COMMENTARY Domain interactions in protein kinase C C. J. PEARS and P. J. PARKER Imperial Cancer Research Fund, PO Box 123, Lincoln's Inn Fields, London WC2A 3PX, UK Introduction Protein kinase C (PKC) consists of a family of serine/ threonine protein kinases which are presumed to play a major role in signal transduction in response to the production of diacylglycerol (reviewed by Nishizuka, 1988; Parker et al. 1989). Originally denned as a kinase dependent on phospholipid and calcium for activity, PKC was found to be the major cellular receptor for the tumour- promoting phorbol esters, and the addition of either phorbol esters or diacylglycerol to the enzyme in vitro increases the affinity for phospholipid and calcium such that the kinase is fully active at physiological concen- trations of these effectors. Subsequent isolation of cDNA clones encoding at least nine different PKC proteins from a minimum of seven genes (Parker et al. 1986; Coussens et al. 1986; Ono et al. 1987; Coussens et al. 1987; Ono et al. 1989a; Osada etal. 1990) has revealed the existence of two subclasses of PKC: PKC-cv, -/S and -y, that contain four regions of high sequence homology (Ci_ 4 ), and PKC-5, -e, -£ and -t], that are lacking the second of these conserved domains (Fig. 1). Expression in mammalian cells of the cDNA clones for this latter class has indicated that these enzymes are independent of calcium for activation in vitro Fig. 1. A schematic comparison of the isoforms of PKC. The regions conserved between the various isoforms of PKC are represented schematically along with their relative separation. The Ci region contains two copies of a cysteine-rich repeat (only one copy is found in PKC-f) responsible for phorbol ester/diacylglycerol binding; the C2 region is thought to confer calcium dependence and is missing from the more recently identified isoforms. The C3 and C4 regions contain sequences conserved in all kinases. The V o domain defines an extended amino-terminal variable region present only in the PKC-<5, -e, • t, and -77 proteins. Journal of Cell Science 100, 683-686 (1991) Printed in Great Britain © The Company of Biologists Limited 1991 and for translocation to the membrane in the intact cell, suggesting that the C2 region is responsible for the calcium dependence of PKC-a-, -ji and -y (Ono et al. 1989a; Osada et al. 1990; Schaap et al. 1989; Olivier and Parker, 1991). C 2 -related sequences are found in other proteins such as GAP (Vogel et al. 1988) and the synaptic vesicle protein p65 (Perin et al. 1990), and a related region in phospholipase A 2 has been implicated in translocating the protein to the membrane in a calcium-dependent manner (Clark et al. 1991). The Ci domain contains two copies of a cysteine-rich repeat that appear to be the sequences required for binding phorbol esters (Ono et al. 19896; Cazaubon et al. 1990). It remains to be clarified whether PKC-£, which only has one copy of this repeat motif (Ono et al. 1989a), responds to phorbol esters or to diacylglycerol, as this enzyme has yet to be purified. Similarly the proto- oncogene c-raf is a more distantly related family member (Ishikawa et al. 1986) that contains only one cysteine repeat and is not known to bind phorbol esters or to respond directly to changes in cellular phospholipid. The C-terminal C3 and C 4 regions of PKC are present in all species and contain sequences conserved in all protein kinases (Hanks et al. 1988). The distribution of the conserved regions within the PKC molecule has given rise to the idea of a two-domain structure for PKC with the N-terminal regulatory domain being linked to the C-terminal catalytic region by a flexible hinge (V 3 ). The regulatory sequences act to inhibit the activity of the intact enzyme, as partial proteolysis of the protein leads to the generation of a 30 kDa fragment which retains phorbol ester-binding activity and a 50 kDa constitutively active kinase fragment (see, for example, Huang et al. 1989; Lee and Bell, 1986). For PKC-o-, -/S and -e, the activating cleavages by trypsin (Young et al. 1988; Schaap et al. 1990) and/or calpain (Kishimoto et al. 1989) have been mapped to the V 3 variable region, consistent with this sequence forming a flexible hinge on the surface of the protein between the two domains. Consistent with the independent behaviour of these two domains, ex- pression of the C-terminal sequences alone in mammalian cells gives rise to functional PKC kinase activity, as assessed by an increased level of expression of a reporter gene from a phorbol ester-inducible promoter (Hata et al. 1989). Similarly, expression of the amino-terminal domain in E. coli leads to the production of a functional phorbol ester-binding protein (see, for example, Ono et al. 19896). Key words: protein kinase C, signal transduction, domain interaction. 683