Donate Help Contact The AHA Sign In Home
American Heart Association
Circulation Research
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Circulation Research. 2004;95:162-169
Published online before print June 10, 2004, doi: 10.1161/01.RES.0000134924.89412.70
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
95/2/162    most recent
01.RES.0000134924.89412.70v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Knowles, H. J.
Right arrow Articles by Harris, A. L.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Knowles, H. J.
Right arrow Articles by Harris, A. L.
Related Collections
Right arrow Cardiovascular Pharmacology
Right arrow Angiogenesis
Right arrow Ischemic biology - basic studies
Right arrow Other Vascular biology
Right arrow Other Research
(Circulation Research. 2004;95:162.)
© 2004 American Heart Association, Inc.

Molecular Medicine

Novel Mechanism of Action for Hydralazine

Induction of Hypoxia-Inducible Factor-1 {alpha}, Vascular Endothelial Growth Factor, and Angiogenesis by Inhibition of Prolyl Hydroxylases

Helen J. Knowles, Ya-Min Tian, David R. Mole, Adrian L. Harris

From the Cancer Research UK Molecular Oncology Laboratory (H.J.K., A.L.H.), Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford; and the Henry Wellcome Building of Genomic Medicine (Y.-M.T., D.R.M.), University of Oxford, Roosevelt Drive, Oxford, UK.

Correspondence to Prof Adrian L Harris, Cancer Research UK Molecular Oncology Laboratory, Weatherall Institute of Molecular Medicine, John Radcliffe Hospital, Oxford OX3 9DS, UK. E-mail harrisa{at}cancer.org.uk


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
The vasodilator hydralazine, used clinically in cardiovascular therapy, relaxes arterial smooth muscle by inhibiting accumulation of intracellular free Ca2+ via an unidentified primary target. Collagen prolyl hydroxylase is a known target of hydralazine. We therefore investigated whether inhibition of other members of this enzyme family, namely the hypoxia-inducible factor (HIF)-regulating O2-dependent prolyl hydroxylase domain (PHD) enzymes, could represent a novel mechanism of action. Hydralazine induced rapid and transient expression of HIF-1{alpha} and downstream targets of HIF (endothelin-1, adrenomedullin, haem oxygenase 1, and vascular endothelial growth factor [VEGF]) in endothelial and smooth muscle cells and induced endothelial cell-specific proliferation. Hydralazine dose-dependently inhibited PHD activity and induced nonhydroxylated HIF-1{alpha}, evidence for HIF stabilization specifically by inhibition of PHD enzyme activity. In vivo, hydralazine induced HIF-1{alpha} and VEGF protein in tissue extracts and elevated plasma VEGF levels. In sponge angiogenesis assays, hydralazine increased stromal cell infiltration and blood vessel density versus control animals. Thus, hydralazine activates the HIF pathway through inhibition of PHD activity and initiates a pro-angiogenic phenotype. This represents a novel mechanism of action for hydralazine and presents HIF as a potential target for treatment of ischemic disease.


Key Words: cardiovascular disease • iron chelation • ischemia • endothelial cells


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Hydralazine is a vasodilator used to treat severe hypertension, congestive heart failure, myocardial infarction, and preeclampsia.1,2 It is thought to reduce peripheral resistance directly by relaxing the smooth muscle cell layer in arterial vessels. The precise mechanism of action remains unknown but potentially involves an altered Ca2+ balance in vascular smooth muscle cells, whereby inhibition of Ca2+ release from the sarcoplasmic reticulum3 prevents contraction mediated by Ca2+-dependent ATPases, kinases,4 or ion channels.5 This disrupted Ca2+ balance is proposed to be secondary to either a membrane hyperpolarization event6 or an elevation of intracellular cGMP levels.7 However, no direct target of hydralazine has been identified explaining this vascular effect. During treatment of congestive heart failure, concomitant administration of hydralazine with isosorbide dinitrate prevents early development of nitrate tolerance and reduces long-term mortality.8,9 In this context, hydralazine is thought to inhibit activation of a membrane-associated oxidase responsible for the increased superoxide production that causes nitrate tolerance.10

The range of physiological, cellular, and molecular actions of hydralazine might be explained if a primary target molecule(s) could be identified. One long-established target is protocollagen prolyl hydroxylase (CPH), a member of the Fe2+-dependent and 2-oxoglutarate (OG)-dependent dioxygenase family. Hydralazine is thought to complex with enzyme-bound Fe2+, inhibiting CPH activity11 and preventing the posttranslational hydroxylation of collagen prolyl residues essential to form stable collagen fibers.12 Other members of this family include the regulators of hypoxia-inducible factor-{alpha} (HIF-{alpha}) subunits, prolyl hydroxylase domain (PHD) enzymes 1 to 3.13,14 Under normoxic conditions, these enzymes posttranslationally hydroxylate HIF-{alpha}, targeting it for interaction with the von Hippel–Lindau (pVHL) E3 ubiquitin ligase complex and rapid proteasomal degradation.13,14 The PHD enzymes have an absolute requirement for dioxygen and are therefore inactive under hypoxia, allowing HIF accumulation and activation of pathways regulating processes such as angiogenesis and metabolic adaptation.15

We have investigated effects of hydralazine on the HIF system. We present evidence that hydralazine rapidly and transiently induces HIF-1{alpha} protein and downstream targets of HIF via inhibition of the PHD enzymes and stimulates specific endothelial cell (EC) proliferation in culture and rapid neo-angiogenesis in vivo. We have identified a novel target molecule for hydralazine and a potential mechanism of action for its therapeutic role clinically.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Cell Culture
Human umbilical vein EC (HUVEC), human dermal microvascular EC, and human vascular smooth muscle cells (HVSMC) (Cambrex Bioscience, Wokingham, UK) were maintained in MCDB with 20% fetal bovine serum, L-glutamine (2 µmol/L), heparin (5 IU/mL) and EC growth supplement (50 µg/mL). MDA 468 breast carcinoma cells were maintained in DMEM with 10% fetal bovine serum, L-glutamine, penicillin (50 IU/mL), and streptomycin sulfate (50 µg/mL). pVHL-deficient renal carcinoma cells (RCC4 VA) and VHL stable transfectants (RCC4 VHL) were cultured in DMEM and maintained in selection with 500 µg/mL G418. Hypoxic exposure (0.1% O2, 5% CO2, balance N2) was in a Heto-Holten CellHouse 170 (RS Biotech, Irvine, Scotland). Reagents were hydralazine HCl, CoCl2, DFO (Sigma, Poole, UK), and the proteasome inhibitor MG132 (Affiniti Research Products, Exeter, UK). Dimethyloxalylglycine (MMOG) was a gift from C. Schofield (Dyson Perrins Laboratory, Oxford, UK).

