Selection of a potential GPCR as a target for a preclinical drug discovery effort involves a number of factors viewed from both business and scientific perspectives. This section deals with the scientific criteria used for the selection of a target. It should be stated that a target should obviously fit a company's business plan including alignment with the therapeutic expertise of the company and the capabilities of other preclinical facilities (chemical libraries, high throughput screening, pharmacokinetics, etc.) that will assist in the drug discovery effort.
In order to illustrate some of the scientific criteria involved in a GPCR drug discovery effort, the Figure 3.2 flowchart will be used. In this scheme, the first part of the drug discovery effort requires the prioritization, selection, and compound library screening of GPCR targets. The second phase involves lead optimization involving rational drug design.12 Important considerations related to GPCR prioritization and discovery research include bioinformatic approaches, receptor localization, draggability, and target validation; the lead optimization phase of drug discovery involves the iterative processes of in vitro and in vivo target-based pharmacological screening and broad-based in vivo pharmacological screening. Each of these points will be discussed in greater detail below and illustrated with specific examples of GPCRs.
As depicted in the Figure 3.2 flowchart, this stage of the GPCR drug discovery process can be roughly grouped into five subsets including those dealing directly with the selection of a particular GPCR target (bioinformatics, receptor localization, druggability, and target validation) and the initial processes in discovering chemical entities that may interact with the receptor including the cloning, expression, and high throughput screening of a compound library to identify GPCR target hits amenable to further compound optimization. This latter step is covered in greater depth in other chapters.
There are fewer iterative steps at this stage as compared to the lead optimization stage; however, in some instances it is appropriate to utilize hits obtained from the high throughput screening phase in appropriate in vivo experimental models expected to reflect the activity of the GPCR in order to validate the target if no transgenic animal or RNA interference data are available. The results from this approach using early high throughput hits should be interpreted with great caution due to the general lack of potency and selectivity of such early hits.
220.127.116.11 Bioinformatics Approaches for GPCR Discovery
Many GPCR sequences were obtained by the use of homology search analyses such as the pairwise alignment tool known as BLAST1314 and degenerate primer pairs in reverse transcriptase/polymerase chain reaction experiments utilizing RNA from tissues of interest.415 However, these approaches are best suited for closely related GPCRs and are not likely to identify unique GPCRs with disparate structures and sequences.
With the cloning of the human genome, the genomic sequences for all GPCRs became available but the problem of identifying unique GPCRs from this mass of information remained. Takeda et al.16 described the identification of 178 GPCRs from the human genome based upon homology sequence search strategies assuming the absence of introns (many Family A GPCRs lack introns) within the GPCR gene. However, this approach will not detect the many intron-containing GPCRs. Others4,517 employed the hidden Markov model1819 — a computational method that provides a statistical representation of the genomic sequence data, weighting the most conserved regions and allowing for insertions and/or deletions within these regions. Although not guaranteed to identify all GPCRs, additional power can be added to this approach by searching against amino acid sequences derived from the DNA sequence (using all three reading frames in both orientations).4,517
Such strategies led to the identification of 26 previously unidentified human GPCRs and 83 mouse GPCRs.5 The identification, cloning, expression, and coun-terscreening of species homologs of human GPCRs are especially important because a number of GPCRs have profound species-dependent pharmacological profile differences that could exert major impacts upon in vivo testing of compounds in preclinical (primarily rodent) models. For example, the a2A-adrenergic,20 histamine H3,21,22 and H423 receptors exhibit profound pharmacological differences between rat and human homologs of the receptor, necessitating the discovery of compounds with balanced activities in both species in order to more easily interpret results from in vivo studies.
GPCR drugs are often noted for their high degrees of selectivity. At the molecular level, this selectivity is due in large part to the high affinity of the drug for the GPCR target. At the in vivo level, where the response of a whole animal or human is measured, the selectivity of the drug can also be attributed to the specific localization of the GPCR and the relative concentration of the drug in that compartment. Thus, one criterion for the selection of a GPCR for a drug discovery effort is its localization. A GPCR with discrete localization to a target of interest (brain frontal cortex for cognition enhancers, dorsal root ganglia for pain, heart for cardiovascular disorders, etc.) with minimal expression in other tissues would be preferable to a GPCR ubiquitously expressed throughout the body where activation or blockade could result in receptor-mediated side-effect liabilities.
In addition, one needs to consider the localization and relative expression level (up- or down-regulated) of the GPCR in the disease state or states in which it is thought to play a role. A variety of approaches have been taken to determine the tissue expression patterns of GPCRs including (in order of precision) (1) in situ hybridization for the detection of the binding of antibodies to the GPCR proteins and/or oligonucleotide probes to tissues, (2) reverse transcriptase/polymerase chain reaction (RT/PCR) of RNA from discrete tissues, and (3) DNA microarrays.24
In an interesting experiment examining GPCR localization, Vassilatis et al.5 described the expression of 100 GPCRs in 26 mouse tissues (9 central and 17 peripheral) using the RT/PCR approach. They were able to show that most GPCRs are expressed in multiple tissues with clearly delineated patterns of expression. However, some GPCRs were expressed only in central tissues (histamine H3 and galanin type 2 receptors), whereas others (cysteinyl leukotriene 2 and CCR3 chemo-kine receptors) were expressed only in peripheral tissues. Interestingly, over 93% of GPCRs were detected in the brain, with the largest number of genes detected in the hippocampus. Additional in situ hybridization experiments using cRNA probes were in good agreement with the RT/PCR results.
Such expression work can be expanded to an even larger scale by using DNA microarrays in which oligonucleotide GPCR chips that contain all human (or other species) GPCR sequences are hybridized with RNA from a number of tissues from control, disease, drug-treated, and other test cases.24 Cluster analyses can then be used to identify patterns of gene expression that may then provide insight into the potential physiological role, side-effect liability, and specificity of the GPCR.4
Was this article helpful?