Constitutive and basal secretion from the endocrine cell line, AtT-20

If empty secretory granule membranes were present in the variant cells, and absence of the dense core was because of the lack of specific contents (as opposed to free GAG chains) to be packaged in the membranes, then secretory granules might be induced in the variant by introducing a protein normaily packaged into vesicles of the regulated pathway. To test this possibility, trypsinogen, a protein that is sorted into regulated granules was expressed in HYA.15.6 variant cells by DNA-mediated transfection. Trypsinogen is targeted in transfected wild type cells to the same regulated secretory compartment that houses the endocrine hormones, ACTH, and insulin (Burgess et al., 1985 (Burgess et al., , 1987 Orci et al., 1987) . To test for the formation of a regulated pathway in trypsinogen transfected AtT-20 variants, cells were labeled for 16-20 h in medium containing [35S]methionine or cysteine, chased for several intervals, and then stimulated to release the labeled material in the regulated pathway (Moore and Kelly, 1985; Burgess et al., 1985) . In agreement with previous reports, trypsinogen secretion from wild type cells was characteristic of the regulated secretory pathway (Burgess et al., 1985; Moore et al., 1983b) . There was a slow release of labeled trypsinogen in the absence of stimulation. Only 13 % of the total labeled trypsinogen was released in the first chase period and the secretion rate fell off with a half time of 5 h. A 5.3-fold increase in release of trypsinogen was obtained when the secretagogue 8-Br-cAMP was present during the last chase period (Fig. 4 A) . In contrast, trypsinogen secretion from the HYA.15.6.T.3 variant was much more rapid. 90% of the trypsinogen was released during the first chase interval, consistent with a half time of secretion of ~,,40 rain. The amount of trypsinogen that was secreted was not increased by exposure to 8-Br-cAMP (Fig. 4 B) . Expression of the regulated secretory protein, trypsinogen, by the variant cells does not, therefore, induce formation of the regulated secretory pathway or reveal the presence of empty secretory granules, lacking content proteins. :The unstimulated release of trypsinogen by the variant, corresponding to a half time of secretion of *40 min, is considerably faster than unstimulated release by wild type AtT-20 (t~ = 5 h). The rate of trypsinogen release from unstimulated variant cells was compared to the rate of constitutive secretion of immunoglobulin chains by wild type cells. In plasma cells, secretion of antibodies has been shown to be Constitutive, with no evidence for their storage or their stimulated release (Tartakoff and Vassalli, 1978; Tartakoff et al.,i 1977) . Antibodies are convenient constitutive secretory proteins to study because the genes for the immunoglobulin heavy and light chains expressed by many plasma cell tumor lines have been cloned. A cell line of particular interest, the MPC-11 plasmacytoma, expresses both IgG2b heavy chain and r light chain but over-expresses the latter. The r light chains not assembled with heavy chain are secreted freely into the medium (Laskov and Scharff, 1970) , making r chain a good candidate for a simple, soluble product to use as a constitutive secretory marker in nonlymphoid cells such as AtT-20 cells. AtT-20 cells transfected with the MPC11 r light chain gene made high levels of r chain, and secreted it with kinetics that were identical to that of immunoglobulin secretion in normal plasma cell lines. In transf