This is the third in a series of Forum interviews with PRF’s eight new science advisors for 2014-2015.
Howard Fields, MD, PhD, is a professor of neurology and physiology and director of the Wheeler Center for the Neurobiology of Addiction at the University of California, San Francisco (UCSF), US. Fields was a founder of the UCSF Pain Management Center and has made major contributions to understanding and treating neuropathic pain, and to understanding mechanisms of pain modulation and placebo analgesia. His recent work has centered on the problem of addiction, and he has begun to delineate the molecular and cellular circuitry of drug reward. He is a member of the Institute of Medicine and the American Academy of Arts and Sciences. Fields spoke with Neil Andrews, PRF executive editor, by telephone to discuss how working in pain research drew him to the field of addiction, the work his lab is undertaking now, and his outlook for the future of pain and addiction research. Below is an edited transcript of their conversation.
Why did you become a pain researcher?
It was a combination of serendipity, being very struck by seeing people in pain, and thinking that I could try to understand the pain and do something about it. I went to Stanford Medical School, and after I graduated I did internal medicine training at Bellevue Hospital in New York. The patient I remember the most during that time was a young woman who had causalgia in her foot, and nobody knew what to do. Then I was drafted and served in the army for three years (from 1967-1970) during the Vietnam War, as a research neurologist at the Walter Reed Army Institute of Research. I was working with a neurosurgeon who was seeing soldiers with nerve injuries and pain. Someone in our group had heard a lecture, about the gate control hypothesis that Ronald Melzack and Patrick Wall published in 1965, suggesting that soldiers’ pain could be treated by stimulating large-diameter fibers. We tried this in a number of patients who had post-traumatic neuralgia, and we saw dramatic results.
I knew I had to understand this, so I developed a correspondence with Pat Wall and he invited me to the first ever pain conference, which he organized with John Bonica—the conference was held in Issaquah, a small town near Seattle, in 1973, and led to the founding of the International Association for the Study of Pain. Here I was, just this kid, who knew virtually nothing, and I was meeting the pioneers and reading everything they published. At that point I realized that pain research was for me.
About this time I finished my clinical neurology training at the Boston City Hospital Service of Harvard Medical School, and then I went to the University of California, San Francisco, where I joined the faculty and set up a lab to do pain research. I had the good fortune that among the first people who came to work in my lab were Jon Levine and Allan Basbaum, who are both professors now at UCSF and major contributors to the field of pain research.
How, as a pain researcher, did you become interested in addiction and reward research?
It had a lot to do with having Allan and Jon as colleagues in my lab. At the time, we were studying a medial pain pathway running from the spinal cord to the reticular formation in the brainstem and then to the thalamus. Allan had started to use a new autoradiographic technique using tritiated amino acids to trace the anatomy of pain pathways. He said to me that he was looking in the spinal cord and found a pathway running in the reverse direction from the brainstem reticular formation through the dorsolateral funiculus of the spinal cord and terminating in the superficial layers of the dorsal horn. This was a pathway that people didn't really think existed, and here it was—he had found it and it was unequivocal.
We thought this might be a pain-modulating pathway that mediated morphine analgesia, and hypothesized that if we lesioned this pathway in the dorsolateral funiculus of the mid-thoracic region of the spinal cord, we should still see morphine analgesia in the forelimbs but not in the hind limbs, and sure enough, that is what we found; these were experiments in animals. With electrical stimulation of neurons in the site of origin of the descending pathway, we also showed inhibition of dorsal horn neurons that responded to noxious stimulation. We then went to see if this pathway existed in people. Jon teamed up with folks in the dental school and found that the opioid receptor antagonist naloxone enhanced pain in people who had their wisdom teeth removed. We also went on to show that we could block the placebo effect with naloxone.
