Human Molecular Genetics, 2002, Vol. 11, No. 1 43-58
© 2002 Oxford University Press
Acute regression of advanced and retardation of early aortic atheroma in immunocompetent apolipoprotein-E (apoE) deficient mice by administration of a second generation [E1–, E3–, polymerase–] adenovirus vector expressing human apoE
Centre for Biomedical Sciences, School of Biological Sciences, Royal Holloway University of London, Egham, Surrey TW20 0EX, UK, 1Department of Medicine, Royal Free and University College Medical School, London NW3 2PF, UK, 2Department of Pediatrics, Division of Genetics, Duke University Medical Center, Durham, NC 27710, USA and 3GlaxoSmithkline, Research and Development, Medicines Research Centre, Stevenage, Hertfordshire SG1 2NY, UK
Received September 4, 2001; Revised and Accepted November 2, 2001.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Apolipoprotein E (apoE) is a 34 kDa glycoprotein with multiple actions that help protect against the development of atherosclerosis. Here, we have assessed the atheroprotective potential of an [E1–, E3–, polymerase–] adenovirus vector expressing human apoE, comparing intramuscular and intravenous (liver-directed) injections in hypercholesterolaemic apoE-deficient mice (apoE–/–). Intramuscular injections resulted in low expression of apoE and afforded no protection against atherogenesis. In contrast, 3 and 7 days after intravenous injections into young (6–8-week-old) apoE–/– mice, plasma levels of apoE were elevated and were accompanied by reductions in plasma cholesterol and normalization of lipoprotein profiles. Thereafter, plasma apoE was still detectable up to day 70, but gradually declined, although no humoral immune response was evoked, and there was a return to dyslipoproteinaemia. High levels of the vector genome were still present in livers of treated animals at 70 days, implying that decrease in apoE expression was due to cellular shutdown of the cytomegalovirus promoter. Importantly, liver-directed apoE gene transfer to these young mice retarded progression of atherosclerosis by 38% (treated, 8.21 ± 1.05%; untreated, 13.26 ± 0.98%, P < 0.05), during the 70 day study period. Moreover, when 10-month-old apoE–/– mice with advanced atherosclerosis were treated with the adenovirus vector, there was clear regression of aortic lesion area by 1 month [24.3 ± 1.7% compared to 40.7 ± 2.6% in baseline controls (P < 0.002)]. We conclude that the stability of the adenovirus vector genome in the livers of intravenously treated animals provides an ideal platform to evaluate liver-specific promoters for sustained transgene expression and control of atherosclerotic lesion pathology.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Apolipoprotein E (apoE) is a 34 kDa plasma glycoprotein with multiple actions that help protect against the development of atherosclerosis. It is an important constituent of atherogenic remnant lipoprotein particles, including lipolysed chylomicrons and very low density lipoproteins (VLDL), and functions as the ligand for the receptor-mediated clearance of these particles by the liver for excretion (1,2). ApoE has several beneficial actions at atherosclerotic lesion sites that include cholesterol efflux via
LpE, an immature HDL particle in which apoE is the sole protein component (3–5), local anti-oxidant (6,7), anti-platelet (8) and anti-inflammatory functions (9). Genetic deficiency or the presence of dysfunctional forms of apoE usually lead to hypercholesterolaemia and premature atherosclerosis (10). The generation of apoE–/– mice has allowed the study of the pathology and the development of gene-based therapies for atherosclerosis, as the animal develops severe hypercholesterolaemia and atherosclerotic lesions similar to those found in humans (11,12). Restoration of apoE expression by the transplantation of normal bone marrow cells into apoE–/– mice results in increased lipoprotein clearance, reduction of cholesterol to normal lipidaemic levels and protection against the development of atherosclerotic lesions (13,14). Protection against atherosclerosis has been achieved in apoE–/– mice by the gene transfer of human apoE, using retroviral, adenoviral and plasmid expression vectors (15–22). Intravenous injection of adenovirus vectors containing apoE cDNA to apoE–/– mice, have demonstrated correction of lipoprotein and cholesterol levels, as well as inhibition of atherosclerotic lesion development (16–20). Conventional adenovirus vectors are constructed by substituting the E1 region of the adenovirus genome with the transgene cassette of interest. These vectors are prone to low level expression of viral genes still present in the adenovirus vector which can lead to the generation of an immune response directed against adenovirus-transduced cells, which ultimately results in the loss of transgene expression (23,24).
