In this study, we found that loss of the circadian clock gene Bmal1 causes disruptions throughout the growth hormone (GH) axis, from hepatic gene expression to production of urinary pheromones and pheromone-dependent behavior. First, we show that Bmal1 knockout (KO) males elicit reduced aggressive responses from WT males and secrete lower levels of major urinary proteins (MUPs); however, we also found that a liver-specific KO of Bmal1 (Liver-Bmal1-KO) produces a similar reduction in MUP secretion without a defect in aggressive behavior, indicating that the decrease in elicited aggression arises from another factor. We then shifted our investigation to determine the cause of MUP dysregulation in Bmal1 KO animals. As the pulse pattern of GH drives sexually dimorphic expression of hepatic genes including MUPs, we examined GH pulsatility. We found that Bmal1 KO males have a female-like pattern of GH release, while Liver-Bmal1-KO mice are not significantly different from either WT or Bmal1 KO. Since differential patterns of GH release regulate the transcription of many sexually dimorphic genes in the liver, we then examined hepatic gene transcription in Bmal1 KO and Liver-Bmal1-KO mice. We found that while some female-predominant genes increase, there was no decrease in male-predominant genes in the Bmal1 KO, and little change in the Liver-Bmal1-KO. We also found disrupted serum IGF-1 and liver Igf1 mRNA in the Bmal1 KO mice, which may underlie the disrupted GH release. Overall, our findings differentiate between GH-pulse-driven and circadian-driven effects on hepatic genes, and the functional consequences of altered GH pulsatility.

Mice were handled daily for four weeks prior to the day of assay to acclimate to sampling stress. Mice were also provided enrichment including nestlets and a small house provided within the cage. To measure endogenous growth hormone pulses, we collected 4 µl of blood through the tail vein every 10 minutes for six hours. The blood was collected by making a small nick at the tail tip and then a capillary tube was held gently to the tail to collect blood into a capillary tube (Drummond microcaps 40 μL) and the end was sealed with Critoseal. If a scab formed during the course of the experiment, it was gently removed with sterile gauze to renew blood flow. The volume of blood collected over the experiment was under 150 μl, and fluid replacement was not necessary. Blood was collected from ZT 4-10, during the light phase. Capillary tubes were incubated at room temperature for 20 minutes, followed by centrifugation at 2000 x g for 20 minutes and then the serum was removed and frozen with serum matrix buffer until assay. Serum was assayed for growth hormone concentration with a Luminex Magpix. Briefly, 1.5 μL of serum was added to 23.5 μL of serum matrix buffer provided in the kit and incubated with antibody beads overnight at 4 degrees C°. GH was measured by Luminex assay (Milipore MPTMAG-49K) according to manufacturer instructions, with an inter- and intra-assay variation of 14.3% and 8.2%, respectively.)

Data are presented as ng/ul GH. These are the data generated after 5-parameter logistic curve fitting by the Luminex software. We have also included the standard curve for each animal. Because plates were assayed with two animals per plate, and one standard curve, identical standard curves indicate both animals were run on the same plate. Standard curves were run in duplicate (displayed horizontally), whereas time point samples were in singlicate due to the low serum volumes required by serial sampling.