X Chen

Centre hospitalier de l'Université de Montréal (CHUM), Montréal, Quebec, Canada

Are you X Chen?

Claim your profile

Publications (12)42.87 Total impact

  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Clinical and animal studies have shown that treatment with angiotensin-converting enzyme (ACE) inhibitors or angiotensin II (Ang II) receptor antagonists slows the progression of nephropathy in diabetes, indicating that Ang II plays an important role in its development. We have reported previously that insulin inhibits the stimulatory effect of high glucose levels on angiotensinogen (ANG) gene expression in rat immortalized renal proximal tubular cells (IRPTCs) via the mitogen-activated protein kinase (p44/42 MAPK) signal transduction pathway. We hypothesize that the suppressive action of insulin on ANG gene expression might be attenuated in renal proximal tubular cells (RPTCs) of rats with established diabetes. Two groups of male adult Wistar rats were studied: controls and streptozotocin (STZ)-induced diabetic rats at 2, 4, 8 and 12 weeks post-STZ administration. Kidney proximal tubules were isolated and cultured in either normal glucose (i.e. 5 mM) or high glucose (i.e. 25 mM) medium to determine the inhibitory effect of insulin on ANG gene expression. Immunoreactive rat ANG (IR-rANG) in culture media and cellular ANG mRNA were measured by a specific radioimmunoassay and reverse transcription-polymerase chain reaction assay respectively. Activation of the p44/42 MAPK signal transduction pathway in rat RPTCs was evaluated by p44/42 MAPK phosphorylation employing a PhosphoPlus p44/42 MAPK antibody kit. Insulin (10(-7) M) inhibited the stimulatory effect of high glucose levels on IR-rANG secretion and ANG gene expression and increased p44/42 MAPK phosphorylation in normal rat RPTCs. In contrast, it failed to affect these parameters in diabetic rat RPTCs. In conclusion, our studies demonstrate that hyperglycaemia induces insulin resistance on ANG gene expression in diabetic rat RPTCs by altering the MAPK signal transduction pathway.
    Journal of Endocrinology 03/2002; 172(2):333-44. · 4.06 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Ascorbic acid (vitamin C) and alpha-tocopherol (vitamin E) have antioxidant properties that could improve redox-sensitive vascular changes associated with hypertension. We determined whether vitamins C and E influence vascular function and structure in hypertension by modulating activity of NADPH oxidase and superoxide dismutase (SOD). Adult stroke-prone spontaneously hypertensive rats (SHRSP) were divided into 3 groups: control (C; n=6), vitamin C-treated (vit C, 1000 mg/day; n=7), and vitamin E-treated (vit E, 1000 IU/day; n=8). All rats were fed 4% NaCl. Blood pressure was measured weekly. After 6 weeks of treatment, the rats were killed, and mesenteric arteries were mounted as pressurized preparations. Vascular O(2)(-) generation and NADPH oxidase activity were measured by chemiluminescence. Vascular SOD activity and plasma total antioxidant status (TAS) were determined spectrophotometrically. Blood pressure increased from 212+/-7 to 265+/-6 mm Hg in controls. Treatment prevented progression of hypertension (vit C, 222+/-6 to 234+/-14 mm Hg; vit E, 220+/-9 to 227+/-10 mm Hg). Acetylcholine-induced vasodilation was improved (P<0.05), and media-to-lumen ratio was reduced (P<0.05) in the treated rats. O(2)(-) was lower in vitamin-treated groups compared with controls (vit C, 10+/-4 nmol. min(-1). g(-1) dry tissue weight; vit E, 9.6+/-3.5 nmol. min(-1). g(-1) dry tissue weight; C, 21+/-9 nmol. min(-1). g(-1) dry tissue weight; P<0.05). Both vitamin-treated groups showed significant improvement (P<0.01) in TAS. These effects were associated with decreased activation of vascular NADPH oxidase (vit C, 46+/-10; vit E, 50+/-9; C, 70+/-16 nmol. min(-1). g(-1) dry tissue weight, P<0.05) and increased activation of SOD (vit C, 12+/-2; vit E, 8+/-1; C, 4.6+/-1 U/mg; P<0.05). Our results demonstrate that vitamins C and E reduce oxidative stress, improve vascular function and structure, and prevent progression of hypertension in SHRSP. These effects may be mediated via modulation of enzyme systems that generate free radicals.
