Abstract. Intracellular targeting may enable protein kinases with broad substrate-specificities, such as multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase) to achieve a selectivity of action in vivo. We have examined the intracellular targeting of three/i-CaM kinase isoforms. The ~-CaM kinase isoform is targeted to the nucleus in transfected cells while the 8^-and 6c-CaM kinase isoforms are cytosolic/cytoskeletal. A chimeric construct of a-CaM kinase containing the 6B-CaM kinase variable domain is rerouted to the nucleus while the native tx-CaM kinase and chimeras of ix-CaM kinase which contain the ~^-or 6c-CaM kinase variable domains are retained in the cytoplasm. Using site-directed mutagenesis, we have defined a nuclear localization signal (NLS) within an l 1-amino acid sequence, likely inserted by alternative splicing, in the variable domain of ~-CaM kinase. Isoform-specific nuclear targeting of CaM kinase is probably a key mechanism in the selective regulation of nuclear functions by CaM kinase.CaM kinase is a multimer that can be composed of several isoforms. We find that when cells express two different isoforms of CaM kinase, cellular targeting is determined by the ratio of the isoforms. When an excess of the cytoplasmic isoform of CaM kinase is coexpressed along with the nuclear isoform, both isoforms are localized in the cytoplasm. Conversely an excess of the nuclear isoform can reroute the cytoplasmic isoform to the nucleus. The nuclear isoform likely coassembles with the cytosolic isoform, to form a beteromultimeric holoenzyme which is transported into the nucleus. These experiments demonstrate isoformspecific targeting of CaM kinase and indicate that such ~argeting can be modified by the expression of multiple isoforms of the enzyme.ROWTH factors, neurotransmitters and hormones typically modify intracellular processes throughout the cell with high specificity and speed. Their action is often mediated by protein kinases with broad substrate specificities that enable them to coordinate multiple responses. However, the ability of such kinases to phosphorylate many proteins, including non-physiological substrates in vitro, suggests that there must be mechanisms for refining their target specificity in situ. We have examined this issue by studying multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase or CaM kinase H), 1 a major intracellular mediator of Ca 2+ action (for review see Hanson and Schulman, 1992b). Our working model is a modification

Multifunctional Ca2+/calmodulin-dependent protein kinase (CaM kinase) participates in diverse calcium signaling pathways in neurons. The alpha- and beta-CaM kinase isoforms are neuron-specific and highly abundant in rat brain. The variable domain of CaM kinase is a potential site for the generation of isoform diversity by alternative spicing of its N- and/or C-terminal segments. We used specific PCR primers which span the variable domain of either alpha- or beta-CaM kinase and isolated three new isoforms from rat brain, namely alpha B-, beta e- and beta'e-CaM kinase. alpha beta-CaM kinase contains 11 amino acids, likely inserted by alternative splicing, at the C-terminal segment of the variable domain. This insertion introduces a nuclear localization signal (NLS) that targets alpha B-CaM kinase to the nucleus of transfected cells; alpha-CaM kinase is excluded from the nucleus. The mRNA and the protein corresponding to this isoform are detected only in the diencephalon/midbrain regions. We have also identified two alternatively spliced isoforms of beta-CaM kinase that lack the 24 amino acid sequence at the N-terminal segment of the variable domain. Alternative splicing of these two isoforms occurs with a three base pair shift of the 3'-splice site. Our analysis shows that these new beta-CaM kinase isoforms are expressed primarily in early developmental stages, and we therefore term them beta e - (embryonic) and beta' e-CaM kinase. Recombinant alpha B-, beta e and beta' e-CaM kinase expressed in COS-7 cells exhibit characteristic Ca2+/calmodulin-dependent protein kinase activity and autophosphorylation.

