A ubiquitous type of enzyme that modifies proteins and regulates their function by catalyzing the addition of a phosphate group

In biochemical language, protein kinases (PKs) transfer the terminal phosphoryl group of the compound adenosine triphosphate (ATP) to an amino acid in the target protein. It is estimated that as much as 3% of all the genes code for PKs (Hunter 1994; Venter et al.2001).

After proteins are produced on the ribosomal machinery in the cell by translation from their corresponding messenger RNA (mRNA, "protein synthesis), they are still subjected to a variety of post-translational modifications, which regulate their function. These post-translational modifications can switch cellular activity from one state to another. PKs are the most ubiquitous agents of post-translational modification in all tissues. The superfamily of PKs is classified into a number of families (Hanks and Hunter 1995). A major meta"crite-rion in this classification is the target amino acid: serine/threonine, or tyrosine. Most phosphorylation sites on proteins involve serine and threonine residues, and only about 0.1% involve tyrosine ('dual specificity' kinases phosphorylate both serine/threonine and tyrosine). PKs phosphorylate other proteins, but in many cases can also undergo autophosphorylation and regulate their own activity. The multiple families of serine/threonine kinases include PKs regulated by cyclic nucleotides, e.g. cyclic adenosine monophosphate (cAMP)-dependent PK (PKA); diacylglycerol-activated/phospholipid-dependent PKs (PKC), e.g. "calcium-dependent PKCs; calcium/calmodulin dependent PKs, e.g. multifunctional calcium/calmodulin dependent PK (CaMK); and mitogen-activated PKs (MAPKs). Other subfamilies of serine/threonine kinases are also known. Protein tyrosine kinases are conventionally classified into "receptor tyrosine kinas-es, which are associated with the cell membrane, and nonreceptor tyrosine kinases. Each of these protein tyrosine kinase families is further classified into subfamilies.

The biochemistry and molecular biology of PKs is complex. In the context of memory mechanisms, some generalizations and examples can however be made. PKs respond, either directly or indirectly, to extracellular "stimuli. This means that they fit to serve as components of the molecular "acquisition or "retrieval machinery in neurons (see also "ion channel, "reduction). PKs can switch the cell from one functional state to another, and some types regulate differentiation and growth. This implies that they fit to serve as triggers for "consolidation, memory "phase shifts and long-term memory. Some types of PKs can be converted into a "persistently active form that is autonomous of the activating signal. This implies that these PKs can serve as molecular information storage devices in neurons, and retain activity dependent information over time. A few examples will illustrate the aforementioned generalizations.

1. PKA. Here the most detailed data so far are from "Aplysia, although data from other invertebrates

(e.g. Müller 2000) and from mammals (e.g. Abel et al. 1997) are also abundant. In the circuits that encode defensive reflexes in the sea hare, facilitatory interneu-rons that mediate the sensitizing stimulus release serotonin and other neuromodulators that bind to "receptors on the sensory neurons. These receptors activate the enzyme adenylyl cyclase, generating the second messenger cAMP ("intracellular signal transduction cascade). PKA is composed of two types of subunits, catalytic (C) and regulatory (R) (Figure 57). In the holoenzyme (i.e. C + R), R masks the enzymatic active site on C. When cAMP binds to R, it dissociates it from C and activates the latter. C phosphorylates substrate proteins in the "synapse, culminating in changes in ionic conductances, "neurotransmitter release, and an overall enhanced efficacy of the sensory-to-motor synapse. This results is synaptic facilitation, which is taken to be the cellular correlate of behavioural "sensiti-zation in "simple systems. In the above process, PKA, as part of the cAMP signal transduction cascade, fulfils multiple roles. It is part of an acquisition machinery that creates the short-term trace (probably by phosphorylation of selected types of "ionic channels); it is also a component of the consolidation machinery that

Fig. 57 A simplified scheme of the structure and function of the cAMP-dependent PK (PKA). The enzymatic complex is composed of two types of proteinic subunits, catalytic (C) and regulatory (R). R masks the enzymatic active site on C (upper configuration in the scheme). Binding of the intracellular messenger cAMP to R, dissociates R from C (middle configuration), and renders the latter free to phosphorylate its substrate proteins in the cell (bottom configuration). Cellular processes that degrade R or lower it affinity to C could hence lead to the accumulation of an autonomous, persistently active C. This could sustain use-dependent changes in the synapse for many hours (Buxbaum and Dudai 1989; Chain et al. 1999).