    Hypertension 09/2001; 38(3 Pt 2):606-11. · 6.87 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: We previously demonstrated that high glucose activates angiotensinogen (ANG) expression and that insulin inhibits this activation. The present studies aim to investigate whether insulin regulates ANG gene expression in kidney proximal tubular cells at the transcription level via interaction of the putative insulin-response element (IRE) with its binding protein(s) in the 5'-flanking region of the ANG gene. Fusion genes containing various lengths of the 5'-flanking region of the rat ANG gene fused to a human GH (hGH) gene as reporter were constructed and transiently introduced into rat immortalized renal proximal tubular cells (IRPTCs). The expression of the fusion genes was monitored by the amount of immunoreactive hGH secreted into the medium as assayed by a specific RIA for hGH. Insulin inhibited the expression of pOGH (rANG N-1498/+18), pOGH (rANG N-1120/+18) and pOGH (rANG N-882/+18) but not pOGH (rANG N-854/+18), pOGH (rANG N-820/+18), pOGH (rANG N-688/+18) and pOGH (rANG N-53/+18) in high-glucose (i.e. 25 mM) medium. Site-directed mutagenesis of nucleotides N-874 to N-867 (5' CCC GCC CT 3') in the 5'-flanking region of the rat ANG gene abolished the response to insulin. Insulin also inhibited the expression of the fusion gene containing the DNA fragment ANG N-882 to N-855 inserted upstream of the ANG gene promoter (N-53/+18), but had no effect on a mutant of N-882 to N-855. Gel mobility shift assays revealed that the labeled putative rat ANG-IRE motif (N-878 to N-864, 5' CCT TCC CGC CCT TCA 3') was bound to the nuclear proteins of IRPTCS: This binding was displaced by unlabeled ANG-IRE and IRE of human glyceraldehyde phosphate dehydrogenase but not by mutants of ANG-IRE and IRE of the rat glucagon gene. Southwestern blotting analysis revealed that the labeled putative ANG-IRE motif bound to a major nuclear protein with an apparent molecular mass of 48 kDA: Finally, high glucose levels enhanced 48-kDa nuclear protein expression and induced an additional 70-kDa nuclear protein expression in IRPTCs, as revealed by Southwestern blotting. Insulin inhibited both 48- and 70-kDa nuclear proteins expression induced by high glucose levels. Its inhibitory effect was reversed by the presence of PD98059 (an inhibitor of mitogen-activated protein kinase, MAPK) but not by wortmannin (an inhibitor of phosphatidylinositol 3- kinase). These studies demonstrate that insulin action on ANG gene expression is at the transcriptional level. The molecular mechanism (s) of insulin action is mediated, at least in part, via interaction of the functional IRE with unidentified 48- and 70- kDa nuclear proteins in the rat ANG gene and is MAPK dependent.
    Endocrinology 07/2001; 142(6):2577-85. · 4.72 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Studies have shown that high levels of glucose and angiotensin II (Ang II) stimulate hypertrophy and the expression of matrix protein genes in mouse proximal tubular cells in vitro. The present study tested the hypothesis that blockade of the renin-angiotensin system (RAS) inhibits the stimulatory effect of high levels of glucose on the expression of the renal angiotensinogen (ANG) gene and the formation of Ang II and subsequently attenuates the induction of hypertrophy in kidney proximal tubular cells. Immortalized rat proximal tubular cells (IRPTC) were cultured in monolayer. The levels of expression of rat ANG and ANG mRNA in the IRPTC were quantified by specific radioimmunoassays for rat ANG (RIA-rANG) and by a reverse-transcription polymerase chain reaction (RT-PCR) assay, respectively. Hypertrophy of IRPTC was analyzed by flow cytometry (FACScan) and cellular protein assay. Our studies showed that losartan (an Ang II (AT(1))-receptor blocker), perindopril and captopril (inhibitors of angiotensin-converting enzyme) blocked the stimulatory effect of a high level of glucose (i.e. 25 mM) on the expression of the rat ANG gene and hypertrophy in IRPTC but not by the Ang II (AT(2))-receptor blocker. Our studies indicate that the blockade of RAS is effective in inhibiting the stimulatory effect of hyperglycemia on the expression of the ANG gene and hypertrophy in IRPTC, supporting the notion that the local formation of intrarenal Ang II may play a role in the development of renal hypertrophy during early diabetes.
