Migratory activity of human breast cancer cells is modulated by differential expression of xanthine oxidoreductase

Abstract

Xanthine oxidoreductase (XOR) may exert an important, but poorly defined, role in the pathogenesis of breast cancer (BC). Loss of XOR expression was linked to aggressive BC, and recent clinical observations have suggested that decreasing XOR may be functionally linked to BC aggressiveness. The goal of the present investigation was to determine whether the decreased XOR observed in clinically aggressive BC was an intrinsic property of highly invasive mammary epithelial cells (MEC). Expression of XOR was investigated using HC11 mouse MEC, HB4a and MCF-10A normal human MEC, and several human mammary tumor cells including MCF-7 and MDA-MB-231. Consistent with clinical observations, data shown here revealed high levels of XOR in normal HC11 and MCF-10A cells that was markedly reduced in highly invasive mammary tumor cells. The contribution of XOR to tumor cell migration in vitro was investigated using MDA-MB-231 and MCF-7 cells and clonally selected derivatives of HC11 that exhibit either weak or strong migration in vitro. We observed that over-expression of an XOR cDNA in MDA-MB-231 and in HC11-C24, both possessing weak XOR expression and high migratory capacity, inhibited their migration in vitro. Conversely, pharmacological inhibition of XOR in MCF-7 and HC11-C4, both possessing high XOR expression and weak migratory capacity, stimulated their migration in vitro. Further experiments suggested that XOR derived ROS mediated this effect and also modulated COX-2 and MMP levels and function. These data demonstrate a functional link between XOR expression and MEC migration and suggest a potential role for XOR in suppressing BC pathogenesis.

FIGURE 1. Expression of XOR activity and protein level in normal mouse and human MEC and in human BC cells

A, the cells indicated were grown for five days after plating in standard rich medium. Cells were harvested, extracts prepared, and total XOR activity was determined by spectroscopic assay of uric acid formation. Data are pMoles of uric acid/min/mg of protein. The mean and SD of six determinations (hatched bars) are shown. Oxypurinol was included in separate reactions at 150uM to confirm specificity (black bars) of urate generation. B, XOR protein level was determined from duplicate samples of four selected cells by western immunoblot, and β-Actin expression was used to control for loading of the gels.

FIGURE 2. Post-confluent induction of XOR activity in mouse and human MEC

The indicated cells were grown to confluency and cells were harvested then and every 24hrs for three days (0, 24, 48, and 72hrs post-confluent). XOR activity was determined from whole cell extracts, and data show the mean and standard deviation of six determinations at each time point. Western immunoblots were run for each of the cells, blots were repeated three times, and representative blots are shown for each cell and included as inset photopgraphs. A, HC11 mouse MEC; B, HB4a human MEC; C, MCF-7 human carcinoma; D, MDA-MB-231 human carcinoma. D, NECA activation of XOR in HC11 and MDA-MB-231 cells. HC11 and MDA-MB-231 cells were grown to confluency in six well trays, treated with 50uM NECA, and XOR activity was determined from whole cells extracts 48 hours later. Data show the mean and standard deviation of six independent determinations.

FIGURE 2. Post-confluent induction of XOR activity in mouse and human MEC

The indicated cells were grown to confluency and cells were harvested then and every 24hrs for three days (0, 24, 48, and 72hrs post-confluent). XOR activity was determined from whole cell extracts, and data show the mean and standard deviation of six determinations at each time point. Western immunoblots were run for each of the cells, blots were repeated three times, and representative blots are shown for each cell and included as inset photopgraphs. A, HC11 mouse MEC; B, HB4a human MEC; C, MCF-7 human carcinoma; D, MDA-MB-231 human carcinoma. D, NECA activation of XOR in HC11 and MDA-MB-231 cells. HC11 and MDA-MB-231 cells were grown to confluency in six well trays, treated with 50uM NECA, and XOR activity was determined from whole cells extracts 48 hours later. Data show the mean and standard deviation of six independent determinations.

