Bisubstrate analog probes for the insulin receptor protein tyrosine kinase: molecular yardsticks for analyzing catalytic mechanism and inhibitor design

Bisubstrate analogs have the potential to provide enhanced specificity for protein kinase inhibition and tools to understand catalytic mechanism. Previous efforts led to the design of a peptide-ATP conjugate bisubstrate analog utilizing aminophenylalanine in place of tyrosine and a thioacetyl linker to the gamma-phosphate of ATP which was a potent inhibitor of the insulin receptor kinase (IRK). In this study, we have examined the contributions of various electrostatic and structural elements in the bisubstrate analog to IRK binding affinity. Three types of changes (seven specific analogs in all) were introduced: a Tyr isostere of the previous aminophenylalanine moiety, modifications of the spacer between the adenine and the peptide, and deletions and substitutions within the peptide moiety. These studies allowed a direct evaluation of the hydrogen bond strength between the anilino nitrogen of the bisubstrate analog and the enzyme catalytic base Asp and showed that it contributes 2.5 kcal/mol of binding energy, in good agreement with previous predictions. Modifications of the linker length resulted in weakened inhibitory affinity, consistent with the geometric requirements of an enzyme-catalyzed dissociative transition state. Alterations in the peptide motif generally led to diminished inhibitory potency, and only some of these effects could be rationalized based on prior kinetic and structural studies. Taken together, these results suggest that a combination of mechanism-based design and empirical synthetic manipulation will be necessary in producing optimized protein kinase bisubstrate analog inhibitors.     

Mechanism-based design of a protein kinase inhibitor

Protein kinase inhibitors have applications as anticancer therapeutic agents and biological tools in cell signaling. Based on a phosphoryl transfer mechanism involving a dissociative transition state, a potent and selective bisubstrate inhibitor for the insulin receptor tyrosine kinase was synthesized by linking ATPgammaS to a peptide substrate analog via a two-carbon spacer. The compound was a high affinity competitive inhibitor against both nucleotide and peptide substrates and showed a slow off-rate. A crystal structure of this inhibitor bound to the tyrosine kinase domain of the insulin receptor confirmed the key design features inspired by a dissociative transition state, and revealed that the linker takes part in the octahedral coordination of an active site Mg2+. These studies suggest a general strategy for the development of selective protein kinase inhibitors.     

Design, synthesis, and characterization of an ATP-peptide conjugate inhibitor of protein kinase A

An ATP-peptide conjugate was synthesized as a bisubstrate analogue inhibitor of the serine/threonine kinase protein kinase A. The compound was found to be a linear, competitive inhibitor with respect to ATP substrate, exhibiting a Ki of 3.8 microM. The compound was noncompetitive with respect to peptide substrate. The inhibitor was shown to be selective for protein kinase A versus the closely related protein kinase C as well as tyrosine kinase Csk. This analysis provides new evidence for the dissociative transition state of protein serine/threonine kinases and illustrates a simple method to convert a low affinity peptide substrate to a selective and moderately potent inhibitor for these enzymes.     

Protein tyrosine kinase-substrate interactions

Protein tyrosine kinases (PTKs) are enzymes that catalyze the phosphorylation of tyrosyl residues. They are important in physiological and pathophysiological processes. Protein substrates of PTKs are often difficult to discern, but recently reported methods have helped to identify targets and characterize their structural interactions with kinases. A mechanism-based bisubstrate analog strategy has given X-ray crystallographic insights into how several topical PTKs, including the insulin receptor, Abl and epidermal growth factor receptor, interact with tyrosine-containing peptide substrates. These PTK co-crystal structures reveal both conserved and specialized features of recognition that probably contribute to substrate selection and the individual functions of these key enzymes.     

