HMG-CoA reductase inhibitors (statins) increase endothelial progenitor cells via the PI 3-kinase/Akt pathway

HMG-CoA reductase inhibitors (statins) have been developed as lipid-lowering drugs and are well established to reduce morbidity and mortality from coronary artery disease. Here we demonstrate that statins potently augment endothelial progenitor cell differentiation in mononuclear cells and CD34-positive hematopoietic stem cells isolated from peripheral blood. Moreover, treatment of mice with statins increased c-kit⁺/Sca-1⁺–positive hematopoietic stem cells in the bone marrow and further elevated the number of differentiated endothelial progenitor cells (EPCs). Statins induce EPC differentiation via the PI 3-kinase/Akt (PI3K/Akt) pathway as demonstrated by the inhibitory effect of pharmacological PI3K blockers or overexpression of a dominant negative Akt construct. Similarly, the potent angiogenic growth factor VEGF requires Akt to augment EPC numbers, suggesting an essential role for Akt in regulating hematopoietic progenitor cell differentiation. Given that statins are at least as potent as VEGF in increasing EPC differentiation, augmentation of circulating EPC might importantly contribute to the well-established beneficial effects of statins in patients with coronary artery disease.

Stimulation of neovascularization is a therapeutic goal to rescue tissue from critical ischemia (1). Postnatal neovascularization was attributed mainly to angiogenesis, a process that is mediated by proliferation, migration, and remodeling of preexisting endothelial cells (2–4). However, recent studies provide increasing evidence that postnatal neovascularization does not rely exclusively on sprouting of preexisting vessels, but also involves bone marrow–derived circulating endothelial cells (5–7). A specific subset of leukocytes positive for CD34 was shown to home to sites of ischemia and express endothelial antigens (5, 8). Moreover, endothelial progenitor cells (EPCs) can be grown out of isolated CD133⁺ or CD34⁺ cells in vitro (5, 9, 10) and make a significant contribution to adult blood vessel formation (8). Importantly, injection of isolated CD34⁺ cells or cultivated EPCs accelerates the restoration of blood flow in diabetic mice (11) and enhances neovascularization in vivo (7, 8, 12, 13). In addition, increased neovascularization induced by hematopoietic CD34-positive cells or angioblasts was shown recently to improve cardiac function (14, 15). Therefore, the stimulation of mobilization and/or differentiation of EPCs may provide a useful novel therapeutic strategy to improve postnatal angiogenesis. Proangiogenic growth factors such as VEGF (16) or GM-CSF therapy (17) augmented EPC levels. Moreover, VEGF₁₆₅ gene therapy increased the number of circulating differentiated EPCs in patients with limb ischemia (18). However, the mechanisms, which regulate EPC mobilization and differentiation, are not fully elucidated.

HMG-CoA reductase inhibitors (statins) have been developed as lipid-lowering drugs and are well established to reduce morbidity and mortality from coronary artery disease (19). Besides lipid-lowering, statins are capable of reducing vascular inflammation (20), decreasing platelet aggregation and thrombus deposition (21), and increasing endothelium-derived nitric oxide production (22). Statins have recently been shown to stimulate the growth of new blood vessels in ischemic limbs of normocholesterolemic rabbits (23). To gain insights into the mechanism by which statins improve neovascularization, we investigated the effects of statins on circulating bone marrow–derived endothelial cells and elucidated the underlying signal transduction pathways.

The present study demonstrates that statins augment EPC numbers in vitro and in mice. The increase in EPC levels induced by both statins and VEGF was mediated via the activation of the serine/threonine kinase Akt, which is known to play an important role in endothelial biology and angiogenesis (23, 24).

Statins increase EPCs in vitro. Incubation of isolated human MNCs with atorvastatin dose and time dependently increased the number of differentiated, adherent EPCs from 2.1% ± 0.16% to 5% ± 0.35% of the total number of MNCs in vitro (Figure 1, a and b). Similarly, the HMG-CoA reductase inhibitors mevastatin (1 μM) and simvastatin (1 μM) also augmented the number of EPCs up to 288% ± 17% and 241% ± 102%, respectively (Figure 1c). EPCs were characterized as adherent cells double positive for DiLDL-uptake and lectin binding (5, 18) (Figure 1c). The endothelial phenotype of the EPCs was additionally confirmed by documenting the expression of well-established cell surface markers like vWF, VEGF-receptor 2 (KDR-receptor), VE-cadherin, and CD31 by fluorescence-activated cell sorting (FACS) and immunostaining (Figure 1d and data not shown). The expression pattern of these endothelial cell markers on EPCs was found to be comparable to that observed in human umbilical venous endothelial cells (HUVECs) (data not shown). In addition, the expression and activity of the endothelial nitric oxide synthase in EPCs was demonstrated by immunostaining, Western blot analysis, and DAF-2 staining, respectively (data not shown).

