UCSF synapse November 20, 1997
Volume 42, Number 12

Medical Marijuana: Doing the Science

    In late October, postdoc Ian Meng presented a paper at the Society for Neuroscience meeting in New Orleans, summarized thus in a UCSF press release and newspapers across the world: "a synthetic marijuana-like drug called WIN 55212 enhances the brain's ability to suppress pain in rats, and probably in humans as well."
   This debriefing was conducted in early November, back in the crowded lab on the 7th floor of Medical Sciences that Meng shares with co-investigators Barton Manning, PhD, and Howard Fields, MD, professor of neurology, and nine others.  Meng is 28; he did his undergraduate work at Brown, and got his PhD there.

meng2.jpg (34017 bytes)
Ian Meng in his lab

   Synapse:  Who decided to name a cannabinoid WIN 55212-2?
   IM:  The company that distributes it is called RBI. [Research Biology International.  The compound was synthesized by Sterling Winthrop in 1990.]  Right now there are three good synthetic cannabinoid agonists that act specifically on the CB-1 receptor.  There's been a big breakthrough in the last year, because they now have a specific antagonist, which lets you be sure that the compound is having its effect on the specific receptor.
    Synapse:  In your study, who took the initiative: you or RBI?
    IM:  I did.  Actually, the antagonist is now free from National Institute of Drug Abuse (NIDA).  I just gave them a call and did the paperwork.
    Synapse:  Why is NIDA making the antagonist available?  In case somebody accidentally ingests some marijuana?
    IM:  Well, marijuana is not lethal, so that's not a big concern.  Recently it's been made available because they want more studies done on cannabinoids.  They're actually getting on the bandwagon and starting to support this kind of research.  It's been a big switch.
    Until recently the government was only willing to fund research that asked questions about the possible negative side effects of marijuana.  But there have been so many anecdotal reports of marijuana helping people with a variety of diseases that people in California and Arizona voted to legalize marijuana for medical purposes.   The voters said, "We're going to use it anyway," so all of a sudden the government is saying, "We'd better start funding research to really see what's happening."
    Synapse:  That is news...  Could you define a cannabinoid?
    IM:  The most active ingredient in cannabis is the delta-9 THC molecule, which binds with a certain affinity to a specific receptor in the brain known as the CB-1 receptor.  There have now been two proposed endogenous ligands for this receptor.  So cannabinoids are drugs that bind to the CB-1 receptor.  There are less active ingredients in marijuana which may have some role in its effects.  That's why we need to do studies where we compare the effects of smoking marijuana to the effects of specific agonists.   You do want to know whether there are other components in the actual marijuana influencing the effect.
    Synapse:  Have all the different cannabinoids been isolated?  Wouldn't you want to know what's in the plant first before you start working with synthetics?
    IM:  I think most of that work was done in the 1970s.  For the kinds of studies I do I want to know that a particular drug is acting at a specific receptor.  That's why I use synthetic compounds...  (Takes a 20 ml vial of clear fluid from a desk drawer).  It comes in a powdered form, I mix it in an oil-based solution and give it intravenously, and I look at the activity of single cells in this particular brain region which can modulate pain.  That's the basic paper I presented at the neuroscience meeting.  I found that when I give the cannabinoids intravenously, it will affect a group of cells in one of the brain's pain areas.  It's thought that these cells, when activated, reduce the amount of pain allowed to be transmitted up through the spinal cord to the brain, where sensation occurs.   So when the cells in this area are activated, they inhibit pain.  Morphine activates these cells to inhibit pain; and so -- we found looking at single cells -- do cannabinoids.
    I wanted to show that the activity of these cells was related to the analgesia being produced.  So the second step was to give the drug to awake rats systemically and then do a microinjection into this brain region to inhibit all cell activity.  Under those circumstances, the animals do not show an analgesic effect after the systemic cannabinoid.  The activity of cells in this brain region is necessary for the cannabinoid to have its analgesic effect.
    Synapse:  So you're trying to figure out how cannabinoids work -- you're not trying to prove that they do work.  You recognize the existing evidence.
    IM:  Yes.  People know that they're analgesic.  But unitl fairly recently it hadn't been proven in animal studies that cannabinoids affect sensation.  When you do pain research it's important to differentiate between the motor effects a drug might have versus the sensory effects.   If you give a drug that has a motor effect -- that decreases the rat's activity, for example -- then some observers might say, "That explains the apparent decrease in pain..."  You can't ask an animal , "Do you feel less pain?"  You have to look at certain behaviors.  What I do is called the tail flick.  [Meng has a testing device in which the heat source is a light bulb inside a gray metal box.   The anesthetized rat is placed on the box with its tail in a groove above a hole, so that the radiant heat from the bulb will reach the tail, which the animal then flicks out of the groove.  More time before flick = feeling less pain.]  If you give an analgesic drug, a cannabinoid, or morphine, then the animal will leave its tail on the heat source much longer.  You can also do this tail-flick in another set-up through a glass bottom with freely moving, unanesthetized rats.