Hydroxylated HIF-1{alpha} Antibody
Polyclonal antihydroxylated HIF-1{alpha} antiserum was raised in rabbit against the synthetic peptide CDLEMLAHypYIPMD (HIF-1{alpha} amino acids 558 to 569 with hydroxylated Pro564). Serum was tested for its ability to distinguish hydroxylated and native HIF-1{alpha} in RCC4 VHL cell extract under normoxia, hypoxia, or proteasomal inhibition.

Western Blotting
Cells were homogenized in lysis buffer (6.2 mol/L urea, 10% glycerol, 5 mmol/L DTT, 1% SDS, protease inhibitors), lysate separated by 8% SDS-PAGE, and transferred to PVDF membrane. Primary antibodies were monoclonal antihuman (BD Transduction Laboratories, Lexington, Ky) or antimouse (Novus Biologicals, Littleton, Col) HIF-1{alpha} and anti-ß-tubulin (Sigma). Immunoreactivity was visualized with HRP-linked goat antimouse serum and chemiluminescence.

Polymerase Chain Reactions
Total RNA was extracted in TRI Reagent (Sigma), reverse-transcribed (1st Strand cDNA Synthesis Kit; Roche Diagnostics, Mannheim, Germany) and polymerase chain reaction (PCR) amplified within the linear range using the Expand High Fidelity PCR system (Roche). PCR primers were: adrenomedullin 5'-AAGAAGTGGAATAAGTGGGCT-3' and 5'-TGGCTTAGAAGACACCAGAGT-3'; endothelin-1 5' TCTCTGCTGTTTGTGGCTTGC-3' and 5'-GACCCAAATGATGTCCAGGTG-3'; and haem oxygenase-1 CTACTTCCCAGAAGAGCTGCA-3' and 5'-AGCGGTAGAGCTGCTTGAACT-3'.

Vascular Endothelial Growth Factor and Endothelin-1 Enzyme-Linked Immunosorbent Assay
Culture medium and tissue extracts were assayed using the Quantikine human vascular endothelial growth factor (VEGF) immunoassay or QuantiGlo human endothelin-1 immunoassay (R&D Systems, Minneapolis, Minn).

PHD Enzyme Activity
In the HIF-VHL capture assay, RCC4 VA cell extract was assayed in the presence of hydralazine and 5 µmol/L Fe2+ by measuring the interaction between a bead immunopurified Gal/HIF/VP16 fusion protein and 35S-methionine-labeled VHL-HA as previously described.16 The HIF peptide hydroxylation assay comprised 25 µg HUVEC cell extract pre-incubated with hydralazine (pH 7.0), then assayed for the radioactivity associated with succinate in the presence of 50 000 dpm [5-14C]2-OG (Sigma Radiochemicals, St Louis, Mo) and 100 µmol/L Pro-564 peptide (LDLEALAPYIPADDDFQLRS) as previously described.17

Cell Proliferation Assay
In the cell proliferation assay, 3x103 cells were quiesced overnight in reduced media (2% fetal bovine serum, no EC growth supplement), then treated with hydralazine daily for 3 days. Viable cells were quantified using the CellTiter AQueous One Solution Cell Proliferation assay (Promega, Madison, Wis). VG76e was a gift from R. Bicknell (Cancer Research UK Angiogenesis Laboratory, Oxford, UK).

Hydralazine-Treated Mice
BALBc NuNu mice (n=44) treated with 5 mg/kg hydralazine intravenously or an equal volume of phosphate-buffered saline were euthanized after 30 minutes, 1 hour, or 2 hours. The heart, liver, kidney, spleen, leg muscle, and lungs were immediately excised on ice and snap-frozen in liquid N2. At the time of euthanization, blood samples were taken into EDTA collection tubes and plasma stored at –80°C.

Tissue Extract Preparation
Frozen tissue was pulverized under liquid N2, homogenized in 5-volume ice-cold RIPA buffer (50 mmol/L Tris [pH 8.0], 300 mmol/L NaCl, 0.1% SDS, 1% NP40, 0.5% Na deoxycholate, protease inhibitors) and incubated on ice for 40 minutes. Homogenate was centrifuged for 15 minutes at 14 000g and frozen at –80°C.

Sponge Angiogenesis Assay
C57 black mice (n=28) were implanted subcutaneously with a sterile polyether sponge (15x5x5 mm; Caligen Foam Ltd, Accrington, UK) under the dorsal skin. Equal volumes of hydralazine (5 mg/kg) or phosphate-buffered saline were administered intravenously daily for 20 days. Mice were euthanized on days 6, 14, and 21, and sponges were removed and fixed in formalin. All procedures were performed under the British Home Office Animals (Scientific Procedures) Act 1986, license number 70/4949.

CD34 Immunostaining
Formalin-fixed sections were incubated in 20 µg/mL rat anti-CD34 monoclonal antibody (Abcam, Cambridge, UK). Staining was visualized with biotinylated antirat antibody (Dako) and the Vectastain Elite ABC kit (Vector Laboratories, Burlingame, Calif), then counterstained with hematoxylin. CD34-positive blood vessels within a 10x10 square eyepiece grid were counted in the 3 most angiogenic areas of each section. Sponge infiltration was assessed on H&E stained sections (x4 objective).


