| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1Department of Surgery and Science, 2Department of Pathology, and 4Department of Cardiovascular Medicine, Graduate School of Medical Sciences, Kyushu University, Fukuoka 812-8582; 3Department of Vascular Surgery, Graduate School of Medical Science, Nagoya University, Nagoya 466-8550; 5DNAVEC Research Company Incorporated, Tsukuba, Ibaraki 305-0856, Japan
Submitted 21 November 2002 ; accepted in final form 27 February 2003
 |
ABSTRACT |
|---|
 TOP ABSTRACT MATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
fibroblast growth factor-2; vein graft; vasorelaxation
We (18) recently demonstrated the unexpectedly harmful effects of overexpressed vascular endothelial growth factor (VEGF) 165 by recombinant Sendai virus (SeV)-mediated intramuscular gene transfer in murine critical limb ischemia. These effects were not seen when basic fibroblast growth factor (bFGF/FGF-2) was used. In fact, overexpression of FGF-2 consistently showed a significant therapeutic effect, indicating that both the choice of therapeutic genes and the expression level of angiogenic growth factors may critically affect the clinical outcome of therapeutic angiogenesis (18, 34).
Unfortunately, "acute" models of limb ischemia, which are generated by simple ligation of feeder arteries, are not always relevant models for the types of limb ischemia seen in clinical settings. As reported by a number of experimental studies, a considerable level of blood flow was recovered within a few weeks in these models (4, 21, 30), indicating different blood flow characteristics from clinical "chronic" limb ischemia. This discrepancy may be a cause of the less dramatic clinical outcomes. A preclinical evaluation of therapeutic angiogenesis using more relevant animal models with chronic limb ischemia is greatly needed at this time.
For 20 years, we have classified the flow waveform into five clinical types (0 to IV) and demonstrated that grafts implanted in a type II pattern (without reversed flow) of arterial circulation show a low wall shear stress, which correlates with a high resistance of peripheral circulation (poor runoff) (8, 9, 23, 28). We (8, 9, 23) also clinically showed that autologous vein grafts (VGs) implanted in poor-runoff limbs frequently exhibited late graft failure. In addition, our continuous efforts at establishing dog (20) and rabbit (10, 12) models for poor distal runoff have resulted in experimentally accelerated neointimal hyperplasia (10, 12, 20), accompanied by impaired production of prostacyclin (25, 26).
From these findings, it might be expected that increase of collateral blood perfusion via intramuscular overexpression of FGF-2 would reduce neointimal hyperplasia in chronic limb ischemia. The use of FGF-2 for therapeutic angiogenesis, however, may prove to be a "double-edged sword" in the clinical setting, particularly after bypass grafting; the vascular proliferative disease might be reduced by restoration of blood flow or be stimulated by the FGF-2-mediated proliferation of mesenchymal cells.
To determine the net effect of SeV-mediated intramuscular gene transfer of FGF-2 on VGs, we employed a rabbit poor-runoff model that had shown stably impaired blood perfusion for several months (9). The aims of the current study were to clarify 1) whether FGF-2 gene transfer might also be effective for increased collateral blood flow in chronic limb ischemia; 2) whether FGF-2 gene transfer would reduce or increase neointimal hyperplasia under poor-runoff conditions; and 3) the functional effect of intramuscular FGF-2 gene transfer on endothelium-dependent vasorelaxation, which has been chronically disturbed in experimental VGs (11, 14, 15).
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
The high-titer stocks of recombinant SeVs used in this study were prepared as described previously (7, 13, 17, 29, 32, 33). Virus titer was determined in terms of plaque-forming units (pfu) by hemagglutination assay using chicken red blood cells and was kept at -80°C until use. Full-length human FGF-2 cDNA for construction of SeV-hFGF2 was obtained by reverse transcription PCR using total RNA from human aortic smooth muscle cells (Kurabo; Tokyo, Japan). The PCR products were subcloned into pCRII vector (Invitrogen; Tokyo, Japan), and the whole sequence was confirmed by using a multicapillary DNA sequencer (model CEQ2000; Beckman Coulter KK; Tokyo, Japan).