FIGURE 3. Activation of the human XOR promoter and upstream regulatory DNA in normal mouse and human MEC and in human BC cells

A, B, ten individual clones of the XOR deletion set, spanning the upstream region from −1,900bp to −200bp, were transfected into mouse HC11 or human HB4a MEC respectively. Expression from each deletion was determined 48hrs after transfection. The parent expression vector, pGL3-Basic (pGL3-B) was transfected independently in each series. Note, the B10 designation, for example, corresponds to phXD-B10 and comprises the previously characterized XOR promoter, while B1 corresponds to phXD-B1 and comprises 1.9kbp of upstream regulatory DNA. C, the phXD-B1 clone containing 1.9kbp of XOR upstream regulatory DNA, including the proximal 200bp promoter, was transfected into the indicated cells at 1.0 to 5.0ug of DNA. Cells were co-transfected with 0.1ug of pcDNA3.1(+)HisMycLacZ and sufficient amounts of pGEM4 to bring the total mass of DNA to 5.1ug. Cells were harvested after 48hrs, and luciferase expression and lacZ expression were determined. Data were normalized to the amount of lacZ expression and are shown as normalized relative light units (R.L.U.).

FIGURE 4. Clonally selected HC11-C24 cells exhibit high migration and reduced XOR expression

A, HC11-C4 and HC11-C24 cells were grown on glass coverslips to 80% confluency, treated with TGFβ, and 24hrs later were stained with Alexa Fluor 488 phalloidin, and photographed. B/C, HC11-C4 and HC11-C24 cells were grown as indicated above. Cells were then treated as described in the lane designation below, lysed in RIPA buffer, and western blots performed (C) against α-SMA or β-Actin and quantitated by scanning dosimetry (B). Lane 1, 1 day of growth without TGFβ; lane 2, 9 days of growth without TGFβ; lane 3, 6hrs growth after 5ng/ml TGF-β1; lane 4, 12hrs after TGFβ1; lane 5, 1 day after TGF-β1; lane 6, 2 days after TGF-β1; lane 7, 3 days after TGF-β1; lane 8, 9 days after TGF-β1; lane 9, 12hrs without TGFβ. Band intensity was quantitated using Kodak Imager software, and relative intensities normalized to the signal obtained with β-Actin (B). D/E, HC11-C4 and HC11-C24 were grown to confluency, wounded with a plastic pipette tip, and washed three times in PBS to remove floating cells. Medium was then replaced with standard growth medium containing 10% heat inactivated FCS, and 5.0 ng/ml TGFβ was added or not as indicated. Cells were photographed at 0, 19, and 31hrs after the addition of TGFβ. Migration was quantitated by open surface area calculation (D), and representative photomicrographs are shown for each time point (E). F, HC11-C4 and HC11-C24 were grown to confluency in six well plates, cells were harvested three days later, and XOR activity was determined. Data show the mean and standard deviation of six determinations. G, western immunoblot analysis was performed on the whole cell extracts used in panel F, and three representative lanes are shown for both cell types.

FIGURE 4. Clonally selected HC11-C24 cells exhibit high migration and reduced XOR expression