    I also look at the actual transmission of pain signals through the central nervous system.  I do single cell recording -- looking at the activity of single neurons that are related to pain.  I can actually look at the electrical impulses that travel down neurons to tell me how active a cell is.  By doing that, we've been able to show, it's not just motor effects; this cannabinoid has very specific sensory effects.  It affects the neurons in the pain pathway.
    Synapse:  So where is this research heading?   What happens next?
    IM:  One thing I'm very interested in is, why is there this endogenous cannabinoid system in the body?  It's got to be there for some reason.  And it has to be activated under certain types of conditions.   It's been known for a long time that certain types of stress will activate the endogenous opioid system so the endorphins and those kinds of things can kick in.   Say if somebody gets their arm blown off in a war, they won't feel any pain because this pain-modulating brain center that I record in is activated, and it shuts off pain before it can reach the part of the brain that coordinates sensation.  And that's important for survival because you don't want to be distracted by your pain, you want to get out of there.
    So certain kinds of stress involve opioids.  The cannabinoid system is a very separate system but it activates the same kind of neurons.  One hypothesis is that there are different kinds of stress that activate the cannabinoid system.
    Synapse:  Why do plants contain cannabinoids?   And why do poppies contain opium?  Do you find yourself pondering the big evolutionary questions?
    IM:  And the bark of the willow tree for aspirin...  Yeah, there are a lot of natural substances which have specific actions that reduce pain.  I'm not sure why plants have evolved to make these compounds, but it has really helped scientists gain insight into the way the brain works.  It also makes me believe that the Western scientific and medical community could learn something from people in other parts of the world who use all kinds of herbal and other natural remedies.
    Synapse:  Why did the company want to develop a synthetic cannabinoid in the first place?  Why not study the naturally occurring ones first?  Were they trying to create a patentable molecule?  Or a legal molecule?
    IM:  The real reason is, when you make a synthetic compound it can actually be more potent and more effective.
    Synapse:  Why is that?
    IM:  It can bind to the receptor molecule better.  You can target that receptor.  It can also have a longer duration of action.
    Synapse:  Is that what the makers of WIN 55212-2 did?
    IM:  Actually, I think the WIN compound was discovered by accident.  They were looking for something completely different.
    Synapse:  They weren't tweaking naturally occurring cannabinoids?
    IM:  Not at all.
    Synapse:  And this drug you're working with is a cannabinoid because it binds to the cannabinoid receptor?
    IM:  Yes.  And now they have other synthetic compounds that have been screened for activity at the cannabinoid receptor.
    Synapse:  What makes a molecule want to bond with the cannabinoid receptor?  Is there some chemical group offering a special handshake?
    IM:  Basically, it's got to do with the makeup of the receptor.  You have certain amino acids and certain positive and negative charges which have to match both the shape and the charges on the drug.  So you can do computer models to try to figure out what would be a good synthetic compound.   But those usually aren't that good at predicting what's going to bind.   Normally, it's just making a lot of compounds and screening them [for function by biologic assay] and seeing what works.
    Synapse:  Do you talk to people back at the company?  Are they following your work?  Could your work translate into big bucks for this company?
    IM:  I really don't know.  I've come to this purely through scientific interest and I never even think about that kind of thing.   I probably should.
    Synapse:  When I read the story out of New Orleans, I thought about those chemists in Basel 30 years ago, tweaking the amphetamine molecule to come up with Ritalin.
    IM:  No, we really rely on what the chemists give us.  They give us the tools, and then we can kind of figure out what's happening...
    Synapse:  I have a vested interest in seeing that the appropriate research is done on any drug that offers hope in the treatment of epilepsy.
    IM:  Of course there have been anecdotal reports of people using marijuana to help their epilepsy.  I'm sure it's just a matter of time before scientists start looking at that actual mechanisms by which cannabinoids can control seizures.  It's through this type of research that new and better treatments could result.
    Synapse:  Has your life changed since your paper was published?
    IM:  (laughs)  A little bit.  One part is people calling up and wanting to know how they can get this drug to help them, because there are a lot of people with really severe chronic pain for whom nothing to this point has worked.  So we've gotten some calls like that.  Then you've got reporters calling, wanting to know how it affects the whole political debate.
    Synapse:  And what's your line on that?
    IM:  My line is that it should be legal.   It definitely should be legal for people who need it to help with an illness or a disease like chronic pain or epilepsy.  And cannabis can really help.  Basically the science is just showing that there are very specific mechanisms by which cannabis can help.  People are taking this as a medicine, and for very specific reasons.   It's hard to get that point through.
    Synapse:  We've all had a lifetime of prejudice and propaganda.
    IM:  Absolutely...  It's satisfying to really do the science.

