WOBURN, Mass., and RAMAT-GAN, Israel, Aug. 18 /PRNewswire/ -- Predix Pharmaceuticals Inc., a drug discovery and development company, announced today that peer-reviewed data illustrating the company’s repeated success in using computerized 3D models of GPCR drug targets for blind in silico screening was published in the August 3rd (Vol. 101, No. 31) issue of the Proceedings of the National Academy of Sciences (PNAS). Additional data are presented in the October 1, 2004, issue of the journal Proteins. These publications present some of the most promising examples of the successful utilization of in silico models for GPCR-targeted drug discovery. The article, “G Protein-Coupled Receptors: In Silico Drug Discovery in 3D,” shows that the Company’s proprietary PREDICT(TM) methodology achieved identification of high-quality hits, including promising lead compounds directly out of the in silico screening for multiple drug discovery programs - all utilizing Predix’s computer generated GPCR structures. By integrating this computational technology with standard medicinal chemistry, Predix has created several lead programs and has recently completed Phase I human studies with their most advanced compound.
“We are extremely pleased with the outcome of our research using the PREDICT(TM) 3D technology. The data presented demonstrate that the hits we identified in silico -- and verified in standard biological assays -- serve as high quality lead compounds for drug discovery,” said Oren Becker, PhD, co- founder and Chief Scientific Officer at Predix. “Using our extensive set of proprietary GPCR and other protein structures, Predix was able to optimize a selective 5-HT1A agonist in six months with only 31 compounds synthesized. This work paves the way to a broader application of 3D models in GPCR drug discovery and Predix Pharmaceuticals is leading the way in this important field of research.”
In this study reported in PNAS, the Company applied its PREDICT(TM) technology to model the 3D structure of five different GPCR drug targets in silico and then screened its virtual library of two million compounds against these 3D models. Using proprietary docking and scoring algorithms, Predix then selected the 100 top ranked compounds for each target. These compounds were tested in standard assays, which demonstrated that 15% to 25% of the compounds show potent binding to their respective targets. Many of these hits exhibited a range of attractive drug properties qualifying them as lead compounds for these specific drug discovery programs.
The Predix hit-to-lead and lead-optimization processes (structure-activity determinations) utilize the 3D structure of the protein together with proprietary algorithms which computationally prioritize hits and leads. Using these algorithms, about 10% of molecules recommended in each lead optimization cycle are actually synthesized by medicinal chemists. Thus, the speed and accuracy of the company’s drug discovery platform yields dramatic reductions in cycle times, at least as efficiently as programs where the 3D structure of the target protein is known. Predix has achieved its goals of moving from initiation of a new project GPCR sequence up to submission of an Investigational New Drug filing for entry into the clinic within 24 months for its first program, and is expects to repeat this for two additional programs within the next year.
About G-Protein-Coupled Receptors (GPCRs)
G-protein-coupled receptors are a pharmacologically important protein family with over 800 genes identified to date. These proteins are responsible for communications between the cell and its environment. Pathways involving these receptors are the targets of 50% of all recently launched drugs including those for the treatment of hypertension, cardiac dysfunction, depression, anxiety and pain. Like other membrane-embedded proteins, GPCRs have characteristics that make their native 3D structure extremely difficult to determine experimentally. This difficulty motivates the search for computational methods that can predict their structure with high reliability.
About PREDICT(TM)
PREDICT(TM) is a proprietary three-dimensional GPCR modeling technology that combines the properties of a protein sequence with those of its membrane environment without relying on x-ray crystallographic or other structural methodologies. The PREDICT(TM) algorithm searches through the receptors’ conformation space for the most stable 3D structure of the TM domain of the protein. To ensure the final model represents the most stable conformation, the method simultaneously optimizes several thousand alternative conformations of the receptor. Predix Pharmaceuticals Inc. has an exclusive worldwide license to this technology.
About Predix
Predix is a drug development company integrating computational and medicinal chemistry to rapidly and efficiently create new drugs targeted to GPCRs and ion channels (IC). Leveraging its proprietary technology, Predix has created a diverse clinical and pre-clinical stage drug development pipeline focused in the GPCR and ion channel areas. The Company’s current pipeline includes one program which has completed Phase I clinical trials, two programs in preclinical stage and four additional programs in lead discovery and early lead optimization. Predix expects to retain a minority of its clinical programs in house through Phase 2 and Phase 3 clinical trials, while generating early upfront and milestone revenues, as well as royalties from partnering some of its novel compounds following Phase I safety studies. Predix was incorporated in December, 2000, and initiated its medicinal chemistry and development activities, with its headquarters, in Woburn, Massachusetts, in September, 2002. The Company was founded in Ramat Gan, Israel, and maintains its computational chemistry division there.
Contact: Christine Hayden (781) 376-0821 Ext. 1010 http://www.predixpharm.com/
Predix Pharmaceuticals Inc.
CONTACT: Christine Hayden of Predix Pharmaceuticals Inc., +1-781-376-0821Ext. 1010
Web site: http://www.predixpharm.com/