Animals, Surgical Procedures, and Experimental Protocols
All animal experiments were performed under approved protocols and in accordance with recommendations for the proper care and use of laboratory animals by the Committees for Animal, Recombinant DNA, and Infectious Pathogen Experiments at Kyushu University.
Experiment 1: Protein Expression and Limb Prognosis Analysis Using a Murine Autoamputation Model
Charles River-derived male BALB/c nu/nu mice (5 wk old; KBT Oriental; Tosu, Saga) were used as a murine "autoamputation model;" details of the surgical treatment were as described previously (18). Briefly, the entire left saphenous artery and vein and the left external iliac artery and vein with deep femoral and circumflex arteries and veins were ligated, cut, and excised to set up a murine model of "severe" hindlimb ischemia. For gene transfer, 25 µl of vector solutions were injected into two portions of the thigh muscle soon after completion of the surgical procedures. For protein measurements, the posterior thigh muscles from 36 animals were excised and subjected to immunoassay as described below (see Experiment 2 and Experiment 3). Forty-eight animals were used for limb survival analysis as previously described (18). Limb prognosis was analyzed according to limb salvage index as previously described (18).
Rabbit Experiments
Charles River-derived male Japanese white rabbits (2.5–3.0 kg body wt; KBT Oriental) were used in this study. These animals underwent the following vector administration and/or surgical procedures. Surgical procedures were performed under sterilized conditions; sufficient xylazine (10 mg/kg, Sigma; St. Louis, MO) and ketamine hydrochloride (25 mg/kg; Sigma) were administered, and anesthesia was maintained by an intravenous injection of pentobarbital sodium (30 mg/kg).
Experiment 2: Generation of Poor Distal Runoff Circulation for a Chronic Limb Ischemia Model and Assessment of Time Course of Transgene Expression
Twenty animals were used in this set of experiments. Briefly, under an operating microscope (model OMS-75; TOPCON, Tokyo, Japan), a longitudinal medial incision was made in a lower thigh, and the popliteal, saphenous, and distal femoral artery were exposed in both hindlimbs of the rabbits. The distal femoral artery, the saphenous artery, and the caudal femoral artery were ligated and cut; only the middle muscle branch was preserved. Soon after the poor-runoff operation was finished, the first flow waveform was assessed as described in Laser Doppler blood perfusion image analysis and Direct assessment of collateral blood flow using an ultrasonic transit-time flowmeter. Two weeks later, when it was evident that the hemodynamic status was stable with the development of mild collateralization and an increased resistance to peripheral circulation, all surgically treated animals met the criteria for poor-runoff limb of a mean flow rate of <2.0 ml/min (10, 12). Soon after this confirmation, 109 pfu of SeV-hFGF2 or SeV-luciferase were injected at the thigh and calf muscles as shown in Fig. 4A. At each time point (2 days and 1, 2, and 4 wk after gene transfer), 5 of 20 animals were euthanized by overdose anesthesia, and 1 g (wt/wt) of thigh muscles was sampled from two portions around the injected points. FGF-2 contents in skeletal muscles were determined using Quantikine Immunoassay systems for human FGF-2 (R&D Systems; Minneapolis, MN) according to the manufacturer's instructions, as previously described (18). The protein concentration was determined by Lowry's method (16).
|
Experiment 3: Assessment of Blood Flow
Nineteen animals were used in this study. The poor-runoff limbs were generated in the left hindlimbs as described above, whereas the right hindlimbs were used for controls.
Laser Doppler blood perfusion image analysis. Measurements of the ischemic (left)-to-normal (right) limb blood flow ratio were made using a laser Doppler blood perfusion image (LDPI) analyzer (Moor Instruments; Devon, UK) (18). Before laser scanning was initiated, excess hair in the area of interest was cleared with the use of a depilatory cream, and the animals were placed on a heating plate kept at 37°C for 15 min. To minimize data variables due to ambient light and temperature, the LDPI index was expressed as the ratio of the left (ischemic) to the right (nonischemic) limb blood flow. Two consecutive measurements were done every week over the same region of interest. The average blood perfusion of the ischemic thigh area was calculated.