A, HC11-C4 and HC11-C24 cells were grown on glass coverslips to 80% confluency, treated with TGFβ, and 24hrs later were stained with Alexa Fluor 488 phalloidin, and photographed. B/C, HC11-C4 and HC11-C24 cells were grown as indicated above. Cells were then treated as described in the lane designation below, lysed in RIPA buffer, and western blots performed (C) against α-SMA or β-Actin and quantitated by scanning dosimetry (B). Lane 1, 1 day of growth without TGFβ; lane 2, 9 days of growth without TGFβ; lane 3, 6hrs growth after 5ng/ml TGF-β1; lane 4, 12hrs after TGFβ1; lane 5, 1 day after TGF-β1; lane 6, 2 days after TGF-β1; lane 7, 3 days after TGF-β1; lane 8, 9 days after TGF-β1; lane 9, 12hrs without TGFβ. Band intensity was quantitated using Kodak Imager software, and relative intensities normalized to the signal obtained with β-Actin (B). D/E, HC11-C4 and HC11-C24 were grown to confluency, wounded with a plastic pipette tip, and washed three times in PBS to remove floating cells. Medium was then replaced with standard growth medium containing 10% heat inactivated FCS, and 5.0 ng/ml TGFβ was added or not as indicated. Cells were photographed at 0, 19, and 31hrs after the addition of TGFβ. Migration was quantitated by open surface area calculation (D), and representative photomicrographs are shown for each time point (E). F, HC11-C4 and HC11-C24 were grown to confluency in six well plates, cells were harvested three days later, and XOR activity was determined. Data show the mean and standard deviation of six determinations. G, western immunoblot analysis was performed on the whole cell extracts used in panel F, and three representative lanes are shown for both cell types.

FIGURE 4. Clonally selected HC11-C24 cells exhibit high migration and reduced XOR expression

A, HC11-C4 and HC11-C24 cells were grown on glass coverslips to 80% confluency, treated with TGFβ, and 24hrs later were stained with Alexa Fluor 488 phalloidin, and photographed. B/C, HC11-C4 and HC11-C24 cells were grown as indicated above. Cells were then treated as described in the lane designation below, lysed in RIPA buffer, and western blots performed (C) against α-SMA or β-Actin and quantitated by scanning dosimetry (B). Lane 1, 1 day of growth without TGFβ; lane 2, 9 days of growth without TGFβ; lane 3, 6hrs growth after 5ng/ml TGF-β1; lane 4, 12hrs after TGFβ1; lane 5, 1 day after TGF-β1; lane 6, 2 days after TGF-β1; lane 7, 3 days after TGF-β1; lane 8, 9 days after TGF-β1; lane 9, 12hrs without TGFβ. Band intensity was quantitated using Kodak Imager software, and relative intensities normalized to the signal obtained with β-Actin (B). D/E, HC11-C4 and HC11-C24 were grown to confluency, wounded with a plastic pipette tip, and washed three times in PBS to remove floating cells. Medium was then replaced with standard growth medium containing 10% heat inactivated FCS, and 5.0 ng/ml TGFβ was added or not as indicated. Cells were photographed at 0, 19, and 31hrs after the addition of TGFβ. Migration was quantitated by open surface area calculation (D), and representative photomicrographs are shown for each time point (E). F, HC11-C4 and HC11-C24 were grown to confluency in six well plates, cells were harvested three days later, and XOR activity was determined. Data show the mean and standard deviation of six determinations. G, western immunoblot analysis was performed on the whole cell extracts used in panel F, and three representative lanes are shown for both cell types.

FIGURE 5. Over-expression of recombinant XOR cDNA inhibits migration in HC11-C24 and MDA-MB-231 MEC in vitro

A, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. B, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. Data show the mean and standard deviation of six independent determinations in both panels. C, western immunoblot of Myc-tagged XOR for the XOR activity data shown in panel B. D, Quantitation of migration in HC11-C24. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in the presence or absence of the XOR inhibitor Y-700. The effect on migration was quantitated at 0, 19, and 31hrs later using open surface area determination. E, representative photomicrographs are shown for the migration assay in panel D. F, migration quantitation in MDA-MB-231. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in the presence or absence of Y-700. The effect on migration was quantitated 0, 19, and 31hrs later by open surface area calculation. G, representative photomicrographs are shown for the wound assay in panel E.