A major improvement in the design of adenovirus vectors is the removal of additional sequences in the vector genome. The construction of [E1–, E3–, polymerase–] adenovirus vectors has resulted in a decreased risk of adenovirus-derived gene expression and therefore removes the potential trigger for the immune clearance of adenovirus-transduced cells (25–27). In addition, an immune response against the adenovirus vector can initiate a response against a neoantigenic transgene product, and therefore removal of adenoviral gene expression may allow for stable transgene expression. These vectors also reduce the risk of the generation of replication competent adenovirus (RCA), as the appearance of RCA would require more than one recombinatorial event due to the multiple deletions in the adenovirus vector backbone. Upon intravenous injection of immunocompetent C57BL/6 mice with an [E1–, E3–, polymerase–] adenovirus vector expressing ß-galactosidase, the animals demonstrated prolonged expression and persistence of the adenovirus vector genome for 2 months after the injection, compared to the rapid clearance of a conventional [E1–] adenovirus vector (27). Following the use of the [E1–, E3–, polymerase–] adenovirus vector in this study, a further development in vector design is the deletion of the pre-terminal protein (pTP) gene from the adenovirus genome that results in increased cloning capacity (from 8.5 to 9.5 kb) and an additional recombinatorial event necessary to generate RCA (28).
Previously, we constructed plasmid-based vectors expressing either of the allelic human apoE2 or apoE3 isoforms under the control of the full cytomegalovirus (CMV) enhancer/promoter (21). ApoE3 is the most common fully functional isoform (1,10), whereas in a small percentage of patients homozygosity for apoE2 causes Type III hyperlipoproteinaemia (29). This is characterized by elevated levels of chylomicron and VLDL remnants and intermediate density lipoprotein (IDL) particles, with concomitant increases in cholesterol and triglyceride levels and consequently a predisposition to premature atherosclerosis (29). Following intramuscular administration of the apoE2 and apoE3 plasmid expression vectors into apoE–/– mice, we were able to demonstrate apoE expression 6 days post-injection (21). After 9 months, aortas from animals treated with the apoE2 plasmid demonstrated a 20–30% reduction in atherosclerotic lesions.
In the present study we have constructed an [E1–, E3–, polymerase–] adenovirus vector containing the full CMV enhancer/promoter driving expression of the human apoE3 cDNA (Ad-CMV/apoE) and have performed intramuscular and intravenous (liver-directed) injections into apoE–/– mice. We show reductions in total plasma cholesterol levels, normalization of lipoprotein profiles, together with retardation of early atherosclerotic lesions and acute regression of advanced atherosclerotic lesions, and persistence of the adenovirus genome throughout the course of the study.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Secretion of human apoE by HepG2 and C2C12 cells transduced with Ad-CMV/apoE
Serial dilutions of the [E1–, E3–, polymerase–] adenovirus vector Ad-CMV/apoE containing a transgene cassette of the full CMV enhancer/promoter driving expression of the human apoE3 cDNA (Fig. 1), were used to infect and engineer human hepatoblastoma HepG2 cells and mouse C2C12 myoblast cells to secrete human apoE into the culture supernatant. The adenovirus vector demonstrated a high efficiency of transduction in HepG2 cells, with the supernatant from untreated cultures revealing the presence of endogenously expressed apoE (Fig. 2). Infection of C2C12 myoblast cells resulted in substantially lower levels of apoE accumulating in the culture supernatant when the indicated multiplicity of infections (MOIs) were taken into account.