Characterization of the insulin receptor kinase purified from human placental membranes

The insulin receptor purified from human placenta by sequential affinity chromatography on wheat germ agglutinin- and insulin-Sepharose to near homogeneity retained tyrosine-specific protein kinase activity. This purified insulin receptor kinase specifically catalyzed the incorporation of 32P from [gamma-32P]ATP into not only the beta-subunit of the insulin receptor but also histone H2B, a synthetic peptide which is sequentially similar to the site of tyrosine phosphorylation in pp60src (a gene product of the Rous sarcoma virus) and antibodies to pp60src present in the sera obtained from three rabbits bearing tumors induced by the Rous sarcoma virus. In each case, phosphorylation occurred exclusively on tyrosine residues. Insulin stimulated phosphorylation of these substrates 3- to 5-fold. Kinetic analysis using the synthetic peptide indicated that insulin acted by increasing the Vmax of peptide phosphorylation from about 3.1 to 9.5 nmol X mg-1 of protein X min-1, whereas the value of the Km for the peptide, about 1.5 mM, was not significantly changed. This kinase acted weakly on casein, alpha-S-casein, actin, and a tyrosine-containing peptide analogue of a serine-containing peptide used commonly as a substrate for the cyclic AMP-dependent protein kinases. These data show that the insulin receptor kinase displays specificity toward exogenous substrates similar to the substrate specificity observed for pp60src and the protein kinase activity associated with the receptor for epidermal growth factor. The data suggest that the catalytic sites of these three tyrosine kinases are similar and that insulin activates its receptor kinase by increasing the Vmax.     

Phosphorylation of phospholamban by calcium-activated, phospholipid-dependent protein kinase. Stimulation of cardiac sarcoplasmic reticulum calcium uptake

Ca2+-activated, phospholipid-dependent protein kinase (protein kinase C) is able to catalyze the phosphorylation of phospholamban in a canine cardiac sarcoplasmic reticulum preparation. This phosphorylation is associated with a 2-fold stimulation of Ca2+ uptake by cardiac sarcoplasmic reticulum similar to that seen following phosphorylation of phospholamban by an endogenous calmodulin-dependent protein kinase or by the catalytic subunit of cAMP-dependent protein kinase. Two-dimensional peptide maps of the tryptic fragments of phospholamban indicate that the three protein kinases differ in their selectivity for sites of phosphorylation. However, one common peptide appears to be phosphorylated by all three protein kinases. These findings suggest that protein kinase C may play a role similar to those played by cAMP- and calmodulin-dependent protein kinases in the regulation of Ca2+ uptake by cardiac sarcoplasmic reticulum, and raise the possibility that the effects of all three protein kinases are mediated through phosphorylation of a common peptide in phospholamban.     

Surface-plasmon-resonance-based biosensor with immobilized bisubstrate analog inhibitor for the determination of affinities of ATP- and protein-competitive ligands of cAMP-dependent protein kinase

Interactions between adenosine-oligoarginine conjugates (ARC), bisubstrate analog inhibitors of protein kinases, and catalytic subunits of cAMP-dependent protein kinase (cAPK Calpha) were characterized with surface-plasmon-resonance-based biosensors. ARC-704 bound to the immobilized kinase with subnanomolar affinity. The immobilization of ARC-704 to the chip surface via streptavidin-biotin complex yielded a high-affinity surface (K(D)=16nM). The bisubstrate character of ARC-704 was demonstrated with various ligands targeted to ATP-binding pocket (ATP and inhibitors H89 and H1152P) and protein-substrate-binding domain of Calpha (RIIalpha and GST-PKIalpha) in competition assays. The experiments performed on surfaces with different immobilization levels of ARC-704 produced similar results. The closeness of the obtained affinities of the tested compounds to the inhibitory potencies and affinities of the compounds measured with other methods demonstrates the applicability of the chip with the immobilized biligand inhibitor for the characterization of both ATP- and substrate protein-competitive ligands of basophilic protein kinases.     

