By David N. Leff

When a surgeon prepares to amputate a gangrenous leg, the patient receives a general anesthetic. Under its soporific effects, he or she is oblivious to the acute pain of the procedure.

But afterwards, throughout life, the lost limb sometimes seems to hurt. It's as if the nerves and brain "remember" that agonizing insult to the body.

Those attacks of phantom pain can often be averted if the amputation proceeds under local as well as general anesthesia.

"A general anesthetic alone," explained neuroscientist Allan Basbaum, of the University of California, San Francisco (UCSF), doesn't shut off nerve-delivered pain messages to the spinal cord. So all those bad circuits will be activated by the surgeon."

He continued: "If the amputation is done under a combination of general and local anesthesia to prevent the spinal cord from getting the message, this can actually eliminate surgically induced phantom pain."

Basbaum's paper in today's Nature, dated April 17, 1997, elucidates circuity that might explain this preemptive-anesthesia phenomenon. More broadly, he told BioWorld Today, it explicates "the circuitry and neurochemistry through which injuries to the body change the spinal cord. The long-term changes that occur," Basbaum observed, "are largely maladaptive, and could lead to persistent pain states."

His paper in Nature bears the cryptic title: "NMDA-receptor regulation of substance P release from primary afferent nociceptors." Decoded, it means that substance P (for pain — a neurotransmitter of messages from pain-specialized nerve cells) is unleashed by NMDA, a receptor for the glutamate neurotransmitter. It triggers long-term changes in the spinal cord's pain message center.

Basbaum's laboratory at UCSF focuses on the mechanisms that produce and control pain.

Receptor Feedback Worsens Pain Message

"The most provocative and novel aspect of our study," he said, "was that the pain fiber that transmits information from the site of peripheral injury to the cord makes the NMDA receptor. Basically, the significance of this finding is that there appears to be a positive feedback system which makes pain worse by acting back on it, causing more output from the pain fibers of the neural circuit."

Basbaum continued: "From a clinical perspective, in terms of pharmacology, it's not earth-shattering, because everyone said: 'Fine, if you can come up with a good NMDA receptor antagonist, you're going to have a good pain-relieving drug.'"

But, he pointed out: "Although it's true that there are some very weak NMDA receptor antagonists available, the really strong ones have major side effects and are not clinically available. Their side-effect profile is just not adequate -- and that's an understatement.

"Knowing, as we now do," Basbaum went on, "that the pain fiber feeding upstream to the spinal cord makes the NMDA receptor, suggests a new pharmacological target. It's the same receptor, but in a new place where we didn't know it existed. Eliminating it from that place would be very beneficial."

He proposed: "Because that critical fiber can be accessed peripherally, rather than by way of the cord, it may be possible, if imaginative, to find ways of getting an antisense drug into that fiber."

Basbaum posited a hypothetical scenario: "Let's say you have a pain originating in the knee joint. You would find a way to get the antisense molecule into the axons, the nerve fibers that innervate the knee. Then it gets taken back into the neuronal cell bodies of those axons and blocks it from making its NMDA receptor.

"Targeted drug delivery," he commented, "is one of the best ways to avoid side effects."

Recently, Basbaum and his co-workers have been focusing on substance P itself. This 11-amino-acid peptide is a major neurotransmitter of the unmyelinated axons that carry the pain message to the spinal cord.

"Now we're looking not so much at the NMDA," he said, "but downstream, to find which neurons have the substance P receptor and under what circumstances P interacts with it.

"Now," he observed, "we're talking real pain."

The team inflicted a measured dollop of acute pain on experimental rats, under anesthesia, by pinching their hind paws to simulate arthritis pain. Within seconds, the animals evinced a non-vocal pain reaction consisting of biting, scratching and licking themselves in a head-to-tail direction.

That NMDA stimulus, which mimics an actual acute bodily injury, promptly activated substance P, spreading it into many parts of the cord where normally it doesn't respond.

Injection of the neurotoxin capsaicin (the active ingredient in hot red peppers) selectively wiped out the substance P-containing nerve fibers and relieved the rats' pain. It also largely reversed the spinal cord's pain-circuit alterations. And repeat stimuli to an animal with an already-inflammed hind paw drove P to even wider distribution.

Thus sensitized, the rats reacted with pain even to a non-painful stimulus, such as gentle stroking with a soft brush. "Substance P," Basbaum pointed out, "is remarkably sensitive to mechanical stimuli. This is interesting, because clinically a big problem that pain patients have is tenderness."

Two Receptors Better Than One

"I think it's important," he added, "that a substance-P antagonist alone may not do anything therapeutic. It may take a synergistic combination between the P and NMDA receptors. And that's not an additive synergy, but probably multiplicative. You're really taking a little bit of input and magnifying it."

Comparing acute and persistent pain, Basbaum questioned the latter's evolutionary survival value:

"Acute pain certainly has that value," he observed. "It tells you have an injury. People who can't feel pain are really in big trouble. But what is the value," he asked, "of persistent pain? Once you know that you have cancer, why do you need to feel pain all the time?

"A few people in the field," Basbaum concluded, "talk about persistent pain as a 'memory.' But it's a bad memory. There must be something useful for it, but I honestly don't know what it is." *