|
|
Detection and quantification of plasma apoE levels
Female apoE–/– mice at 6–8 weeks of age were injected with Ad-CMV/apoE either intravenously via the tail-vein at 1 x 1010 or 6 x 1010 virus particles (vp) per animal or intramuscularly with either 2 x 1010 or 9 x 1010 vp per animal. Tail-vein bleeds were taken at 3, 7, 14, 28 and 56 days post-injection, with all animals being killed at day 70. Following the intravenous injections, plasma apoE was readily detectable by western blotting analysis 3 days after treatment in a dose-dependent manner (Fig. 3A). In addition, quantification of circulating apoE levels by a two-antibody sandwich ELISA, demonstrated that mean peak expression levels were reached 3 days post-injection for the high dose and by day 7 for the low dose giving 7.5 ± 0.98 and 0.09 ± 0.03 mg/dl, respectively (plasma apoE levels in normal mice: 5–8 mg/dl) (30) (Fig. 3B). More importantly, plasma apoE was still detectable 70 days after the intravenous injections, giving 1.4 ± 0.49 and 7.0 ± 0.65 µg/dl for the low and high doses, respectively. In the case of the intramuscular injections, the high dose resulted in detectable circulating apoE by western blotting analysis and apoE ELISA after 3 and 7 days (125 and 185 ng/ml, respectively), but not at later time points (Fig. 4). Following the low dose intramuscular injections, circulating apoE was only detected by apoE ELISA 3 days post-injection, and on subsequent bleeds plasma apoE levels fell below the sensitivity limit of the assay (10 ng/ml).
|
|
Total plasma cholesterol levels
Plasma from individual animals was assayed for total cholesterol levels. Clear reductions in total plasma cholesterol were achieved after the intravenous injections of Ad-CMV/apoE compared to the untreated control group (Fig. 5A). In animals given low dose intravenous injections, cholesterol levels decreased dramatically between day 3 (208 ± 46 mg/dl) and day 7 (14 ± 8 mg/dl) after the injections. In the case of the high dose intravenous injections, cholesterol levels fell to 13 ± 3.3 mg/dl compared to those levels found in the untreated control group (range from 361 ± 31 to 470 ± 58 mg/dl), 3 days after the injections. By day 28, plasma cholesterol levels in the low dose intravenous group were no different to the control (332 ± 59 mg/dl), but the high dose intravenous group remained lower (175 ± 42 mg/dl). At 56 and 70 days post-injection, the total plasma cholesterol of animals receiving the high dose injections had returned to levels found in the untreated control group during the course of the study. In the case of both low and high dose intramuscular injections, a reduction in plasma cholesterol was only statistically significant at day 14 after the injections (Fig. 5B).
|
, P < 0.05; 
, P < 0.005; *, P < 0.000001.Lipoprotein distribution and apoE-containing lipoprotein classes
To determine whether the abnormal lipoprotein profile improved in animals treated at 6–8 weeks of age with the Ad-CMV/apoE vector, pooled plasma samples (n = 4) were subjected to 0.8% native agarose gel electrophoresis followed by staining with a lipid-specific stain, Sudan black, to analyse lipoprotein mobility (Figs 6A and 7A). The resulting gels were analysed by densitometry to determine the relative proportions of the lipoprotein classes (Figs 6B and 7B). In untreated apoE–/– mice the majority of lipoproteins were VLDL/IDL particles characterized as a slow pre-ß migrating broad staining band, with HDL running as a fast
migrating minor fraction (Figs 6A and 7A). In contrast, normal C57BL/10 mice have significantly lower proportions of VLDL/IDL with increased levels of circulating HDL (Figs 6A and 7A). Intravenous injection of Ad-CMV/apoE at both low and high doses resulted in complete correction of lipoprotein distribution by 7 and 3 days post-injection, respectively (Fig. 6). Although mice receiving the high intravenous dose showed raised levels of HDL with low levels of VLDL/IDL particles for at least 28 days, there was a continued decline toward pre-treatment levels at days 56 and 70. In the case of intramuscular injections there was no clear evidence of improvement in HDL (Fig. 7).
|
|
Following intravenous injections of 6–8-week-old apoE–/– mice (n = 7) with Ad-CMV/apoE, the location of the expressed apoE protein in the plasma was assessed by separating the lipoprotein classes by 0.8% native agarose gel electrophoresis, then immunoblotting with an antibody against human apoE. Co-localization of apoE with pre-ß migrating lipoprotein classes was evident 5 days after administration of adenovirus vector, in the absence of incorporation of apoE into
migrating (HDL) lipoprotein particles (Fig. 8). Pre-ß migrating lipoprotein particles containing apoE were still evident 56 days after adenovirus vector administration, with apoE levels declining to undetectable levels after 132 days.
