The inhibitory agent, which contains the tetrafluorotyrosine moiety 15, targets the insulin receptor with a Ki of 4 |iM. The rationale for the use of the fluorinated tyrosine analog was based on the presumed mechanism of catalysis. These investigators reasoned that an active site base partially removes the aromatic hydroxyl proton during the transition state of the enzyme-catalyzed phosphoryl transfer reaction from ATP to the acceptor phenol. Presumably, the enzyme stabilizes this partial-negative charge on the phenol/phenoxide during the transition state, which suggests that a tyrosine analog that is negatively charged might be well accommodated within the active site. The four fluorine substituents not only lower the pKa of the phenol, thereby promoting ionization to the phenoxide at physiological pH, but in addition they render the phenoxide less nucleophilic than its natural counterpart. These investigators also prepared the corresponding d-analog 16, which also displays promising inhibitory activity (Ki=20 |iM). Interestingly, although both 15 and 16 serve as competitive inhibitors versus variable peptide substrate, the l-analog directly competes with ATP as well, whereas the d-derivative does not.
Subsequent work by Fry and his colleagues at Parke-Davis confirmed the usefulness of the tetrafluorotyrosine moiety as a nonphosphorylatable analog, in this case for peptides that target the epidermal growth factor receptor (EGFR) (Fry et al. 1994). The phenylalanine-containing "parent" peptide acetyl-Leu-Ala-Glu-Glu-Ser-Ala-Phe-Glu-Glu displays a Ki of 150 |iM, whereas the corresponding l- and d-tetrafluorotyrosine-containing derivatives exhibit relative inhibitory enhancements of threefold and eightfold, respectively. The Parke-Davis group also prepared peptides that contained other l-ty-rosine analogs, including 3-fluorotyrosine, 3-iodotyrosine, and d-tyrosine, but all of these derivatives were ineffective EGFR inhibitors. Curiously, 3-io-dotyrosine was subsequently found to serve as an excellent tyrosine replacement in a cyclic peptide targeting Src (Alfaro-Lopez et al. 1998).
Walsh, Cole, and their colleagues also examined the use of tetrafluorotyrosine as a tyrosine replacement in a C-terminal Src kinase (CSK)-targeted peptide (Cole et al. 1995; Kim and Cole 1998). However, in this case, the pep-tide serves as a substrate, rather than as an inhibitor, for CSK. These results suggest that the applicability of tetrafluorotyrosine as a nonphosphorylat-able tyrosine replacement is kinase-dependent.
Lam and his collaborators have prepared a series of active site-directed peptides that target the Src tyrosine protein kinase (Lou et al. 1997). These investigators employed both d- and l-napthylalanine (Nal) derivatives in place of the phosphorylatable tyrosine moiety in the sequence Gly-Ile-Tyr-Trp-His-His-Tyr. The corresponding phenylalanine derivative was not prepared; however, the d-Tyr was, which gives a measure of the inherent affinity of the peptide for Src. The IC50 for Gly-Ile-d-Tyr-Trp-His-His is 50 ^M, which indicates that the peptide framework is, comparatively speaking, a remarkably effective peptide-based inhibitor. The corresponding Gly-Ile-Nal-Trp-His-His derivative exhibits only a twofold improvement in IC50 relative the d-Tyr analog. However, the doubly substituted Gly-Ile-Nal-Trp-His-His-Nal exhibits an IC50 of 4 ^M, suggesting that the C-terminal Nal is able to access sites outside of the immediate active site region. Interestingly, one of the less effective inhibitors Gly-Ile-Nal-Trp-His-His-Tyr (IC50=27 ^M) proved to be remarkably selective for Src versus other closely related members of the Src kinase family (Lyn and Lck; IC50>1 mM).
One of the difficulties associated with the acquisition of nonphosphory-latable tyrosine surrogates is their synthesis, which typically resorts to the use of achiral starting material. Following a resolution step, the analogs must then be appropriately protected for use in solid phase peptide synthesis. Some of these difficulties have been circumvented by Kim and Cole, who employed the enzyme tyrosine phenol lyase to prepare gram quantities of an assortment of fluorinated tyrosine analogs (Kim and Cole 1998). The Lawrence group has developed a library-driven strategy, which allows one to prepare and subsequently screen a wide assortment of commercially available aryl-containing amines as peptide-based nonphosphorylatable tyrosine analogs (Niu and Lawrence 1997a,b). In spite of the fact that these are pep-tide derivatives, issues related to synthesis, resolution, and protection of these tyrosine substitutes are all bypassed.
Although the most common protein kinase peptide substrates possess a phosphorylatable residue embedded within the interior of the peptide, protein kinases will also phosphorylate peptides containing tyrosine, serine, and threonine moieties appended off the N- or C-terminus of these substrates. For example, Src catalyzes the phosphorylation of Arg-Arg-Arg-Arg-Arg-Leu-Glu-Glu-Leu-Leu-Tyr-amide (the arginine residues are present for assay purposes, not enzyme recognition). C- and N-terminal residues can be readily appended onto the active site-directed peptide after solid phase pep-tide synthesis. This allows one to employ potential tyrosine analogs that are not protected, possess functionality that might not survive the harsh conditions of peptide synthesis, and even lack the standard a-stereocenter. The synthetic strategy utilizes a solid phase peptide synthesis support (Kaiser's oxime resin) that allows the tyrosine analog to be attached to the synthesized peptide in a fashion that simultaneously promotes cleavage from the resin (Kaiser et al. 1989). For example, a wide assortment of phenylethy-lamine derivatives was attached to the C-terminus of a Src active site-directed peptide (Niu and Lawrence 1997a,b).