Figure 1

Statins increase the number of adherent EPCs. (a and b) MNCs were incubated with atorvastatin as indicated, and adherent DiLDL/lectin-positive cells were counted (13). Basal control values represent 388 ± 146 cells/mm². The solvent DMSO had no effect on EPC differentiation. Data are mean ± SEM, n = 4–6. (c) MNCs were incubated with atorvastatin (1 μM), simvastatin (1 μM), or mevastatin (1 μM) for 24 hours, and adherent cells’ DiLDL uptake (red) and lectin binding (green) were assessed. Double positive cells appear yellow in the overlay. Representative images are shown from at least three experiments. (d) MNCs were incubated for 4 days. Adherent cells were analysed for expression of KDR, CD31, VE-cadherin, and vWF by FACS. Dotted lines represent isotype controls. Similar expression patterns were observed after stimulation of MNCs with atorvastatin (1 μM, 24 hours) (data not shown). Representative images from n = 4 experiments are shown. (e) MNCs were incubated with VEGF for 24 hours and adherent DiLDL/lectin-positive cells were counted. Data are mean ± SEM, n = 3–6.

Because the angiogenic growth factor VEGF is known to augment the number of EPCs (16, 18), we compared the statin effect with VEGF. As illustrated in Figure 1e, human recombinant VEGF also stimulated the differentiation of EPCs with a maximal effect at 100 ng/ml. The effect of statins appears to be at least equipotent compared, with VEGF.

To exclude the possibility that the effect of VEGF and statins is due to increased adherence of adult circulating endothelial cells, we isolated CD34⁺ hematopoietic stem cells (HSCs), which presumably represent the EPC precursor cells in humans (5, 12, 26). After separation of CD34-positive cells by immunomagnetic CD34 microbeads from mobilized peripheral blood, the purity of the population was 93.8% (range 82.4–99%) as analysed by FACS (Figure 2a). Addition of VEGF or atorvastatin to the isolated CD34⁺ hematopoietic progenitor cells significantly increased the number of differentiated EPCs (Figure 2b). Consistent with the effect obtained in MNCs, atorvastatin was at least as effective as VEGF (Figure 2b). Taken together, these data demonstrate that statins and VEGF induce differentiation of hematopoietic progenitor cells into EPCs.

Figure 2

Statins and VEGF induce differentiation of CD34⁺ cells. (a) CD34⁺ cells were isolated and purity of CD34⁺ cell fraction was analysed by FACS. Isotype control is represented by a dotted line and CD34 by a solid line. (b) CD34⁺ cells were incubated with atorvastatin (AT; 1 μM) or VEGF (100 ng/ml) for 24 hours, and adherent DiLDL/lectin-positive cells were counted. Data are mean ± SEM, n = 4–7, *P < 0.05 versus control.

Statins increase EPC levels in vivo. To test the in vivo relevance of our findings, mice were fed with simvastatin (20 mg/kg daily) for 3 weeks, and EPC numbers were determined. As shown in Figure 3a, statin treatment led to a more than twofold increase in DiLDL/lectin-positive cells, thus extending the in vitro data. To gain further insights into the effect of statins, the number of multipotent bone marrow HSCs, which are characterized by expression of c-kit (CD117) and sca-1 (26, 27), was analysed. The number of c-kit^high cells was profoundly increased in bone marrow of simvastatin-fed mice (Figure 3b). Moreover, c-kit/sca-1 double positive bone marrow cells were increased from 0.78% ± 0.3% in controls to 8.2% ± 2.6% in simvastatin-treated mice (P ≤ 0.05). In contrast, the overall number of sca-1^high cells did not significantly change (37% ± 6% in control versus 45% ± 14% in simvastatin-fed mice). These data suggest that statins specifically augment multipotent c-kit/sca-1 HSCs in the bone marrow and concomitantly increase the number of differentiated EPCs in the periphery in vivo.