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Hydralazine Induces HIF-1{alpha} in Vascular Cells
HIF-1{alpha} induction by hydralazine was first evaluated in the primary target cells of the vasculature. Fifty to 500 µmol/L hydralazine rapidly induced HIF-1{alpha} protein in HUVEC. The level of induction was concentration-dependent and decayed with time (Figure 1A), although doses >500 µmol/L were toxic (data not shown). After decay of the response, a repeat 100-µmol/L dose re-induced HIF-1{alpha} to a level comparable with the initial response (Figure 1B). This is consistent with instability of the active form of hydralazine in tissue culture media (Figure 1B), as also occurs in blood plasma.18 Presence of small amounts of the metabolizing enzyme N-acetyl transferase 2 may also contribute.19 All cell lines tested demonstrated a comparable time course for HIF-1{alpha} induction, including smooth muscle cells and several cancer cell lines. In all cases, 50 µmol/L hydralazine induced peak HIF-1{alpha} expression 1 to 2 hours after application, which returned to baseline by 16 to 24 hours (Figure 1C). Peak levels of induction were similar to that induced by hypoxia (Figure 1C). We therefore performed subsequent in vitro experiments with 50 µmol/L hydralazine, a nontoxic concentration that induces physiological levels of HIF-1{alpha}.



View larger version (49K):
[in this window]
[in a new window]
  		Figure 1. Hydralazine induces HIF-1{alpha}. A, Hydralazine-induced HIF-1{alpha} in HUVEC. B, Left panel: time-course of HIF-1{alpha} induction with reapplication of hydralazine at 16 hours. B, Right panel: hydralazine added to MCDB media and maintained at 37°C was transferred onto HUVEC after 2 to 22 hours for a 2-hour period. C, Top panel: HIF-1{alpha} protein induced in MDA468 by 2 hours in 50 µmol/L hydralazine (Hy) and by 16 hours in hypoxia (Hx). Graph, Quantified induction of HIF-1{alpha} by 50 µmol/L hydralazine in HUVEC, HVSMC, and MDA468. Each point represents mean±SD from 3 independent experiments standardized against ß-tubulin. D, Reverse-transcription polymerase chain reaction of adrenomedullin (Ad), HO-1, and Et-1 mRNAs induced by 50 µmol/L hydralazine or 16 hours of hypoxia (Hx). E, Hydralazine induced Et-1 secretion in HUVEC [*P<0.01 versus untreated control (–)] to a level approximating hypoxic induction (Hx; *P<0.005). F, Hydralazine induced VEGF secretion in HVSMC (*P<0.05) and MDA468 (*P<0.01) to a level approximating hypoxic induction (Hx; *P<0.005).

Hydralazine Induces HIF Target Genes in Culture
To determine whether hydralazine-induced HIF was transcriptionally active, we examined induction of downstream genes, focusing on those that modulate vascular tone15 and might be involved in vascular actions of hydralazine in vivo. In HUVEC, 50 µmol/L hydralazine induced adrenomedullin, haem oxygenase-1 (HO-1), and endothelin-1 (Et-1) mRNAs to a level approximating hypoxic induction (Figure 1D). Early expression peaked at 4 hours, complementing rapid induction of HIF-1{alpha}. Et-1 and HO-1 expression was maintained for ≥24 hours, suggesting potential positive feedback mechanisms regulating hydralazine-induced genes. At the protein level, significant Et-1 secretion was observed at 16 to 24 hours (Figure 1E, P<0.01); 50 µmol/L hydralazine increased VEGF secretion from HVSMC and MDA468 (Figure 1F), although no effects were observed on either secreted or cell-bound VEGF in HUVEC (data not shown). This is consistent with reports that EC do not usually produce VEGF under normoxic or hypoxic conditions, either in culture20 or in vivo.21 Thus, hydralazine initiated rapid activation of the HIF transcriptional cascade in target cell lines in culture.

Hydralazine Inhibits PHD Enzyme Activity
Hydralazine (Figure 2A) complexes with free iron in solution22 and inhibits CPH activity by forming a putative labile complex with enzyme-bound iron at the active site.11 It might therefore inhibit the iron-dependent PHD enzymes via chelation of the active site iron moiety, stabilizing HIF in a similar manner to DFO.16 Indeed, exogenous free iron inhibited HIF-1{alpha} induction by equimolar concentrations of DFO and hydralazine (Figure 2B). We therefore assayed PHD activity in the presence of hydralazine. Dose-dependent inhibition of PHD enzyme activity was detected in HUVEC cell extract by analysis of HIF peptide-dependent degradation of 2-OG (Figure 2C, light bars). This dose-dependency complements the hydralazine-mediated induction of HIF-1{alpha} also observed in HUVEC (Figures 1A and 2DownC, dark bars), and the results from the HIF-VHL capture assay (Figure 2C, lower panel). It is clear that 50 µmol/L hydralazine is an insufficient dose to maximally inhibit PHD enzyme activity or maximally induce HIF-1{alpha} protein. However, it is nontoxic and sufficient to induce physiological levels of HIF-1{alpha} in target vascular cells and induce downstream components of the pathway (Figure 1).



View larger version (48K):
[in this window]
[in a new window]
  		Figure 2. Hydralazine inhibits PHD enzyme activity. A, Hydralazine (1-hydrazinophthalazine). B, HIF-1{alpha} induction in MDA 468, after 4-hour incubation with 100 µmol/L FeCl2, hydralazine (Hy) or DFO. C, Graph: HIF peptide-specific degradation of [5-14C]2-OG by HUVEC extract in the presence of hydralazine compared with inhibition of PHD activity by 100 µmol/L DFO or the 2-OG analogue MMOG (1 mmol/L) (light bars). Quantified HIF-1{alpha} induction in HUVEC by densitometry against ß-tubulin (dark bars). Each bar represents mean±SD from 3 independent experiments. C, Lower panel: autoradiograph of 35S-methionine-labeled VHL after pull-down by a bead-bound HIF construct in the presence of hydralazine. D, HIF-1{alpha} induction (top panel) in RCC4 VHL cells by proteasomal inhibition (25 µmol/L MG132), hydralazine and MMOG (M, 1 mmol/L). D, Bottom panel: detection of hydroxylated HIF-1{alpha}. AOX indicates anoxia (HIF induced, PHDs inactive), ROX, reoxygenation plus proteasomal inhibition (HIF induced, PHDs active). E, Four hours of 50 µmol/L hydralazine induced HIF-1{alpha} in RCC4 VHL but not RCC4 VA cells.

