Cyclooxygenase-2 expression and PGE₂ concentration in breast tumours of mice after 13 days of treatment (or 14 days after tumour cell injection) with morphine and co-administration with celecoxib. (A) Western blot showing upregulation of COX-2 protein in the region of 72–74 kDa, whereas β-actin is shown as a loading control. The arbitrary densitometric units above the bands show the ratio of COX-2 to β-actin. (B) RT-PCR, showing an upregulation in COX-2 gene expression in the expected region of 583 bp. Bands on the bottom show the equal loading of RNA represented by GAPDH. Arbitrary numeric units show the ratio of COX-2 to GAPDH expression. (C) Prostaglandin E₂ concentration is significantly increased in tumours of mice treated with morphine. Data are shown as mean±s.d. n=3–5 experiments per observation/data shown. (D) Immunofluorescent staining of tumour cryosections using anti-CD31 PE (red), an endothelial cell marker and anti-COX-2 followed by a 2° Ab conjugated with FITC (green). The overlaid image in the right hand panel shows yellow staining due to increased COX-2 in the endothelium in morphine-treated mice. Larger number of non-endothelial cells show COX-2 expression (green) in middle and overlaid panel after morphine treatment. Magnification, × 600.

Effect of morphine and co-administration with celecoxib on tumour angiogenesis and tumour growth, after 13 days of treatment. (A). Tumour cryosections were stained with endothelial cell-specific anti-CD31 PE (red) and nuclear stain, DAPI (blue). The red and blue images of the same area are overlaid to show the tumour vasculature in red and cellular presence represented by blue nuclei. Note the dense, heavily branched and wide blood vessels distributed uniformly in tumours from morphine group. Celecoxib-treated mouse tumours exhibit fewer and collapsing blood vessels and areas with week CD31 staining suggesting degenerating vasculature. This week CD31 staining is accompanied by bald areas without any blue nuclei. However, celecoxib given with morphine did not result in degenerating looking vessels, even though the blood vessels were fewer, less dilated, smaller and accompanied by bald areas devoid of blue nuclei compared to morphine-alone-treated mice. Magnification, × 150. Each figure represents nine sections from three different mouse tumours. (B) Quantitative expression of vasculature in tumour sections is shown as PE-positive pixels (representing red staining for blood vessels), number of nodes (equivalent to branch points), vessel ends (for the number of vessels) and line length (vessel length). Each point is the mean±s.d. of nine observations from three different mouse tumours. (C) Weight of tumours from mice that survived until day 14. Note, morphine treatment increased tumour growth significantly, whereas celecoxib inhibited morphine-induced increase in tumour weight. n=6–18 tumours.

Effect of morphine and co-administration of celecoxib on survival and pain behaviour in SCK tumour-bearing mice. (A) Kaplan–Meier plot for survival of SCK tumour-bearing mice after treatment with morphine and co-administration with celecoxib for 13 days (after 14 days of tumour cell injection). Note, celecoxib alone led to mortality starting at day 4, whereas morphine treatment resulted in highest mortality, which occurred on day 14. Co-administration of both showed highest percent of surviving mice on day 14. n=20 mice/treatment. (B) Behavioural analysis performed for thermal hyperalgesia using Hargreave's apparatus. Right paw of the tumour-bearing leg was tested on each mouse three times at 10 min intervals for PWL. Decrease in duration of heat tolerance shows an increase in thermal perception of pain. Overall, pain increases with time during tumour growth in control as well as morphine- or celecoxib-treated mice. Co-administration of morphine and celecoxib shows the highest heat tolerance or least perception of thermal stimuli. Data are shown as mean±s.d. n=6–20.