Figure 3

Statins increase EPCs and HSCs in mice. (a) Mice were fed with simvastatin for 3 weeks. Splenocytes were isolated and incubated for 4 days. Adherent DiLDL/lectin-positive cells were counted. Data are mean ± SEM, *P < 0.05 versus control. (b) Bone marrow cells were isolated and the number of c-kit^high cells was measured by FACS. Data are mean ± SEM, *P = 0.006 versus control (Mann-Whitney test).

The effects of VEGF and statins on EPCs is mediated via the PI 3-kinase/Akt pathway. Next, we investigated the molecular mechanisms underlying the effects of statins on EPC differentiation. Since statins and VEGF have both been shown to stimulate the PI 3-kinase/Akt (PI3K/Akt) pathway (23, 28), we used the pharmacological PI3K-inhibitors Ly294002 and wortmannin to assess whether the PI3K-pathway is involved. Inhibition of the PI3K-pathway abolished statin- and VEGF-stimulated increase of EPCs (Figure 4, a and b). In contrast, inhibition of another VEGF-stimulated kinase cascade, namely the MAP-kinases ERK-1 and -2, by PD98059 did not significantly alter the differentiation of EPCs (Figure 4, a and b), suggesting that a PI3K rather than a MAP-kinase pathway mediates the effects of statins and VEGF.

Figure 4

Statin- and VEGF-induced EPC differentiation is mediated by the PI3K/Akt pathway. (a and b) MNCs were incubated with atorvastatin (1 μM) or VEGF (100 ng/ml) and the respective inhibitors (Ly294002, 10 μM; wortmannin, 10 nM; PD98059, 10 μM) for 24 hours. Adherent DiLDL/lectin-positive cells were counted. Neither the substances alone nor the solvents had any toxic effect (data not shown). Data are mean ± SEM, n = 3–6, *P < 0.05 versus VEGF or AT. (c) Western blot against phosphorylated Akt (Ser 473) of MNCs incubated with atorvastatin for the respective time points. Equal loading was confirmed by reprobe of the membranes with total Akt. Protein isolation and Western blot analysis was performed as outlined previously (25). A representative experiment is shown (n = 3). (d and e) MNCs were transfected with pcDNA3.1.-GFP as control vector or pcDNA3.1. dominant negative (dom. neg.) Akt and were incubated with atorvastatin (1 μM) or recombinant hVEGF (100 ng/ml) for 24 hours. Adherent DiLDL/lectin-positive cells were counted. Atorvastatin was similarly effective in empty vector–transfected cells compared with GFP. Data are expressed as mean ± SEM, n = 4–6, *P < 0.05 versus control vector plus VEGF, **P < 0.05 versus control vector plus atorvastatin.

To demonstrate that Akt is indeed an essential downstream signaling event, Akt phosphorylation indicative for Akt activity was detected by Western blot analysis. Statins and VEGF potently stimulated Akt phosphorylation in isolated MNCs (Figure 4c and data not shown). The causal contribution of Akt for statin- and VEGF-induced EPC differentiation was finally demonstrated by transient transfection of the MNCs with a dominant negative Akt construct. The transfection efficiency was about 50%, as determined by GFP-fluorescent cells (Figure 4d). Overexpression of the dominant negative Akt inhibited the increase of EPCs induced by statins or VEGF (Figure 4e), clearly demonstrating that Akt is causally involved in EPC differentiation.

Finally, we attempted to characterize the molecular signal transduction pathway of statins to modulate EPC differentiation. The inhibition of HMG-CoA reductase, which mediates the cholesterol-lowering effect of statins, also reduces the formation of the metabolic products geranylgeranylpyrophosphat (GGPP) (22). Blockade of geranylgeranylation, in turn, abolishes Rho-kinase activation and increases eNOS mRNA stability (Figure 5a) (22, 29). Therefore, we determined the effect of mevalonate, the end product of the HMG-CoA reductase, and the further downstream intermediate GGPP. Mevalonate reversed the inhibitory effect of statins, whereas GGPP led only to a minor decrease in atorvastatin-induced EPC differentiation (Figure 5b), suggesting that the downstream signaling of GGPP is less important to mediate the statin effects. Since statins exert an inhibitory effect on the Rho kinase, we also determined whether blocking of Rho kinase by the pharmacological inhibitor HA 1077 would increase EPC numbers. However, HA 1077 did not augment EPC numbers (Figure 5b). In accordance with these findings, the eNOS inhibitor N^G-mono-methyl-L-arginine did not inhibit statin-induced increases in EPC numbers (Figure 5b), thus excluding a potential role of the Rho kinase/endothelial NO synthesis pathway.

