The woman who cannot feel pain, and the gene behind it

Researchers at University College London identified the mutation in a Scottish woman who has never needed a painkiller, in a gene now called FAAH-OUT.

A hand operation that should have hurt

In the case report that opened this story, a 66-year-old woman presented to Raigmore Hospital in Inverness, Scotland, for a trapeziectomy with ligament reconstruction and tendon interposition, an operation on the base of the thumb that is known to be painful afterwards. Her thumb joint was severely degenerated by osteoarthritis, and she described it as painless. She received a general anaesthetic and an axillary nerve block. After the operation her pain score was 0 out of 10 until discharge the next day. The only analgesic she received in hospital was a single gram of intravenous paracetamol in the recovery room. When the nerve block wore off, she still asked for nothing.

The anaesthetist, Devjit Srivastava, noticed that she showed no pain from pinching or from manipulation of her intravenous cannula, and began asking questions. At 65 she had undergone a hip replacement for a joint so degenerated that the lack of pain was itself inconsistent with the X-rays; she had taken paracetamol on the first two postoperative days because staff encouraged her to, and scored 0 out of 10 throughout except for one score of 1. A single oral dose of morphine had caused two days of vomiting. She had had varicose vein operations and dental procedures without analgesia. She reported a lifetime of cuts and burns that she often noticed only when she smelled her own skin burning, and that healed quickly with little scarring. She ate Scotch bonnet chilli peppers and described a short, pleasant glow.

She was referred to James Cox and the pain genetics group at University College London, with collaborators at Oxford. UCL's account of the case gives the referral year as 2013, and gives her name as Jo Cameron.

An enzyme that clears a natural cannabinoid

The gene the geneticists eventually reached was already well known. Fatty acid amide hydrolase, FAAH, is the enzyme that breaks down anandamide, a lipid signalling molecule that acts on the same CB1 receptor as the active constituent of cannabis. In 2001 Benjamin Cravatt and colleagues reported in PNAS that mice engineered to lack FAAH had brain anandamide levels 15 times higher than normal, showed reduced pain sensation, and lost that reduced sensation when given the CB1 blocker SR141716A. The paper proposed FAAH as a drug target for pain.

The idea was tested and it failed. In 2012 the irreversible FAAH inhibitor PF-04457845 was given at 4 milligrams daily to patients with knee osteoarthritis in a randomised, placebo- and naproxen-controlled crossover trial of 74 patients. The drug reduced FAAH activity by more than 96 percent and raised four of the enzyme's substrates substantially, but the difference from placebo on the WOMAC pain score was 0.04 points, against 1.13 points for naproxen. The trial was stopped at interim analysis for futility. In 2016 a report in the New England Journal of Medicine described a phase 1 study of another FAAH inhibitor, BIA 10-2474, in which the drug produced an acute neurological syndrome in three of four volunteers who consented to publication, starting on the fifth day at the 50 milligram dose. One volunteer became brain dead. The mechanism was never established. By the time the Scottish patient reached the laboratory, FAAH was a target with a failed efficacy trial and a fatality against it.

What her genome showed

Exome sequencing of the patient, her mother and her two children produced a short list of candidate mutations, none of which explained the phenotype. A genome-wide scan for copy number changes did. It found a heterozygous microdeletion of about 8 kilobases on chromosome 1, beginning about 4.7 kilobases downstream of the 3' end of FAAH. The deleted region did not contain FAAH itself. It contained a stretch of DNA annotated as a pseudogene, a presumed non-functional copy.

Cloning experiments showed otherwise. The region produced a transcript of 2.845 kilobases, expressed in dorsal root ganglia and brain among other tissues, which the authors named FAAH-OUT. The patient had also inherited a common variant in FAAH itself, C385A (rs324420), which reduces the enzyme's expression and activity and whose A allele has a population frequency of 26 percent. The combination, a hypomorphic FAAH on one chromosome and the deletion on the other, was proposed as the cause.

Blood measurements, performed at the University of Calgary, supported the proposal. Anandamide averaged 2.0 picomoles per millilitre in the patient against 1.2 in controls; palmitoylethanolamide 113.1 against 43.4; oleoylethanolamine 17.3 against 5.1. All three are substrates of FAAH. 2-Arachidonoylglycerol, which is degraded by a different enzyme, was essentially unchanged at 45 against 42.2. The pattern was what loss of FAAH function would predict.

The findings were published by Abdella Habib, Srivastava, Cox and colleagues in the British Journal of Anaesthesia in 2019.

Pain was not the only thing missing

On the Generalised Anxiety Disorder-7 questionnaire, taken at age 70, the patient scored 0 out of 21. On the Patient Health Questionnaire-9 for depression, she scored 0. She reported never panicking, including during a recent road traffic accident. She also reported lifelong lapses of memory, forgetting words mid-sentence and misplacing keys. Quantitative sensory testing showed reduced sensitivity to noxious heat in both hands and feet.

Each of these traits had a precedent in the FAAH knockout mouse: reduced pain, reduced anxiety-like behaviour, enhanced extinction of fear memories, accelerated skin wound healing, and deficits in short-term memory. A single human had assembled the same list from a single genetic lesion.

How a deletion in junk DNA switches off a gene

The 2019 paper established the association. A 2023 paper in Brain by Hajar Mikaeili, Andrei Okorokov, Cox and colleagues established the mechanism. Using CRISPR-Cas9 to recreate the 8 kilobase deletion in human cells, they showed that removing the region, which contains the first two exons and the promoter of FAAH-OUT, reduced the expression of FAAH. Two mechanisms were at work. The FAAH-OUT transcript, a long non-coding RNA concentrated in the nucleus, normally keeps the FAAH promoter free of DNA methylation; when its transcription is lost, the enzyme DNMT1 methylates the promoter and FAAH is silenced. The deleted region also contains a conserved enhancer element, named FAAH-AMP, that drives FAAH expression directly.

Transcriptome analysis of fibroblasts from the patient found 797 genes turned up and 348 turned down relative to controls. Among them were WNT16, which has been linked to bone regeneration and sits in a pathway involved in wound healing; BDNF, linked to mood regulation; and ACKR3, which regulates opioid peptide levels. The paper offered these as a coherent molecular basis for the three-part phenotype of painlessness, low anxiety and fast healing.

What the story does and does not show

The story rests on one patient. The 2019 report is a case study with four controls for the blood measurements, and the 2023 mechanism was worked out in cell lines and in fibroblasts from the same patient. There is no second person with the FAAH-OUT deletion in the published record, so there is no way yet to know how much of her phenotype is the deletion, how much the common FAAH variant, and how much other genetic background. Her fast wound healing and her memory lapses are self-reported, not measured. The big number on this story's card, zero painkillers, is an approximation: she received paracetamol when it was offered after two operations and a single dose of morphine that she did not tolerate, but she never requested analgesia and her pain scores were zero.

The story does not show that a drug can reproduce her condition. The most direct attempt, pharmacological inhibition of FAAH, did not relieve osteoarthritis pain in a controlled trial and, in a different compound, caused a death. The authors' own proposal is that targeting FAAH-OUT rather than the enzyme might behave differently, and that remains a hypothesis.

What the story does show is that a region of the genome dismissed as a pseudogene regulates a known pain gene by two separate mechanisms, that one woman carries a deletion in it, and that she has lived into her seventies without pain, without anxiety and with wounds that close quickly. The target existed before she was found. She is the evidence that it is worth another look.

No pain, no anxiety, fast healing, from one gene.