| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
(Hypertension. 1997;30:1397-1402.)
© 1997 American Heart Association, Inc.
Articles |
From the Department of Medicine, Divisions of Cardiology (Q.C., R.W.A., P.L., N.I., A.B.H., W.R.T.) and Hematology (H.deL., J.N.W.), Emory University School of Medicine, and Atlanta Veterans Administration Hospital (P.L., W.R.T.), Atlanta, Ga.
Correspondence to W. Robert Taylor, MD, PhD, Emory University School of Medicine, Division of Cardiology, Woodruff Memorial Research Building, Suite 319, 1639 Pierce Dr, Atlanta, Ga 30322.
| Abstract |
|---|
|
|
|---|
Key Words: angiotensin II monocytes macrophages atherosclerosis norepinephrine
| Introduction |
|---|
|
|
|---|
We sought to determine the effect of hypertension on MCP-1 expression in the vascular wall. We analyzed MCP-1 mRNA levels in the aortas of rats made hypertensive by angiotensin II or norepinephrine infusion and compared these with levels in angiotensin IIinfused rats treated with the nonspecific vascular smooth muscle relaxant hydralazine. To gain further mechanistic insight, we mechanically simulated the hypertensive environment in vitro by exposing cultured rat aortic smooth muscle cells to mechanical deformation and measured the effect of cyclic strain on MCP-1 mRNA expression.
| Methods |
|---|
|
|
|---|
Experimental Models of Hypertension and Antihypertensive
Therapy
Male Sprague-Dawley rats (weight, 250 to 300 g) received
angiotensin II (0.75 mg · kg-1 ·
d-1) or norepinephrine (2.8 mg ·
kg-1 · d-1) infusions from implanted
osmotic minipumps for 7 days (Alzet model 2002). Rats were
anesthetized with 10 mg/kg xylazine IP and 80
mg/kg ketamine hydrochloride IP. For the
angiotensin IItreated animals, an osmotic pump containing
angiotensin II dissolved in a solution of 0.15 mol/L
NaCl and 0.01 mol/L acetic acid was placed into the subcutaneous
space. For the norepinephrine experiments, the pumps were
connected to a polyethylene catheter placed in the internal jugular
vein. Control animals were either sham-operated with no pump implanted
or implanted with pumps containing vehicle only. The skin was closed
with surgical staples. Systolic blood pressure was measured at
baseline and before death with a tail-cuff
sphygmomanometer.4 Measurements were done in triplicate,
and a mean value was obtained.
Additional rats were treated with antihypertensive drugs. Losartan, a nonpeptide, competitive antagonist of the angiotensin II AT1 receptor (25 mg · kg-1 · d-1), or the nonspecific vascular smooth muscle relaxant hydralazine (15 mg · kg-1 · d-1), was administered in the drinking water. Antihypertensive drug treatment was initiated 24 hours before pump implantation.
Tissue Harvesting and RNA Preparation
Sprague-Dawley rats were killed 7 days after pump implantation
by CO2 narcosis. A midline ventral incision was made, and
with the heart still beating, an intracardiac injection of 5000 U
heparin was administered. The aorta was then dissected from the renal
arteries to the ascending arch, removed en bloc, and placed in ice-cold
PBS, pH 7.4. Extravascular tissue was removed rapidly with forceps, and
the vessel lumen was flushed sequentially with heparin and PBS. The
specimen was placed in a precooled microcentrifuge tube,
immediately submerged in liquid nitrogen, and transferred to a -70°C
freezer for storage. Aortas were homogenized in TRI Reagent
using a Polytron homogenizer (Brinkmann Instruments).
RNA extraction was achieved using the TRI reagent method as described
previously.5 Then, 10 µg total RNA per specimen was
loaded onto a formaldehyde-containing 1% agarose gel.