Figure 5

Characterization of the statin effect. (a) Schematic illustration of the statin effects on intracellular pathways. (b) MNCs were incubated with atorvastatin (AT; 1 μM) for 24 hours with or without mevalonate (100 μM), GGPP (5 μM), the Rho-kinase inhibitor HA 1077 (10 μM), or N^G-mono-methyl-L-arginin (LNMA, 1mM). DiLDL/lectin-positive cells were counted. Data are expressed as mean ± SEM, P < 0.05 versus AT; n = 3–7.

The data of the present study establish a novel mechanism of action of statins, namely the increase in EPC numbers. Although the proportional contribution of angiogenesis and vasculogenesis to neovascularization of adult tissue remains to be determined, it is well established that EPCs participate in repair following ischemic injury. Thus, increasing the number of circulating EPCs by transplantation of hematopoietic stem cells or by injection of in vitro–differentiated EPCs has been shown to improve neovascularization of ischemic hindlimb (12, 13), accelerate blood flow in diabetic mice (11), and improve cardiac function (14). Therefore, augmentation of circulating EPCs by statin therapy may significantly contribute to the stimulation of neovascularization after tissue ischemia. Recently, in fact, restatin therapy has not only been shown to rapidly enhance coronary blood flow in patients with stable coronary artery disease (30), but also to reduce myocardial ischemia in patients with acute coronary syndromes within a few weeks of treatment (31).

The increase in EPC number by statins is at least equipotent compared with the prototypic angiogenic cytokine VEGF, which is known to promote EPC mobilization and differentiation in vitro and after gene transfer in vivo (5, 16, 18). One may speculate that statins induce the expression of angiogenic growth factors such as VEGF and thereby influence EPC numbers. However, atorvastatin did not affect expression of various angiogenic factors or their receptors in isolated EPCs, as demonstrated by mRNase protection assays (data not shown). Thus, it is unlikely that the effects of statins are mediated via upregulation of angiogenic growth factors. In contrast, the present data suggest that statins — similar to VEGF — can directly induce differentiation of monocytes or CD34-positive hematopoietic precursor cells into EPCs. In vivo, statins further augment the levels of bone marrow hematopoietic precursor cells. Thus, statins may influence not only the differentiation process, but may also modulate mobilization of hematopoietic progenitor cells as shown for VEGF .

VEGF and statins appear to share a common PI3K/Akt signaling pathway to regulate EPC differentiation levels. The protein kinase Akt plays a central role in mature endothelial cells. Activation of Akt promotes endothelial cell survival by inhibiting apoptosis (28, 32), stimulates endothelial NO synthesis (25, 33), and mediates VEGF-induced endothelial cell migration (34, 35). The present study not only extends these findings by demonstrating that Akt regulates differentiation of EPCs, but also identified Akt as a target for statins to modify EPC kinetics, which might importantly contribute to the well-established beneficial effects of statin therapy in patients with myocardial ischemia due to obstructive coronary artery disease.

The question remains how statins can activate the PI3K. In accordance with the data of Kureishi and coworkers (23), we demonstrated that statin-induced activation of Akt in endothelial cells was due to inhibition of mevalonate formation, but largely independent of the downstream product GGPP. Furthermore, inhibition of NO-synthesis did not prevent statin-induced increases in EPC numbers in vitro. This is consistent with the finding that inhibition of the Rho kinase, which mediates the eNOS mRNA stabilizing effect of statins (22) and therefore should mimic the effect of statins on eNOS, did not affect EPC numbers.

In conclusion, the results of the present study demonstrate that statins can induce the differentiation of endothelial progenitor cells and upregulate EPC numbers in vitro and in vivo. In this way, the PI3K/Akt pathway appears to play a major role for statin- and VEGF-induced increases in EPC differentiation. The augmentation of EPC numbers by pharmacological modulation of the signaling pathways, which regulate EPC differentiation, may be a novel strategy to improve neovascularization after ischemia and, thereby, provide a therapeutic concept to improve EPC numbers in patients with coronary artery disease.