These were screened for inhibitory potency, and the lead analog was identified as the dopamine derivative 17 (which, in spite of a p-substituted aromatic alcohol, does not serve as a substrate). A peptide containing the amino acid analog of dopamine, l-dopa (18), was subsequently synthesized and shown to display an inhibitory potency (K=16 |iM) that exceeds the parent phenylalanine-containing peptide by 60-fold.
Cole and his colleagues have reported a high-affinity bisubstrate analog for the insulin receptor protein kinase (IRK) (Parang et al. 2001).
Compound 19 was designed based upon a dissociative mechanism for phosphoryl transfer in the IRK active site. The authors reasoned that the approximately 5 A that separates the aromatic amine nitrogen from the g-thio-phosphate phosphorous roughly recapitulates the distance between acceptor and donor in a metaphosphate-like dissociative mechanism. Unlike the bisubstrate analogs reported to date for the serine/threonine-protein kinases, compound 19 serves as a competitive inhibitor versus both variable ATP and peptide substrate. The Ki of 19 is 370 nM, which corresponds to a binding energy that is roughly equal to the sum of the ATP- and protein-binding site portions of the inhibitor. As one might expect for an inhibitor that contains a peptide sequence targeting IRK, compound 19 is ineffective versus CSK (Ki~40 |iM). These investigators also obtained the crystal structure of the inhibitor bound to the tyrosine kinase domain of IRK. The latter confirmed that the inhibitor is bound in a bisubstrate-like mode with the expected distance between the anilino nitrogen and the g-phosphorous. In addition, the anilino nitrogen is engaged in a hydrogen bond to a key active site Asp residue.
Budde, McMurray, and their collaborators reported an N-myristoylated peptide, myr-Glu-Phe-Leu-Tyr-Gly-Val-Phe-Asp-amide, that serves as an apparent bisubstrate analog for Src (Ramdas et al. 1999). Surprisingly, the cor responding peptide with a free N-terminus is a Src substrate. However, upon acylation the substrate is converted into an inhibitor with the caveat that the acyl group must be lauryl (CnH23CO-) or longer. Clearly, the unexpected structure/activity relationship of this inhibitory species places it in an unusual category in that there is no obvious consolidated nonphosphorylatable tyrosine mimetic present in the peptide framework. Nonetheless, the fatty acyl-peptide serves as a competitive inhibitor with respect to both variable ATP and peptide substrate (poly Glu4Tyr), thereby rendering it, like compound 19, a bisubstrate analog. The nonacylated peptide itself blocks phosphorylation of poly Glu4Tyr with a Ki of 260 ^M via a competitive pattern. Myristic acid also serves as an inhibitor of the Src-catalyzed phosphoryla-tion of poly Glu4Tyr, but in this instance the fatty acid competes with ATP (Ki=35 ^M).
The conjoined myr-Glu-Phe-Leu-Tyr-Gly-Val-Phe-Asp-amide exhibits bisubstrate inhibition with Ki values of 3 ^M (variable Glu4Tyr) and 6 ^M (variable ATP). Consequently, the fatty acyl-peptide is unable to serve as a Src substrate because ATP is unable to bind to the active site in the presence of the myristyl group. Unfortunately, this inhibitory species does not display selectivity against other protein kinases (CSK, PKA, and the FGF receptor). However, it may ultimately be possible to enhance either selectivity or potency by placing the fatty acid moiety at different sites along the peptide chain to minimize the distance between the site of phosphorylation (i.e., the Tyr residue) and the ATP-binding site. These results stand in interesting contrast to earlier work described by Ward and O'Brian (O'Brian et al. 1990; O'Brian et al. 1991; Ward and O'Brian 1993). PKC peptide substrates, upon N-myristoylation, are converted into inhibitors with IC50 values of between 3-10 ^M, depending upon the amino acid sequence. However, unlike the myristoylated peptides that inhibit Src, Ward and O'Brian's PKC inhibitors do not display a competitive pattern versus variable ATP, nor do they serve as competitive inhibitors versus peptide substrate. The authors conclude that their myristoylated peptides bind to a different enzyme form (i.e., the free enzyme) than the peptide substrate (i.e., the enzyme-ATP complex). This accounts for the noncompetitive inhibition pattern versus peptide substrate. In addition, Ward, O'Brian, and their colleagues have suggested that the inhibitory effect conferred by the myristoyl appendage is due, at least in part, to its interaction with the phosphatidylserine cofactor of PKC (O'Brian et al. 1990). However, more recent studies suggest that a myristyl-binding region is located in close proximity to the peptide-binding region of the active site (Zaliani et al. 1998). Consequently, it appears likely that the inhibitory behavior of myristoylated peptides toward PKC is at least partly due to the presence of a near active site hydrophobic region that is able to accommodate the lipophilic fatty acid moiety.
Was this article helpful?
Diabetes is a disease that affects the way your body uses food. Normally, your body converts sugars, starches and other foods into a form of sugar called glucose. Your body uses glucose for fuel. The cells receive the glucose through the bloodstream. They then use insulin a hormone made by the pancreas to absorb the glucose, convert it into energy, and either use it or store it for later use. Learn more...