Northern Blotting and Hybridization
After size-fractionation on a denaturing agarose-formaldehyde
gel, total RNA was transferred to a nylon membrane (MSI, Inc) and
exposed to ultraviolet light to cross-link the RNA. Membranes were
prehybridized for 2 hours at 42°C with prehybridization buffer (50%
[vol/vol] formamide, 1 mol/L NaCl, 5x Denhardt's
solution, 0.5% sodium dodecyl sulfate, 50 mmol/L
Tris buffer, pH 7.4, and 100 µg/mL denatured salmon testes
DNA). Subsequently, membranes were hybridized overnight with a
random-primed 32P-dCTP-labeled rat cDNA probe. Membranes
were then washed two to four times for 15 minutes in 0.5x standard
saline citrate and 0.1% sodium dodecyl sulfate at 55°C and
exposed to x-ray film at -70°C for 24 to 48 hours.
Autoradiographic bands were quantified via densitometry and
normalized to 28S ribosomal RNA.
Immunohistochemistry
Macrophages were identified with a mouse anti-rat
macrophage antibody (ED-1) at a 1:250 dilution. Frozen
paraformaldehyde-fixed tissue sections were thawed and
fixed in acetone for 5 minutes, dried, and rehydrated in PBS. The
primary antibody was applied at the indicated dilution in 1.0% BSA in
PBS and incubated with a biotinylated secondary antibody (horse
anti-mouse IgG at a 1:400 dilution) in PBS containing 1.0% BSA and
2.0% normal horse serum for 30 minutes at room temperature. This was
followed by washing in PBS and incubation with the avidin-biotin enzyme
complex and chromogenic substrate (Vector Red) as described
by the manufacturer.
Stretch Experiments
VSMCs were isolated from the thoracic aortas of rats as
described previously.6 Cells were grown in DMEM
supplemented with 4500 mg/L D-glucose, 2
mmol/L L-glutamine, 100 U/mL penicillin, 100
µg/mL streptomycin, 25 mmol/L HEPES, and 10%
heat-inactivated calf serum. Cells were plated onto
flexible collagen-coated membranes (Flexcell International Corp) and
fed every other day until near confluent. Wells were washed twice with
serum-free DMEM and placed in 0.1% serum DMEM for 72 hours. The
flexible membranes were stretched by vacuum using the Flexercell Strain
Unit (model FX-2000).7 The degree of deformation of the
membrane is a nonlinear function of the degree of negative pressure
applied, and the strain gradient is a nonlinear function of radial, but
not axial, position. A negative pressure of 13 kPa was applied beneath
the membrane (radius, 12.7 mm) for all experiments, corresponding
to a maximum deformation of 20% at
9.5 mm from the center of
the well, decreasing exponentially toward the center and the periphery
of the well. The strain unit was housed in an incubator at 5%
CO2 in humidified air at 37°C.
| Results |
|---|
|
|
|---|
|
After 7 days of angiotensin II infusion, MCP-1 mRNA levels
were measured with the use of quantitative Northern blot
analysis. As can be seen in Fig 2
, there was a significant 259±26%
increase in MCP-1 mRNA expression in rats made hypertensive with
angiotensin II compared with controls.
|
To gain further mechanistic insights into the signaling mechanisms
responsible for increased MCP-1 expression in angiotensin
IIinduced hypertension, we studied the effects of several different
antihypertensive agents on MCP-1 expression in hypertensive rats. As
can be seen in Fig 2
, we found that losartan caused a complete
inhibition of both the hypertensive response (107±10% inhibition) and
the associated increase in MCP-1 expression (95±4% inhibition). These
data are consistent with an AT1 receptormediated
process.
The nonselective vasodilator hydralazine also caused a
significant, but incomplete, inhibition of increased MCP-1 mRNA
expression in angiotensin IItreated rats (56±19%
inhibition). This occurred despite complete normalization of the blood
pressure response (98±9% inhibition; Fig 3
). These data suggest that in addition
to the direct humoral effects of angiotensin II, the
resultant mechanical effects of blood pressure elevation may play an
important role in the regulation of aortic MCP-1 expression.
|
To test this hypothesis, two different approaches were taken. First, we
used another model of hypertension to exclude direct effects of
angiotensin II. Using a norepinephrine model of
hypertension, we found that there was a similar increase in aortic
MCP-1 expression (Fig 4
). Normalization
of blood pressures with hydralazine completely eliminated the
upregulation of MCP-1 expression. We also used an in vitro system to
directly test the effects of cell deformation on MCP-1 expression. We
exposed cultured rat aortic smooth muscle cells to a maximal 20%
cyclic deformation at 1 Hz. Fig 5
shows a
representative Northern blot depicting the time course
of MCP-1 upregulation by mechanical deformation that demonstrates a
rapid increase in MCP-1 expression.
