Skip to Content


No Pain, and No Worries?

The FAAH (fatty acid amide hydrolase) enzyme system has provided a number of headlines over the years. FAAH itself is involved in the brain’s endocannabinoid system – it clears neurotransmitters like anandamide – and a number of other biologically important hydroxyethylamide and acyltaurines. So the potential for inhibitors of it in analgesia has attracted a lot of notice, as does any new mechanism for pain relief that offers a possible way out of the opioid class.

That history comes to mind when reading about this new report of a woman in Inverness who has a mutation affecting the expression of just that enzyme. The story, which has gotten a lot of press this week, is a fascinating one: the patient herself was noticed when she had (notoriously painful) hand surgery but didn’t ask for any pain medication post-operation. All she had was some paracetamol with no further complaints, and the hospital staff immediately flagged that as highly unusual. Examination showed that she was, in fact, extremely insensitive to pain of all sorts. The patient was (of course) aware this, since she had been injured numerous times over the years without particularly noticing until later (“multiple scars around the arms and on the back of her hands”), but she hadn’t been concerned. That, actually, was another thing that turned up on interviews and examinations: she wasn’t all that concerned about things in general, with notable lack of anxiety and fear responses.

Sequencing showed that she had an inactivating mutation (now named FAAH-OUT, which as someone living in the Boston area I can appreciate phonetically). Interestingly, it’s not in the the FAAH gene itself, but represents a short deletion some ways downstream of it. It’s believed that this must be part of a long noncoding RNA whose presence strongly affects the expression of FAAH – and indeed, blood assays showed that the patient had significantly higher levels of the amides that FAAH cleaves.

One might think that lower levels of pain and anxiety would be a good thing, but consider those multiple scars. There’s something a bit worrying about not noticing that you’ve been cut until you see dripping blood (or, as the report notes, walking around with osteoarthritris of the hip and not noticing any particular pain from the “severely degenerated” joint). And more subtly, there’s something a bit worrying about not worrying about any of these things over the years. Too much anxiety is crippling, but a tiny bit of it might come in useful now and then, rather like hot sauce. I use that analogy because the patient also reported being able to eat a mouthful of Scotch Bonnet peppers while only noticing a short-lived “pleasant glow” in the mouth(!) This is not a recommended experiment for those people with a functioning nociception system, although do shoot some video for everyone if you’re still foolish enough to try it. Another interesting effect: she also reported longstanding memory problems (forgetting words in the middle of sentences, losing keys, etc.)

Now, if this story had appeared some years ago, it would have set off a chase similar to the PCSK9 story, I feel sure. But we already knew about FAAH, and about FAAH inhibitors. There are a lot of them, and many of them have gone into clinical trials (here’s a recent one). None have progressed, to my knowledge, despite trials for arthritis pain, anxiety and depression, etc., which is rather interesting considering what one might conclude from the “human knockout” example. These rare human mutations are extremely interesting and important, but they do not necessarily provide a royal road to clinical success, either. And it goes in both directions – the knockout mouse literature is replete with examples of things that should have had an effect but produced apparently normal rodents because other systems compensated as the animals developed, and there are surely any number of humans walking around with unusual deletions who have never been noticed because they are also phenotypically nearly invisible.

I have to note that compound in France that precipitated a clinical trial disaster a few years ago, which was a rather nonselective FAAH inhibitor. That mechanism doesn’t seem to have been what led to the terrible effects on patients, though – witness the variety of other FAAH compounds that have gone into humans without any such incidents. Rather, the lack of selectivity along with an extremely ill-advised approach to pharmacodynamic data and human dosing seems to have been the underlying problem, although the exact mechanism that led to the stroke-like effects is (I believe) still unknown.

Finally, there’s another broadly similar enzyme, monoacylglycerol hydrolase (MAGL) that cleaves another class of endocannabinoid, but this system was not affected in the Inverness patient. MAGL is a story all its own – it came into view as a possible cancer target, has been looked at for its neurological effects (pain, depression etc., as in FAAH), and a functionally homologous enzyme in nematodes has just been reported as a longevity factor. That one was picked up when the Cravatt group’s known MAGL inhibitor probe extended lifespan, despite nematodes not actually having MAGL analog of their own! Inhibitors of this enzyme have been studied pretty intensively, but (as far as I’m aware) none of them have made it out of the clinic. Work is definitely still going on, though.

This whole area is notably difficult to get a handle on, and I haven’t even gotten into the metabolic implications of either enzyme. What you see is several levels of complexity: the enzymes involved each handle several different substrates, often in completely different tissues, and the products that they produce each do several different things. So you can’t expect a simple story to appear when you step in with a small-molecule inhibitor (or when you find a person with a mutation in part of the system, either). Evolution has shown no interest in presenting us with modular, well-isolated, plug-and-play systems, but rather repurposes, tacks on, makes do, and piles up complexity with no regard for those of us trying to untangle the results. . .

