Association of COX-2 gene haplotypes with prostaglandins production in bronchial asthma

Article Outline

• References
• Copyright

To the Editor:

Common single nucleotide polymorphisms (SNPs) are linked within a gene locus and are inherited as several gene variants of different functional properties. The haplotype analysis emerged from genomic studies as an important clinical tool. In the lung, inflammation leads to induction of COX-2 gene and to enhanced production of both proinflammatory and anti-inflammatory eicosanoids. Recently, we reported on a possible functional importance of the promoter polymorphism of COX-2 G₋₇₆₅C in subjects with asthma.¹ Peripheral blood monocytes' capacity for biosynthesis of prostaglandin (PG) E₂ and PGD₂ was increased 10-fold in CC homozygotes compared with GG ones. Similar genotype frequencies of COX-2 G₋₇₆₅C polymorphism were subsequently found in a large Australian study²; however, the authors did not investigate sex associations with asthma. Because COX-2 gene has numerous single nucleotide polymorphisms, and our functional findings of G₋₇₆₅C were in disagreement with previous work by Papafili et al,³ we wondered whether particular haplotypes of COX-2 could better correlate with prostaglandins' biosynthetic capacity in subjects with asthma. Out of several SNP candidates, a T→C transition within the 3′-untranslated region (COX2.8473) seemed particularly interesting. This SNP was found to be associated with increased risk for nonsmall cell lung cancer.⁴ As an explanatory mechanism, the authors proposed stabilization of the mRNA molecule due to disruption by T→C transition of the adenine-uracil-rich motif, conferring a signal for COX-2 transcript degradation.

Subjects with asthma, all women, were studied for capacity of PGE₂ and PGD₂ biosynthesis in peripheral blood monocytes by using a haplotype-based approach. The methods were described previously.¹ In addition to ₋₇₆₅CC homozygotes (n=8) and ₋₇₆₅GG homozygotes (n=9), the peripheral blood monocyte cultures were completed in 8 women with asthma heterozygous for this locus (Table I). Genomic DNA was genotyped for ₋₇₆₅G→C polymorphism as described previously¹ and for COX2.8473 SNP using a similar PCR-RFLP method. Briefly, a 177-bp PCR product was amplified with primers: 5′-GAAATTTTAAAGTACTTTTGAT and 5′-CTTTTACAGGTGATTCTACCC. A mutagenic primer (underlined nucleotide) introduced a restriction site for BclI restriction nuclease; thus, C allele was cut to 156 and 21 bp. Combined genotypes were also ascertained in a sample of 76 individuals without asthma. Haplotype frequencies were calculated by a maximum likelihood method using a Markov chain of 10,000 iterations with 3 bootstrap replicates,⁵ and the number of copies of C-C haplotype was selected as an independent variable for ANOVA tests.

Table I. Clinical characteristics of studied subjects with asthma^∗

In subjects with asthma, the C-C haplotype (locus order G₋₇₆₅C – COX2.8473T→C) had the highest frequency, 41.7%, followed by G-T (33.7%), G-C (18.3%), and C-T (6.3%) because of preselection of ₋₇₆₅CC homozygous subjects. However, when corrected for this selection bias, haplotypes did not differ from the ones in the population sample (C-C, 12.0% vs 12.3%; G-T, 55.9% vs 64.9%; G-C, 30.3% vs 21.3%; and C-T, 1.8% vs 1.5%). The 2 COX-2 loci were in linkage disequilibrium (P < .0001) in controls, suggesting a tight linkage of COX2.8473 T→C SNP with ₋₇₆₅C allele. Linkage disequilibrium between the loci was also significant in the subjects with asthma (P=.0005). Among subjects with asthma, 7 individuals had 2 copies of the C-C haplotype, 7 had 1 copy of C-C, and 11 had other haplotypes.

The mean concentrations of PGs in monocyte cultures and post hoc comparisons of ANOVA for diplotype classes are presented in Table II. ANOVA on log-transformed PG data was highly significant (P < .001) for the number of C-C haplotype copies as an independent variable. Comparison of PG levels in cases with the most contrasting wild-type G-T/G-T (n=4) and variant C-C/C-C genotypes (n=7) revealed a 19.8-fold increase of PGE₂ (P < .001) and a 7.4-fold increase of PGD₂ (P=.024) in unstimulated monocytes with the variant genotype. In LPS-stimulated cells, these differences were respectively 19.6-fold and 11.5-fold (P=.001).