Direct assessment of collateral blood flow using an ultrasonic transit-time flowmeter. Four weeks after gene transfer, a 5-mm flow probe (Transonic Flow Probe 0.5 V; Transonic Systems; Ithaca, NY) was connected to an ultrasonic transit-time flowmeter (USTF, Transonic T201; Transonic Systems), which was applied on the external iliac artery (first assessment) and the deep femoral artery (second assessment). Blood flow rates and flow waveforms were recorded, and the recorded waveforms were traced with a digitizer (K-150; Kanto Denshi; Tokyo, Japan) connected to a personal computer. Regarding the intraluminal veracity profile, changes in the flow of the boundary layer adjacent to the vessel wall reflected the flow waveform patterns; stagnation in the boundary layer was observed under conditions of a gentle-sloping flow waveform, whereas a remarkable fluctuation was observed under a flow waveform pattern with a steep acceleration and deceleration.
Experiment 4: Assessment of Neointimal Hyperplasia
Vein grafting and graft harvest procedures. After the
establishment of poor-runoff limb was confirmed, the femoral artery and vein
were exposed, and an 
1.5-cm segment of the femoral vein was taken for an
autologous reversed vein graft
(10,
12). Graft harvesting was
completed with meticulous care to avoid injuring the graft wall. The harvested
graft was preserved in heparinized saline (5 IU/ml) at room temperature for
about 5 min. The femoral artery, distal to the orifice of the lateral
circumflex femoral artery, was cut and replaced with the harvested autologous
vein in reversed end-to-end fashion with interrupted 10-0 monofilament sutures
while under a surgical microscope. Throughout these procedures, neither
anticoagulant drugs nor antibiotics were used.
Four weeks later, the rabbits were euthanized with a lethal dose of anesthesia. The VGs were then exposed, removed, and perfusion fixed in situ at 100 cmH2O for 15 min. The perfused autologous VG, including anastomosis, was immersed in neutralized 10% formaldehyde overnight at room temperature.
Assessment of neointimal thickness. Perfusion-fixed VGs were mounted in paraffin, and 5-µm-thick tissue sections were subjected to hematoxylin-eosin staining and elastica van Giesons's staining. Four sections were divided from one harvested VG. The degree of neointimal thickness was measured by a MACSCOPE (Mitani; Fukai, Japan) in eight randomly selected points of each section. The average of eight values was considered to be the intimal thickness of one section, and the average of four values was considered to be that of one vein graft. The intimal area of each vein graft was also calculated by computer-assisted quantification using MACSCOPE software.
Experiment 5: Assessment of Endothelial Cell-Dependent Vasorelaxation
At 4 wk after the operation, VGs were sectioned into rings, 3 to 4 mm in
length, and transferred to a blinded examiner. The vessel rings were suspended
by hooks in an organ bath that contained oxygenated Krebs solution of the
following composition (in mM): 121.3 NaCl, 4.7 KCl, 24.7 NaHCO3,
1.2 KH2PO4, 1.2 MgSO4, 2.5 CaCl2,
and 5.77 D-glucose. The bath was bubbled with 95% O2-5%
CO2 and maintained at 37°C. The hooks were connected to a force
transducer (Nihon Kohden; Tokyo, Japan), and the data were stored in a
computer by using MacLab software. After a 40-min equilibration period, the
rings were stimulated with 60 mM KCl every 30 min, and the resting load was
increased in a stepwise manner to obtain the maximal force development. After
the resting loads were obtained, the rings were precontracted with
phenylephrine (10-6 mol/l), and the vasodilator responses to
increasing concentrations of acetylcholine (10-10 to
10-5 mol/l) were examined during a contraction to prostaglandin
F2
(2 x 10-6 mol/l). The
responses were expressed as a percentage of the contractions to prostaglandin
F2.
Statistical Analysis
All values are expressed as means ± SE. The Mann-Whitney U-test was used to analyze all data, with the exception of that for limb survival. For survival analysis, the limb survival rate was analyzed using Kaplan-Mayer's method. The statistical significance of the survival experiments was determined using the log-rank test. Values of P < 0.05 were considered to indicate statistical significance.
|
RESULTS |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Optimization of virus titer using a murine severe limb ischemia model. Before moving to rabbit studies, we first evaluated the optimized dose of SeV-hFGF2 using a murine "autoamputation model" of BALB/c nu/nu mice, which consistently lost their limbs during the 10-day experiment (18). In particular, we focused on the relationship between the expression level of FGF-2 and limb survival.
