Membrane insertion of gap junction connexins: polytopic channel forming membrane proteins

Additional information on the proteolytic processing of Figure 4 . Sucrose gradient analysis of connexin integration into microsomal membranes. Synthetic RNAs encoding c~ and/3~ GJ proteins were translated in the reticulocyte lysate system in the presence of intermediate concentrations of microsomes. Aliquots of each reaction were equilibrated to pH 7.0 or pH 11.5, respectively, and fractionated using a physiological salt or an alkaline sucrose step gradient into a supernatant (S) and a pellet (P) fraction. Proteins were resolved by SDS-PAGE and visualized by fluorography. At pH 7.0, as well as at pH 11.5, the faster migrating processed GJ proteins (c~' and/3,~ were primarily abundant in the pellet fractions (lanes 2 and 4, 14, and 16) , while the unprocessed full-size products (a~ and/3~) were primarily abundant in the supernatant fractions (lanes 1 and 3, and 13 and 15 ). The integral membrane protein acetylcholine receptor subunit c~7 (AChR c~7, lanes 5 and 6, and 17 and 18), the secretory proteins preprolactin (pPL, lanes 7 and 8, 19 and 20) and yeast a-factor (co-factor, lanes 9 and 10, 21 and 22), and the cytoplasmic protein/~-globin (lanes 11 and 12, 23 and 24) were analyzed in parallel. These proteins gave the expected results, namely the B-globin was completely located in the supernatant (lanes H and 23) , and the mature, glycosylated integral membrane protein AChR c~7" (lane 18), but not the mature secretory proteins PL and a-factor* (lanes 20 and 22), was resistant to alkali treatment. the connexins was obtained by characterizing the conditions that promoted the specific cleavage reaction in more detail. Fig. 6 A shows that the proteolytic cleavage of the connexins occurred only if the membrane vesicles were added cotrans-lationaUy ( Fig. 6 A, lane 2) . The cleavage was completely abolished when the microsomes were added after translations were blocked by the addition of RNase (Fig. 6 A, lane 3) . The proteolytic activity was restricted to the lumen of the microsomes as indicated by the fact that only microsomes (Fig. 6 A, lane 5) , and not microsomal supernatants derived from pelleted microsomes (Fig. 6 A, lane 4) , efficiently processed the GJ proteins. Preincubation of the translocation reactions with the serine and cysteine protease inhibitors TPCK, TLCK, E-64, leupeptin, and DFP (each 1 mM final concentration) before the addition of synthetic RNAs had no effect on the cleavage reaction (Fig. 6 B, lanes 3-7) . The potential dependence of the connexin cleavage on the signal recognition particle (SRP) was determined by using microsomes depleted from SRP by a high salt wash in combination with purified SRP. In translation reactions supplemented with SRP-depleted microsomes, only uncleaved connexins were detected (Fig. 6 C, lanes 3 and 7) , while cleaved connexins were generated if the depleted microsomes were supplemented with purified SRP (Fig. 6 C, lanes 4 and 8) . In summary, all of these results indicated a specific, but atypical, proteolytic processing by signal peptidase. A cleavage by signal peptidase was further indicated by the location of the cleavage site in proximity to the NH~ terminus and by the proteolytic processing reaction occurring concomitant to the translation reaction~ as described above.