Researchers are designing new drugs on the basis of their shape and ability to match up like a puzzle piece to a target in the body, writes Dick Ahlstrom
Two Trinity College departments have joined forces to open a new centre where novel drugs can be tested by computer "in silico" long before they are ever prescribed for a human. It should make the development of useful new drugs both faster and safer.
The departments of biochemistry and pharmaceutical chemistry recently launched a Biomolecular Modelling Suite. It allows researchers to model the biochemical interactions between a drug and its target inside the body.
"It is a joint venture operated by both departments to provide hands-on access for our graduates and postgraduates on computational molecular modelling and drug design," explains Dr Mary Meegan, head of the department of pharmaceutical chemistry at TCD. "It allows the modelling of very large molecules such as proteins to be carried out quickly."
The pharmaceutical chemistry department received funding for the suite from the Wellcome Trust.
Cells in the body produce thousands of proteins, the chemicals that allow vital processes to happen. Proteins are made in the cell according to a recipe provided by the genetic code found in genes. Each protein has a specific role in the cell and diseases can arise when a protein is missing or if a faulty genetic code produces a faulty protein that does not perform as required.
Drugs can counteract these problems; either blocking a protein process or enhancing its ability to act. To do this, the drug must be able to reach the target protein, ideally without interfering with any other protein process.
Each protein is unique, but its key attribute is its shape. Shape is everything to a protein and this is dictated by the chemical building blocks - amino acids - that it contains. Cellular processes can be switched on and off when one protein locks into another to make things happen.
Proteins have many functions. Some are "receptors" that sit on the outside edge of the cell like a docking port, waiting to interact with another protein that has the correct shape to connect. Others are enzymes, catalysts that regulate chemical reactions in the cell.
The new modelling suite allows researchers to "see" what is happening at a molecular level between proteins or between proteins and drugs, Meegan explains. This visualisation can be done using a process called X-ray crystallography, which provides the exact molecular shape of a protein sample. The model can use this crystallography data but can also model non-existent drugs or protein shapes on the basis of their chemical content.
"Most drugs interact with proteins, be they enzymes or receptors," she says. "What we can do with this facility is look at the precise part of the drug that is interacting with the protein."
IF THE research group can get the three-dimensional shape of a protein, either from crystallography or by modelling it from its amino acid content, drug designers will have a target shape to try to match. They can model existing drugs or new compounds, adding or deleting pieces to achieve the molecular shape they need.
These modifications can increase the selectivity of a drug, Dr Meegan says, making it target as specific a group of proteins as possible. Equally, they can improve the binding strength between a drug and a protein. They change the structure to make a drug work better, make it less toxic or make it safer for the patient, she says.
"From a medicinal chemistry point of view, it allows you to explain at a precise molecular level how these drugs bind to your target protein or enzyme. It allows you to predict and design a molecule and really guides your design," she explains.
"The area I am interested in is anti-cancer drugs such as tamoxifen." Tamoxifen is a drug that interferes with the activity of oestrogen, a female hormone.
"We are designing molecules that will interact more effectively with the oestrogen receptor," she says. "We put in functional groups that will interact and bind in a specific way. The tools that we have now, particularly the protein modelling, are extremely refined and can give you a good fix on what your protein would look like."
Drug designers can use information about the shape of a receptor protein to screen libraries of drugs, looking for candidates that have the correct shape for the job. "It is not going to tell you in advance how something is going to work," Dr Meegan says. It does greatly shorten the time it takes to identify possible candidates however. "I think the real advantage is in designing new novel drugs or modifying existing drugs."