As shown in Fig. 1A, a virus titer-dependent increase of FGF-2 expression was found in ischemic thigh muscles 2 days after the gene transfer. Subsequent experiments for limb survival demonstrated that a significant limb-salvaging effect was found at 106 pfu; higher doses of 3 x 106 and 1 x 107 pfu showed an optimal therapeutic effect (Fig. 1B). At 5 x 107 pfu, however, the therapeutic effect worsened (Fig. 1B; P < 0.01), and it was lost altogether at 1 x 108 pfu, which corresponded to over 300 ng/g muscle protein of FGF-2 (data not shown).
|
From these findings, we determined that a total of 1 x 109 pfu/limb of SeV-hFGF2 for 3.0- to 3.5-kg rabbits would be roughly equivalent to the optimal titer for 30-g mice.
FGF-2 gene transfer increases blood perfusion in chronically ischemic limbs. We next assessed the time course of transgene expression in thigh muscles with chronic limb ischemia. The vector solution was injected 2 wk after the poor-runoff operation, and left thigh muscles were obtained 2, 7, 14, and 28 days (n = 5, respectively, total 20 animals) after gene transfer. Right thigh muscles that underwent surgery with SeV-luciferase injection were used as controls. As shown in Fig. 2, transgene expression peaked on day 2, declined at 1 wk, and reached the baseline level at 2 wk and later after gene transfer. These findings were comparable with those obtained using the murine ischemia model (18).
|
In the next set of experiments, we assessed the time course of blood flow profiles using LDPI as well as a USTF (Fig. 3A).
|
Typical pixel images of LDPI are presented in
Fig. 3B. At the time
of vector injection (2 wk after poor-runoff surgery), the flow pixel of the
foot pad was consistently reduced in both animals treated with the control
vector (SeV-luciferase) or SeV-hFGF2 (Fig.
3B, top left and bottom left, yellow
arrows). Four weeks later (2 wk after poor-runoff surgery), markedly increased
flow pixels were observed only in the animals treated with SeV-hFGF2
(Fig. 3B, right
bottom, red arrows). The percentage of flow increase, calculated as
described in MATERIALS AND METHODS, was estimated as 150% in
limbs treated with SeV-hFGF2, whereas no significant increase was found in
control animals treated with SeV-luciferase or buffer (phosphate-balanced
saline; PBS) (Fig.
3C).
Although LDPI is an easy, noninvasive, and reliable blood flow detection
system, it can only detect relatively superficial flow conditions; the laser
only reaches 2–3 mm in depth from the skin surface (according to
the supplier's data), suggesting that LDPI might mainly reflect dermal blood
perfusion rather than muscular flow. To obtain further evidence with regard to
the muscular collateral blood flow, we measured arterial blood flow directly
using USTF at 6 wk after surgical treatment
(Fig. 4A). In control
right limbs without surgical treatment, the total blood flow through the deep
femoral artery was only 30% of that through external iliac artery
(Fig. 4B). Our
preliminary study using three untreated animals demonstrated a similar flow
ratio (data not shown). In the case of left poor-runoff limbs, this flow ratio
increased to 58% (mean), indicating increased collateral flow through the
deep femoral artery as a result of physiological collateral development by the
ligation of a superficial femoral artery. This effect was significantly
increased to 71% (mean) by FGF-2 gene transfer to the quadriceps muscle
(P < 0.05), indicating that the collateral flow was enhanced by
FGF-2 overexpression.
These results thus indicate that SeV-mediated gene transfer of human FGF-2 could increase blood perfusion in chronically ischemic limbs.
Effect of Intramuscular FGF-2 Gene Transfer to an Autologous Vein Graft in Chronically Ischemic Limbs
Assessment of neointimal hyperplasia. As a subsequent experiment, we assessed the effect of intramuscular gene transfer of FGF-2 on neointimal hyperplasia of the VGs implanted under poor-runoff condition. In this experiment, an ipsolateral femoral vein was implanted into the superficial femoral artery 2 wk after the poor-runoff operation. At the same time, SeV-hFGF2 was injected in a manner similar to that described above.
