Generation of mice expressing the human glucagon receptor with a direct replacement vector.

The process of evaluating the in vivo efficacy of non-peptidyl receptor antagonists in animal models is frequently complicated by failure of compounds displaying high affinity against the human receptors to show measurable affinity at the corresponding rodent receptors. In order to generate a suitable animal model in which to evaluate the in vivo activity of non-peptidyl glucagon receptor antagonists, we have utilized a direct targeting approach to replace the murine glucagon receptor with the human glucagon receptor gene by homologous recombination. Specific expression of the human glucagon receptor (GR) in the livers of transgenic mice was confirmed with an RNase protection assay, and the pharmacology of the human GRs expressed in the livers of these mice parallels that of human GR in a recombinant CHO cell line with respect to both binding of 125I-glucagon and the ability of glucagon to stimulate cAMP production. L-168,049, a non-peptidyl GR antagonist selective for the human GR shows a 3.5 fold higher affinity for liver membrane preparations of human GR expressing mice (IC50 = 172 +/- 98 nM) in the presence of MgCl2 in marked contrast to the measured affinity of the murine receptor (IC50 = 611 +/- 197 nM) for this non-peptidyl antagonist. The human receptors expressed are functional as measured by the ability of glucagon to stimulate cAMP production and the selectivity of this antagonist for the human receptor is further verified by its ability to block glucagon-stimulated cyclase activity with 5 fold higher potency (IC50 = 97.2 +/- 13.9 nM) than for the murine receptor (IC50 = 504 +/- 247 nM). Thus we have developed a novel animal model for evaluating GR antagonists in vivo. These mice offer the advantage that the regulatory sequences which direct tissue specific and temporal expression of the GR have been unaltered and thus expression of the human gene in these mice remains in the normal chromosomal context.