FIGURE 5. Over-expression of recombinant XOR cDNA inhibits migration in HC11-C24 and MDA-MB-231 MEC in vitro

A, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. B, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. Data show the mean and standard deviation of six independent determinations in both panels. C, western immunoblot of Myc-tagged XOR for the XOR activity data shown in panel B. D, Quantitation of migration in HC11-C24. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in the presence or absence of the XOR inhibitor Y-700. The effect on migration was quantitated at 0, 19, and 31hrs later using open surface area determination. E, representative photomicrographs are shown for the migration assay in panel D. F, migration quantitation in MDA-MB-231. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in the presence or absence of Y-700. The effect on migration was quantitated 0, 19, and 31hrs later by open surface area calculation. G, representative photomicrographs are shown for the wound assay in panel E.

FIGURE 5. Over-expression of recombinant XOR cDNA inhibits migration in HC11-C24 and MDA-MB-231 MEC in vitro

A, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. B, pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in six well plates at the indicated DNA levels, cells were harvested after 24hrs, and oxypurinol inhibitable XOR activity was determined. Data show the mean and standard deviation of six independent determinations in both panels. C, western immunoblot of Myc-tagged XOR for the XOR activity data shown in panel B. D, Quantitation of migration in HC11-C24. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into HC11-C24 in the presence or absence of the XOR inhibitor Y-700. The effect on migration was quantitated at 0, 19, and 31hrs later using open surface area determination. E, representative photomicrographs are shown for the migration assay in panel D. F, migration quantitation in MDA-MB-231. pCMV-Myc-XOR or the pCMV-Myc empty vector were transfected into MDA-MB-231 in the presence or absence of Y-700. The effect on migration was quantitated 0, 19, and 31hrs later by open surface area calculation. G, representative photomicrographs are shown for the wound assay in panel E.

FIGURE 6. Inhibition of XOR stimulates migration in HC11-C4 mouse and MCF-7 human MEC in vitro

A, HC11-C4 mouse MEC were grown to confluency and treated with either oxypurinol (150uM, stippled), allopurinol (150uM, open), Y-700 (1uM, black), or remained untreated (hatched). After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of XOR inhibitors. Migration was quantitated 0, 19, and 31hrs after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. B, representative photomicrographs are shown for each time point. C, MCF-7 human MEC were grown to confluency, treated as above, and migration was quantitated 0, 19, 31, and 48hrs after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. D, representative photomicrographs are shown for each time point.

FIGURE 6. Inhibition of XOR stimulates migration in HC11-C4 mouse and MCF-7 human MEC in vitro

A, HC11-C4 mouse MEC were grown to confluency and treated with either oxypurinol (150uM, stippled), allopurinol (150uM, open), Y-700 (1uM, black), or remained untreated (hatched). After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of XOR inhibitors. Migration was quantitated 0, 19, and 31hrs after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. B, representative photomicrographs are shown for each time point. C, MCF-7 human MEC were grown to confluency, treated as above, and migration was quantitated 0, 19, 31, and 48hrs after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. D, representative photomicrographs are shown for each time point.

FIGURE 7. XOR derived ROS may contribute to suppression of MEC migration in vitro

A, HC11-C4, HC11-C24, and MDA-MB-231 cells were grown to confluency and treated with NECA at the indicated doses. XOR activity was determined 48hrs later. Data show the mean and standard deviation of four repetitions each. B, C, HC11-C4 MEC (B) or MDA-MB-231 (C) were grown to confluency and treated with either Y-700, N-acetylcysetine (NAC), or remained untreated. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of Y-700 or NAC, and NECA was added at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. ***, p < 0.001 by Students T-test. D, E HC11-C4 (D) and MDA-MB-231 (E) cells were grown to confluency and treated with NAC at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of NAC at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. F, G HC11-C4 (F) and MDA-MB-231 (G) cells were grown to confluency and treated with urate at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of urate at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays.