As further evidence that HIF-1{alpha} induction is caused by PHD inhibition, we analyzed the hydroxylation status of hydralazine-induced HIF-1{alpha} (Figure 2D). When PHDs are fully active, HIF-1{alpha} is hydroxylated and rapidly degraded. Any HIF-1{alpha} induced because of PHD inhibition should therefore be in the nonhydroxylated form. As shown in Figure 2D, hydralazine-induced HIF-1{alpha} was nonhydroxylated, demonstrating the same level of background antibody reactivity as in untreated control cells and during complete PHD inactivation (anoxia and MMOG). Also, hydralazine prevented hydroxylation of hypoxia-induced HIF-1{alpha} caused by PHD reactivation on reoxygenation (Figure 2D).

To eliminate other possible mechanisms of HIF-1{alpha} induction, we assayed effects of hydralazine in pVHL-deficient RCC4 VA cells that express HIF constitutively because of a defect in targeting the protein for degradation. Fifty micromoles per liter hydralazine did not increase basal HIF-1{alpha} levels (Figure 2E), indicating that it induces neither transcription nor translation but modulates HIF-1{alpha} degradation. Also, hydralazine did not interfere directly with the HIF–VHL interaction (data not shown). Together, these data suggest that hydralazine-mediated induction of HIF-1{alpha} can be explained directly by inhibition of PHD enzyme activity.

Hydralazine Stimulates EC Proliferation
Considering the vascular functions of hydralazine-induced genes, we hypothesized that hydralazine might stimulate EC proliferation. Fifty to 100 µmol/L hydralazine induced significant proliferation of HUVEC and HDMEC (Figure 3). Higher concentrations resulted in toxicity. This pro-proliferative effect was EC-specific. Hydralazine-induced EC proliferation (but not endogenous growth) was inhibited by the NO synthase inhibitor L-NAME (Figure 3 and data not shown). Both HIF23 and Et-124 induce NO release from EC. Although not itself a growth factor, NO mediates autocrine-positive feedback of Et-1 expression25 and induces HO-1 transcription and mRNA stabilization,26 possible explanations for its apparent pro-proliferative effect and the prolonged induction of these mRNAs (Figure 1D). In concordance with Figure 1 data, EC proliferation was not inhibited by anti-VEGF neutralizing antibody at concentrations that inhibit proliferation mediated by 10 ng/mL VEGF (Figure 3 and data not shown).



View larger version (40K):
[in this window]
[in a new window]
  		Figure 3. Hydralazine induces EC proliferation. Fifty to 100 µmol/L hydralazine (**P<0.01 versus untreated controls) and VEGF (*P<0.05) induced proliferation of HUVEC. VG76e (neutralizing VEGF antibody) did not inhibit hydralazine-mediated proliferation, whereas L-NAME attenuated this effect (*P<0.05 versus 50 µmol/L hydralazine). Hydralazine induced proliferation of HDMEC (*P<0.05), but not HVSMC, and inhibited proliferation of MDA468 (*P<0.05).

In agreement with reports on the selective toxicity of iron chelators toward cancer cells,27 hydralazine concentrations that were pro-proliferative in EC and demonstrated no effect in HVSMC significantly inhibited MDA468 cell growth (Figure 3). Unexpectedly, higher concentrations did not produce this effect. It is possible that the 250-µmol/L dose induced sufficient HIF-1{alpha} to compensate in some manner for toxic effects of iron chelation; however, because lower doses were generally used, this effect was not pursued.

Hydralazine Induces HIF-1{alpha} and VEGF In Vivo
In vivo analyses of hydralazine-mediated vasodilation describe substantial effects on vascular tone, blood flow, and tissue perfusion parameters.28–30 These effects show rapid onset after intravenous administration and are transient; for example, peak effects on normal murine tissue perfusion occur 1 to 2 hours after injection.28 This rapid and transient vascular response is strikingly similar to the hydralazine-induced HIF response we observed in culture. We therefore sought to determine whether HIF, and its downstream targets, are activated at early time points in vivo. Five milligrams per kilogram IV hydralazine28,29 stabilized HIF-1{alpha} protein in normal murine lung, heart, liver, and spleen (Figure 4A through 4F and Tables 1 and 2Down). This occurred 30 minutes after injection in lung and peaked at 1 hour in the majority of tissues tested. It is of note that we observed HIF-1{alpha} stabilization in organs of <50% of the mice. This was not caused by gender or age-related differences and may be a result of individual variation in response rate and/or to difficulties associated with preparation of tissue extracts for such a labile factor as HIF-1{alpha}. Despite this, significant HIF-1{alpha} induction was observed in 4 of 6 tissues (Table 1). We next assayed VEGF levels in tissue extracts and blood plasma. Combining all time points, VEGF was significantly increased in lung (P<0.03) and liver (P<0.04, Table 1), whereas VEGF induction in blood plasma was significant at 2 hours (P<0.05; Table 2). It was therefore evident that hydralazine rapidly activates the HIF pathway in normal tissues in vivo. In parallel with the extensively reported effects on vascular tone, this suggests a potential role for components of this pathway in mediating the physiological response to hydralazine.



View larger version (92K):
[in this window]
[in a new window]
  		Figure 4. Hydralazine induced HIF-1{alpha} in vivo. HIF-1{alpha} protein (top panel) was induced ≥30 minutes after intravenous injection of hydralazine; (A) lung, (B) liver, (C) spleen, (D) heart, (E) kidney, and (F) leg muscle. Bottom panel=ß-tubulin.