|
|
| Discussion |
|---|
|
|
|---|
The results presented here may help to explain previous studies demonstrating an arterial inflammatory response in hypertension. Clozel et al9 have shown that in the cerebral vasculature of spontaneously hypertensive rats, monocyte/macrophage infiltration of the arterial wall occurs during the development of hypertension. Similar results have been seen in the Dahl salt-sensitive model of hypertension10 and in spontaneously hypertensive rats.11 In this study, we have shown a marked inflammatory response of the arterial wall in hypertension that is characterized by marked infiltration of the adventitia with inflammatory cells. Our findings of upregulated MCP-1 expression provide a potential mechanism for the recruitment of monocytes. Additional mechanisms involving the upregulation of adhesion molecules are also likely to be involved in the development of the inflammatory response. The ultimate importance of a monocytic infiltration of the vascular wall in hypertension remains to be determined; however, monocytes/macrophages have been implicated in the regulation of vascular proliferation and vascular remodeling. The recruitment and differentiation of monocytes within the vascular wall may be a pivotal step in the development of both adaptive and maladaptive changes in hypertensive arteries.
The prominence of the monocyte/macrophage infiltration of the adventitia is indeed striking in that it resembles the response of the adventitia that is seen in arteritis, in atherosclerosis,12 and after balloon injury.13 In atherosclerosis, the degree of adventitial infiltration corresponds with more advanced vascular lesions.12 Although the presence of a cause-and-effect relationship between adventitial inflammation and intimal vascular disease in atherosclerosis has yet to be demonstrated, it has been shown that adventitial injury or removal results in endothelial cell loss and the subsequent development of neointimal hyperplasia.14,15 We observed only occasional inflammatory cells that appeared to be adherent to the endothelial surface. Given the techniques used here, it would not be possible to accurately assess whether this was altered in the hypertensive animals.
Previous reports have documented an increased presence of MCP-1 in human and experimental atherosclerotic plaques,2,3 with most of these studies identifying VSMCs and macrophages as the major sources.3 Cell culture studies have determined that MCP-1 accounts for a large portion of the monocyte chemoattractant activity produced by VSMCs.16 Thus, VSMC-derived MCP-1 is likely to play a major role in the early pathogenesis of vascular lesions and might logically be expected to be increased in disorders known to influence the development of atherosclerosis. Indeed, this has already been shown in a model of hypercholesterolemia.3 We have shown a significant upregulation of MCP-1 mRNA in the aortic walls of rats made hypertensive by angiotensin II. Because hypertension is a well known risk factor for the development of atherosclerosis in humans, the present study has implications for understanding this relationship on a fundamental level.
Considerable controversy exists regarding the relative importance of humoral versus hemodynamic stimuli as the predominant signal for the adverse effects of hypertension on arterial wall. Alderman et al17 found that humans with hypertension characterized by high circulating levels of renin and angiotensin II have three times the myocardial infarction risk of hypertensive patients without high renin levels, a relationship that persisted after controlling for conventional risk factors. Rajagopalan et al18 recently demonstrated that rats with angiotensin II, but not catecholamine-, induced hypertension have impaired endothelium-dependent vasodilatation and increased vascular superoxide production. Both of these studies suggest that the blood pressure elevation induced by angiotensin II (ie, the mechanical stimulus) is not as important as other actions of angiotensin II in mediating the deleterious effects of hypertension on the vasculature. However, studies using aortic coarctation models of hypertension have shown that the arterial segment proximal to the coarctation, and thus subject to increased hemodynamic stress, is more prone vascular hypertrophy than the low-pressure segment below the stenosis,19 suggesting a more important role for hemodynamic forces in the vascular pathology associated with high renin hypertension. In the present study, we reduced blood pressure in angiotensin IIinfused rats using either the specific angiotensin II receptor antagonist losartan or the nonspecific vasodilator hydralazine. This enabled us to compare the effects of blood pressure lowering via angiotensin II antagonism with vasorelaxation in the presence of persistent angiotensin II signaling. Both strategies were effective in inhibiting MCP-1 induction, although the inhibition by hydralazine was incomplete despite normalization of blood pressure. Therefore, angiotensin II may have a direct effect on MCP-1 expression, as suggested by others.20 In the norepinephrine-treated animals, we found that not only did norepinephrine cause a similar increase in MCP-1 expression but also normalization of the blood pressure with hydralazine completely inhibited the increased MCP-1 expression. Thus, in regard to MCP-1 mRNA regulation in hypertension, we can conclude that both humoral and mechanical mechanisms appear to be involved and that in the specific case of angiotensin IIinduced hypertension, elevated blood pressure and its consequences appear to be as important as the nonpressor effects of angiotensin II.
