2 INOSITOL 1,4,5-TRISPHOSPHATE BINDING-SITES IN RAT BASOPHILIC LEUKEMIA-CELLS - RELATIONSHIP BETWEEN RECEPTOR OCCUPANCY AND CALCIUM-RELEASE

Quantal calcium release is a novel parardigm for second messenger sign al transduction which provides spatial and temporal control of calcium release from intracellular stores by inositol 1,4,5-trisphosphate (In sP(3)). We have proposed a mechanism to account for this phenomenon [K indman, L. A., & Meyer, T. (1993) Biochemistry 32, 1270-1277], which h ypothesized the existence of five channels, each with a different affi nity for InsP(3). As a direct test of this hypothesis, InsP(3) binding to microsomes from RBL cells was examined under conditions similar to those used for calcium release. Scatchard analyses performed under a variety of conditions indicates the presence of high affinity (K-D = 0 .9 +/- 0.3 nM) and low affinity (K-D = 47 +/- 5 nM) InsP(3) binding si tes. The low affinity sites are more prevalent, constituting 82 +/- 5% of the total. Both sites are identified in the presence and absence o f MgATP. Moreover, both sites are selective for InsP(3) over InsP(4), though high concentrations of InsP(4) displace InsP(3) from each site (with inhibition constants of 16 and 267 nM InsP(4), respectively). Th e relative abundance of the two InsP(3) binding sites is Ca2+ dependen t. An increase in Ca2+ from 0.1 to 0.5 mu M results in the apparent co nversion of a portion of the low affinity sites into high affinity sit es. Ca2+ (0.5 mu M) also increased the K-D of the low affinity InsP(3) binding site. Given the presence of both high and low affinity InsP(3 ) binding sites, two simple mathematical models describing both the ki netics of calcium release and quantal calcium release from RBL cells w ere developed. Each model assumes that the two types of InsP(3) recept ors interact randomly to form five different calcium channels (i.e., t wo homotetramers and three heterotetramers), with a distribution refle ctive of the relative abundance of the two binding sites. In the first model, binding of three or four molecules of InsP(3) to any of the fi ve channel types is sufficient to open the channel. In the second mode l, InsP(3) binding to two or three low affinity binding sites only wil l open the channel. This latter model predicts that channels composed of three or four high affinity InsP(3) binding sites do not contribute to Ca2+ release. Given the Ca2+ dependence of the conversion between low affinity and high affinity InsP(3) binding sites, the latter model predicts the desensitization of some of the channels following elevat ion of cytosolic Ca2+. Neither model requires cooperativity, consisten t with the lack of cooperativity in the InsP(3) binding assays.