FIGURE 7. XOR derived ROS may contribute to suppression of MEC migration in vitro

A, HC11-C4, HC11-C24, and MDA-MB-231 cells were grown to confluency and treated with NECA at the indicated doses. XOR activity was determined 48hrs later. Data show the mean and standard deviation of four repetitions each. B, C, HC11-C4 MEC (B) or MDA-MB-231 (C) were grown to confluency and treated with either Y-700, N-acetylcysetine (NAC), or remained untreated. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of Y-700 or NAC, and NECA was added at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. ***, p < 0.001 by Students T-test. D, E HC11-C4 (D) and MDA-MB-231 (E) cells were grown to confluency and treated with NAC at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of NAC at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. F, G HC11-C4 (F) and MDA-MB-231 (G) cells were grown to confluency and treated with urate at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of urate at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays.

FIGURE 7. XOR derived ROS may contribute to suppression of MEC migration in vitro

A, HC11-C4, HC11-C24, and MDA-MB-231 cells were grown to confluency and treated with NECA at the indicated doses. XOR activity was determined 48hrs later. Data show the mean and standard deviation of four repetitions each. B, C, HC11-C4 MEC (B) or MDA-MB-231 (C) were grown to confluency and treated with either Y-700, N-acetylcysetine (NAC), or remained untreated. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of Y-700 or NAC, and NECA was added at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. ***, p < 0.001 by Students T-test. D, E HC11-C4 (D) and MDA-MB-231 (E) cells were grown to confluency and treated with NAC at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of NAC at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays. F, G HC11-C4 (F) and MDA-MB-231 (G) cells were grown to confluency and treated with urate at the indicated doses. After one hour, cells were subjected to wounding, washed, and the medium refreshed in the presence of urate at the indicated doses. Migration was quantitated 19hrs (HC11-C4) or 48hrs (MDA-MB-231) after wounding using surface area calculation. Data show the mean and standard deviation of six independent assays.

FIGURE 8. XOR contributes to COX-2 regulation in carcinoma and normal MEC

A, The indicated MEC were grown to confluence and harvested every 24hrs over the subsequent 72hrs. Whole cell extracts were prepared and used for western blot analysis of COX-2 and β-Actin, which was used to standardize loading of the gels. All samples were run in triplicate, and representative blots are shown for each. B, HC11-C24 and MDA-MB-231 were transfected in six well plates with 4.0 ug of either pCMV-Myc vector (lanes1 and 2) or pCMV-Myc-XOR (lanes 3 and 4). Whole cell extracts were prepared 48 hours after transfection and western blots were analyzed for COX-2 expression and β-Actin. Duplicate samples were analyzed for each, and XOR protein levels were quantitated by scanning dosimetry and normalized to the β-Actin signal. C, HC11 and MCF-7 cells were plated in six well plates at 35% confluency. After 24hrs cells were treated with either oxypurinol or allopurinol at 150uM. Whole cell extracts were prepared 24hrs later and analyzed for expression of COX-2 and β-Actin. Band intensity was quantitated by scanning dosimetry. Samples were prepared in triplicate and representative blots are shown.

FIGURE 8. XOR contributes to COX-2 regulation in carcinoma and normal MEC

A, The indicated MEC were grown to confluence and harvested every 24hrs over the subsequent 72hrs. Whole cell extracts were prepared and used for western blot analysis of COX-2 and β-Actin, which was used to standardize loading of the gels. All samples were run in triplicate, and representative blots are shown for each. B, HC11-C24 and MDA-MB-231 were transfected in six well plates with 4.0 ug of either pCMV-Myc vector (lanes1 and 2) or pCMV-Myc-XOR (lanes 3 and 4). Whole cell extracts were prepared 48 hours after transfection and western blots were analyzed for COX-2 expression and β-Actin. Duplicate samples were analyzed for each, and XOR protein levels were quantitated by scanning dosimetry and normalized to the β-Actin signal. C, HC11 and MCF-7 cells were plated in six well plates at 35% confluency. After 24hrs cells were treated with either oxypurinol or allopurinol at 150uM. Whole cell extracts were prepared 24hrs later and analyzed for expression of COX-2 and β-Actin. Band intensity was quantitated by scanning dosimetry. Samples were prepared in triplicate and representative blots are shown.

