No abstract

Somatostatin (SST) exerts direct antiproliferative effects in tumor cells, triggering either growth arrest or apoptosis. The cellular actions of SST are transduced through a family of five distinct somatostatin receptor subtypes (SSTR1-5). Whereas growth inhibition has been reported to follow stimulation of protein tyrosine phosphatase via SSTR2 or inhibition of Ca2+ channels via SSTR5 in heterologous expression systems, the subtype selectivity for signaling apoptosis has not been investigated. The tumor suppressor protein p53 and the protooncogene product c-Myc regulate cell cycle progression (growth factors present) or apoptosis (growth factors absent). The p53-induced G1 arrest requires induction of p21, an inhibitor of cyclin-dependent kinases, whereas apoptosis requires induction of Bax. c-Myc is capable of abrogating p53-induced G1 arrest by interfering with the inhibitory action of p21 on cyclin-dependent kinases. We have, therefore, investigated the regulation of p53, p21, c-Myc, and Bax and cellular apoptosis in relation to cell cycle progression in CHO-K1 cells stably expressing individual human SSTR1-5. We demonstrate that apoptosis is signaled uniquely through human SSTR3 and is associated with dephosphorylation-dependent conformational change in wild-type (wt) p53 as well as induction of Bax. The induction of wt p53 occurs rapidly and precedes the onset of apoptosis. We show that the increase in wt p53 is not associated with the induction of p21 or c-Myc when octreotide-induced apoptosis becomes evident, suggesting that such apoptosis does not require G1 arrest and is not c-Myc dependent. These findings provide the first evidence for hormonal induction of wt p53-associated apoptosis via G protein-coupled receptor in a subtype-selective manner.

Recent reports (Raynor et al) have claimed the identification of potent somatostatin (SST) agonists exhibiting binding affinities of 1-2 pM and up to 30,000 fold binding selectivity for several of the 5 cloned sstr subtypes. These conclusions, however, are based on binding comparisons of sstr subtypes from different species expressed in different cell lines and studied with different radioligands. To eliminate the effect of species and/or methodological variations, we have investigated agonist selectivity of 32 synthetic SST analogs for all 5 hsstrs stably expressed in CHO-K1 cells under identical binding conditions. We show that hsstr2, 3, 5 react potently with hexapeptide as well as cyclic and linear octapeptide analogs and belong to a similar sstr subclass. hsstr1 and 4 react poorly with these analogs and belong to a separate subclass. The present generation of SST analogs exhibit a modest-50 fold increase in binding potency compared to SST-14 for 2 subtypes (hsstr2, 3), and relative selectivity for only 1 subtype (hsstr2) which is at best only 35 fold. The potency and degree of selectivity of these analogs is several orders of magnitude less than that reported earlier and suggests the need for caution in using these compounds as putative superagonists or subtype selective compounds for any of the individual sstrs.