An agonist is a drug which binds to and activates a receptor.  An antagonist prevents the agonist from binding.  A ligand is a compound, endogenous or synthetic, that binds to a receptor.

meng.jpg (40242 bytes)

Q&A with Howard Fields

    Howard L. Fields, MD, PhD is vice-chair and professor, Neurology and professor, Physiology at UCSF.
    Synapse:  How could you have been in the field for 30 years and not realized that marijuana had medical applications?
    HF:  It never occurred to me.  I assumed that the drugs we had were good enough but underutilized.  Meaning morphine, codeine...  The opioid-containing medications tend to be underutilized.  Patients -- and doctors -- are deathly afraid that they'll become addicted.  I've know a lot of people who have used marijuana, and I've used it myself in the past, and was never really aware that it was analgesic.  Now people are coming along and saying it seems to be very good as an analgesic.  If they're right, that's great.
    To a layman it seems odd that when we haven't studied the naturally occurring cannabinoids, we're studying a synthetic.
    You can't study the naturally occurring ones very well.
    Why not?
    You don't know the concentration of the effective agent.   It's different from batch to batch.  With the synthetics you know exactly how much you're administering.  And it's a simpler compound.  Now you could say "Why not take the natural product, see if you get analgesic effect and then block it with the antagonist."
    Yes, you could...
    We now know two of the receptors for cannabis.  And we know that you've got a substance that's present in the natural product and acts at that receptor.  So if it gives you pain relief, you know that at least one agent in the natural product does, and you have an idea about how it does it.  That leaves two problems.  One: there might be something else in the natural product other than what we know about and that what we know acts at the cannabinoid receptor that might add to its effectiveness; or there might be something there that detracts from its effectiveness.   Or there might be things that produce side effects or are toxic.  So in my mind you're always better off with a known molecule when you're doing research.
    What does it mean to say that a plant contains 400 cannabinoids?
    It's able to synthesize 400 slightly different molecules, all of which act at the cannabinoid receptor...  Your brain probably makes a dozen endogenous opioids.  And the opium resin probably contains seven or eight that have slightly different actions on the body.
    Wouldn't you assume that these actions would be modulating or synergistic?
    They might be.
    People who've taken marinol say it knocks them on their butt in a way that marijuana doesn't.
    It's pure and it's in high concentration.
    And it doesn't have the modulating effects of the other cannabinoids or trace elements.
    But we don't know that there are such things...  I just read a paper today where they found another endogenous cannabinoid in the brain: 2AG.  So the field is just exploding.  The question is, what's its normal function?
    What's your hyposthesis?
    I don't have one yet.  I'm not an expert.  I hope to become one.  The beauty of this whole thing is not so much that we're going to come up with another drug to treat pain, but when all is said and done, we're going to know more about what makes people the way they are.  The whole trick is, keep an open mind.  You never know when the next insight's going to come.  This new cannabinoid has very powerful anti-memory action.  So you have to ask "Why does the brain have an anti-memory molecule?"
    Obviously there are things too painful to recall.
    Or maybe forgetting is part of learning.  Maybe its an endogenous antidepressant.  We're going to learn something about memory that we never would have suspected if it hadn't been for some serendipitous discovery thousands of years ago that the hemp plant was fun to smoke.  What's really exciting to me is that our knowledge is really exploding; maybe we'll come up with a drug that will improve people's memory.  Maybe it'll be a treatment for Alzheimer's disease.  The main thing is, we've got a lever here for understanding the brain, and only good can come of it as far as I'm concerned.