View this table:
[in this window]
[in a new window]
  		Table 1. Hydralazine Induces HIF-1{alpha} and VEGF in Normal Murine Tissues


View this table:
[in this window]
[in a new window]
  		Table 2. Hydralazine Elevates Plasma VEGF Levels

Hydralazine Is Proangiogenic In Vivo
Considering the pro-proliferative effect on EC, we hypothesized that hydralazine might induce angiogenesis in vivo. Daily 5 mg/kg hydralazine intravenously increased stromal cell infiltration into subcutaneously implanted sponges (8.9±1.4% infiltration versus 4.9±0.9%) in association with significantly increased blood vessel density (160±52 vessels/mm2 versus 32±12, *P<0.02) at 6 days compared with controls (Figure 5A through 5C). No significant difference in either parameter was observed at 14 and 21 days (Figure 5D). The apparent reduced blood vessel density in the hydralazine-treated group at later time points was caused by 2 factors. First, the highest blood vessel concentration within the sponge always occurred at the leading edge of infiltration. As the degree of infiltration increased, the volume of the leading edge decreased, resulting in a reduced volume of tissue containing high-density vasculature. Second, at later time points, remodeling of the vasculature occurred to form fewer, larger vessels (compare Figure 5C and 5D). It was therefore evident that in this model, chronic administration of hydralazine accelerated the angiogenic response, resulting in an acute period of elevated tissue vascularization.



View larger version (88K):
[in this window]
[in a new window]
  		Figure 5. Hydralazine induced angiogenesis in vivo. A, Hydralazine-induced sponge vascularization (solid line; 6 days *P<0.02, n=6) vs phosphate-buffered saline control (dashed line). Day 6 shown in B and C. CD34-immunostained blood vessels (arrows) infiltrating sponges (S) in hydralazine-treated animals versus controls. LE indicates leading edge of infiltration). Objectives x20 (B) and x40 (C) are shown. D, Day 14 showing development of the vasculature to form larger vessels [arrows, compare (B) and (D) (both x20 objective)].


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
We have demonstrated rapid and transient induction of HIF-1{alpha} by hydralazine in culture and in murine systems. No innate cell-type specificity was evident in hydralazine-mediated induction of HIF, although downstream consequences showed such effects. Hydralazine stimulated VEGF production in smooth muscle cells and cancer cell lines but not in EC. Conversely, it stimulated EC proliferation but not that of other cell types. It is recognized that EC do not generally induce VEGF,20,21 but VEGF production from the adjacent smooth muscle cell layer and other cells in vivo could modify EC function. Additionally, activation of the HIF transcriptional cascade upregulated a series of EC growth factors. Adrenomedullin,31 endothelin-1,32 and VEGF33 all induce EC proliferation and promote angiogenesis in vivo. HO-1 protects EC function under oxidative stress by catalyzing conversion of haem to biliverdin, which is cytoprotective versus reactive oxygen species, and to CO, which mediates vasodilation.34 Additionally, all these molecules induce cGMP35–38 and are therefore candidates for the intermediate molecule(s) proposed to regulate the hydralazine-induced accumulation of cGMP initiating smooth muscle relaxation in vivo.7

We have demonstrated that rapid stabilization and activation of the HIF system also occurs in vivo using standard doses28,29 modeling the clinical effect.30,39 HIF levels were sufficient to elevate tissue and plasma VEGF significantly 2 hours after injection, in concordance with maximal effects on vascular tone seen at 2 to 4 hours in murine models28,29 and humans.30,39 Thus, the time course of induction of the HIF system in vivo is compatible with that of established vascular effects of hydralazine. It is also of note that the target organs showing HIF induction by hydralazine are the same as those in which it is induced by hypoxia, namely the liver, heart, lungs, spleen, and kidney.40,41 Added to the in vitro data, this suggests the HIF system as a candidate for the effector pathway mediating actions of hydralazine in vivo.

We have identified the HIF{alpha}-regulating PHD enzymes as molecular targets for hydralazine. Hydralazine inhibited PHD enzyme activity in cell-free assays and prevented hydroxylation (and subsequent degradation) of HIF-1{alpha} in cultured cells. This started at a 50 µmol/L concentration with only small inhibition of PHD activity using either cell extracts or recombinant proteins (10% inhibition by HIF–VHL capture assay). It is worth noting that the iron concentrations required for the assays could affect the chelation capacity of hydralazine, especially at lower doses. Although 5 µmol/L (capture assay) is within the upper range of cellular free iron concentration, the 50 µmol/L iron in the 2-OG degradation assay might affect chelation. However, a clear dose response was evident in the 2-OG degradation assay and, using standard doses, stabilization of HIF-1{alpha} has been observed in animals, suggesting hydralazine is a sufficiently powerful chelator to bind enzyme-associated iron under physiological conditions and reduce PHD activity.

It is of interest to consider whether rapid activation of the HIF system in vascular cells could mediate the clinical effects of hydralazine. We have obtained direct evidence of inhibition of PHD activity, induction of HIF-1{alpha} and downstream genes, and pro-proliferative effects of hydralazine on HUVEC. These effects were greater with doses >50 µmol/L, but these are not necessary for induction of HIF protein or proliferation. In vivo there is increasing evidence of an association between endothelial dysfunction, hypertension, and cardiovascular disease. Impaired EC-dependent vasorelaxation caused by loss of NO activity in vessel walls can be caused by factors including abnormal regulation of endothelial NOS and NO degradation by reactive oxygen species.42 There is no documented evidence linking hydralazine with endothelial NOS and, unlike other hydrazine derivatives, hydralazine cannot be oxidized to release NO directly.43 However, hydralazine-mediated HIF pathway activation in EC may release NO indirectly (eg, via Et-1) and genes such as VEGF also modulate vascular tone, potentially explaining its acute vasodilatory effect.

We have also identified a novel effect of hydralazine that could explain the chronic benefits of long-term administration. Hydralazine induced EC proliferation in vitro and in vivo, where chronic systemic administration at doses modeling the clinical administration initiated a pro-angiogenic phenotype. In myocardial ischemia patients, recombinant VEGF has been shown to improve perfusion and coronary vessel density.44 Similarly, adrenomedullin infusion enhances coronary blood flow and cardiac function.45 It also promotes re-endothelialization and is antiapoptotic and cardioprotective in transgenic mice.46 Although we have only measured VEGF levels in vivo, it is likely that this angiogenic effect is mediated by several different growth factors. This is supported by data that mice overexpressing VEGF alone develop inflammation and vascular leakage, in contrast to the leakage-resistant hypervascularity induced by HIF-1{alpha} overexpression.47 Induction of an angiogenic phenotype may therefore represent a novel mechanism whereby chronic hydralazine administration reduces mortality in congestive heart failure.