Translocation of protein kinases with broad substrate specificities between different subcellular compartments by activation of signaling pathways is an established mechanism to direct the activity of these enzymes toward particular substrates. Recently, we identified two isoforms of Ca 2؉ /calmodulin-dependent protein kinase II (CaM kinase II), which are targeted to the nucleus by an alternatively spliced nuclear localization signal (NLS). Here we report that cotransfection with constitutively active mutants of CaM kinase I or CaM kinase IV specifically blocks nuclear targeting of CaM kinase II as a result of phosphorylation of a Ser immediately adjacent to the NLS of CaM kinase II. Both CaM kinase I and CaM kinase IV are able to phosphorylate this Ser residue in vitro, and mutagenesis studies suggest that this phosphorylation is both necessary and sufficient to block nuclear targeting. Furthermore, we provide experimental evidence that introduction of a negatively charged residue at this phosphorylation site reduces binding of the kinase to an NLS receptor in vitro, thus providing a mechanism that may explain the blockade of nuclear targeting that we have observed in situ.Phosphorylation and dephosphorylation reactions control a myriad of signal transduction processes within the cell including cell growth and differentiation, metabolic pathways, and gene expression. The specificity of some kinases mediating these reactions is attained by a strict substrate specificity that limits the action of these dedicated kinases to a single or limited number of potential targets. Other kinases, however, are able to phosphorylate a large number of proteins in vitro, so the in vivo specificity of these kinases must occur through a different mechanism. Examples of these multifunctional or general protein kinases include protein kinase A, protein kinase C, and the Ca 2ϩ /calmodulin-dependent protein kinase (CaM kinase) 1 family consisting of CaM kinase I, CaM kinase II, and CaM kinase IV (reviewed in Refs. 1 and 2). All of these kinases are able to phosphorylate nuclear transcription factors in vitro at sites that either activate or repress gene expression and so all of these kinases have the capacity, at least theoretically, to alter cellular phenotype through changes in protein expression. Over the past decade, the evidence that this actually occurs in vivo is becoming increasingly strong; for example, there is now abundant evidence that protein kinase A mediates activation of the cyclic AMP response element-binding protein through phosphorylation of a key Ser residue (3, 4).Nuclear localization of a kinase is necessary for phosphorylation of nuclear proteins such as transcription factors, although there are examples of transcription factors that are activated in the cytoplasm and then translocate to the nucleus (5, 6). The ability of the catalytic subunit of protein kinase A to be released from cytoplasmic tethering and then passively diffuse into the nucleus, where it can phosphorylate nuclear proteins such as cyclic AMP ...

Here we review our current results studying B cells as APC and the mechanisms by which processed antigen is transported to and held on the cell surface for recognition by the specific T cell along with the MHC class II molecules. These studies were carried out using the globular protein cytochrome c as antigen for which the T-cell antigenic determinant was localized to a C-terminal 10-amino acid peptide fragment. For certain analyses, native cytochrome c or antigenic peptide fragments were covalently coupled to antibodies directed toward B-cell surface structures, allowing the targeting of antigen to the APC surface. Our findings indicate that all B cells function as APC and that the APC function is not differentially regulated in defined B-cell subpopulations. Using cytochrome c-antibody conjugates, it was shown that the surface Ig plays two significant roles in augmenting the B-cell APC function following antigen binding: signalling for enhanced APC function and concentrating antigen for subsequent internalization and processing. Both IgM and IgD appear to function identically in facilitating antigen processing in both immune and nonimmune B-cell populations. Furthermore, the surface Ig does not appear to be specially differentiated to function in concentrating antigen, as antigen artificially bound to other B-cell surface structures including MHC class I and class II molecules is also effectively presented. Lastly, evidence is presented that a previously described B-cell activating factor activity is strongly associated with the membranes of activated but not unactivated helper T cells, providing a mechanism by which the T-cell helper function can be focused on the specific antigen-presenting B cell. Concerning the mechanism by which processed antigen is presented at the B-cell surface, evidence is presented suggesting a role of peptide-binding chaperone proteins which may function to transport peptide to the APC surface and facilitate its association with the appropriate Ia. One candidate protein, PBP72/74, is described which binds peptides but not native antigens, is a member of the hsp70 family and appears to play a role in antigen presentation by the ability of antisera raised against it to block APC functions. Peptide-antibody conjugates were used to explore the spacial restrictions on MHC-restricted peptide presentation and it was shown that peptides covalently coupled to antibodies specific for Ig, class I or class II molecules are effective antigens in vitro even in the absence of processing.(ABSTRACT TRUNCATED AT 400 WORDS)

Helper T cells recognize fragments of antigen bound to the class II molecules on the surface of antigen-presenting cells. Naturally processed antigenic fragments have been isolated from the class II molecules and shown to be heterogeneous in length, ranging from 13 to 25 residues, and to vary at both the N and C termini. A 15-residue peptide in an extended conformation is predicted to fit in an open peptide-binding cleft of the class II molecules. Thus, the longer peptides observed bound to class II presumably have regions which reside outside the cleft. It is not known if the additional length contributes significantly to T cell activation. We have carried out a systematic analysis of the antigenicity of peptides of increasing length beyond the minimally defined T cell antigenic peptide. Here we show that the full functional activities of peptides representing the major antigenic determinant of the protein antigen, cytochrome c, minimally require that the peptides be 23 amino acids long. The long peptides do not require processing and are presented by purified class II molecules incorporated into synthetic membranes, indicating that such peptides associate directly with class II and require no additional cellular machinery for presentation. We also show that a hybrid peptide, 51 residues in length, containing a 29-residue cytochrome c peptide and a "promiscuous" peptide of tetanus toxoid, is more antigenic than the 23-residue peptide alone and significantly, does not require processing. Thus, the additional peptide length, although not predicted to bind in the peptide-binding groove of the MHC class II molecule, has a significant impact on the ability of the peptides to stimulate T cell responses maximally.