November 1, 2005 (Vol. 25, No. 19)
Translational Research Studies at the Molecular Level
Our ability to treat and cure disease requires a deep understanding of the underlying mechanisms. As biological systems are complex, in depth knowledge from different scientific disciplines is required.
Although many new technologies, such as combinatorial chemistry and high throughput screening, have been developed during the last decade and applied within the pharmaceutical industry, the number of new drugs reaching the market has not increased.
Genomics and proteomics research has not yet provided many new targets, and the attrition rate within drug discovery programs is still above 90%. The introduction of new and expensive technologies has not lowered the attrition rates and has lead to an increased cost in discovery and development.
It is now estimated that almost $1.2 billion is required to bring a new drug to the market. This number includes all the failures on the way from selecting a target to clinical research.
One of the main problems may be the progression of disease research from human biology into test systems and later back to human biology again. Successful transfer between human biology and other test systems requires a deep understanding of disease mechanisms at all levels, from organs to molecules.
Unless the disease mechanism is truly understood before drug development starts, the risk remains high that the model-system-based process will develop a successful drug doing the wrong job.
Bridging the Gap
Translational research is trying to bridge the gap between model systems and human biology. It covers a wide range of disciplines including clinical research, biomarker discovery, medical imaging, use of transgenic animals, and mapping of pathways.
Selection of the right targets for drug discovery is one of the key decisions in pharmaceutical and biotechnology research and development. Starting with the wrong target will generate a lot of expense without the possibility to make a drug in the end.
A target can be wrong for many reasons: not relevant to the disease, different in animal systems compared to human, subunit composition is different in various cells, or the assay does not reflect the true situation in the diseased tissue.
Clues as to a target’s involvement in disease (target validation) can be obtained by using different in vitro and in vivo approaches, including analysis of the pattern of expression of the target in normal and diseased tissue, gene modulation using molecular biology techniques, studies using different target modulators, genetic studies, functional tests in cells and tissues, and transgenic animal models.
One of the major problems in working with drug targets is determining whether the different models used are comparable and truly predictive (assay validation). To form a solid basis for mechanistic understanding and thus for “go/no-go” decisions, the fully functional target must be studied.
Mechanisms at the molecular level can be studied by many methods. Biochemical assays are used to understand activity of proteins in vitro. Binding of small molecules to proteins can be studied by a number of techniques, including calorimetry, fluorescence microscopy, chromatography, and spectroscopy.
The structure of proteins can be determined by x-ray crystallography and NMR. However, many of these methods require the purification of proteins, which means that studies are performed outside the proper context.
Ion Channels at the Molecular Level
Ion channels are membrane proteins of key physiological and pharmacological importance and are widely regarded as attractive drug targets. However, only a fraction of known ion channels are exploited commercially or are part of drug discovery programs. One of the reasons has been sub-optimal screening technologies. This is now changing with the development of high throughput patch-clamp technologies.
Ion channels are made up of different subunits and interact with other protein modulators. Differences in subunit composition and modulator interactions can alter the properties of the channel and its interaction with ligands, for instance, molecules developed as drug candidates. It is therefore important to understand the subunit composition in diseased cells and also to understand the composition in humans and model systems before designing assays and choosing animals to work with.
There are currently no standard technologies to determine the subunit composition of any protein complex in situ without the need for extracting and purifying the sample. A method that could look at the structure directly in cells would be the optimal way of determining subunit composition of protein complexes in cells.
Electron Microscopy (EM) has the ability to reach a resolution that will identify different protein subunits in their true biological environment inside the cell. For inorganic material, EM can reach a resolution beyond 1.
For biological samples this is not possible since the sample is destroyed in the electron beam, and resolution is limited by the need to use a low dose, which gives low signal to noise. Developments in the field of protein tomography, such as improved instrumentation, new algorithms for image reconstruction, and better sample preparation, are pushing the resolution for biological samples downward.
It is now possible to look at individual protein molecules inside a cell at a resolution where subunits can be located and domain movements seen. Together with labeling by antibodies the possibility of studying subunit composition of ion-channels inside different cells is now a reality.
Protein Tomography is a tool that scientists can use to close the gap between model systems and human biology at the molecular level for an important class of proteins. This technology is also applicable for other protein classes enabling in situ studies of oligomerization of G protein coupled receptors, flexibility of nuclear receptors and kinases, and the interaction between antibody and antigens. Protein Tomography looks at individual macromolecules, enabling the comparison of different molecules of the same protein inside a cell.
Ion-channels serve as an example of the importance of understanding biological mechanisms at the molecular level. For a drug discovery program the ability to define subunit composition of ion-channels in diseased cells can guide the design of assay development and selection of model systems.
Summary
It is becoming clear that our understanding of complex biological systems is not good enough to be exploited for drug discovery by only applying automated and high throughput systems.
There is a strong need for experiments that couple model systems to human biology and in vitro studies to experiments in vivo. It may be that this has been forgotten in the development of novel technologies or that automation of biological experiments has moved us away from the in vivo situation.
It is reasonable to believe that we now will see a move toward experiments and technologies that produce more quality data rather than quantity. This is becoming painfully obvious in the pharmaceutical industry where there is now a negative correlation between investment in R&D and productivity.
