COX-3: the Acetaminophen Target Finally Revealed

It has been known for years that non-steroidal anti-inflammatory drugs (NSAIDs) such as aspirin, ibuprofen, and acetaminophen, provide relief from fever, pain, and inflammation through their actions on cyclooxygenase (COX) enzymes.1 Two COX isozymes, COX-1 and -2, were first identified in the early 1990’s as the catalysts for an important step in prostaglandin biosynthesis.2 Although both enzymes have similar functions, their temporal and spatial expression patterns are very different.3 COX-1 is constitutively expressed in many somatic cell types and is considered a “housekeeping” enzyme with roles in such processes as vascular hemostasis and gastroprotection.4 In contrast, COX-2 expression is primarily induced by factors such as endotoxins, cytokines, and growth factors.5 COX-2 is expressed at sites of inflammation and produces prostaglandins that mediate inflammatory and pain sensation responses.6 COX involvement in inflammation, pain, and a variety of diseases has inspired researchers to investigate the actions of NSAIDs on these enzymes. Although many advances have been made over the last 10 years in understanding the pain relief and anti-inflammatory mechanisms of aspirin, ibuprofen, and the new COX-2 inhibitors, the mechanism of acetaminophen action has remained elusive.7,8

Finally, identification of a new isozyme, COX-3, suggests that it is the target for acetaminophen.9 COX-3 was discovered by Northern analysis of canine cerebral cortex RNA using a COX-1 cDNA probe. The COX-1 probe unexpectedly illuminated a band at 2.6 kb, labeling a transcript later confirmed to be COX-3, an alternate splice variant of COX-1 in which intron 1 is retained (Figure 1). Interestingly, intron 1 is not only present in canine, human, and murine versions of COX-3, but it is conserved in length and sequence in these species as well. While COX-3 retains all of the important catalytic and structural features of COX-1 and -2, it is likely that intron 1 is responsible for the deviant enzymatic properties of COX-3 perhaps via subtle alterations in structure, glycosylation state, and/or expression.9

Figure 1. Structure of COX-1, -2, and -3 genes.
[Adapted from Chandrasekharan, N.V. et al. (2002) Proc. Natl. Acad. Sci. USA 99:13926.]

Thus far, little is known about the temporal regulation of COX-3 expression. However, it has been known for decades that acetaminophen inhibits COX activity in canine brain homogenates more than in spleen homogenates.8 This suggests a much more strict spatial regulation of COX-3 than either of the other COX enzymes. Northern analysis has since shown that COX-3 is found in greatest abundance in the canine and human cerebral cortex.9 COX-3 activity appears to be selectively inhibited by acetaminophen as well as a few other analgesic and antipyretic NSAIDs.7 Further, unlike other NSAIDs, acetaminophen is capable of crossing the blood-brain barrier allowing it to reach concentrations in the brain sufficient to inhibit COX-3. All of these lines of evidence strongly implicate COX-3 as the target of acetaminophen action and partly explain the long-standing mystery of why it is often more efficacious against headache and fever than some of the other NSAIDs.9 Further investigation in this area will lead not only to a better understanding of the role of COX enzymes in pain, fever, inflammation, and disease but also to more specific and efficient treatment of ailments involving these enzymes.


  1. Smith, W.L. et al. (2000) Ann. Rev. Biochem. 69:145.
  2. Hinz, B. & K. Brune (2002) J. Pharmacol. Exp. Ther.300:367.
  3. Kam, P.C.A. & A.U-L. See (2000) Anaesthesia 55:442.
  4. Katori, M. & M. Majima (2000) Inflamm. Res. 49:367.
  5. Kulkarni, S.K. et al. (2000) Methods Find. Exp. Clin.Pharmacol. 22:291.
  6. Urban, M.K. (2000) Orthopedics 23:S761.
  7. Botting, R.M. (2000) Clin. Infect. Dis. 31:S202.
  8. Flower, R.J. & J.R. Vane (1972) Nature 240:410.
  9. Chandrasekharan, N.V. et al. (2002) Proc. Natl. Acad.Sci. USA 99:13926.