So you can see how our early work was morphing from studying pain pathways to studying opioids. Then the issue became, if opioids are powerful pain relievers, but they are also producing reward, then there is this problem that the most powerful painkillers are addicting. I wondered whether there was a way to get around this problem—if the pain-relieving effect of opioids could be retained while the addictive quality could be avoided. That is what got me to where I am right now.
Let’s talk more about that—where you are now. Are there any guiding principles that you follow to direct your lab's current research?
When people come to the lab, I tell them the thing they have to keep in mind is that
all of our experiments, in one way or another, must get us closer to our goal of figuring out how to treat addiction.
What specific projects is your lab focused on now?
One of the projects we are working on is to understand how opioids control the reward pathway. The reward pathway includes neurons in the mid-brain that project to the nucleus accumbens, and this pathway overlaps significantly with the distribution of dopamine-containing neurons. A major focus now, using in vitro electrophysiology, is to learn how opioid drugs like morphine affect the firing of dopamine neurons—to understand the synaptic mechanism.
As a corollary, we are also working to understand how opioid actions on these neurons compare with the actions of endogenous opioids. Endogenous opioids act at several receptors fairly equally. For instance, enkephalin is equally effective at the mu and the delta receptors, whereas morphine is relatively selective for the mu receptor. The questions are, Under normal circumstances, what is the endogenous ligand acting at the mu receptor? And is there something different about plant-derived or synthesized drugs that could lead them to become addicting, whereas endogenous opioids are probably not addicting?
We are also studying awake, behaving animals. We are looking at the firing patterns of neurons in the targets of the dopamine neurons and want to determine how those neurons are activated by natural rewards, and then figure out what opioids do to the firing patterns.
Our overall strategy with these projects is to define a neural circuit that is involved in drug reward. We believe that this strategy will lead to a cure for addiction, or at least to the ability to develop very powerful analgesics that are much less likely to cause addiction.
What is the most unexpected thing you have found recently?
My colleague Elyssa Margolis and I have made a discovery that, hopefully, will change the way people think about how drugs like morphine produce reward. Very simply put, most people believe that morphine activates the reward pathway by inhibiting inhibitory neurons, thereby disinhibiting or exciting the neurons that they connect to. The classic model is a neuron whose neurotransmitter is gamma-aminobutyric acid (GABA). GABA is the most common inhibitory neurotransmitter, and GABA neurons connect with dopamine neurons. Mu agonists inhibit GABA neurons, which results in disinhibition of dopamine neurons, and that is how reward is produced. The most surprising and interesting thing we have discovered recently is that morphine directly excites dopamine neurons and does so by opening a calcium channel, called the P/Q channel. These results are in a paper just published last month in the Journal of Neuroscience (Margolis et al., 2014).
You mentioned that endogenous opioids generally are not addicting—what are some of the ways in which addictive drugs differ from endogenous opioids in how they act on the reward system?
If you look at the normal functioning of the brain, taking, as an example, eating behavior, which is probably the most common behavior that produces a reward in a hungry animal, normally you will eat a certain amount and then become full; you experience satiety. Gradually, eating becomes less and less pleasant and then you stop eating until you are hungry again. But with the way that drugs work, you don't necessarily achieve that satiety effect. For example, people who are alcoholics will drink a huge amount and never have that feeling of having had enough.
Again the question is, What is different about how endogenous opioids act that may make them more likely to produce satiety? As I mentioned earlier, morphine is relatively selective for the mu receptor, but there are at least four well-accepted opioid receptors, including the mu, delta, kappa, and orphanin receptors. It turns out that drugs acting at and selective for the kappa receptor actually produce a sense of satiety, so what is different about the kappa receptor and the mu receptor? And is there something different about endogenous opioids, perhaps acting non-selectively, or maybe there is release of kappa agonists, such as dynorphin, along with an endogenous mu ligand?