Hypoxia is fundamental to many ischemic diseases including stroke, pulmonary hypertension, cancer, myocardial ischemia, and chronic lung disease. HIF mediates the pathophysiological response to hypoxia in several such conditions and modulation of its activity has been proposed as a therapeutic approach.15 Such a strategy raises questions regarding possible effects of systemic activation of the HIF system, which may enhance erythropoiesis and elevate glycolytic and angiogenic capacity in areas besides the target organ. None of these side effects is described for hydralazine. This may be due to the transient nature of hydralazine effects, because of rapid metabolism of the active drug, making it difficult to detect and determine pharmacokinetic properties clinically.18 One side effect of hydralazine is a lupus erythematosus-like syndrome characterized by antinuclear autoantibodies and connective tissue lesions. This could possibly represent activation of HIF downstream genes, such as collagen and fibronectin,48 but is rare at the lower doses of hydralazine currently used. Target specificity and low chronic toxicity are key in the design of novel agents activating HIF. We have demonstrated that an agent acting through rapid, short-term activation of the HIF pathway is clinically viable, suggesting that intermittent activation of HIF with rapid metabolism of the activator can result in vascular specificity.

In summary, we have identified a potential novel mechanism for the therapeutic effect of hydralazine in cardiovascular therapy. First, that hydralazine transiently activates the HIF system by inhibiting PHD enzyme activity. Second, that hydralazine promotes angiogenesis. Development of novel agents activating HIF could therefore be feasible for treatment of ischemic disease, with the potential to monitor target specificity through plasma markers of HIF such as VEGF and endothelin-1.


*    Acknowledgments
 
The work was supported by the Wellcome Trust and Cancer Research (UK). The authors are grateful to Professor Peter J. Ratcliffe for many helpful discussions.


