QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Simmons, D. L. Articles by Tomsik, J. Search for Related Content PubMed PubMed Citation Articles by Simmons, D. L. Articles by Tomsik, J.

LETTERS TO THE EDITOR

Comments on "Acetaminophen and the Cyclooxygenase-3 Puzzle: Sorting out Facts, Fictions, and Uncertainties"

Brigham Young University, Provo, Utah

Received August 11, 2005; accepted September 19, 2005.

In the review by Kis et al. (2005), the authors state that their aim was "to point out critical theoretical and methodological limitations of the COX-3 studies that led several investigators to scientifically questionable conclusions." As one laboratory whose work was addressed, we note that, in pursuing their goal, the authors make inaccurate statements regarding our study (Chandrasekharan, 2002) that could misinform readers.

Kis et al. (2005) imply that our Western blots, demonstrating the synthesis of canine COX-3 in insect cells [see Fig. 3 from Chandrasekharan et al. (2002)], were faultily interpreted. To do this, they misstate that our anti-COX-3 antibody was made against only the human oligopeptide corresponding to the first 12 amino acids of the predicted human COX-3. In reality, our antibody, as clearly described in our study, was raised against a mixture of human and mouse peptides corresponding to the first 12 amino acids encoded by exon-1 and the retained intron-1 in each species. These were mixed together in a 50:50 ratio and administered to rabbits. Kis et al. (2005) note that only the first seven amino acids of these peptides are highly conserved with dogs. We state this clearly in our study and define the sequence that is conserved (see page 13928, third paragraph). However, unsubstantiated by any data, Kis et al. (2005) proclaim that "it is very unlikely that the antibody that was produced against the N-terminal part of the human COX-3 sequence will recognize the canine COX-3." They further state that "Critical analysis of [our] results is difficult because [we] published only cropped parts of the Western blots of proteins from insect cells without corresponding molecular weight markers, and these blots show several putative specific and nonspecific bands." And finally, they state that "the expression level of the canine COX-3 construct in Sf9 cells is uncertain because the authors used anti-human COX-3 antibody to detect the canine variant." The specific purpose of the Western blots referred to by Kis et al. (2005) [see Fig. 3 from Chandrasekharan et al. (2002)] was to demonstrate the effect of tunicamycin on COX-1, COX-3, and PCOX-1a glycosylation. Because of space limitations, and to enlarge the region of COX proteins for better viewing of glycosylated forms, the three Western blots in Fig. 3 from our study were indeed cropped. However, they accurately show the abundant COX-1, COX-3, and PCOX-1a proteins and their response to tunicamycin. The statements of Kis et al. (2005) regarding the specificity of our antibody toward canine COX-3 and PCOX-1a are based purely on speculation and lack foundation in fact. To allay concerns raised by Kis et al. (2005) that COX-1, PCOX-1a, or COX-3 were not highly overexpressed in the described experiments or specifically detected by our anti-COX-1 monoclonal or anti-COX-3 antisera, the complete blots are shown herein as Fig. 1. This is one of many blots that assessed expression of COX-3 and PCOX-1a in these cells.

View larger version (67K): [in this window] [in a new window]   Fig. 1. Immunoblots of insect Sf9 cells overexpressing canine COX-3, PCOX-1a, and murine COX-1. Blots were probed with anti-COX-3 polyclonal sera (A) and COX-1 monoclonal antibody (B). Uninfected cells or cells infected with wild-type baculovirus are denoted by C and WT, respectively. Cells were treated with tunicamycin (T+) or without (T–). Molecular weight markers (in kilodaltons) are denoted by M.

These assertions are not correct. We state in our study (page 13928, third paragraph) the results of our PCR analyses of our insect cell studies that concluded that our cells were infected and properly expressing COX-3 and PCOX-1a with intron-1 retained. Moreover, these studies did not use "transfection" of constructs, as asserted by Kis et al. (2005), but infection with recombinant baculovirus, as is clearly described in our study. Upon infection with the respective recombinant baculoviruses, these cells initiate massive synthesis of recombinant COX-1, COX-3, and PCOX-1a at amounts visible on Coomassie Blue-stained gels and clearly detected by our specific antibodies [see Fig. 3 from Chandrasekharan et al. (2002) or Fig. 1 herein]. These levels are orders of magnitude higher than achievable by transfection. For these reasons, this system has been used by the pharmaceutical industry in its analyses of COX-1 and COX-2 inhibition by NSAIDs. In addition, infected cells are cytologically discernible from nontransfected cells.

We, of course, have done numerous mock infections of insect cells with wild-type baculovirus (see Fig. 1 herein, for example), which exhibit negligible prostaglandin synthesis. We did not include these data in this experiment because, as Fig. 3 from Chandrasekharan et al. (2002) clearly demonstrates, expression of PCOX-1a (an excellent control for the effects of viral infection) in insect cells produces negligible COX activity, as does expression of COX-1 or COX-3 in the presence of tunicamycin. As noted in our study, tunicamycin has been demonstrated by others to block N-linked glycosylation of COX-1, resulting in a loss of enzyme activity—a result we duplicated in this experiment. Finally, Fig. 5 and Table 1 from our study demonstrate that the prostaglandin synthesis activity in Fig. 3 is clearly inhibited by an array of known COX inhibitors, including aspirin, demonstrating that it is a cyclooxygenase.