Hydralazine may have chemical properties in addition to its vasodilator capacity. Munzel et al21 recently demonstrated that hydralazine is capable of inhibiting membrane NADH oxidase activity in arterial segments. Thus, it is possible that administration of hydralazine in these experiments was in effect administration of an antioxidant. This is potentially important because aortic membraneassociated NADH oxidase is upregulated in angiotensin IIinduced hypertension22 and MCP-1 expression is upregulated by reactive oxygen species.23 However, our experiments showing a direct upregulation of MCP-1 expression by mechanical strain and our studies with the norepinephrine-treated rats that do not produce excess superoxide24 argue against this possibility.
A cardinal feature of hypertension is increased pressure and strain on
the arterial wall, which is transmitted to
endothelial and smooth muscle cells. Recent advances in
cell biology have allowed investigators to study cells cultured in
mechanical environments that more closely mimic the in vivo
situation.7,25 Cyclic strain, a force experienced by cells
in the arterial wall, induced a brisk upregulation of MCP-1
mRNA that was sustained for
4 hours. This finding is
consistent with our in vivo experiments and implicates
mechanical forces in the upregulation of this chemokine.
In summary, hypertension is associated with a marked inflammatory response and an upregulation of MCP-1 mRNA in the aortic wall. Elevated blood pressure, and the resultant mechanical strain on the aortic smooth muscle cells, appears to be as important as direct angiotensin II signaling in the upregulation of MCP-1, thus underscoring the importance of hemodynamic forces in angiotensin IIinduced hypertension. MCP-1 may subsequently play a central role in the development of hypertensive vascular disease through its effect on the recruitment of monocytes into the vascular wall. In addition, the known association between MCP-1 and atherosclerotic lesions provides potential mechanistic insights into the molecular basis for hypertension as a risk factor for the development of atherosclerosis.
| Selected Abbreviations and Acronyms |
|---|
|
| Acknowledgments |
|---|
Received July 9, 1997; first decision July 15, 1997; accepted July 15, 1997.
| References |
|---|
|
|
|---|
B
activation, monocyte chemoattractant protein-1 expression, and
macrophage infiltration in a rabbit model of early accelerated
atherosclerosis. Circulation. 1997;95:15321541.This article has been cited by other articles:
![]() |
J. L. M. Madrigal, J. C. Leza, P. Polak, S. Kalinin, and D. L. Feinstein Astrocyte-Derived MCP-1 Mediates Neuroprotective Effects of Noradrenaline J. Neurosci., January 7, 2009; 29(1): 263 - 267. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J Allan, J. Beattie, and D. J Flint Epithelial injury induces an innate repair mechanism linked to cellular senescence and fibrosis involving IGF-binding protein-5 J. Endocrinol., November 1, 2008; 199(2): 155 - 164. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. Klanke, N. Cordasic, A. Hartner, R. E. Schmieder, R. Veelken, and K. F. Hilgers Blood pressure versus direct mineralocorticoid effects on kidney inflammation and fibrosis in DOCA-salt hypertension Nephrol. Dial. Transplant., November 1, 2008; 23(11): 3456 - 3463. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. Schober Chemokines in Vascular Dysfunction and Remodeling Arterioscler. Thromb. Vasc. Biol., November 1, 2008; 28(11): 1950 - 1959. [Abstract] [Full Text] [PDF] |
