Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

How are medicines formed?

  1. Jul 21, 2010 #1
    I'm not a science student. So, please keep your replies simple so you can be understood.

    How are medicines formed? Is it a trial and error method at the expense of some poor animals' lives? There are many medicines which were not actually invented for the function they are now known for. Some of the examples I can think of offhand are: Minoxidil and Viagra. Some are completely accidental discoveries. The discovery of Penicillin was a quirk of nature.

    When we make a 100-storey building, we know what to do. Use reinforced-concrete, steel, broad base, etc. In case of medicines, how do they proceed? One procedure I can think of is that they move from simple case to complex, from one relation to another. Garlic can naturally lower the blood pressure. So can onion and perhaps many other vegetables. They would isolate, pinpoint the chemical which is most active in lowering the pressure and synthetically manufacture it. Then, they would combine different other such chemicals known for this effect into one medicine.

    By the way, can you please tell me which medicine or vaccine in the past was extracted from a tree's bark? I don't remember what disease it cured but it was a famous discovery. Perhaps, someone drank water from a pond beside that tree and came to know that the beneficial effects were the result of natural extraction of leaves, bark which had fallen in the pond. Thanks for your time and any help you can offer.

  2. jcsd
  3. Jul 21, 2010 #2
    You pretty much have the gist of it. Most of the compounds found in medicines have their analogs in nature (like the opioids-morphine et al-which are analogs to heroin but were developed to have less abuse potential, euphoric effects and more pain relieving potential etc). Many are just happy accidents, like the penicillins etc. There is no real step by step method to developing new drugs. It takes probably an equal amount of intelligence, creativity, hard work and know-how to find, isolate, synthesize etc. Drugs do have to go through several sets of trials to be deemed safe and approved for human use, but that is more about proving the safety vs efficacy of the compound once it has already be been discovered/developed. Also it should be stated that many drugs are just chemically modified versions of older ones and may, for instance, be active longer in the body.

    I think the medicine you are thinking of is aspirin (acetylsalicylic acid) and it came from the bark of the white willow tree (which was chewed, if I'm not mistaken, and used as a pain alleviating substance in olden times). It follows the same kind of story (this is from memory so don't hold it against me) wherein chemists were able to isolate and study the specific compound and then were able to make a synthetic version. It is a lot more complicated than this but you asked to keep it simple. I'm sure wikipedia will have tons more info on the history of aspirin.
  4. Jul 21, 2010 #3


    User Avatar

    Staff: Mentor

    You mean aside from aspirin? (salicylic acid from Willow bark)
  5. Jul 22, 2010 #4


    User Avatar
    Science Advisor

    Interesting word I just discovered, pharmacognosy, "the study of medicines derived from natural sources":

    EDIT: Quinine, the stuff in tonic water, was also discovered in the bark of a tree. The Quecha used the bark of the cinchona to cure malaria:
    Last edited: Jul 22, 2010
  6. Jul 23, 2010 #5


    User Avatar
    Science Advisor

    Trial and error, but not purely by trial and error anymore.

