Peptidic and Nonpeptidic Mimicry: An Ultimate Approach for Drug Development

25 February, 1999

Peptidic and Nonpeptidic Mimicry: An Ultimate Approach for Drug Development

Competitive R&D requires ingenuity in the concept and approach, and major improvenemts in productivity are needed that may now come through the use of new technologies, including combinatorial chemistry, robotic high-throughput screening, advances in medical genetics and the application of bioinformatics. In particular, recent advances in the concept and technology of molecular modeling may lead to innovative and perhaps ultimate approaches to the development of new drugs. As an increasing number of 3D structures of proteins and their complexes with other molecules are elucidated (as of 20 November, 1998, 8919 entries deposited at Protein Data Bank), atomic interaction at the active site (pharmacophore) may be utilized to conceive agonists and antagonists of the molecule (ligand) in both natural and truncated forms. These agonists and antagonists are likely to be first of native peptide sequences. With new methods such as phage-display, non-native sequences of similar or even higher affinity are being discovered. One of the serious problems of peptide ligands is obviously the degradation by peptidases upon use. Physico-chemical data of the essential sequence and the coordinate file at PDB gives the details of atomic orientation of the pharmacophore that in turn offers the leverage of screening compounds deposited in the compound library (300,000-600,000 level) at major institutions such as National Cancer Institute (USA) Developmental Therapeutics Program and major pharmaceutical companies. After the first hit from traditional screening and/or computer-assisted design, combinatorial chemistry and high-throughput technology are employed for expanded derivatives with high affinity and enhanced efficacy. Described below are some of the recent success in this line. Most remarkably, some of the peptidominetics are of non-peptide in nature, that is, being expected to be more advantageous in therapeutic use in the resistance to degradation, in permeability and in possible oral administration.
Peptidomimetics

EPO Mimetic (see elsewhere for more recent and extensive description): Erythropoietin (EPO), used for stimulating production of red blood cells, is a major product in biotechnology and there has been an increasing interest in molecular mimicry with EPO potency for obviously huge commercial implications. Wrighton NC et al (1996: uid=9629199) employed phage display where random peptides are to be exposed on coat proteins of filamentous phage. A library of random peptide-phage was allowed to bind to and subsequently eluted from the extracellular domain of EPO receptor in the screening system. They used weak-binding system to first fish out EPO domain-weak-binding (Kd 10 mM) CRIGPITWVC as the consensus sequence. With further clever manipulation for selecting the higher affinity peptides, a 20-amino acid peptide (GGTYSCHFGPLTWVCKPQGG) with an affinity (Kd) of 200 nM, compared to 200 pM for EPO was isolated, the sequence of which does not actually exist in the native EPO. The EPO receptor requires dimerization for activation, much like the growth hormone receptor. Indeed, the crystal structure at 2.8 A resolution of a complex of this mimetic agonist peptide with the extracellular domain of EPO receptor revealed that a peptide dimer induces an almost perfect twofold dimerization of the receptor (Livnah O et al, 1996: uid=8662530). This 20-amino acid peptide has a b-sheet structure and is stabilized by the C-C disulfide bond.

TNF Mimetic: An successful example of mimetic antagonist peptide that blocks the TNF-[alpha]-mediated cytotoxicity (Chirinos-Rojas CL et al, 1998: uid=99036427) is a recent case of success in the hunt for peptidomimetic antagonists. A 15-mer phage-displayed peptide library was used to identify potential TNF-[alpha] antagonists. After direct interaction of recombinant human TNF-[alpha] with the library, four randomly selected phage clones were shown to inhibit in a dose-dependent fashion both mouse and human TNF-[alpha]-induced cytotoxicity in vitro. Sequencing of the 4 clones revealed a common amino acid sequence that does not bear any structural similarity to the known primary structures of the extracellular domains of either 55-kDa or 75-kDa TNF receptors. This sequence was synthesized and shown to bind to the recombinant human TNF-[alpha] to inhibit both recombinant mouse and human TNF-[alpha]-induced cytotoxicity in vitro in a dose-dependent manner. These findings support the potential use of phage-displayed random peptide libraries for the identification of novel low molecular antagonistic molecules.

Peptide Aptamer Against Cyclin-Dependent Kinase 2 (Cdk2): Understanding the network that operates inside cells requires the dissection of interactions among network members. In particular, ligand-protein binding, enzymatic protein phosphorylation and protein-protein interactions are important in signal transduction, gene expression and cell differentiation. Barak A et al (1998: uid=99045674) have reported a peptide aptamer that distinguishes among such interactions. This aptamer binds to human Cdk2 and inhibits its kinase activity. The aptamer, pep8, has distinct substrate specificity in that it inhibits the action of Cdk2 kinase only on one specific substrate, histone H1, but not on another substrate, retinoblastoma protein (Rb). Naturally occurring inhibitors, such as p21Cip1, inhibit the activity of Cdk2 on all its substrates. The Pep8 (20-residue peptide) was designed, in a combinatorial library of 20-residue peptides, to mimic the recognition function of the complementarity-determining regions of monoclonal antibodies against different epitopes on the Cdk2 and those displayed by the active-site loop of E. coli thioredoxin. Yeast two-hybrid system (Finley RL Jr and Brent R, 1994: uid=95108082) reveals individual binary interactions between proteins in two-dimensional matrices, and it was used to isolate the aptamer. The aptamer competitively binds to Cdk2 at or near its active site. In fact cell cycle was retarded when cytomegarovirus-pep8 transfected Saos-2 cells.