    Experimental nephrology 01/2001; 9(2):109-17.
  • [Show abstract] [Hide abstract]
    ABSTRACT: The present studies investigated whether the effect of high levels of glucose on angiotensinogen (ANG) secretion and gene expression in kidney proximal tubular cells is mediated at least in part via the activation of p38 mitogen-activated protein kinase (p38 MAPK). Rat immortalized renal proximal tubular cells (IRPTCs) were cultured in monolayer. The levels of immunoreactive rat ANG (IR-rANG) secreted into the medium and the levels of cellular ANG messenger RNA were determined by a specific RIA for rat ANG and a RT-PCR assay, respectively. Phosphorylation of cellular p38 MAPK was determined by Western blot analysis using the Phospho Plus p38 MAPK antibody kit. High levels of glucose (i.e. 25 mM) and phorbol 12-myristate 13-acetate (PMA; 10(-7) M) increased the secretion of IR-rANG and cellular ANG messenger RNA as well as phosphorylation of p38 MAPK in IRPTCs. This stimulatory effect of high levels of glucose and PMA was blocked by SB 203580 (a specific inhibitor of p38 MAPK), but not by SB 202474 (a negative control of SB 203580). High levels of D-sorbitol or 2-deoxy-D-glucose (i.e. > or = 35 mM) also stimulated the phosphorylation of p38 MAPK, but did not stimulate ANG secretion or gene expression. GF 109203X (an inhibitor of protein kinase C) blocked the stimulatory effect of high levels of glucose and PMA on ANG gene expression, whereas it did not block the effect of high levels of glucose, sorbitol, or 2-deoxy-D-glucose on p38 MAPK phosphorylation in IRPTCs. These studies demonstrate that the stimulatory effect of a high level of glucose (25 mM) on ANG gene expression in IRPTCS may be mediated at least in part via activation of p38 MAPK signal transduction pathway and is protein kinase C independent.
    Endocrinology 12/2000; 141(12):4637-46. · 4.72 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: To investigate the molecular mechanism(s) of action of catecholamines on the expression of the angiotensinogen (ANG) gene in kidney proximal tubular cells, we used opossum kidney (OK) cells with a fusion gene containing the 5'-flanking regulatory sequence of the rat ANG gene fused with a human growth hormone (hGH) gene as a reporter, pOGH (rANG N-1498/+18), permanently integrated into their genomes. The level of expression of the ANG-GH fusion gene was quantified by the amount of immunoreactive-hGH (IR-hGH) secreted into the medium. The addition of norepinephrine (NE), isoproterenol (a beta1/beta2-adrenergic receptor (AR) agonist) and iodoclonidine (an alpha2-AR agonist) stimulated the expression of the ANG-GH fusion gene in a dose-dependent manner, whereas the addition of epinephrine and phenylephrine (alpha1-AR agonist) had no effect. The stimulatory effect of NE was blocked by the presence of propranolol (beta-AR blocker), atenolol (beta1-AR blocker), yohimbine (alpha2-AR blocker), Rp-cAMP (an inhibitor of cAMP-dependent protein kinase AI & AII) and staurosporine (an inhibitor of protein kinase C), but was not blocked by ICI 118, 551 (beta2-AR blocker) and prazosin (alpha1-AR blocker). The addition of a combination of isoproterenol and iodoclonidine or a combination of 8-Bromo-cAMP (8-Br-cAMP) and phorbol 12-myristate (PMA) synergistically stimulated the expression of the ANG-GH fusion gene as compared to the addition of isoproterenol, iodoclonidine, 8-Br-cAMP or PMA alone. Furthermore, the addition of NE, 8-Br-cAMP or PMA stimulated the expression of pOGH (rANG N-806/-779/-53/+18), a fusion gene containing the putative cAMP responsive element (CRE, ANG N-806/-779) upstream of the ANG promoter (ANG N-53/+18) in OK cells, but had no effect on the expression of fusion genes containing the mutant of the CRE. Gel mobility shift assays revealed that the ANG-CRE binds with the DNA-binding domain (bZIP254-327) of the cAMP-responsive binding protein (CREB). The binding of the labeled ANG-CRE to CREB (bZIP254-327) was displaced by unlabeled ANG-CRE and the CRE of the somatostatin gene but not by the mutants of the ANG-CRE. Finally, NE stimulated the phosphorylation of CREB in OK cells. These studies demonstrate that the molecular mechanism(s) of NE action on the expression of the ANG gene in OK cells may be mediated via both the PKA and PKC signalling pathways and via the phosphorylation of CREB. The phosphorylated CREB then interacts with the CRE in the 5'-flanking region of the ANG gene and subsequently stimulates the gene expression.