Four weeks after graft implantation, VGs were subjected to histological examination. As shown in Fig. 5A, apparently decreased neointimal thickness was found in the VGs treated with FGF-2. Two means of histological quantification, namely, ocular cytometer-mediated measurement of thickness and computer-assisted neointimal square, confirmed the reduced thickness (P < 0.002) and area (P < 0.002) of the neointima in the VGs treated with FGF-2 compared with those with the control virus (SeV-luciferase).
|
FGF-2 gene transfer reverses disturbed vasorelaxation mediated by regenerative endothelial cells. We previously demonstrated the markedly disturbed production of endothelium-derived relaxing factor in VGs, resulting in disturbed vasorelaxation and impaired production of cGMP (14, 15, 25). We also demonstrated that supplementation of L-arginine (24) as well as gene transfer of endothelial cell nitric oxide (NO) synthase (19, 22) could suppress neointimal formation in VGs, indicating that disturbed production of NO from regenerative endothelial cells (ECs) may play an important role during the neointimal formation of VGs. Because FGF-2 can promote NO production from ECs (31), we expected that FGF-2 gene transfer might have a beneficial effect on the regenerative ECs of VGs. Thus we assessed the EC-dependent vasorelaxation of VGs 4 wk after implantation.
As shown in Fig. 6,
untreated control veins showed a dose-dependent relaxation in response to
acetylcholine, whereas VGs treated with SeV-luciferase showed no response
(Fig. 6A). VGs
implanted under muscular beds overexpressing FGF-2 showed 30% of the
relaxation of those seen in control veins, which effect was significantly
diminished in the presence of
N
-nitro-L-arginine, an NO synthase
inhibitor (Fig. 6B,
P < 0.01). These results indicate that FGF-2 gene transfer could
ameliorate NO-mediated endothelial vasodilator functions; this beneficial
effect of FGF-2 gene transfer might be involved in its suppressive effects on
neointimal hyperplasia of VGs.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1–2 wk, a
result comparable to that previously obtained using a murine model
(17); 2) SeV-mediated
overexpression of FGF-2 was effective in increasing relatively superficial
blood perfusion by LDPI, as well as collateral blood flow assessed by USTF, in
the chronic hindlimb ischemia model of consistently impaired blood flow;
3) intramuscular gene transfer of FGF-2 resulted in suppressed
neointimal formation in VGs implanted under poor-runoff arterial circulation;
and 4) intramuscular overexpression of FGF-2 significantly restored
disturbed vasorelaxation of regenerative ECs in VGs. This is the first report
to describe the therapeutic effect of FGF-2 gene transfer not only for
chronically ischemic limbs, but also for neointimal hyperplasia of VGs.
First, we should emphasize the utility of recombinant SeV, a powerful tool
for in vivo gene transfer in various organs
(7,
13,
17,
18,
29,
32,
33), in treatments of critical
limb ischemia. As shown in Fig.
1, for example, we optimized the relationships among the dose of
SeV-hFGF2, transgene expression, and limb-salvaging effects using a murine
limb ischemia model, which resulted in limb amputation. Our unpublished study,
using this model with 30 g body wt, demonstrated that more than 300 µg of
plasmid DNA expressing human FGF-2 driven by CMV IE promoter (pcDNA3) were
required to achieve the transgene expression and therapeutic effect equivalent
to that obtained by 106 pfu of SeV-hFGF2. At that time, an
injection of 100 µg of plasmid DNA did not show any limb-salvaging effect.
In clinical trials, roughly estimated, >600 mg of the plasmid would be
necessary to treat 60-kg patients at an equivalent rate. However, this amount
of plasmid DNA may be harmful in clinical practice, because CpG DNA including
bacterial plasmid evokes a strong innate immune response via toll-like
receptor 9 (5). In fact, Alton
et al. (1) reported in their
double-blind gene therapy clinical trial for treatment of cystic fibrosis that
pulmonary nebulization of 40 mg of plasmid resulted in transient flulike
symptoms accompanied with an increased serum level of proinflammatory
cytokines. In the case of recombinant SeV, a similar expression level could be
achieved by 106 pfu in mice. In the present study, we used
109 pfu of SeV-hFGF2 for 3.0- to 3.5-kg rabbits (3 x
108 pfu/kg body wt), which was roughly equivalent to 107
pfu for 30-g mice and 2 x 1010 pfu for 60-kg patients. Our
recent safety study using nonhuman primates demonstrated that only a mild
(maximum value = 20.5 pg/ml) and transient (up to day 1) increase of
serum interleukin-6 was detected in some animals intramuscularly receiving 5
x 109 pfu/kg of SeV-FGF2 (unpublished observations),
suggesting the potential for far less inflammatory reactions to therapeutic
doses of SeV-hFGF2 than to those of plasmid DNA. From the data including these
findings, our clinical protocol of therapeutic angiogenesis using SeV-hFGF2
has passed the review of our Institutional Review Board and is now under
consideration of the Japanese Governmental Review Board.
Although less attention has been paid to this subject, we believe that a chronic limb ischemia model showing sustained impairment of limb blood supply may be important for predicting the clinical outcome of therapeutic angiogenesis. In the early phase after initial surgery on mice to induce limb ischemia, dramatic changes in the protein levels of various angiogenic factors were found in ischemic muscles (27); the findings in these models did not always correspond to the situations of patients with critical limb ischemia. In fact, physiological recovery of blood flow has been demonstrated in most of the literature. As we demonstrated previously (10) and in the present study, the rabbit models of poor-runoff consistently show sustained impairment of blood flow in the hindlimbs, suggesting that they could provide a more relevant model for predicting the clinical outcome. The relevance of the rabbit model of poor runoff will be one of the foci of our clinical study.
A remaining question is "How does overexpressed FGF-2 restore EC-dependent relaxation of VGs?" Although FGF-2 has been shown to regulate NO release (31) and vascular tone (35), no significant recombinant FGF-2 was detected at the time of the tension study, as shown in Fig. 2. A preliminary study was performed for inhibition of NO production via chronic administration of NG-nitro-L-arginine methyl ester (L-NAME) to rule out FGF-2-mediated vasorelaxation; however, an Evans blue dye study demonstrated that EC coverage was greatly impaired in this group compared with those without L-NAME, indicating that this study was not appropriate for evaluation. In fact, our findings may suggest that early supplementation of FGF-2 might be beneficial for regenerative ECs; we are currently investigating how to test this hypothesis. Another, perhaps more up-to-date viewpoint, is that circulating endothelial progenitor cells (EPCs) may make a contribution to EC regeneration of VGs. A recent study demonstrated that a FGF receptor-1 (FGFR1)-positive subpopulation in human CD34 ± cells could enrich the EPCs, and the stimulation of the receptor using recombinant FGF-2 induced EC differentiation, indicating that FGF-2 plays a significant role in the recruitment and differentiation of EPCs (2). Further studies will be needed to determine whether EPCs help restore the ability of VGs to regenerate EC function.
In summary, we obtained evidence indicating that FGF-2 gene transfer is effective not only for treating chronically ischemic limbs, but also for reducing neointimal hyperplasia of VGs via restoration of EC function. These findings may suggest that the indications for intramuscular administration of SeV-hFGF2 should be extended to include not only limb salvage, but also late graft failure.
|
ACKNOWLEDGMENTS |
|---|
This work was supported by a grant promoting Basic Scientific Research in a Medical Frontier from the Organization for Pharmaceutical Safety and Research.
|
FOOTNOTES |
|---|
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
|
REFERENCES |
|---|
|
TOPABSTRACTMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
This article has been cited by other articles:
|
|
 |
|
  R. Fukunaga, K. Hirano, M. Hirano, N. Niiro, J. Nishimura, Y. Maehara, and H. Kanaide Upregulation of proteinase-activated receptors and hypercontractile responses precede development of arterial lesions after balloon injury Am J Physiol Heart Circ Physiol, November 1, 2006; 291(5): H2388 - H2395. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. Kaneko, Y. Yonemitsu, T. Fujii, M. Onimaru, C.-H. Jin, M. Inoue, M. Hasegawa, T. Onohara, Y. Maehara, and K. Sueishi A free radical scavenger but not FGF-2-mediated angiogenic therapy rescues myonephropathic metabolic syndrome in severe hindlimb ischemia Am J Physiol Heart Circ Physiol, April 1, 2006; 290(4): H1484 - H1492. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| Visit Other APS Journals Online |