Aptamers that break specific protein-protein interactions ,and those that interact with specific alleles of gene products, should be useful as dominant agents in the analysis of complex biological systems, and may find therapeutic use.

Nonpeptidomimetics

G-CSF Mimetic: Granulocyte colony-stimulating factor (G-CSF) for the production and activation of granulocytes is used to treat human neutropenias as in cancer chemotherapy. G-CSF binding triggers homodimerization of its receptor that leads to activate protein tyrosine kinases, JAK1 and JAK2, which then phosphorylates tyrosine of receptor-associated STATs (signal transducers and activators of transcription). The STATs dimerize, translocate to the nucleus and bind to specific DNA sequences for transcriptional regulation of responsive genes. It is crucial to figure out successful screening systems to find the first hit for the target. Here is an ingenious approach taken by Tian SS et al (1998: uid=98322350) to isolate a nonpeptidyl small chemical compound (SB 247464) that is a G-CSF mimetic. The screening relied on a G-CSF-responsive construct consisting of 4 copies of a synthetic STAT-binding element linked to a minimal promoter and the reporter gene (luciferase). Murine myeloid cell line NFS60, stably transfected with the construct (clone 4B6), expressed a 20-fold increase in luciferase activity in response to G-CSF. At the optimal concentration (1 uM), SB 247464 had an efficacy of 30% of G-CSF in luciferase assay. To raise neutrophil counts to ca 4-fold in normal mice (sc injection), SB 247464 was used at 30 mg/kg that was equivalent to 50 ug/kg of G-CSF. This compound binds to extracellular domain of G-CSF receptor that is in fact distinct from that for G-CSF binding. The compound had no effect on cell lines that contain STAT-responsive luciferase activated by erythropoietin, interferon alpha, or interferon gamma.

Somatostatin Receptor Subtype Mimetics: Somatostatin inhibits hormone secretion by the pituitary and gastrointestinal tract, acts as a neuromodulatory peptide in the central nervous system, and is implicated as an inhibitor of tumor growth through somatostatin receptors. Somatostatin occurs naturally in 2 major 14- and 28-amino acid forms (ss-14 and ss-28). ss-14 is C-C cyclic AGXKQFFTrp(8)Lys(9)TFTSX. Five somatostatin receptors (sst-1 to sst-5) have been cloned from human tissues. All 5 receptors are members of the G protein-linked receptor family. The Trp(8)-Lys(9) is important for high-affinity binding of somatostatin. Clinically, potent analogs such as octreotide (SMS 201-955 or Sandostatin) are used for the treatment of acromegaly and certain endocrine tumors (Weckbecker G et al, 1993: uid=94294478).

The concept that there are recurring structural units termed "privileged structures" in many of receptor ligands of biologic origin (Evans BE et al, 1988: uid=89055725) led to an idea of synthesizing oft-present double-ring structures in a combinatorial library of 20 diamines, 20 amino acids, and 79 amines and fished out a lead compound that was further elaborated to have high affinity and selectivity for sst-2. One of the compounds, L-054,522, binds to human sst-2 with an apparent dissociation constant of 0.01nM and 3,000-fold selectivity when compared against the other somatostatin receptors (Yang L et al, 1998: uid=98393732). This discovery prompted further search for compounds selective for each subtype of human somatostatin receptor (Rohrer SP et al, 1998: uid=9784130). As the Trp(8)-Lys(9) of ss-14 is necessary for high-affinity binding, the pharmacophore created by the dipeptide was given priority together with molecular constraints of distance geometry based on NMR data to randomly assemble compounds from chemical collection of 200,000. The side chains of residues Tyr(7)Try(8)Lys(9) were derivatized in combinatorial libraries. Combinatorial library #1 was synthesized with 79 aromatic moieties, 20 Trp moieties, 20 diamine moieties, namely to have 79 x 20 x 20 = 130,000 compounds. This library was expanded further such as in the combination of 147 x 22 x 21 =350,000 or in that an aryl indoles were used in stead as the aromatic moieties. Five combinatorial libraries have after all produced somatostatin receptor subtype-selective compounds with up to 6,000-fold selectivity to respective receptor subtypes. The dissociation constants (nM) compared with that of ss-14 were L-797,791 for sst-1: 1.4/0.4 for ss-14; likewise, L-779,976 for sst-2: 0.05/0.04; L-796,778 for sst-3: 24/0.7; L-803,087 for sst-4: 0.7 /1.7; L-817,818 for sst-5: 0.4/2.3. Thus combinatorial chemistry is quite powerful for finding and refining new chemical entitites specific and selective to a chosen target. It is also indicated that some of the above-described subtype-selective compounds seem to differentially mediate the inhibition of hormone release (Yang L et al, 1998: uid=98393732). This approach should be useful to develop drugs for specific pathological conditions and to prescribe individual patients based on their pharmacogenomic backgrounds. The nature of being nonpeptide is also expected to be advantageous for oral administration and good permeability.　

Please send your comments and ideas through e-mail form at this website.

(go back to top)

Home | Site Map | Contact Us | About iHumans | Alliance and R&D Collaboration | Seach Executives and Scientists | Job Openings and Opportunities | Health Care | News and Commentaries | ニホン語のページ (Japanese Pages) |