We published a paper a few years ago showing that, whereas the selective activation of the mu receptor, with a mu agonist, in the reward pathway leads animals to drink more alcohol, if you use a selective delta opioid receptor agonist, they will actually drink less. Philip Portoghese, at the University of Minnesota, has developed bivalent compounds that activate the mu receptor but, at the same time, they block the delta receptor. What is interesting is that if you block the delta receptor at the same time that you activate the mu receptor, you don't observe tolerance to the morphine end of the molecule—there is this interaction between the two receptors. He has also published papers showing that these same bivalent compounds are less rewarding, so they are not as addicting. That is a very exciting finding suggesting that the negative interaction between the mu and the delta receptor could be exploited to prevent the escalation of drug dose that might be a precursor to addiction. We are working actively in this area.
What other findings stand out from your work thus far?
One question that has been of interest to us is the personality features that put people at risk for addiction. One of those features that may also put them at risk for chronic pain is a personality trait called impulsivity or disinhibition. We have been working to understand cortical control of subcortical neurons and how this relates to disinhibition.
The main finding from that work, which was a collaborative project with Vania Apkarian’s group, is that we uncovered a very specific circuit from the prefrontal cortex to the nucleus accumbens. This was very interesting because this same pathway is one of the earliest markers of a shift from acute to chronic pain. This work shows the importance of what I would call decision-making circuits rather than reward circuits—you make a decision about whether you are going to escape from pain. In order to make that decision and use information that you have learned through experience, you need this connection between the cortex and the nucleus accumbens.
One of the more interesting developments over the last 10 years or so, which started with work by Lino Becerra and David Borsook, and continued with the more recent work from Vania’s group, is that there is a growing realization that some of the earliest responses to noxious stimuli are in these decision circuits—circuits that used to be thought of as reward circuits but that are activated by painful stimuli. There is a constant battle between whether you should respond to pain or whether you should go about other activities. How is that decision made, and how are opioids involved? Those are questions that are gaining a lot of ground in the field of pain research, and I think it is a great development. And it is probably where my current work applies most directly to the pain research field.
Finally, I have also collaborated with Frank Porreca, and Frank has shown that when you relieve pain by a variety of methods there is increased release of dopamine in the nucleus accumbens, and this is associated with place preference. That is very consistent with this new model of decision pathways.
What are some of the important unknowns in the field of addiction and reward?
A question that keeps coming up concerns an endogenous opioid peptide that was discovered by James Zadina, now at Tulane University, that has only four amino acids and is a mu-selective peptide. What is interesting about that peptide is that, in contrast to the other endogenous opioid peptides, the precursor molecule for it is unknown. What is the precursor molecule? Is it a breakdown product of some other peptide? If not, where and when is it released?
One general problem is that all endogenous opioids come from large precursor molecules, and it is very difficult to detect how these precursor molecules are cleaved under normal circumstances. That turns out to be very important, because exactly how they are cleaved is going to determine the receptors they act at and their affinity for and efficacy at those different receptors. As far as I know, hardly anyone is working on this problem, and it is one of the most important unknown areas in the field.
As someone who is interested in addiction, reward, and decision making, and how opioids are involved with those processes, what is your perspective on the so-called epidemic of opioid abuse and addiction?
I think I hold the minority view on this issue. I don't doubt that many people are abusing opioids, and there is no question that many of them are overdosing. But where I really part company with most others is that I don't believe that there is any significant risk of addiction in people who have never had a drug addiction problem, and who go to a physician with a primary complaint of pain and then are started on opioids for the first time. What has happened is that people who are abusing opioids were already abusing some other drug, and they have now changed the drug that they use.
The sad thing is that many individuals who don't have much of a risk of addiction are not getting the pain relief that they need because people are afraid that they are going to create an addict. My attitude is that this almost never happens. What happens a lot is that people who really don't have a pain problem will go to a doctor and say they have a pain problem and receive a prescription, but they were already addicts, and they never had a pain problem to begin with.
Looking beyond your own work, where does pain research need to go?