In our study, we report a 65-kDa protein in human aorta detected by our anti-COX-3 sera that we "postulate[d] is COX-3" (page 13931, second paragraph). However, we also noted that this protein was smaller than predicted from the human COX-3 DNA sequence, "suggesting that hypoglycosylation or other differences existed between the 65-kDa protein and COX-1" (page 13931, paragraph 1). We did not, as Kis et al. (2005) claim, explain this protein with a "hypothesis that a ribosomal frameshift correct[ed] the sequence into the regular open reading frame of the COX-1, leading to the synthesis of a peptide that is similar to COX-1 except that it is 31 amino acids longer in length."

Instead, in the paragraph before that describing the 65-kDa protein, we address the clear problem associated with the presence of a frameshift in human intron-1. We state: "However, COX-3 in human will require further experimentation because some of the published sequences differ by one nucleotide in intron-1 and hence are out of frame. These may constitute genuine polymorphisms or sequencing errors. Alternatively, intron-1 may be out of frame in humans, requiring other mechanisms such as ribosomal frame shifting to produce a functional COX-3 protein" (emphasis added).

Snipes et al. (2005) have recently done studies on COX-3 in rodents where a frameshift in intron-1 is also present. Their findings strongly affirm in rodents the most important finding we reported for canine COX-3 and PCOX-1a [see Fig. 3 from Chandrasekharan et al. (2002)]. That is, COX-1 mRNA retaining intron-1 escapes the nucleus and is translated. This is the most important property needed for cells to synthesize canine COX-3 and PCOX-1a, as we reported.

The facts are that our study accurately reports the cloning, structure, and characterization of two previously unknown COX-1 variants retaining intron-1 that are expressed at high-enough levels in canine brain to be readily detected by Northern blots. Overexpression of these intron-1 containing variants in insect cells produces proteins that retain their signal peptides. Like COX-1 and COX-2, these proteins are translocated into the lumen of the endoplasmic reticulum where they are N-glycosylated. Both COX-3 and PCOX-1a are specifically detected by our anti-COX-3 antibody. One of the two variants, COX-3, possesses cyclooxygenase activity, and, to date, is the only known COX splice variant to exhibit this property. Acetaminophen, aspirin, and competitively acting NSAIDs were studied with regard to their inhibitory action on canine COX-3 and murine COX-1 and COX-2. These experiments yielded the results shown in our publication and are an accurate representation of the activity of these enzymes and their response to the drugs tested. Our anti-COX-3 antibody detected low-abundance proteins in human aorta that were competed with cognate peptides as described.

Footnotes

doi:10.1124/jpet.105.094169.

ABBREVIATIONS: COX, cyclooxygenase; kDa, kilodalton(s); PCOX, partial cyclooxygenase; PCR, polymerase chain reaction; PGE₂, prostaglandin E₂.

References

Chandrasekharan NV, Dai H, Roos KLT, Evanson NK, Tomsik J, Elton TS, and Simmons DL (2002) COX-3, a cyclooxygenase-1 variant inhibited by acetaminophen and other analgesic/antipyretic drugs: cloning, structure and expression. Proc Natl Acad Sci USA 99: 13926–13931.[Abstract/Free Full Text]

Chandrasekharan NV and Simmons DL (2004) The cyclooxgenases. Genome Biol 5: 241.[CrossRef][Medline]

Kis B, Snipes JA, and Busija DW (2005) Acetaminophen and the COX-3 puzzle: sorting out facts, fictions and uncertainties. J Pharmacol Exp Ther 315: 1–7.[Abstract/Free Full Text]

Simmons DL (2003) Variants of cyclooxygenase-1 and their roles in medicine. Thromb Res 110: 265–268.[CrossRef][Medline]

Simmons DL, Botting RM, and Hla T (2004) Cyclooxygenase isozymes: the biology of prostaglandin synthesis and inhibition. Pharmacol Rev 56: 387–437.[Abstract/Free Full Text]

Snipes JA, Kis B, Shelness GS, Hewett JA, and Busija DW (2005) Cloning and characterization of cyclooxgenase-1b (putative cyclooxygenase-3) in rat. J Pharmacol Exp Ther 313: 668–676.[Abstract/Free Full Text]

This article has been cited by other articles:

				E. M. Weinheimer, B. Jemiolo, C. C. Carroll, M. P. Harber, J. M. Haus, N. A. Burd, J. K. LeMoine, S. W. Trappe, and T. A. Trappe Resistance exercise and cyclooxygenase (COX) expression in human skeletal muscle: implications for COX-inhibiting drugs and protein synthesis Am J Physiol Regulatory Integrative Comp Physiol, June 1, 2007; 292(6): R2241 - R2248. [Abstract] [Full Text] [PDF]

				B. Kis, J. A. Snipes, and D. W. Busija Response to Comments on "Acetaminophen and the Cyclooxygenase-3 Puzzle: Sorting out Facts, Fictions, and Uncertainties" J. Pharmacol. Exp. Ther., December 1, 2005; 315(3): 1415 - 1416. [Full Text] [PDF]

This Article Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Simmons, D. L. Articles by Tomsik, J. Search for Related Content PubMed PubMed Citation Articles by Simmons, D. L. Articles by Tomsik, J.