    Molecular and Cellular Biochemistry 09/2000; 212(1-2):73-9. · 2.33 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To investigate the molecular mechanism(s) of insulin action on the expression of the angiotensinogen (ANG) gene in kidney proximal tubular cells, we constructed a fusion gene, pOGH (hANG N-1064/+27), containing the 5'-flanking regulatory sequence of the human ANG gene fused with the human growth hormone (hGH) gene as a reporter and stably integrated the fusion gene into the opossum kidney (OK) cell genomes. The level of expression of pOGH (hANG N-1064/+27) was quantified by the amount of immunoreactive hGH secreted into the medium. The addition of a high level of D(+)-glucose (25 mM) or phorbol 12-myristate 13-acetate (PMA, 10(-7) M) stimulated the expression of the fusion gene in OK cells. The stimulatory effect of glucose (25 mM) was blocked by insulin and tolrestat (an inhibitor of aldose reductase). Tolrestat also inhibited the increase of cellular DAG and PKC activity stimulated by 25 mM glucose. While insulin did not affect the cellular DAG and PKC activity, it did block the stimulatory effect of high glucose (25 mM) and PMA on the expression of the fusion gene. Finally, PD98059 (an inhibitor of mitogen-activated protein kinase kinase (MEK)) enhanced the stimulatory effect of high levels of glucose and blocked the inhibitory effect of insulin on the expression of the fusion gene as well as on the phosphorylation of MEK and mitogen-activated protein kinase (MAPK). In contrast, Wortmannin (an inhibitor of phosphatidylinositol-3-kinase) did not block the inhibitory effect of insulin on the ANG gene expression. These studies demonstrate that the action of insulin, blocking the stimulatory effect of a high level of D(+)-glucose (25 mM) on the ANG gene expression is mediated, at least in part, via the 5'-flanking region of the ANG gene and MAPK signal transduction pathway.
    Journal of Renin-Angiotensin-Aldosterone System 07/2000; 1(2):166-74. · 2.29 Impact Factor
  • [Show abstract] [Hide abstract]
    ABSTRACT: Tyrosine kinases may regulate Ang II-induced vascular contraction. However the specific kinases involved are unknown. This study investigated the regulatory role of c-Src in Ca2+ and contractile responses mediated by Ang II in human vascular smooth muscle cells (VSMC). Cultured VSMC derived from resistance arteries from healthy humans were studied. Intracellular free Ca2+ concentration ([Ca2+]i) was measured by fura 2, c-Src phosphorylation was determined by Western blot, and inositol 1,4,5-trisphosphate (IP3) production was determined by radioimmunoassay. Contractile responses were examined in intact arteries mounted as pressurized systems. Ang II rapidly increased c-Src activity (4-5 fold increase). PP2, the selective Src inhibitor, but not PP3, the inactive analogue, blocked this effect. Ang II induced a biphasic [Ca2+]i response (Emax=636±123 nmol/L, pD2=8.2±0.61). PP2 but not PP3 attenuated (p
    01/2000;
  • J.-B. Park, X. Chen
    [Show abstract] [Hide abstract]
    ABSTRACT: We investigated effects of vitamins C and E on blood pressure elevation, vascular remodeling and endothelial function in salt-loaded stroke-prone SHR (SHRSP). The role of oxidative stress was also assessed in these processes. 16 week-old SHRSP (n = 16) on a high salt diet (4% NaCl) were randomly divided into 3 groups: control (C), Vit C (1000 mg/d) and Vit E (1000 IU/d). Systolic blood pressure (SBP) and plasma antioxidant status (spectrophotometric assay system) were assessed weekly. 