30 comments on “No Pain, and No Worries?”

  1. Colts Fan says:

    This reminds me of stories I have heard about former Indianapolis Colts coach Tony Dungy, and his son Jordan. Jordan Dungy suffers from a different condition (I believe) called CIPA (congenital insensitivity to pain with anhidrosis), linked to an NTRK1 mutation. In addition to not feeling pain, he also cannot perspire normally. The boy underwent over 30 surgeries when he was young to treat injuries he sustained because he couldn’t feel pain and never learned what things he couldn’t do.

    1. Derek Lowe says:

      Interesting – the report on the Inverness woman specifically mentioned that she perspired normally, so they were aware of that possible connection, it seems.

    2. Imaging guy says:

      NTRK1 is also called high affinity receptor for nerve growth factor (NGF) and an anti NGF antibody (tanezumab) has recently been found to relieve pain in osteoarthritis and low back pain.

  2. Imaging guy says:

    SCN9A is the another gene involved in pain sensation as human knockout of this gene does not feel pain (1). This gene was also believed to be like PCSK9 (2) and antibodies directed against it are assumed to relieve pain. I don’t know how they fared in clinical trials.

  3. Road says:

    They didn’t even show reduced levels of FAAH or FAAH-activity (simple Western blot or substrate assay would be easy). The whole story seems like a huge rush to ascribe an anectodal phenotype to a poorly characterized non-coding mutation. Pretty shoddy science all-around. That, along with the fact that highly potent FAAH inhibitors have been in humans dozens of times with no such analgesic consequences, makes it seem most unlikely that her phenotype is due to this mutation.

    1. Dara says:

      Doesn’t the increased levels of anandamide in her blood strongly suggest inactivity of FAAH? There doesn’t necessarily have to be reduced levels of the enzyme. An enzyme activity assay would have been beneficial I agree

  4. ScientistSailor says:

    Another case where the human KO is not replicated by pharmacological inhibition is NaV1.7. Companies are still beating that dead horse, despite the target having been invalidated…

    1. Bob says:

      Nav1.7 has not been clinically “invalidated”. Simply, potent and selective inhibitors have yet to be identified, and perhaps never will.

      1. ScientistSailor says:

        Do you believe amyloid hypothesis?

        1. b says:

          Sure, it seems to play a fairly profound role in Prion diseases

    2. MrXYZ says:

      Nav1.7 is still a valid pain target although people are certainly not looking at it in as simplistic fashion as they did. Many of the post-mortems of the clinical trials (as well as some of the more recent basic science and pre-clinical studies) suggest that the failure to show clinical efficacy may have been more related to the specific MOA and PK of the actual molecules rather than a problem with the target itself. Unfortunately, these same studies put some pretty strict requirements on what type of Nav1.7 inhibitor would make a good analgesic, which may not be achievable.

      Since this blog was just talking about scandals at Duke, I will note that a lab at Duke published an analgesic Nav.17 antibody a number of years ago. A number of companies tried to replicate the data. No one was ever able to reproduce the data (I suspect at least a dozen companies tried to reproduce the experiments). The original paper was never retracted.

      1. Imaging guy says:

        Are you talking about this paper?
        “A monoclonal antibody that targets a NaV1.7 channel voltage sensor for pain and itch relief”, Cell. 2014 Jun 5;157(6):1393-404. doi: 10.1016/j.cell.2014.03.064

        1. MrXYZ says:

          Correct, that’s the paper. They filed a patent on their antibodies and quite a number of companies tried to reproduce their data. One of these attempts was written up here:

          The Duke lab then followed up on this finding in this paper

          The party line from the Duke lab was that they did have a hybridoma cell line (that they no longer have) that produced an actual Nav1.7 antibody (that they no longer have) but that the sequence they pulled out was incorrect. While this does happen (sequencing hybridomas can be tricky), the Duke lab comes off as either dishonest or incompetent. I know that sounds harsh but a lot of effort was wasted by companies trying to reproduce their poor data.

      2. Rock says:

        Just like the rodent knockouts, human knockouts are not as simple as they may seem. Work from John Wood’s lab in the UK had shown that almost 200 genes were changed in nav 1.7 knockout mice at the DRG level. That included an upregulation of enkephalins. Administering naloxone to a human knockout subject restored their pain sensation as did the knockout mice.

        1. Dunkin says:

          True, but it’s odd that the follow-up clinical trials that Wood et al cited, combos of Nav1.7 inh + opioid (e.g. naloxone), seem to have never materialized. If co-dosing a Nav1.7 inh enabled a much lower dose of opioid, with equivalent efficacy as a high dose, that could be big. But has that been demonstrated in the clinic? Perhaps they’re still sitting on the clinical data: Nav1.7 remains yet another target where companies go in, but never come out the other side.

