Previous results have shown that the pattern of GnRH pulses (amplitude and frequency) can differentially regulate expression of gonadotropin subunit cytoplasmic messenger RNA (mRNA) concentrations. The present study examined the effect of GnRH pulses on a, LH-β and FSH-β transcription rates as determined by nuclear runoff transcription assay. GnRH pulses (saline to controls) were given to castrate, testosteronereplaced male rats, and the rate of subunit gene transcription was measured in isolated pituitary nuclei. The effect of GnRH treatment duration was examined by giving GnRH pulses (25 ng/pulse at 30-min intervals) for 1, 4, or 24 h. The basal transcription rates [expressed as parts per million (ppm)] were 82 ± 25 for α; 39 ± 19 for LH-β and 27 ± 6 ppm for FSH-β, and transcription rates of all 3 subunits were elevated at 1 h (3–5-fold vs. saline controls). After 4 h of GnRH pulses, α and FSH-β transcription rates were reduced vs. 1 h, but LH-β mRNA synthesis rate was maintained. At 24 h, the α transcription rate was still increased (66%), but LH-β and FSH-β transcription rates had fallen to basal levels despite the continuing pulsatile GnRH stimulus. The second experiment investigated the effect of the duration of GnRH pulses (25 ng/pulse, every 30 min for 4 h or 24 h), on cytoplasmic subunit mRNA concentrations to assess if the initial 4-h increase in transcription rate would induce a rise in cytoplasmic mRNAs. After 4 h of GnRH pulses, a and LH-β mRNAs were unchanged, but FSH-β mRNA had increased by 36% (P < 0.05) compared to controls. All 3 subunit mRNAs were increased (- 2-fold) by 24 h of GnRH pulses. Administering GnRH pulses for 4 h followed by 20 h of saline pulses did not incresae α mRNA; LH-β was slightly increased (P < 0.05), but FSH-β mRNA concentrations were similar to levels seen after 24 h of continued GnRH pulses. The third experiment examined the effects of a continuous GnRH infusion and different GnRH pulse frequencies on gonadotropin subunit transcription rates. GnRH (25 ng/pulse) was given at intervals of 8, 30, or 120 min for 4 h (saline to controls). The continuous GnRH infusion (200 ng/h) did not increase the transcription rate of any of the three subunit mRNAs. α-subunit transcription rate was increased 2.7- or 4-fold by GnRH pulses given every 8 or 30 min, respectively. LH-β mRNA synthesis was only increased (3-fold) by 30-min GnRH pulses.In contrast, FSH-β transcription was elevated (2-fold) by 120-min pulses. These results show that GnRH stimulates gonadotropin subunit gene transcription, and a pulsatile signal is required for GnRH to be effective. The duration of β-subunit transcription appears to be limited in the model used. Neither LH nor FSH-β transcription was increased at 24 h despite the continuation of a pulsatile GnRH stimulus. The frequency of the pulsatile GnRH stimulus can selectively regulate gonadotropin subunit gene transcription with faster frequencies increasing a and LH-β and slower pulses FSH-β. This suggests that alterations in GnRH pulse frequency may be one mechanism involved in enabling a single GnRH to produce both the coordinate and divergent increases in gonadotropin subunit gene expression that occur in normal physiology. (Endocrinology128: 509–517,1991)