    Today, a typical drug is developed roughly as follows:
    1) Basic research on the disease leads to some discovery, e.g. that a certain enzyme X is involved in the disease. Typically it's all quite complicated and all you have is a guess as to what it's doing.
    2) A postulated drug mechanism. For instance, the idea is postulated that de-activating enzyme X might cure the disease or alleviate the symptoms. This might be shown, for instance, from experiments on 'knock-out mice', where the gene that codes for this enzyme has been 'knocked-out' (deactivated). Obviously we can't change the genes of the patient, but you might be able to chemically inhibit the enzyme.
    3) Finding candidate substances for inhibitors. This is a whole 'art' in itself, but thing you look at are chemicals which are somewhat similar to the chemical that usually binds to the enzyme. In essence you're looking for something to 'plug the keyhole'. There's a whole laundry list of requirements for this: It needs to 'fit' the enzyme, and it needs to fit that one specifically (or you have lots of side effects). It can't fit too well either, or the drug might permanently disable the thing (which is also unwanted). It has to be relatively non-toxic. That includes any substances it might be metabolized into by the body. It has to be able to get where it needs to go in the body. (especially tricky for things that need to target the brain, as there's a 'blood-brain barrier) It has to survive being metabolized long enough to do what it needs to do, and so on.
    4) Screening the candidates. This is basically trial-and-error. The methods of finding and screening candidates are constantly getting better. But the first step is to synthesize these thousands of compounds and test them against the enzyme (in a test tube).
    5) Next step is to see if the candidate can do its job in an actual cell. (and there are other tests involved such as P450 assays, which test if the main metabolic enzyme Cytochrome P450 breaks it down)
    (By now you've gone from thousands to perhaps dozens of candidates, and somewhere around 4-5 the patent application for the drug would be submitted)
    6) Now you get to the actual animal testing. There are literally millions of different compounds in an organism, each one which might have adverse interactions with the compound in question. There's just no way to predict this, and won't be for the foreseeable future.
    7) http://en.wikipedia.org/wiki/Clinical_trial#Phases". Now that there's at least a reasonable chance it's non-toxic and works, you start the human trials. This is a very strict procedure and takes years. About at the same time process chemists would start working on how to synthesize the substance industrially; because there's a huge difference between manufacturing milligrams and kilograms of something.
    8) If it makes it through the clinical trials (which takes 8 years or longer), you get to go to market, and a new drug is born.

    So obviously there's more luck than skill involved. A pharmaceutical researcher would probably do well to see 5 drugs make it to market in an entire career. In a way, things are getting both 'stupider' and 'smarter' at the same time. Advances in computational chemistry have enabled us to predict candidate substances better. On the other hand, advances in combinatorial chemistry (so-called microassays and such) have enabled us to test many more substances much faster.

    The glory days of pharmacognosy are largely over. In the old days (i.e. a few decades ago), we simply had no way of knowing or predicting which substances would bind to proteins/enzymes, and had to look to nature for analogues and candidates. And we had far fewer means of even determining which substances in the body were involved. Given the explosive growth of biochemistry, biotechnology and related fields, our knowledge and methods of finding drugs are improving daily.

    (BTW guys, it's salicylic acid that's from the willow. Aspirin is acetylsalicylic acid, the acetyl group was added to make it less acidic and easier on your stomache. It's still pretty rough though.)
    Last edited by a moderator: Apr 25, 2017
  7. Aug 1, 2010 #6
    Thank you, everyone.

    Alxm, special thanks to you for such a detailed reply. I don't know how I missed it, I just read it now.

    By the way, I was wondering that isn't 'trial and error' in itself an essential part of chemistry? Chemistry isn't that much straightforward and predictable science as compared to physics. Please let me know your opinion.
  8. Aug 1, 2010 #7


    User Avatar
    Science Advisor

    Well if it was purely trial-and-error we wouldn't really need chemists! :biggrin:
    But yes, basically chemical theory (as in the level of theory used by most working chemists) is largely qualitative and not quantitative. E.g. an organic chemist might predict correctly that a certain scenario would have two competing reactions producing the products A and B respectively. He can usually predict which one of A or B will be preferred, to varying extents. But he can't give an actual number. It could be 99%-1% or 80%-20%. Experiment will have to tell.

    Some things can be simply and fairly accurately predicted quantitatively, such as reaction energies, by calculating heats of formation and such. But things like reaction rates cannot. Doing so requires quantum-chemical calculations, which is a project in itself for a theoretical chemist, and not something your average chemist will do.

    Basically for the first 50 years or so after quantum mechanics, theoretical chemistry was mostly limited to providing these qualitative models for the practicing chemists, with orbital theory, Paulings "The Nature of the Chemical Bond", and so on. But due to the rapid increase in computing power, in recent decades this has begun to change, and the theorists have more and more been able to perform calculations and make useful predictions rather than just explaining the existing experimental results.

    Just as an example, the organic chemistry department at my own university recently created a chair and hired a professor to do theoretical calculations full-time.
Share this great discussion with others via Reddit, Google+, Twitter, or Facebook