We would like to thank Marga Müller-Adorgan and Meike Stahmer for technical help. This study was supported by the Deutsche Forschungsgemeinschaft (SFB 335, B6) and by a cooperation with the Max-Planck Institute for Physiological and Clinical Research, Department of Experimental Cardiology (Bad Nauheim, Germany). Alexandra Aicher was supported by the KGF e.V. Frankfurt/Main (Bone Marrow Transplantation/Gene Therapy Association).

See the related Commentary beginning on page 365.

1. Isner, JM, Asahara, T. Angiogenesis and vasculogenesis as therapeutic strategies for postnatal neovascularization. J Clin Invest 1999. 103:1231-1236.
   View this article via: JCI PubMed CrossRef Google Scholar
2. Folkman, J. Angiogenesis in cancer, vascular, rheumatoid and other disease. Nat Med 1995. 1:27-31.
   View this article via: PubMed CrossRef Google Scholar
3. Risau, W. Mechanisms of angiogenesis. Nature 1997. 386:671-674.
   View this article via: PubMed CrossRef Google Scholar
4. Carmeliet, P, Jain, RK. Angiogenesis in cancer and other diseases. Nature 2000. 407:249-257.
   View this article via: PubMed CrossRef Google Scholar
5. Asahara, T, et al. Isolation of putative progenitor endothelial cells for angiogenesis. Science 1997. 275:964-967.
   View this article via: PubMed CrossRef Google Scholar
6. Shi, Q, et al. Evidence for circulating bone marrow-derived endothelial cells. Blood 1998. 92:362-367.
   View this article via: PubMed Google Scholar
7. Asahara, T, et al. Bone marrow origin of endothelial progenitor cells responsible for postnatal vasculogenesis in physiological and pathological neovascularization. Circ Res 1999. 85:221-228.
   View this article via: PubMed Google Scholar
8. Crosby, JR, et al. Endothelial cells of hematopoietic origin make a significant contribution to adult blood vessel formation. Circ Res 2000. 87:728-730.
   View this article via: PubMed Google Scholar
9. Gehling, UM, et al. In vitro differentiation of endothelial cells from AC133-positive progenitor cells. Blood 2000. 95:3106-3112.
   View this article via: PubMed Google Scholar
10. Bhattacharya, V, et al. Enhanced endothelialization and microvessel formation in polyester grafts seeded with CD34(+) bone marrow cells. Blood 2000. 95:581-585.
    View this article via: PubMed Google Scholar
11. Schatteman, GC, Hanlon, HD, Jiao, C, Dodds, SG, Christy, BA. Blood-derived angioblasts accelerate blood-flow restoration in diabetic mice. J Clin Invest 2000. 106:571-578.
    View this article via: JCI PubMed CrossRef Google Scholar
12. Murohara, T, et al. Transplanted cord blood-derived endothelial precursor cells augment postnatal neovascularization. J Clin Invest 2000. 105:1527-1536.
    View this article via: JCI PubMed CrossRef Google Scholar
13. Kalka, C, et al. Transplantation of ex vivo expanded endothelial progenitor cells for therapeutic neovascularization. Proc Natl Acad Sci USA 2000. 97:3422-3427.
    View this article via: PubMed CrossRef Google Scholar
14. Kocher, AA, et al. Neovascularization of ischemic myocardium by human bone-marrow-derived angioblasts prevents cardiomyocyte apoptosis, reduces remodeling and improves cardiac function. Nat Med 2001. 7:430-436.
    View this article via: PubMed CrossRef Google Scholar
15. Kawamoto, A, et al. Therapeutic potential of ex vivo expanded endothelial progenitor cells for myocardial ischemia. Circulation 2001. 103:634-637.
    View this article via: PubMed Google Scholar
16. Asahara, T, et al. VEGF contributes to postnatal neovascularization by mobilizing bone marrow-derived endothelial progenitor cells. EMBO J 1999. 18:3964-3972.
    View this article via: PubMed CrossRef Google Scholar
17. Takahashi, T, et al. Ischemia- and cytokine-induced mobilization of bone marrow-derived endothelial progenitor cells for neovascularization. Nat Med 1999. 5:434-438.
    View this article via: PubMed CrossRef Google Scholar