Fig. 57 A simplified scheme of the structure and function of the cAMP-dependent PK (PKA). The enzymatic complex is composed of two types of proteinic subunits, catalytic (C) and regulatory (R). R masks the enzymatic active site on C (upper configuration in the scheme). Binding of the intracellular messenger cAMP to R, dissociates R from C (middle configuration), and renders the latter free to phosphorylate its substrate proteins in the cell (bottom configuration). Cellular processes that degrade R or lower it affinity to C could hence lead to the accumulation of an autonomous, persistently active C. This could sustain use-dependent changes in the synapse for many hours (Buxbaum and Dudai 1989; Chain et al. 1999).

stabilizes the long-term trace (see also *CREB). PKA has also been suggested as a candidate information-storage device in the neuron. One of the *immediate early genes induced in consolidation codes for the enzyme ubiquitin hydrolase, which enhances specific proteolysis (breakdown of proteins). A major target of ubiquitin hydrolase is the R subunit of PKA. In the absence of R, C becomes persistently active and autonomous. This mechanism can sustain experience-dependent changes in the synapse for at least 24 h (Chain et al. 1999; for other mechanisms whereby the alteration in the R/C ratio can lead to persistent activation of PKA, see Buxbaum and Dudai 1989).

Studies of *classical conditioning in the *honeybee provide another illustration of the postulated role of PKA in triggering the consolidation of neuronal information in the context of learning. Conditioning of the bee to extend its tongue (proboscis) in response to an odour that is associated with sucrose solution, is correlated with activation of PKA in the antennal lobes, the functional analogues of the mammalian olfactory bulbs (Müller 2000). Only multiple-trial conditioning, which induces long-term memory, but not single-trial conditioning, which induces only short-term memory, leads to prolonged activation of the kinase. Inhibition of the PKA during training blocks long-term memory. Mimicry of the prolonged PKA activation, by a biochemical trick that increases the level of cAMP in the antennal lobes, when combined with a single conditioning trial, is sufficient to induce long-term memory. Hence in this study the *methods of observation, intervention, and mimicry combine to provide evidence that the activity of PKA is correlated in vivo with the induction of long-term memory, is necessary for long-term memory, and is sufficient to trigger the long-term registration of the sensory information in memory.

2. CaMKII. This kinase, highly concentrated in synapses, is implicated in multiple facets of neuronal *plasticity and growth (Braun and Schulman 1995). CaMKII is capable of autophosphorylation, which reduces its dependency on calcium, and produces an autonomous kinase. The enzyme can thus be considered as a molecular switch that becomes persistently activated following a transient calcium burst, i.e. a cellular information storage device (Saitoh and Schwartz 1985). In *long-term potentiation (LTP), CaMKII phosphorylates and enhances the responsiveness and the trafficking into the synapse of AMPA *glutamatergic receptors (Barria et al. 1997; Hayashi et al. 2000), and regulates AMPA receptor synthesis (Nayak et al. 1998). Shifts in the availability of the autonomous enzyme alter the responsiveness of synapses. An increased level of autonomous CaMKII was found to favour long-term depression over LTP, providing an appealing mechanism for regulating *metaplasticity (Chapman et al. 1995; Mayford et al. 1995). And mice expressing a mutant, autonomous CaMKII were found to be defective in certain types of LTP, and incapable of forming stable place *maps in the *hippocampus (Rotenberg et al. 1996).

3. MAPKs. The function of the MAPKs involves sequential activation of several cytoplasmic kinases, resulting in transmission of regulatory signals from the cell surface to the nucleus (Seger and Krebs 1995). MAPKs are involved in response to growth factors, and have been shown to be required for the formation of long-term memory (Martin et al. 1997b; Berman et al. 1998). In this process MAPK functions in concert with PKA (Kornhauser and Greenberg 1997; Martin et al. 1997b).

4. Tyrosine kinases.These are involved in many facets of regulation of differentiation and growth (Schlessinger and Ullrich 1992). Ample evidence attests to their role in synaptic plasticity and learning. For example, mutation in a nonreceptor tyrosine kinase, fyn, impaired LTP, *maze learning, yet also hippocam-pal development (Grant et al. 1992). In the behaving rat, tyrosine phosphorylation of the glutamatergic NMDA receptor was shown to correlate with LTP (Rosenblum et al. 1996) as well as with taste learning (Rosenblum et al. 1997).

By no means are the above examples exhaustive. Many other PKs are implicated in plasticity, learning, and memory, for example, members of the PKC family (e.g. Thomas et al. 1994). Judging by their multiple, ubiquitous roles in cellular function, and by the extensive cross-talk of networks of PKs in the cell (*coincidence detector), the involvement of PKs in multiple phases of learning and memory is expected to be the rule rather than the exception. Protein phos-phatases should also not be forgotten (Mulkey et al. 1993; Winder et al. 1998). Whatever goes up also comes down, and phosphatases are enzymes that undo what the kinases do. Protein phosphatases may inhibit memory formation in the first seconds or minutes in the life of an *engram, and, furthermore, erase cellular traces of immediate- and short-term memories. They should definitely provide interesting targets for modern analogues of the legendary *lotus.

Selected associations: CREB, Consolidation, Intracellular signal transduction cascades, Plasticity, Reduction

0 0

Post a comment