||||
![]() |
T.-D. Liao, X.-P. Yang, Y.-H. Liu, E. G. Shesely, M. A. Cavasin, W. A. Kuziel, P. J. Pagano, and O. A. Carretero Role of Inflammation in the Development of Renal Damage and Dysfunction in Angiotensin II-Induced Hypertension Hypertension, August 1, 2008; 52(2): 256 - 263. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. Weiss and W. R. Taylor Deoxycorticosterone Acetate Salt Hypertension in Apolipoprotein E-/- Mice Results in Accelerated Atherosclerosis: The Role of Angiotensin II Hypertension, February 1, 2008; 51(2): 218 - 224. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. A. Elmarakby, J. E. Quigley, J. J. Olearczyk, A. Sridhar, A. K. Cook, E. W. Inscho, D. M. Pollock, and J. D. Imig Chemokine Receptor 2b Inhibition Provides Renal Protection in Angiotensin II Salt Hypertension Hypertension, December 1, 2007; 50(6): 1069 - 1076. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. Maiellaro and W. R. Taylor The role of the adventitia in vascular inflammation Cardiovasc Res, September 1, 2007; 75(4): 640 - 648. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. J. Haurani and P. J. Pagano Adventitial fibroblast reactive oxygen species as autacrine and paracrine mediators of remodeling: Bellwether for vascular disease? Cardiovasc Res, September 1, 2007; 75(4): 679 - 689. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Riou, B. Mees, B. Esposito, R. Merval, J. Vilar, D. Stengel, E. Ninio, R. van Haperen, R. de Crom, A. Tedgui, et al. High Pressure Promotes Monocyte Adhesion to the Vascular Wall Circ. Res., April 27, 2007; 100(8): 1226 - 1233. [Abstract] [Full Text] [PDF] |
||||
![]() |
K. R. Stenmark, N. Davie, M. Frid, E. Gerasimovskaya, and M. Das Role of the Adventitia in Pulmonary Vascular Remodeling Physiology, April 1, 2006; 21(2): 134 - 145. [Abstract] [Full Text] [PDF] |
||||
![]() |
P. Pauletto and M. Rattazzi Inflammation and hypertension: the search for a link Nephrol. Dial. Transplant., April 1, 2006; 21(4): 850 - 853. [Full Text] [PDF] |
||||
![]() |
A. Tedgui and Z. Mallat Cytokines in Atherosclerosis: Pathogenic and Regulatory Pathways Physiol Rev, April 1, 2006; 86(2): 515 - 581. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. G. Lakoski, M. Cushman, W. Palmas, R. Blumenthal, R. B. D'Agostino Jr, and D. M. Herrington The Relationship Between Blood Pressure and C-Reactive Protein in the Multi-Ethnic Study of Atherosclerosis (MESA) J. Am. Coll. Cardiol., November 15, 2005; 46(10): 1869 - 1874. [Abstract] [Full Text] [PDF] |
||||
![]() |
N. Ishizaka, K. Saito, I. Mori, G. Matsuzaki, M. Ohno, and R. Nagai Iron Chelation Suppresses Ferritin Upregulation and Attenuates Vascular Dysfunction in the Aorta of Angiotensin II-Infused Rats Arterioscler. Thromb. Vasc. Biol., November 1, 2005; 25(11): 2282 - 2288. [Abstract] [Full Text] [PDF] |
||||
![]() |
D. M. Shah Role of the renin-angiotensin system in the pathogenesis of preeclampsia Am J Physiol Renal Physiol, April 1, 2005; 288(4): F614 - F625. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. Gruden, G. Setti, A. Hayward, D. Sugden, S. Duggan, D. Burt, R. E. Buckingham, L. Gnudi, and G. Viberti Mechanical Stretch Induces Monocyte Chemoattractant Activity via an NF-{kappa}B-Dependent Monocyte Chemoattractant Protein-1-Mediated Pathway in Human Mesangial Cells: Inhibition by Rosiglitazone J. Am. Soc. Nephrol., March 1, 2005; 16(3): 688 - 696. [Abstract] [Full Text] [PDF] |