Multi-PK antibodies: Powerful analytical tools to explore the protein kinase world

Diverse biological events are regulated through protein phosphorylation mediated by protein kinases. Some of these protein kinases are known to be involved in the pathogenesis of various diseases. Although 518 protein kinase genes were identified in the human genome, it remains unclear how many and what kind of protein kinases are expressed and activated in cells and tissues under varying situations. To investigate cellular signaling by protein kinases, we developed monoclonal antibodies, designated as Multi-PK antibodies, that can recognize multiple protein kinases in various biological species. These Multi-PK antibodies can be used to profile the kinases expressed in cells and tissues, identify the kinases of special interest, and analyze protein kinase expression and phosphorylation state. Here we introduce some applications of Multi-PK antibodies to identify and characterize the protein kinases involved in epigenetics, glucotoxicity in type 2 diabetes, and pathogenesis of ulcerative colitis. In this review, we focus on the recently developed technologies for kinomics studies using the powerful analytical tools of Multi-PK antibodies.     

Phosphorylation of synthetic peptide analogs of rabbit cardiac troponin inhibitory subunit by the cyclic AMP-dependent protein kinase.

A series of synthetic peptide analogs of the cardiac troponin inhibitory subunit (TN-1) phosphorylation site sequence, Arg12-Pro-Ala-Pro-Ala-Val-Arg18-Arg19-Ser20-Asp21-Arg22-Ala, have been tested as substrates for the catalytic subunit of the cyclic AMP-dependent protein kinase (EC 2.7.1.37, ATP:protein phosphotransferase). As substrates, these peptides were generally inferior to the pyruvate kinase analog peptide Leu-Arg-Arg-Ala-Ser-Leu-Gly or its COOH-terminal amide analog. Replacing Arg-19 with alanine had only a minor effect on the kinetics of phosphorylation of the TN-1 peptide analog. In contrast, replacement of Arg-22 and Arg-18 with alanine resulted in marked enhancement and reduction of the Vmax, respectively. The results of this study have demonstrated that synthetic peptide analogs of the local phosphorylation site sequences of natural substrates may differ widely in their capacity to act as substrates for this protein kinase. In the case of the TN-1 peptide analogs, the contribution of the 4 arginine residues can be distinguished in terms of their influence on the kinetics of phosphorylation.     

Okadaic acid mimics the action of insulin in stimulating protein kinase activity in isolated adipocytes. The role of protein phosphatase 2a in attenuation of the signal

Treatment of adipocytes with okadaic acid (a specific inhibitor of type 1 and 2a protein phosphatases) resulted in a rapid 8-10-fold stimulation of cell extract myelin basic protein (MBP) kinase activity (t1/2 = 10 min) and kinase activity toward a synthetic peptide RRLSSLRA (S6 peptide) (t1/2 = 5 min). Insulin brought about a smaller stimulation of these two activities (t1/2 = 2.5 min). MBP kinase activity from cells treated with okadaic acid or insulin was resolved by anion exchange chromatography into two well defined peaks; S6 peptide kinase activity was less well resolved. The two partially purified MBP kinases were inactivated by the protein tyrosine phosphatase CD45 or by protein phosphatase 2a (PP-2a). In contrast, partially purified S6 peptide kinase activity was inactivated only by PP-2a or protein phosphatase 1 (PP-1). Furthermore, a 38-kDa protein which co-eluted with one peak of MBP kinase and a 42-kDa protein which co-eluted with the other peak of MBP kinase were phosphorylated on tyrosine after treatment with okadaic acid. These findings illustrate several important points concerning regulation of MBP and S6 peptide kinases. First, these protein kinases are regulated by phosphorylation, and, second, in the absence of hormonal stimuli their activities are strongly suppressed by protein phosphatases. Lastly, the increased tyrosine phosphorylation accompanying the activation of MBP kinases following okadaic acid treatment suggests a role for PP-2a in events that are mediated by tyrosine phosphorylation.     

Src family protein tyrosine kinases induce autoactivation of Bruton's tyrosine kinase.

