional profile mediating this phenotypic effect, we monitored changes in testicular gene expression at postnatal days (PNDs) 15 and 25 in Akt1+/+, Akt1+/-, and Akt1-/- testes following exposure to 0.01% PTU allowing us to determine changes in gene expression due to 1.) genotype effects; 2.) exposure effects; and 3.) genotype-by-exposure interactions. Early PTU-dependent gene changes included genes involved in lipid metabolism, spermatid differentiation, meiosis and adhesion. Early Akt1-dependent effects were associated with germ cell development, spermatid development and differentiation, and sperm motility. By PND25, the Akt1 gene-environment interaction had pronounced effects on genes associated with Sertoli cell (SC) differentiation and claudin-associated junctional formation suggesting delayed formation of the blood-testis barrier (BTB). To confirm these observations, biotin tracer experiments demonstrated a permeable blood-testis barrier in PTU-exposed PKB/Akt1-/- tubules as late as PND25 compared to PTU-exposed Akt1+/+ seminiferous tubules. Transmission electron microscopy demonstrated altered SC morphology, aberrant SC localization, and disorganized actin bundle formation. Taken together, loss of Akt1 coupled with postnatal exposure to the neonatal goitrogen, PTU, in the testis contributes to a transcriptional profile associated with impaired integrity of the blood-testis barrier. In summary, the Akt1-/- mouse represents a potentially important model to study BTB formation and reassembly in response to male reproductive toxicants and the various signaling networks which mediate these responses. The microarray analysis employed a balanced factorial design, with one chip created for each of three mice under each experimental condition: mouse genotype (Akt1+/+, Akt1+/-, and Akt1-/-), exposure (PTU and Control), and age (sacrificed at PND15 and 25), with a total of 36 mice and 36 chips. MoGene 1.0 st v1 arrays were used for the samples from PND 15 and Mouse Genome 430 2.0 arrays were used for the samples from PND 25. One chip, PTU-exposed, Akt1-/-, PND25, was determined to not be of high quality and was not included in the analysis or provided here.