*    Footnotes
 
Original received June 9, 2003; resubmission received April 21, 2004; revised resubmission received May 14, 2004; accepted May 27, 2004.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Gomberg-Maitland M, Baran DA, Fuster, V. Treatment of congestive heart failure: guidelines for the primary care physician and the heart failure specialist. Arc Intern Med. 2001; 161: 342–352.[Abstract/Free Full Text]
  2. Duley L, Henderson-Smart DJ. Drugs for treatment of very high blood pressure during pregnancy (Cochrane Review). Cochrane Database Syst Rev. 2002; CD001449.
  3. Ellershaw DC, Gurney AM. Mechanisms of hydralazine induced vasodilation in rabbit aorta and pulmonary artery. Br J Pharmacol. 2001; 134: 621–631.[CrossRef][Medline] [Order article via Infotrieve]
  4. Jacobs M. Mechanism of action of hydralazine on vascular smooth muscle. Biochem Pharmacol. 1984; 33: 2915–2919.[CrossRef][Medline] [Order article via Infotrieve]
  5. Bang L, Nielsen-Kudsk JE, Gruhn N, Trautner S, Theilgaard SA, Olesen S-P, Beosgaard S, Aldershvile J. Hydralazine-induced vasodilation involves opening of high conductance Ca2+-activated K+ channels. Eur J Pharmacol. 1998; 361: 43–49.[CrossRef][Medline] [Order article via Infotrieve]
  6. Hermsmeyer K, Trapani A, Abel PW, Worcel M. Effect of hydralazine on tension and membrane potential in the rat caudal artery. J Pharmacol Exp Ther. 1983; 227: 322–326.[Abstract/Free Full Text]
  7. Wei S, Kasuya Y, Yanagisawa M, Kimura S, Masaki T, Goto K. Studies on endothelium-dependent vasorelaxation by hydralazine in porcine coronary artery. Eur J Pharmacol. 1997; 321: 307–314.[CrossRef][Medline] [Order article via Infotrieve]
  8. Cohn JN, Archibald DG, Ziesche S, Franciosa JA, Harston WE, Tristani FE, Dunkman WB, Jacobs W, Francis GS, Flofr KH et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure. Results of a Veterans Administration Cooperative Study. N Engl J Med. 1986; 314: 1547–1552.[Abstract]
  9. Gogia H, Mehra A, Parikh S, Raman M, Ajit-Uppal J, Johnson JV, Elkayam U. Prevention of tolerance to hemodynamic effects of nitrates with concomitant use of hydralazine in patients with chronic heart failure. J Am Coll Cardiol. 1995; 26: 1575–1580.[Abstract]
  10. Munzel T, Kurz S, Rajagopalan S, Thoenes M, Berrington WR, Thompson JA, Freeman BA, Harrison DG. Hydralazine prevents nitroglycerin tolerance by inhibiting activation of a membrane-bound NADH oxidase—a new action for an old drug. J Clin Invest. 1996; 98: 1465–1470.[Medline] [Order article via Infotrieve]
  11. Bhatnagar RS, Rapaka SSR, Liu TZ, Wolfe SM. Hydralazine-induced disturbances in collagen biosynthesis. Biochim Biophys Acta. 1972; 271: 125–132.[Medline] [Order article via Infotrieve]
  12. Murad S, Tajima S, Pinnell SR. A paradoxical effect of hydralazine on prolyl and lysyl hydroxylase activities in cultured human skin fibroblasts. Arch Biochem Biophys. 1985; 241: 356–363.[CrossRef][Medline] [Order article via Infotrieve]
  13. Epstein ACR, Gleadle JM, McNeill LA, Hewitson KS, O’Rourke J, Mole DR, Mukherji M, Metzen E, Wilson MI, Dhanda A, Tian YM, Masson N, Hamilton DL, Jaakkola P, Barstead R, Hodgkin J, Maxwell PH, Pugh CW, Schofield CJ, Ratcliffe PJ. C. elegans EGL-9 and mammalian homologues define a family of dioxygenases that regulate HIF by prolyl hydroxylation. Cell. 2001; 107: 42–54.
  14. Bruick KB, McKnight SL. A conserved family of prolyl-4-hydroxylases that modify HIF. Science. 2001; 294: 1337–1340.[Abstract/Free Full Text]
  15. Semenza GL. Hypoxia-inducible factor 1: control of oxygen homeostasis in health and disease. Pediat Res. 2001; 49: 614–617.[Medline] [Order article via Infotrieve]
  16. Jaakkola P, Mole DR, Tian Y-M, Wilson MI, Gielbert J, Gaskell SJ, Kriegsheim AV, Hebestreit HF, Mukherji M, Schofield CJ, Maxwell PH, Pugh CW, Ratcliffe PJ. Targeting of HIF to the von Hippel-Lindau ubiquitylation complex by O2-regulated prolyl hydroxylation. Science. 2001; 292: 468–472.[Abstract/Free Full Text]
  17. D’Angelo G, Duplan E, Boyer N, Vigne P, Frelin C. Hypoxia up-regulates prolyl hydroxylase activity. A feedback mechanism that limits HIF-1 responses during reoxygenation. J Biol Chem. 2003; 278: 38183–38187.[Abstract/Free Full Text]
  18. Reece PA, Cozamanis I, Zacest R. Kinetics of hydralazine and its main metabolites in slow and fast acetylators. Clin Pharmacol Ther. 1980; 28: 769–778.[Medline] [Order article via Infotrieve]
  19. Sim E, Payton M, Noble M, Minchin R. An update on genetic, structural and functional studies on arylamine N-acetyltransferases in eucaryotes and procaryotes. Human Mol Genet. 2000; 9: 2435–2441.[Abstract/Free Full Text]
  20. Minchenko A, Bauer T, Salceda S, Caro J. Hypoxic stimulation of vascular endothelial growth factor expression in vitro and in vivo. Lab Invest. 1994; 71: 374–379.[Medline] [Order article via Infotrieve]
  21. Marti HH, Risau W. Systemic hypoxia changes the organ-specific distribution of vascular endothelial growth factor and its receptors. Proc Natl Acad Sci U S A. 1998; 95: 15809–15814.[Abstract/Free Full Text]
  22. Al-Falahi H, May PM, Roe AM, Slater RA, Trott WJ, Williams DR. Metal-binding by pharmaceuticals. Part 4. A comparative investigation of the interaction of metal ions with hydralazine, prizidilol and related compounds. Agents Actions. 1984; 14: 113–120.[CrossRef][Medline] [Order article via Infotrieve]
  23. Palmer LA, Semenza GL, Stoler MH, Johns RA. Hypoxia induces type II NOS gene expression in pulmonary artery endothelial cells via HIF-1. Am J Physiol. 1998; 274: L212–L219.[Medline] [Order article via Infotrieve]
  24. Gattullo D, Pagliaro P, Marsh NA, Losano G. New insights into nitric oxide and coronary circulation. Life Sci. 1999; 65: 2167–2174.[CrossRef][Medline] [Order article via Infotrieve]
  25. Saito S, Hirata Y, Imai T, Marumo F. Autocrine regulation of the endothelin-1 gene in rat endothelial cells. J Cardiovasc Pharmacol. 1995; 26: S84–S87.[Medline] [Order article via Infotrieve]
  26. Hartsfield CL, Alam J, Cook JL, Choi AMK. Regulation of heme oxygenase-I gene expression in vascular smooth muscle cells by nitric oxide. Am J Physiol. 1997; 273: L980–L988.[Medline] [Order article via Infotrieve]
  27. Le NTV, Richardson DR. The role of iron in cell cycle progression and the proliferation of neoplastic cells. Biochim Biophys Acta. 2002; 1603: 31–46.[Medline] [Order article via Infotrieve]
  28. Fisker RV, Horsman MR, Overgaard J. Hydralazine-induced changes in tissue perfusion and radiation response in a C3H mammary carcinoma and mouse normal tissues. Acta Oncologica. 1991; 30: 641–647.[Medline] [Order article via Infotrieve]
  29. Honess DJ, Bleehen NM. Comparative effects of hydralazine on perfusion of KHT tumour, kidney and liver and renal function in mice. Int J Radiat Oncol Biol Phys. 1992; 22: 953–961.[Medline] [Order article via Infotrieve]
  30. Leier CB, Magorien RD, Desch CE, Thompson MJ, Unverferth DV. Hydralazine and isosorbide dinitrate: comparative central and regional haemodyamic effects when administered alone or in combination. Circulation. 1981; 634: 102–109.
  31. Oehler MK, Hague S, Rees MCP, Bicknell R. Adrenomedullin promotes formation of xenografted endometrial tumours by stimulation of autocrine growth and angiogenesis. Oncogene. 2002; 21: 2815–2821.[CrossRef][Medline] [Order article via Infotrieve]
  32. Salani D, Taraboletti G, Rosano L, Di Castro V, Borsotti P, Giavazzi R, Bagnato A. Endothelin-1 induces an angiogenic phenotype in cultured endothelial cells and stimulates neovascularisation in vivo. Am J Path. 2000; 157: 1703–1711.[Abstract/Free Full Text]
  33. Farrara N. Role of vascular endothelial growth factor in physiologic and pathologic angiogenesis: therapeutic implications. Semin Oncol. 2002; 29: S10–S14.
  34. Deramaudt BMJM, Braunstein S, Remy P, Abraham NG. Gene transfer of human haem oxygenase into coronary endothelial cells potentially promotes angiogenesis. J Cell Biochem. 1998; 68: 121–127.[CrossRef][Medline] [Order article via Infotrieve]
  35. Ali N, Yousufzai SYK, Abdel-Latif AA. Activation of particulate guanylate cyclase by adrenomedullin in cultured SV-40 transformed cat iris sphincter smooth muscle (SV-CISM-2) cells. Cellular Signalling. 2000; 12: 491–498.[CrossRef][Medline] [Order article via Infotrieve]
  36. Duckers HJ, Boehm M, Trye AL, Yet SF, San H, Park JL, Webb CR, Lee ME, Nabel GJ, Nabel EG. Heme oxygenase-1 protects against vascular constriction and proliferation. Nat Med. 2001; 7: 693–698.[CrossRef][Medline] [Order article via Infotrieve]
  37. Ding K-H, Latimer AJ, Abdel-Latif AA. Activation of particulate guanylate cyclase by endothelins in cultured SV-40 transformed cat iris sphincter smooth muscle cells. Life Sci. 1999; 64: 161–174.[Medline] [Order article via Infotrieve]
  38. Kroll J, Waltenberger J. A novel function of VEGF receptor-2 (KDR): rapid release of nitric oxide in response to VEGF-A stimulation in endothelial cells. Biochem Biophys Res Commun. 1999; 30: 636–639.
  39. Toppozada MK, Darwish E, Barakat AA. Management of severe preeclampsia detected in early labor by prostaglandin A1 or dihydralazine infusions. Am J Obstet Gynecol. 1991; 164: 1229–1232.[Medline] [Order article via Infotrieve]
  40. Stroka DM, Burkhardt T, Desbaillets I, Wenger RH, Neil DAH, Bauer C, Gassmann M, Candinas D. HIF-1 is expressed in normoxic tissue and displays an organ-specific regulation under systemic hypoxia. FASEB J. 2001; 15: 2445–2453.[Abstract/Free Full Text]
  41. Yu AY, Frid MG, Shimoda LA, Wiener CM, Stenmark K, Semenza GL. Temporal, spatial and oxygen-regulated expression of hypoxia-inducible factor-1 in the lung. Am J Physiol. 1998; 275: L818–L826.[Medline] [Order article via Infotrieve]
  42. Cai H, Harrison DG. Endothelial dysfunction in cardiovascular diseases. The role of oxidant stress. Circ Res. 2000; 87: 840–844.[Abstract/Free Full Text]
  43. Rehse K, Shahrouri T. New NO donors with antithrombotic and vasodilating activities, part 24. Hydrazine derivatives. Arch Pharm Pharm Med Chem. 1998; 331: 308–312.[CrossRef]
  44. Henry TD, Rocha-Singh K, Isner JM, Kereiakes DJ, Giordano FJ, Simons M, Losordo DW, Hendel RC, Bonow RO, Eppler SM, Zioncheck TF, Holmgren EB, McCluskey ER. Intracoronary administration of recombinant human vascular endothelial growth factor to patients with coronary artery disease. Am Heart J. 2001; 142: 872–880.[CrossRef][Medline] [Order article via Infotrieve]
  45. Nagaya N, Goto Y, Satoh T, Sumida H, Kojima S, Miyatake K, Kangawa K. Intravenous adrenomedullin in myocardial function and energy metabolism in patients after myocardial infarction. J Cardiovasc Pharmacol. 2002; 39: 754–760.[CrossRef][Medline] [Order article via Infotrieve]
  46. Chao J, Kato K, Zhang JJ, Dobrzynski E, Wang C, Agata J, Chao L. Human adrenomedullin gene delivery protects against cardiovascular remodelling and renal injury. Peptides. 2001; 22: 1731–1737.[CrossRef][Medline] [Order article via Infotrieve]
  47. Elson DA, Thurston G, Huang LE, Ginzinger DG, McDonald DM, Johnson RL, Arbeit JM. Induction of hypervascularity without leakage or inflammation in transgenic mice overexpressing hypoxia-inducible factor-1{alpha}. Genes Dev. 2001; 15: 2520–2532.[Abstract/Free Full Text]
  48. Pfander D, Cramer T, Schipani E, Johnson RS. HIF-1 alpha controls extracellular matrix synthesis by epiphyseal chondrocytes. J Cell Sci. 2003; 116: 1819–1826.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Exp PhysiolHome page
C. Liu, G. M. Balanos, M. Fatemian, T. G. Smith, K. L. Dorrington, and P. A. Robbins
Effects of hydralazine on the pulmonary vasculature and respiratory control in humans
Exp Physiol, January 1, 2008; 93(1): 104 - 114.
[Abstract] [Full Text] [PDF]