People are now looking at chronic pain as a learning problem, and that is shifting the focus from the periphery and the spinal cord to the forebrain, and this is the great unknown area in pain research—what the role of the cortex is in the development or non-development of chronic pain after injury. One of the problems in current pain research is an almost complete absence of research in awake, behaving animals. That has got to be an area of further development. It is not easy to do, and there is hardly anybody in the pain field who is doing it, whereas there are quite a few people who are doing it in other fields of neuroscience. It just hasn't penetrated the psyche of the pain research community to any great extent, and that is a big problem.
Do you think that will change?
It has to. People are going to figure out everything they need to know about the primary afferent nociceptor and the dorsal horn, and then what are they going to do? They have to do something. Someone is going to figure out that pain researchers need to be doing this type of work, and if I were 34 instead of 74, that is for sure what I would be doing.
What words of advice do you have for young investigators just starting out in pain research?
First, if everybody is working on a problem, then don't work on it. Second—and Allan Basbaum has always said this—don't do pain research without spending some time with the clinicians who are talking to and taking care of pain patients. Then, try to connect the dots between what patients are saying and what you are doing. If you let patients and their clinical problems be your guide, it is hard to go wrong.
Do you have any advice for researchers who want to go into the field of addiction and reward?
My advice is the same: do not follow the herd, and talk to patients so that you can find out from them if what you are trying to explain is something that is critical to the problems they have.
What’s different about pain research compared to addiction research?
What’s different is that in pain research, there is a stimulus, you can show that the stimulus is painful, you know the pain pathway, and, if you cut the pain pathway the pain goes away, and if you stimulate the pain pathway, then the pain comes back. That is a very reliable experimental approach that can account for most of the psychophysics of pain. But in addiction research, there is nothing like that. An addiction is a behavioral and psychological construct, and it is much more difficult to know if the experiment you are doing is relevant to the problem, whereas in pain research, you almost always know. The nice thing for me is that I am comfortable in both areas.
Considering the difficult funding environment for scientific research these days, do you recommend scientific research as a viable career option?
My single piece of advice: do it if you love it. If you love it, and you have the talent, then you are probably going to succeed at it. When I was in medical school I realized that I could go into clinical practice and make a very comfortable and secure living, or I could do research and live with uncertainty. I thought about what I would do if I went into practice and made a lot of money, and realized I would do research—if the answer is obvious, why bother with something else? Even with the limitations on funding, what is unknown is so much greater than what is known, and what people are looking for is good ideas. You might be the one with the good idea.
That is very hopeful and optimistic. Thank you for speaking with PRF.
Thank you. I am very optimistic—is there any point in being pessimistic?
PRF Related Content:
See "Related Content" box in the right column of this page.
Baliki MN, Mansour A, Baria AT, Huang L, Berger SE, Fields HL, Apkarian AV.
J Neurosci. 2013 Oct 9;33(41):16383-93.
Mitchell JM, O'Neil JP, Janabi M, Marks SM, Jagust WJ, Fields HL.
Sci Transl Med. 2012 Jan 11;4(116):116ra6.
Navratilova E, Xie JY, Okun A, Qu C, Eyde N, Ci S, Ossipov MH, King T, Fields HL, Porreca F.
Proc Natl Acad Sci U S A. 2012 Dec 11;109(50):20709-13.
Baliki MN, Petre B, Torbey S, Herrmann KM, Huang L, Schnitzer TJ, Fields HL, Apkarian AV.
Nat Neurosci. 2012 Jul 1;15(8):1117-9.
Levine JD, Gordon NC, Fields HL.
Lancet. 1978 Sep 23;2(8091):654-7.
Levine JD, Gordon NC, Jones RT, Fields HL.
Nature. 1978 Apr 27;272(5656):826-7.
Basbaum AI, Clanton CH, Fields HL.
Proc Natl Acad Sci U S A. 1976 Dec;73(12):4685-8.
Other Forum Interviews with PRF's 2014-2015 Science Advisors: