A perifusion system for studying the luteinizing hormone surge.

Perifusion, an ex vivo method for studying dynamic responses of living cells, has been used in studies of receptor-mediated secretion to produce a deeper, integrated understanding of responses to the timing and patterns of hormonal inputs. The preovulatory luteinizing hormone (LH) surge is driven by massive secretion of the hypothalamic peptide gonadotropin-releasing hormone (GnRH) and is coordinated with changes in circulating levels of ovarian steroids. With a technologic advancement, perifusion can be a useful tool for studying the mechanisms underlying the LH surge. This work had a twofold purpose: first, to develop a perifusion system able to mimic hormonal patterns, observed in sheep, that conduce the LH surge; and, second, to use the system to produce a laboratory model of the surge. Toward the first goal, a computer-automated system for delivering multiple, independently varying hormonal inputs was designed and developed. A previously measured GnRH surge indicated that to reproduce patterns with 90% accuracy, it must yield a 500-fold dynamic range and a frequency response of 17 mHz. Factory specifications for the employed pumps suggested that the system would meet these requirements. Next, six attempts to produce a reliable model of the LH surge were performed. With an LH surge defined as the occurrence over at least 4 h of continuously elevated LH above the 95% confidence limit of its mean pre-surge concentration, one attempt to produce LH surges was successful under 4 of 5 steroidal conditions. No subsequent attempts succeeded and, thus, lead to a re-evaluation of underlying assumptions. Analysis of fluid delivery revealed a subtle caveat in the architecture of the pump controllers which supported volumetric algorithms but constrained flow-oriented algorithms. This limited developed algorithms to 30-fold dynamic range. Although the system falls short of requirements for producing a reliable model of the LH surge, three significant achievements can be identified. First, the system operates properly within its limits and thus can be used to study responses to compatible endocrine signals; second, a design that overcomes pump controller problems is proposed; and, third, the preliminary results suggest that achieving an LH surge in perifusion is possible.PhDAnimal PhysiologyApplied SciencesBiological SciencesBiomedical engineeringUniversity of Michigan, Horace H. Rackham School of Graduate Studieshttp://deepblue.lib.umich.edu/bitstream/2027.42/131473/2/9909933.pd