18. Kalka, C, et al. Vascular endothelial growth factor(165) gene transfer augments circulating endothelial progenitor cells in human subjects. Circ Res 2000. 86:1198-202.
    View this article via: PubMed Google Scholar
19. Maron, DJ, Fazio, S, Linton, MF. Current perspectives on statins. Circulation 2000. 101:207-213.
    View this article via: PubMed Google Scholar
20. Bustos, C, et al. HMG-CoA reductase inhibition by atorvastatin reduces neointimal inflammation in a rabbit model of atherosclerosis. J Am Coll Cardiol 1998. 32:2057-2064.
    View this article via: PubMed CrossRef Google Scholar
21. Lacoste, L, et al. Hyperlipidemia and coronary disease. Correction of the increased thrombogenic potential with cholesterol reduction. Circulation 1995. 92:3172-3177.
    View this article via: PubMed Google Scholar
22. Laufs, U, Liao, JK. Post-transcriptional regulation of endothelial nitric oxide synthase mRNA stability by Rho GTPase. J Biol Chem 1998. 273:24266-24271.
    View this article via: PubMed CrossRef Google Scholar
23. Kureishi, Y, et al. The HMG-CoA reductase inhibitor simvastatin activates the protein kinase Akt and promotes angiogenesis in normocholesterolemic animals. Nat Med 2000. 6:1004-1010.
    View this article via: PubMed CrossRef Google Scholar
24. Dimmeler, S, Zeiher, AM. Akt takes center stage in angiogenesis signaling. Circ Res 2000. 86:4-5.
    View this article via: PubMed Google Scholar
25. Dimmeler, S, et al. Activation of nitric oxide synthase in endothelial cells via Akt-dependent phosphorylation. Nature 1999. 399:601-605.
    View this article via: PubMed CrossRef Google Scholar
26. Takakura, N, et al. A role for hematopoietic stem cells in promoting angiogenesis. Cell 2000. 102:199-209.
    View this article via: PubMed CrossRef Google Scholar
27. Weissman, IL. Stem cells: units of development, units of regeneration, and units in evolution. Cell 2000. 100:157-168.
    View this article via: PubMed CrossRef Google Scholar
28. Gerber, HP, et al. Vascular endothelial growth factor regulates endothelial cell survival through the phosphatidylinositol 3′-kinase/Akt signal transduction pathway. Requirement for Flk-1/KDR activation. J Biol Chem 1998. 273:30336-30343.
    View this article via: PubMed CrossRef Google Scholar
29. Laufs, U, Liao, JK. Targeting rho in cardiovascular disease. Circ Res 2000. 87:526-528.
    View this article via: PubMed Google Scholar
30. Baller, D, et al. Improvement in coronary flow reserve determined by positron emission tomography after 6 months of cholesterol-lowering therapy in patients with early stages of coronary atherosclerosis. Circulation 1999. 99:2871-2875.
    View this article via: PubMed Google Scholar
31. Schwartz, GG, et al. Effects of atorvastatin on early recurrent ischemic events in acute coronary syndromes: the MIRACL study. A randomized controlled trial. JAMA 2001. 285:1711-1718.
    View this article via: PubMed CrossRef Google Scholar
32. Dimmeler, S, Assmus, B, Hermann, C, Haendeler, J, Zeiher, AM. Fluid shear stress stimulates phosphorylation of Akt in human endothelial cells: involvement in suppression of apoptosis. Circ Res 1998. 83:334-342.
    View this article via: PubMed Google Scholar
33. Fulton, D, et al. Regulation of endothelium-derived nitric oxide production by the protein kinase Akt. Nature 1999. 399:597-601.
    View this article via: PubMed CrossRef Google Scholar
34. Dimmeler, S, Dernbach, E, Zeiher, AM. Phosphorylation of the endothelial nitric oxide synthase at Ser 1177 is required for VEGF-induced endothelial cell migration. FEBS Lett 2000. 477:258-262.
    View this article via: PubMed CrossRef Google Scholar
35. Morales-Ruiz, M, et al. Vascular endothelial growth factor-stimulated actin reorganization and migration of endothelial cells is regulated via the serine/threonine kinase Akt. Circ Res 2000. 86:892-896.
    View this article via: PubMed Google Scholar