6 weeks after treatment rats were killed. Vascular structure (media:lumen ratio) and endothelial function (ACH-induced vasodilation) were assessed in mesenteric arteries. Vascular •O2− production was measured in aortic vessels using lucigenin (5 μM). SBP increased progressively from 204 ± 9.4 to 246 ± 5.6 mmHg in C group. Progression of hypertension was prevented in Vit C (210 ± 5 mmHg) and Vit E (209 ± 11 mmHg) groups. Ach-induced dilation was significantly improved in the treated groups. Media:lumen ratio was reduced (p < 0.001) in Vit E group (6.8 ± 0.8%) vs C group (12.3 ± 0.1%0. Total antioxidant status was significantly improved (p < 0.05) in the Vit C (1.3 ± 0.2 mM) and Vit E (1.68 ± 0.4 mM) groups compared with C (0.87 ± 0.1 mM). •O2− production was significantly lower in both Vit-treated groups compared with the untreated group.Vits C and E improved endothelial dysfunction in small arteries and prevented the progression of hypertension in salt-loaded SHRSP. Vit E also corrected vascular remodeling. These effects were associated with improved antioxidant status and reduced vascular oxidative stress. Thus beneficial effects of antioxidant vitamins in vascular damage associated with hypertension are related, in part, to alterations in vessel redox state.
    American Journal of Hypertension - AMER J HYPERTENS. 01/2000; 13(4).
  • [Show abstract] [Hide abstract]
    ABSTRACT: The present study aimed to investigate the molecular mechanism(s) of insulin action on angiotensinogen (ANG) secretion and gene expression in kidney proximal tubular cells exposed to high levels of glucose. Immortalized rat proximal tubular cells (IRPTC) were cultured in monolayer. The levels of rat ANG and ANG messenger RNA in the IRPTC were quantified by a specific RIA for rat ANG (RIA-rANG) and by an RT-PCR assay. Insulin inhibited the stimulatory effect of a high level of glucose (25 mM) and phorbol 12-myristate 13-acetate, an activator of protein kinase C) on the secretion of ANG and the expression of the ANG messenger RNA in IRPTC. This inhibitory action of insulin on the ANG secretion and gene expression was blocked by PD98059 (an inhibitor of mitogen-activated protein kinase kinase) but not by Wortmannin (an inhibitor of phosphatidylinositol-3-kinase). PD98059 was effective in inhibiting the phosphorylation of MEK 1/2 and p44/42 MAP kinase in IRPTC stimulated by insulin. These studies demonstrate that insulin prevents the stimulatory effect of high levels of glucose on the expression of the renal ANG gene in IRPTC, at least in part, via the MAPK kinase signal transduction pathway, subsequently inhibiting the activation of the local renal renin-angiotensin system.
    Endocrinology 12/1999; 140(11):5285-92. · 4.72 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: beta-adrenoceptors are present in the renal proximal tubules. We have previously reported that isoproterenol stimulates the accumulation of intracellular cAMP and the expression of the angiotensinogen (ANG) gene in opossum kidney (OK) proximal tubular cells via the beta 1-adrenoceptor. We hypothesized that the molecular mechanism(s) of action of isoproterenol on the expression of the ANG gene is mediated via the interaction of the phosphorylated cAMP-responsive element binding protein (CREB) and the cAMP-responsive element (CRE; that is, ANG N-806/-779) in the 5'-flanking region of the rat ANG gene. The fusion genes containing the putative ANG-CRE of the rat ANG gene inserted upstream of the rat ANG basal promoter (ANG N-53/+18) fused to a human growth hormone (hGH) gene as reporter were stably cotransfected, with or without the plasmid containing the cDNA for 43 kDa CREB, into the OK cells. The effect of various agonists and antagonists of adrenoceptors on the expression of the fusion genes was evaluated by the amount of immunoreactive hGH secreted into the culture medium. The interactions of OK cellular nuclear protein(s) with the ANG N-806/779 were determined by gel mobility shift assays and by Southwestern and Western blot analysis. The addition of isoproterenol, forskolin, or 8-Bromo-cAMP (8-Br-cAMP) stimulated the expression of pOGH (ANG N-806/-779/-53/+18) by 135, 150, and 160%, respectively, but not mutants of the ANG N-806/-779. The stimulatory effect