FIGURE 9. XOR modulates MMP protein levels and function in human carcinoma cells

A, MDA-MB-231 cells were grown to confluency in six well plates and treated with either oxypurinol or Y-700. After 72hrs of growth whole cell extracts were prepared and analyzed by western blot for expression of MMP1, MMP3, and β-Actin. Band intensities of the mature, active MMP1 and MMP3 were quantitated by scanning dosimetry and normalized to the actin signal. B, MDA-MB-231 cells were grown as above and treated with either Y-700 at a dose of 1.0uM, NAC at a dose of 1mM, or untreated (control). NECA was added 1hr later at a dose of 50uM. Whole cell extracts were prepared 24hrs later and western blots analyzed for COX-2, MMP1, and β-Actin. All samples were run in duplicate. Band intensities (top panel) were quantitated by scanning dosimetry for the gels shown in the lower panel and normalized to the control signal. *** p< 0.02 by Students t-Test. C, Quantitation of migration in the human carcinoma cell line BT20. Cells were grown to confluency in 12 well trays, wounded, and treated as indicated with either the XOR inhibitor Y-700, Y-700 in combination with MMP inhibitor I (150uM), or Y-700 in combination with MMP inhibitor III (150nM). Migration was quantitated 0 and 24hrs after wounding using open surface area calculation. Data show the mean and standard deviation of six determinations for each treatment. D, representative photomicrographs are shown for each treatment group shown in panel C.

FIGURE 9. XOR modulates MMP protein levels and function in human carcinoma cells

A, MDA-MB-231 cells were grown to confluency in six well plates and treated with either oxypurinol or Y-700. After 72hrs of growth whole cell extracts were prepared and analyzed by western blot for expression of MMP1, MMP3, and β-Actin. Band intensities of the mature, active MMP1 and MMP3 were quantitated by scanning dosimetry and normalized to the actin signal. B, MDA-MB-231 cells were grown as above and treated with either Y-700 at a dose of 1.0uM, NAC at a dose of 1mM, or untreated (control). NECA was added 1hr later at a dose of 50uM. Whole cell extracts were prepared 24hrs later and western blots analyzed for COX-2, MMP1, and β-Actin. All samples were run in duplicate. Band intensities (top panel) were quantitated by scanning dosimetry for the gels shown in the lower panel and normalized to the control signal. *** p< 0.02 by Students t-Test. C, Quantitation of migration in the human carcinoma cell line BT20. Cells were grown to confluency in 12 well trays, wounded, and treated as indicated with either the XOR inhibitor Y-700, Y-700 in combination with MMP inhibitor I (150uM), or Y-700 in combination with MMP inhibitor III (150nM). Migration was quantitated 0 and 24hrs after wounding using open surface area calculation. Data show the mean and standard deviation of six determinations for each treatment. D, representative photomicrographs are shown for each treatment group shown in panel C.

FIGURE 10. A possible mechanism for the contribution of XOR to modulation of MEC migratory activity and to COX-2 and MMP expression

The postulated situation obtained in cells exhibiting high XOR and weak COX-2 expression is illustrated. Xanthine, NADH, and nitrites are the three substrates most likely used by XOR in MEC. Allopurinol (Allo), Oxypurinol (Oxy), and Y-700 are XOR specific inhibitors. XOR can generate the ROS O^−.₂ and H₂O₂ during catalysis. NO can be generated directly from XOR catalysis during nitrite metabolism, contributing to RNS generation. Urate will be formed by XOR following xanthine oxidation. We posit that XOR derived ROS, RNS, or urate may contribute to modulation of COX-2. When COX-2 levels are sufficiently high, it is imagined that COX-2, possibly through and ROS based mechanism, stimulates MMP1/MMP3 levels contributing to migratory activity. Modulation of COX-2 by high level XOR activity is then imagined to reduce MMP1/MMP3 and migratory activity. AA, arachidonic acid; PGE, prostaglandin; NAC, N-acetylcysteine.