Table II. COX-2 haplotypes and prostaglandins concentrations in monocyte cultures^§

Compound genotype	PGE2	PGD2	PGE2-LPS	PGD2-LPS
C-C/C-C (n=7)	0.79 ± 0.19‡/‡	0.048 ± 0.03†/NS	1.68 ± 0.25†/†	0.113 ± 0.03‡/NS
C-C/other (n=7)	0.24 ± 0.03NS	0.024 ± 0.04∗	1.05 ± 0.30†	0.073 ± 0.03†
Other/other (n=11)	0.07 ± 0.03	0.05 ± 0.03	0.19 ± 0.09	0.013 ± 0.0

There was uniformly strong linear correlation between the number of C-C haplotype copies and capacity for biosynthesis of PGE₂ (log-transformed data, Pearson r=0.80; P < .001) and PGD₂ (r=0.69; P=.002). Stimulation of monocytes with LPS resulted in a very similar correlation for PGE₂ (r=0.80; P < .001) and PGD₂ (r=0.78; P < .001). The C-C haplotype had no effect on magnitude of PG induction by LPS.

We analyzed haplotypes of 2 functional SNP candidates within COX-2 gene. One of the loci— G₋₇₆₅C —was previously studied by us in opposite homozygotes with asthma. We now expanded this group, adding 8 subjects heterozygous for this SNP. In these heterozygotes, PG biosynthetic capacity of monocytes was in between CC and GG homozygotes (data not shown). Haplotypes of the 2 SNPs were reconstructed, and PG levels in monocyte cultures were reanalyzed using arbitrary classes of compound genotype. The haplotype variant C-C was expected¹, ⁴ to associate with the highest activity of COX-2. Indeed, there was a very strong linear correlation between the number of C-C haplotype copies and PG levels in monocyte cultures. Because our subjects with asthma were selected for functional studies of COX-2 polymorphisms by G₋₇₆₅C genotypes, estimated haplotype frequencies are not representative for a random sample of subjects. The haplotypes of COX-2 were in linkage disequilibrium both in a group of controls without asthma and in subjects with asthma. The difference in PG production by peripheral blood monocytes of subjects with asthma with extreme compound genotypes G-T/G-T and C-C/C-C was striking. Our data suggest an additive effect of C-C COX-2 haplotype on gain of function, because the difference between the extreme compound genotypes (C-C vs G-T) in PGE₂ levels was by 65% greater than the one between opposite homozygotes ₋₇₆₅CC and ₋₇₆₅GG.¹ The presented haplotype approach was validated by several pharmacogenomic studies⁶ and identified the COX-2 variant related to overproduction of prostaglandins. No complete study on the interaction between the 3′ untranslated region COX2.8473T→C polymorphism and promoter G₋₆₇₅C variants was possible because of a low frequency of some haplotypes. However, altered promoter activity of COX-2 gene related to G₋₇₆₅C SNP seemed to be predominant but additive with stabilization of the mRNA transcript as a result of COX2.8473 T→C. Because COX-2 activity plays a pivotal role in inflammation, this overproducing genetic C-C variant of the gene seems worthy of study in several human pathologies that, apart from asthma, are mediated by the inducible activity of the isoenzyme.

Back to Article Outline

References 

1. Szczeklik W, Sanak M, Szczeklik A. Functional effects and gender association of COX-2 polymorphism G_-765C in bronchial asthma. J Allergy Clin Immunol. 2004;114:248–253
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (124 KB)
   • CrossRef
2. Shi J, Misso NL, Duffy DL, Thompson PJ, Kedda M-A. A functional polymorphism in the promoter region of the cyclooxygenase-2 gene is not associated with asthma and atopy in an Australian population. Clin Exp Allergy. 2004;34:1714–1718
   • View In Article
   • MEDLINE
   • CrossRef
3. Papafili A, Hill MR, Brull DJ, McAnulty RJ, Marshall RP, Humphries SE, et al. Common promoter variant in cyclooxygenase-2 represses gene expression: evidence of role in acute-phase inflammatory response. Arterioscler Thromb Vasc Biol. 2002;22:1631–1636
   • View In Article
   • CrossRef
4. Campa D, Zienolddiny S, Maggini V, Skaug V, Haugen A, Canzian F. Association of a common polymorphism in the cyclooxygenase-2 gene with risk of non-small cell lung cancer. Carcinogenesis. 2004;25:229–235
   • View In Article
   • MEDLINE
   • CrossRef
5. Schneider S, Roessli D, Excoffier L. Arlequin Ver. 2.00: A Software for Population Genetics Data Analysis. 2002. Available at: http://anthropologie.unige.ch/arlequin/. Accessed December 15, 2004.
   • View In Article
6. Ahmadi KR, Weale ME, Xue ZY, Soranzo N, Yarnall DP, Briley JD, et al. A single nucleotide polymorphism tagging set for human drug metabolism and transport. Nat Genet. 2005;37:84–89
   • View In Article
   • MEDLINE