of isoproterenol was blocked in the presence of propranolol, Rp-cAMP, and atenolol, but not by the presence of stauro-sporine, U73122, and ICI 118,551. Transient transfection of the plasmid containing the cDNA for the catalytic subunit of protein kinase A further enhanced the stimulatory effect of 43 kDa CREB on the expression of the fusion gene. The gel mobility shift assays revealed the the nuclear protein(s) of OK cells binds to the radioactive-labeled ANG N-806/-779. The binding of the labeled ANG N-806/-779 to the OK cell nuclear protein(s) was displaced by unlabeled ANG N-806/-779, but not by the CRE of the somatostatin gene, the CRE of the tyrosine amino-transferase gene, or the mutants of the ANG N-806/-779. Southwestern blot analysis revealed that the labeled ANG N-806/-779 binds to two nuclear species of 43 and 35 kDa proteins. Western blot analysis, however, revealed that rabbit polyclonal antibodies against the 43 kDa CREB interacted with only the 43 kDa molecular species but not with the 35 kDa species. These studies demonstrate that the stimulatory effect of isoproterenol on the expression of the ANG gene may be mediated, at least in part, via the interaction of the phosphorylated CREB and the CRE in the 5'-flanking region of the rat ANG gene. The novel 35 kDa nuclear protein that is immunologically different from the 43 kDa CREB may also play a role in the expression of the ANG gene.
    Kidney International 06/1999; 55(5):1713-23. · 8.52 Impact Factor
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: To identify the nuclear protein(s) that interact with the putative cAMP response element (CRE) of the rat angiotensinogen (ANG) gene (i.e. nt 806-779 upstream of the transcriptional start site), mouse liver nuclear proteins were prepared for the present studies. The DNase 1 footprinting protection analysis revealed that nt -799/-788 in the 5'-flanking region of the rat ANG gene are protected by the mouse liver nuclear protein. Gel mobility-shift assays revealed that the addition of the unlabelled DNA fragment, ANG nt -806/-779 competed effectively with the binding of the labelled ANG nt -806/-779 to the mouse liver nuclear proteins but the addition of unlabelled mutants of ANG nt -806/-779 were only weakly effective in competing with the labelled ANG nt -806/-779. The addition of unlabelled CRE of the somatostatin (SOM) gene and the CRE of the tyrosine aminotransferase (TAT) gene was also ineffective in competing with the labelled ANG nt -806/-779. Southwestern blot analysis revealed that the labelled ANG nt -806/-779 interacted with two mouse liver nuclear proteins with apparent molecular masses of 52 and 43 kDa, whereas the labelled SOM-CRE, TAT-CRE and the CRE of the phosphoenolpyruvate carboxykinase (PEPCK) gene interacted with one molecular species of 43 kDa. The binding of the labelled ANG nt -806/-779 to the 52 kDa protein was effectively competed for by the addition of unlabelled ANG nt -806/-779 but not by unlabelled SOM-CRE, TAT-CRE and PEPCK-CRE. Finally, Western blot analysis revealed that polyclonal antibodies against the CRE-binding protein (CREB) interacted with the mouse liver nuclear 43 kDa protein but not with the 52 kDa protein. These studies demonstrate that the CRE of the rat ANG gene (ANG nt -806/-779) interacts with the 43 kDa CREB and a novel 52 kDa protein from mouse liver. The novel 52 kDa protein is immunologically distinct from the 43 kDa CREB. These studies suggest that the 52 kDa protein might have a role in the expression of the hepatic ANG gene.
    Biochemical Journal 03/1998; 329 ( Pt 3):623-9. · 4.65 Impact Factor

Publication Stats

309 Citations
42.87 Total Impact Points

Institutions

  • 2001
    • Centre hospitalier de l'Université de Montréal (CHUM)
      Montréal, Quebec, Canada
  • 1998–2001
    • Université de Montréal
      • Center for Mathematical Research
      Montréal, Quebec, Canada