Endogenous substrates of sphingosine-dependent kinases (SDKs) are chaperone proteins: heat shock proteins, glucose-regulated proteins, protein disulfide isomerase, and calreticulin.

Protein kinases whose activity is detectable only in the presence of sphingosine (Sph) or N,N'-dimethyl-Sph (DMS), but not in the presence of 15 other sphingolipids, phospholipids, and glycerolipids tested (Megidish, T., et al. (1995) Biochem. Biophys. Res. Commun. 216, 739-747), have been termed "sphingosine-dependent kinases" (SDKs). We showed previously that a purified SDK (termed "SDK1") phosphorylates a specific Ser position of adapter/chaperone protein 14-3-3 isoforms beta, eta, and zeta but not tau or sigma (Megidish, T., et al. (1998) J. Biol. Chem. 273, 21834-45). In this study we found the following: (i) other SDKs with different substrate specificities are present in cytosolic and membrane extracts of mouse Balb/c 3T3 (A31) fibroblasts. (ii) The activation of these SDKs is specific to D-erythro-Sph and its N-methyl derivatives, the effect of L-threo-Sph or its N-methyl derivatives is minimal, and nonspecific cationic amphiphiles have no effect at all. An SDK separated as fractions "TN31-33" phosphorylated a 50 kDa substrate which was identified as calreticulin, as well as two endogenous substrates with molecular mass 58 and 55 kDa, both identified as protein disulfide isomerase (PDI). This SDK, which specifically phosphorylates calreticulin and PDI, both molecular chaperones found at high levels in endoplasmic reticulum, is tentatively termed "SDK2". Another SDK activity was copurified with glucose-regulated protein (GRP) and heat shock proteins (HSP). One GRP substrate had the same amino acid sequence as GRP94 (synonym: endoplasmin); another HSP substrate had the same amino acid sequence as mouse HSP86 or HSP84, the analogues of human HSP90. An SDK activity separated and present in "fraction 42" from Q-Sepharose chromatography specifically phosphorylated GRP105 (or GRP94) and HSP68 but did not phosphorylate PDI or 14-3-3. This SDK is clearly different from other SDKs in its substrate specificity and is tentatively termed "SDK3". Interestingly, substrates of all these SDKs so far identified are molecular chaperones or adapters capable of binding to enzymes and key molecules involved in signal transduction, maintaining tertiary structure of bioactive molecules, or maintaining cellular homeostasis in response to environmental stress. Thus, the essential role of Sph and DMS is to activate molecular chaperones, thereby providing a link to the mechanism by which SDK activity regulates cellular homeostasis and signal transduction.