Home page
 Cancer Res.Home page
D. C. Birle and D. W. Hedley
Suppression of the Hypoxia-Inducible Factor-1 Response in Cervical Carcinoma Xenografts by Proteasome Inhibitors
Cancer Res., February 15, 2007; 67(4): 1735 - 1743.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
K. Takeda and G.-H. Fong
Prolyl Hydroxylase Domain 2 Protein Suppresses Hypoxia-Induced Endothelial Cell Proliferation
Hypertension, January 1, 2007; 49(1): 178 - 184.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
M. Zampino, M. Yuzhakova, J. Hansen, R. D. McKinney, P. H. Goldspink, D. L. Geenen, and P. M. Buttrick
Sex-related dimorphic response of HIF-1{alpha} expression in myocardial ischemia
Am J Physiol Heart Circ Physiol, August 1, 2006; 291(2): H957 - H964.
[Abstract] [Full Text] [PDF]

Home page
 Am. J. Physiol. Heart Circ. Physiol.Home page
S. Philipp, L. Cui, B. Ludolph, M. Kelm, R. Schulz, M. V. Cohen, and J. M. Downey
Desferoxamine and ethyl-3,4-dihydroxybenzoate protect myocardium by activating NOS and generating mitochondrial ROS
Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H450 - H457.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
P. Brassard, F. Amiri, G. Thibault, and E. L. Schiffrin
Role of Angiotensin Type-1 and Angiotensin Type-2 Receptors in the Expression of Vascular Integrins in Angiotensin II-Infused Rats
Hypertension, January 1, 2006; 47(1): 122 - 127.
[Abstract] [Full Text] [PDF]

Home page
 HypertensionHome page
P. Brassard, F. Amiri, and E. L. Schiffrin
Combined Angiotensin II Type 1 and Type 2 Receptor Blockade on Vascular Remodeling and Matrix Metalloproteinases in Resistance Arteries
Hypertension, September 1, 2005; 46(3): 598 - 606.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
95/2/162    most recent
01.RES.0000134924.89412.70v1
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow Request Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar