The effects of VEGF-A on atherosclerosis, lipoprotein profile, and lipoprotein lipase in hyperlipidaemic mouse models.

AIMS: The role of vascular endothelial growth factor (VEGF-A) in atherogenesis has remained controversial. We addressed this by comparing the effects of adenoviral VEGF-A gene transfer on atherosclerosis and lipoproteins in ApoE(-/-), LDLR(-/-), LDLR(-/-)ApoE(-/-), and LDLR(-/-)ApoB(100/100) mice. METHODS AND RESULTS: After 4 weeks on western diet, systemic adenoviral gene transfer was performed with hVEGF-A or control vectors. Effects on atherosclerotic lesion area and composition, lipoprotein profiles, and plasma lipoprotein lipase (LPL) activity were examined. On day 4, VEGF-A induced alterations in lipoprotein profiles and a significant negative correlation was observed between plasma LPL activity and VEGF-A levels. One month after gene transfer, no changes in atherosclerosis were observed in LDLR(-/-) and LDLR(-/-)ApoB(100/100) models, whereas both ApoE(-/-) models displayed increased en face lesion areas in thoracic and abdominal aortas. VEGF-A also reduced LPL mRNA in heart and white adipose tissue, whereas Angptl4 was increased, potentially providing further mechanistic explanation for the findings. CONCLUSION: VEGF-A gene transfer induced pro-atherogenic changes in lipoprotein profiles in all models. As a novel finding, VEGF-A also reduced LPL activity, which might underlie the observed changes in lipid profiles. However, VEGF-A was observed to increase atherosclerosis only in the ApoE(-/-) background, clearly indicating some mouse model-specific effects.

Diet intervention reduces uptake of αvß3 integrin-targeted PET tracer 18F-galacto-RGD in mouse atherosclerotic plaques.

BACKGROUND: Expression of α(v)ß(3) integrin has been proposed as a marker for atherosclerotic lesion inflammation. We studied whether diet intervention reduces uptake of α(v)ß(3) integrin-targeted positron emission tomography tracer (18)F-galacto-RGD in mouse atherosclerotic plaques. METHODS AND RESULTS: Hypercholesterolemic LDLR(-/-) ApoB(100/100) mice on high-fat diet for 4 months were randomized to further 3 months on high-fat diet (high-fat group, n = 8) or regular mouse chow (intervention group, n = 7). Intima-media ratio describing plaque burden was comparable between intervention and high-fat groups (2.0 ± 0.5 vs 2.3 ± 0.8, P = .5). Uptake of (18)F-galacto-RGD in the aorta was lower in the intervention than high-fat group (%ID/g 0.16 vs 0.23, P < .01). Autoradiography showed 35% lower uptake of (18)F-galacto-RGD in the atherosclerotic plaques in the intervention than high-fat group (P = .007). Uptake of (18)F-galacto-RGD in plaques correlated with uptake of (3)H-deoxyglucose and nuclear density, which was lower in the intervention than high-fat group (P = .01). Flow cytometry demonstrated macrophages expressing α(v) and ß(3) integrins in the aorta. CONCLUSIONS: Uptake of (18)F-galacto-RGD in mouse atherosclerotic lesions was reduced by lipid-lowering diet intervention. Expression of α(v)ß(3) integrin is a potential target for evaluation of therapy response in atherosclerosis.

Uptake of 68gallium in atherosclerotic plaques in LDLR-/-ApoB100/100 mice.

BACKGROUND: Atherosclerosis is a chronic inflammatory disease of artery wall characterized by infiltration of monocytes into subendothelial space and their differentiation into macrophages. Since rupture-prone plaques commonly contain high amounts of activated macrophages, imaging of the macrophage content may provide a useful tool for the evaluation of plaque vulnerability. The purpose of this study was to explore the uptake of 68gallium (68Ga) in atherosclerotic plaques in mice. METHODS: Uptake of ionic 68Ga was investigated in atherosclerotic LDLR-/-ApoB100/100 and C57BL/6N control mice at 3 h after injection. The ex vivo biodistribution of the 68Ga was assessed and autoradiography of aortic cryosections was defined. In vivo imaging of 68Ga was performed using a small animal positron emission tomography PET/CT scanner. RESULTS: Our results revealed that the uptake of 68Ga-radioactivity was higher in atherosclerotic plaques than in healthy vessel wall (ratio 1.8 ± 0.2, p = 0.0002) and adventitia (ratio 1.3 ± 0.2, p = 0.0011). The autoradiography signal co-localized with macrophages prominently as demonstrated by Mac-3 staining. In both mice strains, the highest level of radioactivity was found in the blood. CONCLUSIONS: We observed a moderate but significantly elevated 68Ga-radioactivity uptake in the aortic plaques of atherosclerotic mice, especially at the sites rich in macrophages. While the uptake of 68Ga was promising in this animal model, the slow blood clearance may limit the usability of 68Ga as a PET tracer for clinical imaging of atherosclerotic plaques.

Silencing of either SR-A or CD36 reduces atherosclerosis in hyperlipidaemic mice and reveals reciprocal upregulation of these receptors.

AIMS: Macrophage scavenger receptor A (SR-A) and class B scavenger receptor CD36 (CD36) contribute to foam cell formation and atherogenesis via uptake of modified lipoproteins. So far, the role of these scavenger receptors has been studied mainly using knockout models totally lacking these receptors. We studied the role of SR-A and CD36 in foam cell formation and atherogenesis by RNA interference (RNAi)-mediated silencing, which is a clinically feasible method to down-regulate the expression of these receptors. METHODS AND RESULTS: We constructed lentivirus vectors encoding short hairpin RNAs (shRNAs) against mouse SR-A and CD36. Decreased SR-A but not CD36 expression led to reduced foam cell formation caused by acetylated low-density lipoprotein (LDL) in mouse macrophages, whereas the uptake of oxidized LDL was not altered. More importantly, silencing of SR-A upregulates CD36 and vice versa. In LDL receptor-deficient apolipoprotein B100 (LDLR(-/-)ApoB(100/100)) mice kept on a western diet, silencing of either SR-A or CD36 in bone marrow cells led to a marked decrease (37.4 and 34.2%, respectively) in cross-sectional lesion area, whereas simultaneous silencing of both receptors was not effective. CONCLUSION: Our results suggest that silencing of either SR-A or CD36 alone reduces atherogenesis in mice. However, due to reciprocal upregulation, silencing of both SR-A and CD36 is not effective.

Evaluation of alphavbeta3 integrin-targeted positron emission tomography tracer 18F-galacto-RGD for imaging of vascular inflammation in atherosclerotic mice.

BACKGROUND: (18)F-Galacto-RGD is a positron emission tomography (PET) tracer binding to alpha(v)beta(3) integrin that is expressed by macrophages and endothelial cells in atherosclerotic lesions. Therefore, we evaluated (18)F-galacto-RGD for imaging vascular inflammation by studying its uptake into atherosclerotic lesions of hypercholesterolemic mice in comparison to deoxyglucose. METHODS AND RESULTS: Hypercholesterolemic LDLR(-/-)ApoB(100/100) mice on a Western diet and normally fed adult C57BL/6 control mice were injected with (18)F-galacto-RGD and (3)H-deoxyglucose followed by imaging with a small animal PET/CT scanner. The aorta was dissected 2 hours after tracer injection for biodistribution studies, autoradiography, and histology. Biodistribution of (18)F-galacto-RGD was higher in the atherosclerotic than in the normal aorta. Autoradiography demonstrated focal (18)F-galacto-RGD uptake in the atherosclerotic plaques when compared with the adjacent normal vessel wall or adventitia. Plaque-to-normal vessel wall ratios were comparable to those of deoxyglucose. Although angiogenesis was not detected, (18)F-galacto-RGD uptake was associated with macrophage density and deoxyglucose accumulation in the plaques. Binding to atherosclerotic lesions was efficiently blocked in competition experiments. In vivo imaging visualized (18)F-galacto-RGD uptake colocalizing with calcified lesions of the aortic arch as seen in CT angiography. CONCLUSIONS: (18)F-Galacto-RGD demonstrates specific uptake in atherosclerotic lesions of mouse aorta. In this model, its uptake was associated with macrophage density. (18)F-Galacto-RGD is a potential tracer for noninvasive imaging of inflammation in atherosclerotic lesions.

Uptake of inflammatory cell marker [11C]PK11195 into mouse atherosclerotic plaques.

PURPOSE: The ligand [(11)C]PK11195 binds with high affinity and selectivity to peripheral benzodiazepine receptor, expressed in high amounts in macrophages. In humans, [(11)C]PK11195 has been used successfully for the in vivo imaging of inflammatory processes of brain tissue. The purpose of this study was to explore the feasibility of [(11)C]PK11195 in imaging inflammation in the atherosclerotic plaques. METHODS: The presence of PK11195 binding sites in the atherosclerotic plaques was verified by examining the in vitro binding of [(3)H]PK11195 onto mouse aortic sections. Uptake of intravenously administered [(11)C]PK11195 was studied ex vivo in excised tissue samples and aortic sections of a LDLR/ApoB48 atherosclerotic mice. Accumulation of the tracer was compared between the atherosclerotic plaques and non-atherosclerotic arterial sites by autoradiography and histological analyses. RESULTS: The [(3)H]PK11195 was found to bind to both the atherosclerotic plaques and the healthy wall. The autoradiography analysis revealed that the uptake of [(11)C]PK11195 to inflamed regions in plaques was more prominent (p = 0.011) than to non-inflamed plaque regions, but overall it was not higher than the uptake to the healthy vessel wall. Also, the accumulation of (11)C radioactivity into the aorta of the atherosclerotic mice was not increased compared to the healthy control mice. CONCLUSIONS: Our results indicate that the uptake of [(11)C]PK11195 is higher in inflamed atherosclerotic plaques containing a large number of inflammatory cells than in the non-inflamed plaques. However, the tracer uptake to other structures of the artery wall was also prominent and may limit the use of [(11)C]PK11195 in clinical imaging of atherosclerotic plaques.

Prepartum and postpartum open-field behavior and maternal responsiveness in mice bidirectionally selected for open-field thigmotaxis.

The authors examined pre- and postpartum open-field (OF) behavior and maternal responsiveness in mice that they bidirectionally selected for OF thigmotaxis. The authors tested 40 female mice under 3 conditions: prepartum OF, postpartum OF, and a pup retrieval test. In both OF conditions, the high OF thigmotaxis (HOFT) mice were more thigmotactic but explored and reared less than the low OF thigmotaxis (LOFT) mice, indicating that the HOFT mice were more emotional. In the postpartum condition, the HOFT mothers also defecated more and ambulated less than the LOFT mothers. The increase in grooming after parturition was more conspicuous among the LOFT mothers than among the HOFT mothers. The LOFT mothers were also more attracted to their pups in the OF, but the retrieval test did not show any substantial line differences. The results suggested that the line difference in emotionality was more pronounced during lactation than during pregnancy, although parturition exerted no effect on thigmotaxis.

Severe coronary artery stenoses and reduced coronary flow velocity reserve in atherosclerotic mouse model: Doppler echocardiography validation study.

OBJECTIVE: Genetically modified hyperlipidemic mice are increasingly used as an animal model of atherosclerosis, but their coronary artery disease remains poorly characterized. Furthermore, non-invasive tools to detect functional consequences of coronary lesions remain to be tested in mice. Coronary flow velocity reserve (CFVR) by transthoracic Doppler echocardiography provides a hemodynamic measure of coronary artery stenosis severity in humans. Thus, we applied Doppler echocardiography in atherosclerotic mice to study the relationship between CFVR and histologically determined coronary artery narrowing. METHODS: Atherosclerotic LDLR/ApoB48 double knockout mice of 58-72 weeks age (n=12) and age-matched C57BL/6 mice (n=5) were studied. CFVR was measured in anesthetized mice by Doppler echocardiography in the middle left coronary artery (LCA) during adenosine-induced maximal vasodilatation. Histopathology of proximal and middle LCA was studied in serial tissue sections. RESULTS: All LDLR/ApoB48 double knockout mice had atherosclerotic lesions in the proximal, but not in the middle LCA causing various degrees of luminal narrowing (30-97%). No lesions were found in controls. Compared with controls, CFVR was significantly reduced in the atherosclerotic mice (2.3+/-0.5 vs. 1.7+/-0.5, p=0.02). There was a negative correlation between CFVR and the amount of luminal narrowing (r=-0.91, p=0.001). Average CFVR was consistently lower in mice that had >or=70% than <70% stenosis (1.3+/-0.1, n=7 vs. 2.2+/-0.4, n=5, p=0.0002). CONCLUSIONS: LDLR/ApoB48 double knockout mice are characterized with histologically severe coronary artery narrowings. Reduced CFVR is a consistent feature of these lesions. Doppler echocardiography of coronary artery flow can be used to detect flow-limiting stenosis in living atherosclerotic mice.

Increased atherosclerotic lesion calcification in a novel mouse model combining insulin resistance, hyperglycemia, and hypercholesterolemia.

No mouse model is currently available where the induction of type 2 diabetes on an atherosclerotic background could be achieved without significant concomitant changes in plasma lipid levels. We crossbred 2 genetically modified mouse strains to achieve a model expressing both atherosclerosis and characteristics of type 2 diabetes. For atherosclerotic background we used low-density lipoprotein receptor-deficient mice synthetizing only apolipoprotein B100 (LDLR(-/-) ApoB(100/100)). Diabetic background was obtained from transgenic mice overexpressing insulin-like growth factor-II (IGF-II) in pancreatic beta cells. Thorough phenotypic characterization was performed in 6- and 15-month-old mice on both normal and high-fat Western diet. Results indicated that IGF-II transgenic LDLR(-/-)ApoB(100/100) mice demonstrated insulin resistance, hyperglycemia, and mild hyperinsulinemia compared with hypercholesterolemic LDLR(-/-)ApoB(100/100) controls. In addition, old IGF-II/LDLR(-/-)ApoB(100/100) mice displayed significantly increased lesion calcification, which was more related to insulin resistance than glucose levels, and significantly higher baseline expression in aorta of several genes related to calcification and inflammation. Lipid levels of IGF-II/LDLR(-/-)ApoB(100/100) mice did not differ from LDLR(-/-)ApoB(100/100) controls at any time. In conclusion, type 2 diabetic factors induce increased calcification and lesion progression without any lipid changes in a new mouse model of diabetic macroangiopathy.

Adenovirus-mediated gene transfer of human vascular endothelial growth factor-d induces transient angiogenic effects in mouse hind limb muscle.

We evaluated the therapeutic potential of adenovirus (Ad)-mediated human vascular endothelial growth factor-D (hVEGF-D) gene delivery in mice. Hind limbs of hypercholesterolemic mice ( n = 120) were injected with AdhVEGF-D, AdhVEGF-A, control AdLacZ (all at 1x10(11)viral particles) or saline. Animals were killed at 4, 7, 14, 28, and 42 days. Newly formed vessels were characterized for their quantity, sprouting, angiogenic versus lymphangiogenic phenotype, and arterial versus venous phenotype by endothelial enzymes markers, pericyte coverage, and electron microscopy. Perfusion was measured by power Doppler ultrasound and edema by magnetic resonance imaging (MRI). AdhVEGF-D induced significant formation of new blood vessels, which featured lumenal enlargement, branching, and sprouting. Branching originated mainly from arterioles. The highest vessel density was present on days 4-7 and the effect lasted up to 28 days. Endothelial marker enzyme activity indicated the predominance of arterial capillaries and arterioles. Forty percent of the neovessels were positive for desmin, indicating that VEGF-D increased pericyte coverage. However, branching vessels were highly positive for smooth muscle actin pericyte marker but negative for desmin. Maximal perfusion was measured during the first week after AdhVEGF-D gene transfer. Ultrastructural analysis showed endothelial cells enriched with vesiculo-vacuolar organelles and cytoplasmic protrusions. Modest lymphangiogenic activity was also detected, which could contribute to the relatively low level of edema detected by MRI. In conclusions, AdhVEGF-D has a strong angiogenic effect and a modest lymphangiogenic effect in mouse skeletal muscle. VEGF-D also increases the presence of pericytes/smooth muscle cells in neovessels. AdhVEGF-D is a potential new agent for the induction of therapeutic vascular growth in skeletal muscle.

Short and long-term effects of hVEGF-A(165) in Cre-activated transgenic mice.

We have generated a transgenic mouse where hVEGF-A(165) expression has been silenced with loxP-STOP fragment, and we used this model to study the effects of hVEGF-A(165) over-expression in mice after systemic adenovirus mediated Cre-gene transfer. Unlike previous conventional transgenic models, this model leads to the expression of hVEGF-A(165) in only a low number of cells in the target tissues in adult mice. Levels of hVEGF-A(165) expression were moderate and morphological changes were found mainly in the liver, showing typical signs of active angiogenesis. Most mice were healthy without any major consequences up to 18 months after the activation of hVEGF-A(165) expression. However, one mouse with a high plasma hVEGF-A(165) level died spontaneously because of bleeding into abdominal cavity and having liver hemangioma, haemorrhagic paratubarian cystic lesions and spleen peliosis. Also, two mice developed malignant tumors (hepatocellular carcinoma and lung adenocarcinoma), which were not seen in control mice. We conclude that long-term uncontrolled hVEGF-A(165) expression in only a limited number of target cells in adult mice can be associated with pathological changes, including possible formation of malignant tumors and uncontrolled bleeding in target tissues. These findings have implications for the design of long-term clinical trials using hVEGF-A(165) gene and protein.

Non-specific binding of [18F]FDG to calcifications in atherosclerotic plaques: experimental study of mouse and human arteries.

PURPOSE: [(18)F]FDG has been used as an inflammation marker and shown to accumulate in inflammatory atherosclerotic plaques. The aim of this study was to investigate the uptake and location of [(18)F]FDG in atherosclerotic plaque compartments. METHODS: The biodistribution of intravenously administered [(18)F]FDG was analysed in atherosclerotic LDLR/ApoB48 mice (n=11) and control mice (n=9). Digital autoradiography was used to detect the ex vivo distribution in frozen aortic sections. In vitro binding of [(18)F]FDG in human atherosclerotic arteries was also examined. RESULTS: The uptake of [(18)F]FDG was significantly higher in the aorta of atherosclerotic mice as compared with the control mice. Autoradiography of excised arteries showed higher [(18)F]FDG uptake in the plaques than in the healthy vessel wall (mean ratio +/-SD 2.7+/-1.1). The uptake of [(18)F]FDG in the necrotic, calcified sites of the advanced atherosclerotic lesions was 6.2+/-3.2 times higher than that in the healthy vessel wall. The in vitro studies of human arterial sections showed marked binding of [(18)F]FDG to the calcifications but not to other structures of the artery wall. CONCLUSION: In agreement with previous studies, we observed [(18)F]FDG uptake in atherosclerotic plaques. However, prominent non-specific binding to calcified structures was found. This finding warrants further studies to clarify the significance of this non-specific binding in human plaques in vivo.

Twenty-three generations of mice bidirectionally selected for open-field thigmotaxis: selection response and repeated exposure to the open field.

We examined: (a) the response to bidirectional selection for open-field (OF) thigmotaxis in mice for 23 generations and (b) the effects of repeated exposure (during 5 days) on different OF behaviors in the selectively bred high OF thigmotaxis (HOFT) and low OF thigmotaxis (LOFT) mice. A total of 2049 mice were used in the study. Prior to the testing in the selection experiment, the mice were exposed to the OF apparatus for approximately 2 min on each of 4 consecutive days. Thus, the selection was based on the scores registered on the 5th day after the four habituation periods. The HOFT mice were more thigmotactic than the LOFT mice in almost each generation. The HOFT mice also tended to rear less than the LOFT mice, which was explained by the inverse relationship between emotionality and exploratory tendencies. The lines did not generally differ in ambulation. Sex differences were found in thigmotaxis, ambulation, and rearing. In the repeated exposure experiment, the development of nine different OF behaviors across the 5 days of testing was addressed. Both lines ambulated, explored, and reared most on the 1st, 4th, and 5th days. Grooming and radial latency decreased and thigmotaxis increased linearly across the testing days. Line differences were found in ambulation, exploration, grooming, and rearing, while sex differences were manifested in ambulation and exploration. The line difference in thigmotaxis was evident only on the 5th day. Temporal changes were partially at variance with the general assumptions. OF thigmotaxis was found to be a powerful characteristic for producing two diverging lines of mice.

Gene transfers of vascular endothelial growth factor-A, vascular endothelial growth factor-B, vascular endothelial growth factor-C, and vascular endothelial growth factor-D have no effects on atherosclerosis in hypercholesterolemic low-density lipoprotein-receptor/apolipoprotein B48-deficient mice.

BACKGROUND: The role of vascular endothelial growth factors (VEGFs) in large arteries has been proposed to be either vasculoprotective or proatherogenic. Because VEGF family members are used for human therapy, it is important to know whether they could enhance atherogenesis. We tested the effects of the members of the VEGF gene family on atherogenesis in LDL-receptor/apolipoprotein (apo) B48 double-knockout (LDLR/apoB48) mice using systemic adenoviral gene transfer. METHODS AND RESULTS: Six groups of LDLR/apoB48-deficient mice (n=110) were kept 3 months on a Western-type diet. After 6 weeks of diet, mice were injected via tail vein with recombinant adenoviruses expressing VEGF-A, -B, -C, or -D or LacZ (1 x 10(9) PFU) or rhVEGF-A protein (2 microg/kg) and euthanized 6 weeks later. Also, older mice (n=36) were injected after 4 months on the diet and euthanized 6 weeks later (total time on the diet, 22 weeks) to evaluate the effects of gene transfers on the development of more mature lesions. Aortas were analyzed for the presence of macroscopic lesions, cross-sectional lesion areas, neovascularization, and cellular composition of the lesions. All groups had equivalent plasma cholesterol and triglyceride levels. Gene transfers with recombinant adenoviruses or administration of rhVEGF-A protein had no statistically significant effects on en face atherosclerotic lesions in the aorta, cross-sectional lesion area, neovascularization, or cellular composition of the lesions. CONCLUSIONS: This study shows no proatherogenic effects of adenovirus-mediated gene transfers of VEGF-A, -B, -C, or -D in the LDLR/apoB48-deficient hypercholesterolemic mice, in which lipoprotein profile and atherosclerosis closely resemble those in human disease.

Mice selectively bred for open-field thigmotaxis: life span and stability of the selection trait.

In 2 experiments, the authors examined 69 mice selectively bred for high or low levels of open-field (OF) thigmotactic behavior (high open-field thigmotaxis [HOFT] and low open-field thigmotaxis [LOFT], respectively). They found that the strains differed in defecation during the 60-min exposure to the OF. Furthermore, the strains differed with regard to their life spans: The more thigmotactic HOFT mice lived longer than the LOFT mice. The strains were not differentiated by food intake or excretion. The strain difference in thigmotaxis was not age dependent, and it persisted in the home-cage condition as well. Neither the location (center or wall) of the starting point nor the shape (circular or square) of the OF arena affected the difference in wall-seeking behavior between the two strains. The authors concluded that the difference in thigmotaxis (or emotionality) between the HOFT and LOFT mice is a stable and robust feature of these animals.

Crossfostering in mice selectively bred for high and low levels of open-field thigmotaxis.

The main purpose of this research was to investigate whether the difference in open-field (OF) thigmotaxis between mice selectively bred for high and low levels of wall-seeking behavior originated from genetic or acquired sources. Unfostered, infostered, and crossfostered mice were compared in two experiments in which the effects of strain, sex, and fostering on ambulation, defecation, exploration, grooming, latency to move, radial latency, rearing, thigmotaxis, and urination were studied. These experiments revealed that OF thigmotaxis was unaffected by the foster condition and thus genetically determined. The selected strains of mice also diverged repeatedly with regard to exploration and rearing. The findings are in line with the previously described existence of an inverse relationship between emotionality and exploration.

Long-term lowering of plasma cholesterol levels in LDL-receptor-deficient WHHL rabbits by gene therapy.

Lentiviral vectors encoding rabbit low-density lipoprotein receptor (LDLR) or green fluorescent protein (GFP) under the control of a liver-specific promoter (LSP) were used for intraportal gene transfer into the liver of hypercholesterolemic LDLR-deficient Watanabe Heritable Hyperlipidemic rabbits. In vitro cell culture analysis demonstrated functionality of the LSP-LDLR vector in mediating increased degradation of LDL in transduced liver cells. Twenty-five rabbits were each injected with 1 x 10(9) infectious virus particles into the portal vein. Liver biopsy samples were collected 4 weeks after the gene transfer and the rabbits were followed up for 2 years. Histological and RT-PCR analyses showed the expression of GFP and LDLR transgenes in the biopsy samples. Clinical chemistry and histological analyses revealed normal liver function and morphology during the 2-year follow-up with no safety issues. LSP-LDLR-treated rabbits demonstrated an average of 14 +/- 7% decrease in serum cholesterol levels during the first 4 weeks, 44 +/- 8% decrease at 1 year, and 34 +/- 10% decrease at the 2-year time point compared to the control rabbits. This study demonstrates the safety and potential benefits of the third-generation liver-specific lentiviral vectors in the treatment of familial hypercholesterolemia using direct intraportal liver gene therapy without the need for liver resection.

Feline immunodeficiency virus and retrovirus-mediated adventitial ex vivo gene transfer to rabbit carotid artery using autologous vascular smooth muscle cells.

We have developed an ex vivo gene transfer technique to rabbit arterial wall using autologous smooth muscle cells (SMCs). SMCs were harvested from rabbit ear artery, transduced in vitro with vesicular stomatitis virus G-glycoprotein pseudotyped retrovirus or feline immunodeficiency virus (FIV) and returned to the adventitial surface of the carotid artery using a periadventitial silicone collar or collagen sheet placed around the artery. Beta-galactosidase (lacZ) and human apolipoprotein E3 (apoE3) cDNAs were used as transgenes. After retrovirus-mediated gene transfer of lacZ the selected cells implanted with high efficiency and expressed lacZ marker gene at a very high level 7 and 14 days after the operation. The level of lacZ expression decreased thereafter but was still detectable 12 weeks after the gene transfer, and was exclusively localized to the site of cell implantation inside the collar. Utilizing FIV vector expressing apoE3, low levels of apoE were measured from serum collected from a low-density lipoprotein receptor deficient Watanabe heritable hyperlipidemic rabbits 1 month after the gene transfer. The physiological effect of apoE expression was detected as transiently elevated serum cholesterol levels. The results indicate that the model can be used for high efficiency local gene transfer in arteries, e.g. during vascular surgery. The model is also valuable for studying expression, stability and safety of new gene transfer vectors and their expression products in vivo.

Adeno-associated virus-mediated gene transfer of a secreted decoy human macrophage scavenger receptor reduces atherosclerotic lesion formation in LDL receptor knockout mice.

Macrophage scavenger receptors (MSR) promote atherosclerotic lesion formation, and modulation of MSR activity has been shown to influence atherosclerosis. Soluble receptors are effective in inhibiting receptor-mediated functions in various diseases. We have generated a secreted macrophage scavenger receptor (sMSR) that consists of the bovine growth hormone signal sequence and the human MSR A I extracellular domains. sMSR reduces degradation of atherogenic modified low-density lipoproteins and monocyte/macrophage adhesion on endothelial cells in vitro. To test long-term effects of sMSR, atherosclerosis-susceptible LDLR knockout mice were transduced via the tail vein with an adeno-associated virus (AAV) expressing sMSR or control enhanced green fluorescent protein (EGFP), and a Western-type diet was started. Gene transfer caused a temporary elevation in alkaline phosphatase and aspartate amino transferase values without a change in C-reactive protein. sMSR protein was detected in the plasma of the transduced mice by a specific ELISA 6 months after the gene transfer. AAV-mediated sMSR gene transfer reduced atherosclerotic lesion area in the aorta by 21% (P < 0.05) compared to EGFP-transduced control mice. Even though eradication of established disease was not possible, atherosclerotic lesion formation could be modified using AAV-mediated gene transfer of the decoy sMSR.

Vascular endothelial growth factor-D expression in human atherosclerotic lesions.

OBJECTIVE: Vascular endothelial growth factor-D (VEGF-D) is a recently characterized member of the VEGF family, but its expression in atherosclerotic lesions remains unknown. We studied the expression of VEGF-D and its receptors (VEGFR-2 and VEGFR-3) in normal and atherosclerotic human arteries, and compared that to the expression pattern of VEGF-A. METHODS: Human arterial samples (n=39) obtained from amputation operations and fast autopsies were classified according to the stage of atherosclerosis and studied by immunohistochemistry. The results were confirmed by in situ hybridization and RT-PCR. RESULTS: We found that while VEGF-A expression increased during atherogenesis, VEGF-D expression remained relatively stable only decreasing in complicated lesions. In normal arteries and in early lesions VEGF-D was mainly expressed in smooth muscle cells, whereas in complicated atherosclerotic lesions the expression was most prominent in macrophages and also colocalized with plaque neovascularization. By comparing the staining profiles of different antibodies, we found that proteolytic processing of VEGF-D was efficient in the vessel wall. VEGFR-2, but not VEGFR-3, was expressed in the vessel wall at every stage of atherosclerosis. CONCLUSIONS: Our results suggest that in large arteries VEGF-D is mainly expressed in smooth muscle cells and that it may have a role in the maintenance of vascular homeostasis. However, in complicated lesions it was also expressed in macrophages and may contribute to plaque neovascularization. The constitutive expression of VEGFR-2 in arteries suggests that it may be one of the principal mediators of the VEGF-D effects in large arteries.