||||
![]() |
H. M. Dourron, G. M. Jacobson, J. L. Park, J. Liu, D. J. Reddy, M. L. Scheel, and P. J. Pagano Perivascular gene transfer of NADPH oxidase inhibitor suppresses angioplasty-induced neointimal proliferation of rat carotid artery Am J Physiol Heart Circ Physiol, February 1, 2005; 288(2): H946 - H953. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Liu, A. Ormsby, N. Oja-Tebbe, and P. J. Pagano Gene Transfer of NAD(P)H Oxidase Inhibitor to the Vascular Adventitia Attenuates Medial Smooth Muscle Hypertrophy Circ. Res., September 17, 2004; 95(6): 587 - 594. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Ishibashi, K.-i. Hiasa, Q. Zhao, S. Inoue, K. Ohtani, S. Kitamoto, M. Tsuchihashi, T. Sugaya, I. F. Charo, S. Kura, et al. Critical Role of Monocyte Chemoattractant Protein-1 Receptor CCR2 on Monocytes in Hypertension-Induced Vascular Inflammation and Remodeling Circ. Res., May 14, 2004; 94(9): 1203 - 1210. [Abstract] [Full Text] [PDF] |
||||
![]() |
F. Kuwahara, H. Kai, K. Tokuda, M. Takeya, A. Takeshita, K. Egashira, and T. Imaizumi Hypertensive Myocardial Fibrosis and Diastolic Dysfunction: Another Model of Inflammation? Hypertension, April 1, 2004; 43(4): 739 - 745. [Abstract] [Full Text] [PDF] |
||||
![]() |
W. Ni, S. Kitamoto, M. Ishibashi, M. Usui, S. Inoue, K.-i. Hiasa, Q. Zhao, K.-i. Nishida, A. Takeshita, and K. Egashira Monocyte Chemoattractant Protein-1 Is an Essential Inflammatory Mediator in Angiotensin II-Induced Progression of Established Atherosclerosis in Hypercholesterolemic Mice Arterioscler. Thromb. Vasc. Biol., March 1, 2004; 24(3): 534 - 539. [Abstract] [Full Text] |
||||
![]() |
K. Tokuda, H. Kai, F. Kuwahara, H. Yasukawa, N. Tahara, H. Kudo, K. Takemiya, M. Koga, T. Yamamoto, and T. Imaizumi Pressure-Independent Effects of Angiotensin II on Hypertensive Myocardial Fibrosis Hypertension, February 1, 2004; 43(2): 499 - 503. [Abstract] [Full Text] [PDF] |
||||
![]() |
G. J. Blake, N. Rifai, J. E. Buring, and P. M Ridker Blood Pressure, C-Reactive Protein, and Risk of Future Cardiovascular Events Circulation, December 16, 2003; 108(24): 2993 - 2999. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. Hayashidani, H. Tsutsui, T. Shiomi, M. Ikeuchi, H. Matsusaka, N. Suematsu, J. Wen, K. Egashira, and A. Takeshita Anti-Monocyte Chemoattractant Protein-1 Gene Therapy Attenuates Left Ventricular Remodeling and Failure After Experimental Myocardial Infarction Circulation, October 28, 2003; 108(17): 2134 - 2140. [Abstract] [Full Text] [PDF] |
||||
![]() |
M. Higashi, H. Shimokawa, T. Hattori, J. Hiroki, Y. Mukai, K. Morikawa, T. Ichiki, S. Takahashi, and A. Takeshita Long-Term Inhibition of Rho-Kinase Suppresses Angiotensin II-Induced Cardiovascular Hypertrophy in Rats In Vivo: Effect on Endothelial NAD(P)H Oxidase System Circ. Res., October 17, 2003; 93(8): 767 - 775. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. Liu, F. Yang, X.-P. Yang, M. Jankowski, and P. J. Pagano NAD(P)H Oxidase Mediates Angiotensin II-Induced Vascular Macrophage Infiltration and Medial Hypertrophy Arterioscler. Thromb. Vasc. Biol., May 1, 2003; 23(5): 776 - 782. [Abstract] [Full Text] [PDF] |
||||
![]() |
|