• Question: Once you've chosen a specific disease to target, how do you go about designing a chemical compound to treat it with?

    Asked by 972prmc53 to Alex, Laura, Lesley, Richard, vediacan on 13 Jun 2015.
    • Photo: Richard Prince

      Richard Prince answered on 13 Jun 2015:

      Hi 972prmc53,
      The first thing to do is to come up with an idea about how to attack the disease. For example, if you are treating a disease like asthma where the airways constrict, you want a drug that either blocks the proteins involved in constriction, or activates proteins that cause relaxation of the airways and opens them up again.
      The next stage would usually to set up an assay to test if compounds have the kind of activity you are looking for. Pharmaceutical companies have libraries of millions of compounds that they can test in this way and they have to use robots to do the experiments in a process called high through-put screening (HTS). HTS robots can test up to 100000 compounds per day!
      If they find some compounds that look promising, they start thinking about how to modify them to make them even better and make them more specific to the intended target – this is really important because it helps avoid side effects. Next, they would test the best compounds in animal models of the disease to make sure they work and in healthy animals to make sure that the drugs are safe. This is a controversial area and people are working hard to replace animals in drug development, but at the moment the alternatives aren’t really good enough.
      If the drug passes all the tests in the lab then in might be tested in healthy humans (this is the kind of work that Alex does, so he should be able to give you some really good insights here). This is called a phase 1 clinical trial. The people involved are usually healthy young males – they are given a low dose of the drug and the researchers look to see what side effects occur and how the volunteers’ bodies metabolise the drug. Finally, the drug will be tested in patients with the disease in question (phase 2 and 3 clinical trials). If it passes these tests, it will then be licensed so that it can be prescribed to patients. This is a long (10-12 years) and expensive process (£550 million)!
      Great question btw!
      Richard

    • Photo: Laura Newton

      Laura Newton answered on 13 Jun 2015:

      Hey!

      First you have to find a protein that is responsible for a disease which can take a few years in itself. When you have one then there are two main ways to try and find a good drug to target it and ‘fix’ it.

      You could test thousands and thousands of molecules that have already been made, against the protein or cells to see if they work. This is called ‘high throughput screening.’ If you find a molecule that works (called the ‘lead compound’) then you make small changes to it and keep testing to see which changes make it better.

      Or you can try and find the specific structure of the protein – so you can see exactly what shape it is. This can be very hard but if you manage it then you can look at the shape of it and design drug molecules that are the right shape to fit into the protein. This is done with computer programs (which is called molecular modelling) which predict how the protein will interact with a drug to find the best shaped drug. Then you make the molecule with chemistry. This whole approach is called ‘structure based drug design’ and can mean that you don’t waste time testing lots of random compounds but it can be very difficult to find the protein structure.

      Then when you have some drugs that might work you have to test them in animals (called ‘pre-clinical testing’) and then humans (called ‘clinical trials’) to make sure that they are safe and they actually work. Again, this bit can be very hard and take a long time. A lot of drugs that worked in a dish don’t work in an actual body or have too many side effects so they don’t make it all the way.

      Thanks for your question!
      Laura

    • Photo: Lesley Pearson

      Lesley Pearson answered on 15 Jun 2015:

      Hi!

      What a great question!

      Firstly, once we have the disease we want to target, we need to know a great deal more about it in order for a drug to have an effect.

      If it is a disease caused by an external organism (like Tuberculosis, which is caused by a bacteria, or Malaria, which is caused by a single celled parasite which we call a protozoan) then what we really want is a process that the organism uses in order to infect us, or survive in the human body, that we don’t use. That way, we can design a drug to ‘turn off’ that process in the disease causing organism without harming ourselves.

      If it is a disease caused by something in our bodies’ natural systems going wrong (like Rheumatoid Arthritis or Diabetes) then we need to know what exactly has gone wrong.

      In both of these cases it is likely that we are going to be looking for a protein called an enzyme. Once we have the target, we need to design a way to measure its activity-that’s where I come in! We will isolate that particular enzyme and set up its ideal conditions (which can be very complicated and differ greatly between different enzymes) and quite often use a colour change reaction to show that the enzyme is working. This process allows us to test many thousands of new chemicals that our chemists have made that they think might be a good starting place for a drug. Often we use vast stores of these that have already been made. These are called compound libraries. None of the chemicals in these libraries are ready to become drugs yet, but they may form the begining of a drug.

      As well as testing these huge libraries, we can also go to our computational chemists for help! They use all the available information about the target molecule and use computer modelling to see what kind of chemical we would need to stop it from working. We can use this information along with the compound libraries to decide what are the best chemicals to work on. The chemists can make changes to the molecules to make them work better, and also to control how well they are absorbed into the body and how easy they are for our body to get rid of.

    • Photo: Vedia Can

      Vedia Can answered on 15 Jun 2015:

      Hi 972prmc53,

      I think the rest of the guys have beaten me to it in answering this questions. One last advice would be to perform lots of cell viability assays and look at the apoptotic pathways because this will give you a better idea of the signalling mechanisms involved; what part of the compound binds to the area of your target cell and activates it. This will give you a better idea about the design of your compound and allow you to further develop your compound.

      Best Wishes,

      Vedia

Comments