  5. Todd says:

    Why isn’t anandamide a drug?

    1. Road says:

      Because it gets hydrolyzed immediately by FAAH. Half life of minutes, at-best

      1. Foodscientist says:

        It’s like how ghb is a drug, but it imitates the natural gaba(which is not a drug). Another example is adding more gas to your car will not make it faster or more powerful, as long as you have the minnimum amount.

      2. loupgarous says:

        Which could give it a niche in some kinds of surgery, where you want rapid induction of anesthesia and rapid metabolic inactivation. Wasn’t that the selling point of some of the fentanyls?

    2. Legalize it says:

      The THC in cannabis is a drug that binds CB1 (just like anandamide) and is not hydrolyzed by FAAH. Why has there been no chatter on this post about medicinal cannabis or the CB1 receptor? C’mon people!

      Also, if it wasn’t for the wasteful and inefficient ‘war on drugs’ we’d probably know a lot more about the endocannabinoid system by now since research would have been facilitated. Just so damn aggravating.

      1. loupgarous says:

        I agree. There’s very little rhyme or reason, much less consistency, to how molecules become Schedule I Controlled Substances. They are placed in that category, not by scientists and physicians, but by legislators, whose qualifications to judge the danger those molecules pose consist solely of having been elected, and bureaucrats, who may or may not have even those qualifications.

        I just talked a close relative down from an severe adverse reaction – fear, agitation, anger – to an atypical depressant he was prescribed in a drug diversion program. The drug he’s to be diverted from, cannabis, has never been subjected to large scale safety and effectiveness studied so that the need for it to be on Schedule I can be determined. At least two DEA administrative law judges have ruled that cannabis doesn’t belong on Schedule I – and been overruled, which makes you wonder why DEA even bothers paying administrative law judges.

        The reasoning seems to go like this:

        Cannabis is a dangerous drug. How dangerous and in which ways? Here are some studies we’ve paid for. High p-values and small cohorts, you say? Well, we’d have better studies, but it’s hard and expensive to perform large-scale studies on cannabis – because it’s such a dangerous drug.

        This is why more and more state legislatures are making their own decisions on how dangerous cannabis is. Unfortunately, those are purely political decisions, too.

        Cannabis, due to our legal attitudes, exists in a box marked “Don’t Study!” Our decisions regarding it, whether to permit or prohibit its use, are almost guaranteed not to be based on scientific evidence.

  6. MrXYZ says:

    This is a response to Imaging guy above. I was having trouble replying directly to his comment.

    Correct, that’s the paper. They filed a patent on their antibodies and quite a number of companies tried to reproduce their data. One of these attempts was written up here:

    The Duke lab then followed up on this finding in this paper

    The party line from the Duke lab was that they did have a hybridoma cell line (that they no longer have) that produced an actual Nav1.7 antibody (that they no longer have) but that the sequence they pulled out was incorrect. While this does happen (sequencing hybridomas can be tricky), the Duke lab comes off as either dishonest or incompetent. I know that sounds harsh but a lot of effort was wasted by companies trying to reproduce their poor data.

    1. Hap says:

      If something was that important, you’d figure that they’d actually try to keep it. Academia isn’t great for reagent maintenance, but if something’s really important, you would try to keep it safe (and have redundant storage), wouldn’t you?

      “We can’t find our stuff” seems to be an awfully prevalent line in bad or dishonest research. Apparently either all the times that teachers dissed “The dog ate my homework (or cell line)” excuse, they were wrong, or grant agencies should be hiring K-12 teachers during the summer to review grants and perform discipline.

    2. Imaging guy says:

      To MrXYZ
      Thanks. I used to think very highly of that paper and it is in my EndNote. Now I think I have to either delete or tag as non-reproducible. This group from DUKE has published a few crystallography papers in Nature and Science. I wonder whether they are also reproducible. By the way I notice that many crystallography papers get published in Nature and Science. Do people try to reproduce these findings or they are just one-off affairs?

  7. Foodscientist says:

    My mom once had a patient who couldn’t feel pain. She was driving while sitting on one foot. Her seat heater malfunctioned and pretty much cooked her foot.

  8. steve says:

    I like the comment about living in Boston. Hopefully one day you can post about the new adoptive cell immunotherapy that shows so much promise – CAH-T.

  9. Simona says:

    Given the association between weed and psychosis I wish they had given her a psychotic symptom questionnaire too.

  10. Anonymous says:

    Closely linked to the mutant gene found in the 1960s – FAAH-OUT-MAN. That phenotype was long hair, tie-dye clothing and increased cannabis consumption.

  11. Simon Auclair says:

    Simona, do you have references for that?

Leave a Reply

Your email address will not be published. Required fields are marked *

Time limit is exhausted. Please reload CAPTCHA.