Bruton's tyrosine kinase (Btk) is tyrosine phosphorylated and enzymatically activated following ligation of the B-cell antigen receptor. These events are temporally regulated, and Btk activation follows that of various members of the Src family of protein tyrosine kinases, thus raising the possibility that Src kinases participate in the Btk activation process. We have evaluated the mechanism underlying Btk enzyme activation and have explored the potential regulatory relationship between Btk and Src protein kinases. We demonstrate in COS transient-expression assays that Btk can be activated through intramolecular autophosphorylation at tyrosine 551 and that Btk autophosphorylation is required for Btk catalytic functions. Coexpression of Btk with members of the Src family of protein tyrosine kinases, but not Syk, led to Btk tyrosine phosphorylation and activation. Using a series of point mutations in Blk (a representative Src protein kinase) and Btk, we show that Src kinases activate Btk through an indirect mechanism that requires membrane association of the Src enzymes as well as functional Btk SH3 and SH2 domains. Our results are compatible with the idea that Src protein tyrosine kinases contribute to Btk activation by indirectly stimulating Btk intramolecular autophosphorylation.     

Protein kinase C substrate and inhibitor characteristics of peptides derived from the myristoylated alanine-rich C kinase substrate (MARCKS) protein phosphorylation site domain.

The phosphorylation sites in the myristoylated alanine-rich C kinase substrate or MARCKS protein consist of four serines contained within a conserved, basic region of 25 amino acids, termed the phosphorylation site domain. A synthetic peptide comprising this domain was phosphorylated by both protein kinase C and its catalytic fragment with high affinity and apparent positive cooperativity. Tryptic phosphopeptides derived from the peptide appeared similar to phosphopeptides derived from the phosphorylated intact protein. The peptide was phosphorylated by cAMP- and cGMP-dependent protein kinases with markedly lower affinities. In peptides containing only one of the four serines, with the other three serines replaced by alanine, the affinities for protein kinase C ranged from 25 to 60 nM with Hill constants between 1.8 and 3.0. The potential pseudosubstrate peptide, in which all four serines were replaced by alanines, inhibited protein kinase C phosphorylation of histone or a peptide substrate with an IC50 of 100-200 nM with apparently non-competitive kinetics; it also inhibited the catalytic fragment of protein kinase C with a Ki of 20 nM, with kinetics of the mixed type. The peptide did not significantly inhibit the cAMP- and cGMP-dependent protein kinases. It inhibited Ca2+/calmodulin-dependent protein kinases I, II, and III by competing with the kinases for calmodulin. In addition, the peptide inhibited the Ca2+/calmodulin-independent activity of a proteolytic fragment of Ca2+/calmodulin protein kinase II, with an IC50 approximately 5 microM. Thus, the phosphorylation site domain peptide of the MARCKS protein is a high affinity substrate for protein kinase C in vitro; the cognate peptide containing no serines is a potent but not completely specific inhibitor of both protein kinase C and its catalytic fragment.     

Phosphorylation and activation of p40 tyrosine kinase by casein kinase-1

Because examination of regulatory trans-phosphorylations can help elucidate the cellular functions of tyrosyl protein kinases, we have investigated the effects of phosphorylation by casein kinase-1 on the activity of the p40 tyrosyl protein kinase. We find that casein kinase-1 can phosphorylate the p40 tyrosyl kinase on serine and threonine residues, in part on a unique tryptic peptide. The phosphorylation induces a substantial increase in the tyrosyl protein kinase activity of p40, in contrast to most instances in which serine/threonine phosphorylation inhibits activity of tyrosyl protein kinases. These findings raise the possibility that p40 might be part of a protein phosphorylation network in which casein kinase-1 participates.     

Nonphosphorylatable substrate analogs selectively block autophosphorylation and activation of the insulin receptor, epidermal growth factor receptor, and pp60v-src kinases

The receptors for insulin and epidermal growth factor undergo tyrosine autophosphorylation in response to ligand stimulation, while pp60v-src is an unregulated tyrosine kinase. In this report we show that each of the kinases phosphorylates an exogenous peptide that corresponds to the insulin proreceptor sequence 1142-1153. When the kinases were pre-phosphorylated, saturable Michaelis-Menten kinetics were observed. However, when the kinases had not been pre-phosphorylated biphasic kinetics were observed; at progressively higher substrate concentrations (greater than Km) less substrate phosphorylation was seen. Furthermore, when the kinases had not been pre-phosphorylated kinase autophosphorylation was inhibited at high substrate concentrations. On this basis we postulated that the substrate inhibition of substrate phosphorylation resulted directly from substrate inhibition of kinase autophosphorylation. To test this we designed additional peptides to function specifically as inhibitors of the kinases. Each of the 3 tyrosine residues within the substrate sequence were replaced either by 4-methoxyphenylalanine or phenylalanine, residues structurally similar to tyrosine but unable to accept phosphoryl transfer. Both analogs inhibited insulin and epidermal growth factor receptor autophosphorylation, whereas only the Phe-substituted analog inhibited pp60v-src phosphorylation. These data suggest that autophosphorylation of tyrosine residues near the kinase active site is a generalized mechanism for tyrosine kinase activation and that activation can be selectively blocked by substrates and nonphosphorylatable analogs.     

Chemical dissection of the effects of tyrosine phosphorylation of SHP-2

The regulation of the protein tyrosine phosphatase (PTPase) SHP-2 by tyrosine phosphorylation has been difficult to elucidate because of the intrinsic instability of the phosphoprotein. In the past, expressed protein ligation has been used to site-specifically incorporate the phosphotyrosine mimic Pmp (phosphonomethylene phenylalanine) into the two tyrosine phosphorylation sites (542, 580) of SHP-2 one at a time to analyze the effects on catalytic behavior. In this study, we have incorporated two Pmps into the phosphorylation sites simultaneously and examined the effects of double SHP-2 tyrosine phosphorylation. We have found that the Pmp groups show close to additive effects on PTPase stimulation, suggesting dual SH2 domain occupancy. The relative effects of the phosphotyrosine analogue difluoromethylene phosphonophenylalanine (F(2)Pmp) compared to those of Pmp were also examined. It was found that the F(2)Pmp analogue showed slightly enhanced PTPase stimulation compared with the Pmp analogue, consistent with its higher affinity for SH2 domains. Taken together with the bis-Pmp studies, these data suggest that double phosphorylation of the SHP-2 C-terminus could give rise to a 9-fold overall PTPase activation, 30-50% of the value associated with deletion of the SH2 domains. Catalytically inactive forms of phosphorylated SHP-2 proteins were also produced by expressed protein ligation. This allowed for a systematic analysis of intermolecular autodephosphorylation of SHP-2, which revealed how conformational plasticity can modulate phosphotyrosine stability.     

Protein kinase D (PKD): A novel target for diacylglycerol and phorbol esters

A novel serine/threonine protein kinase regulated by phorbol esters and diacylglycerol (named PKD) has been identified. PKD contains a cysteine-rich repeat sequence homologous to that seen in the regulatory domain of protein kinase C (PKC). A bacterially expressed NH₂-terminal domain of PKD exhibited high affinity phorbol ester binding activity (K_d = 35 nM). Expression of PKD cDNA in COS cells conferred increased phorbol ester binding to intact cells. The catalytic domain of PKD contains all characteristic sequence motifs of serine protein kinases but shows only a low degree of sequence similarity to PKCs. The bacterially expressed catalytic domain of PKD efficiently phosphorylated the exogenous peptide substrate syntide-2 in serine but did not catalyse significant phosphorylation of a variety of other substrates utilised by PKCs and other major second messenger regulated kinases. PKD expressed in COS cells showed syntide-2 kinase activity that was stimulated by phorbol esters in the presence of phospholipids. We propose that PKD may be a novel component in the transduction of diacylglycerol and phorbol ester signals.     

Multiple phosphorylation sites in the 165-kilodalton peptide associated with dihydropyridine-sensitive calcium channels

Evidence from electrophysiological and ion flux studies has established that dihydropyridine-sensitive calcium channels are subject to regulation by neurotransmitter-mediated phosphorylation and dephosphorylation reactions. In the present study, we have further characterized the phosphorylation by cAMP-dependent protein kinase and a multifunctional Ca/calmodulin-dependent protein kinase of the membrane-associated form of the 165-kDa polypeptide identified as the skeletal muscle dihydropyridine receptor. The initial rates of phosphorylation of the 165-kDa peptide by both protein kinases were found to be relatively good compared to the rates of phosphorylation of established substrates of the enzymes. Phosphorylation of the 165-kDa peptide by both protein kinases was additive. Prior phosphorylation by either one of the kinases alone did not preclude phosphorylation by the second kinase. The cAMP-dependent protein kinase phosphorylated the 165-kDa peptide preferentially at serine residues, although a small amount of phosphothreonine was also formed. In contrast, after phosphorylation of the 165-kDa peptide by the Ca/calmodulin-dependent protein kinase, slightly more phosphothreonine than phosphoserine was recovered. Phosphopeptide mapping indicated that the two kinases phosphorylated the peptide at distinct as well as similar sites. Notably, one major site phosphorylated by the cAMP-dependent protein kinase was not phosphorylated by the Ca/calmodulin-dependent protein kinase, while other sites were phosphorylated to a high degree by the Ca/calmodulin-dependent protein kinase, but to a much lesser degree by the cAMP-dependent protein kinase. The results show that the 165-kDa dihydropyridine receptor from skeletal muscle can be multiply phosphorylated at distinct sites by the cAMP- and Ca/calmodulin-dependent protein kinases. As the 165-kDa peptide may be the major functional unit of the dihydropyridine-sensitive Ca channel, the results suggest that the phosphorylation-dependent modulation of Ca channel activity by neurotransmitters may involve phosphorylation of the 165-kDa peptide at multiple sites.     

ATP analog specificity of cAMP-dependent protein kinase, cGMP-dependent protein kinase, and phosphorylase kinase

The ATP analog specificities of the homogeneous cGMP-dependent protein kinase and the catalytic subunit of cAMP-dependent protein kinase have been compared by the ability of 27 analogs to compete with ATP in the protein kinase reaction. Although the data suggest general similarities between the ATP sites of the two homologous cyclic-nucleotide-dependent protein kinases, specific differences especially in the adenine binding pocket are indicated. These differences in affinity suggest potentially useful ATP analog inhibitors of each kinase. For example, apparent autophosphorylation of the purified regulatory subunit of the cAMP-dependent protein kinase is blocked by nebularin triphosphate, suggesting that the phosphorylation is catalyzed by trace contamination of cGMP-dependent protein kinase. Some of the ATP analogs have also been tested using phosphorylase b kinase in order to compare this enzyme with the cyclic-nucleotide-dependent enzymes. All three protein kinases have high specificity for the purine moiety of ATP, and lower specificity for the ribose or triphosphate. The similarity between the ATP site of phosphorylase b kinase to that of the cyclic-nucleotide-dependent protein kinases suggests that it is related to them. The ATP analog specificities of enzymes examined in this study are different from those reported for several unrelated ATP-utilizing enzymes.     

Insulin-activated protein kinases phosphorylate a pseudosubstrate synthetic peptide inhibitor of the p70 S6 kinase

p70 S6 kinase, a major insulin-mitogen-activated ribosomal S6 protein kinase in mammalian cells, is activated by phosphorylation of multiple Ser/Thr residues on the enzyme polypeptide. A synthetic peptide, corresponding to a 37-residue segment from the carboxyl-terminal tail of the kinase which resembles the sequence phosphorylated in S6, acts as a competitive inhibitor of p70 S6 kinase without itself being phosphorylated by the enzyme. This synthetic peptide is phosphorylated by an array of protein kinases which are rapidly activated by insulin. Thus, these sequences of p70 S6 kinase constitute a potential autoinhibitory pseudosubstrate site, whose phosphorylation is catalyzed by candidate upstream-activating protein kinases.