|
Analysis of aortic atherosclerotic lesion area
To investigate the effect of human apoE gene transfer upon atherosclerotic lesion development the aortas were removed, dissected en face onto cork beds and stained with Oil-Red-O which specifically identifies the lipid-laden lesions. Upon examination of the aortas from those animals treated at 6–8 weeks of age with Ad-CMV/apoE by intravenous injection and sacrificed 70 days later, significant retardation of early lesion progression was observed compared to untreated controls, with 8.21 ± 1.05% (n = 8) and 13.26 ± 1.87 (n = 4) aortic lesion area, respectively (P = <0.05). Animals given intramuscular injections of Ad-CMV/apoE demonstrated no significant retardation of early atherosclerotic lesion development. Further, upon intravenous injection of Ad-CMV/apoE into apoE–/– mice at 10.5 months of age, followed by analysis 1 month later, a clear regression of advanced aortic lesions was observed (Fig. 9). Comparison of mean lesion area of the Ad-CMV/apoE-treated mice with the baseline and endpoint untreated control groups (which gave aortic lesion areas of 24.3 ± 1.7%, 40.7 ± 2.6% and 37.5 ± 2.5%, respectively) demonstrated significant regression of atherosclerosis of 40% (P <0.002) and 35% (P <0.005), respectively, by analysis of aortic lesion area (Fig. 9A), and this was also evident by gross morphological examination (Fig. 9B). No significant variation in lesion area was observed between the baseline and endpoint untreated control groups (P = 0.43).
|
Humoral immune responses against the apoE transgene product and adenovirus vector
Plasma samples were pooled and used to screen for antibodies against apoE and the adenovirus capsid proteins by western blotting analysis. In the case of anti-apoE antibody detection, the plasmas were screened against partially purified human apoE3. Throughout the course of the study (70 days) intravenous injection of Ad-CMV/apoE did not trigger a humoral immune response against the transgene product under the limits of detection (Fig. 10A), whereas low and high dose intramuscular injections resulted in the detection of anti-apoE antibodies 28–56 days post-injection (Fig. 10B). For the detection of antibodies against the adenovirus particles, plasmas were screened against a Hybond-P blot of Ad-CMV/apoE virus stock. Both intravenous and intramuscular injections resulted in the appearance of antibodies against the adenovirus vector, which could be detected at day 7 and persisted throughout the course of the experiment (Fig. 11).
|
|
Persistence of vector genome following intravenous injection
To assess the stability of the vector genome in vivo, animals were killed 70 days after the injections and DNA was isolated from the livers of apoE–/– mice injected intravenously with recombinant adenovirus. Real-time PCR was performed on known concentrations of the plasmid pAd-CMV/apoE using Ad fibre primers 1 and 2 with Sybr Green I to detect the amplification of the target sequence (Fig. 1). The analysis produced a linear standard curve allowing the quantification of vector copies in the unknown DNA samples derived from livers of the treated animals (Fig. 12A). The real-time PCR analysis demonstrated a dose-dependent level and persistence of vector genome (Fig. 12B). The livers from the two animals treated with 2 x 1010 vp and killed 8 and 16 days after the injections, contained 0.24 and 0.21 copies/cell, respectively, whereas those animals receiving low (1 x 1010 vp, n = 4) and high (6 x 1010 vp, n = 4) dose injections followed by analysis 70 days post-injection, displayed 0.053 ± 0.017 and 0.39 ± 0.067 copies/cell, respectively. The dose-related levels of vector genome in the livers of intravenously injected apoE–/– mice was still maintained irrespective of the time at which the animals were killed. Accordingly, the lowest vector levels were found in animals that received 1 x 1010 vp killed at 70 days, then animals injected with 2 x 1010 vp killed at 8 and 16 days, followed by those injected with 6 x 1010 vp killed at 70 days. The melting temperatures of the amplified products had values within the narrow range 84.59–84.95 ± 0.036°C, confirming the specific amplification and detection of the target sequence within the vector genome. A DNA sample derived from one of the untreated animals used in the analysis resulted in the amplification of product with a melting temperature of 74.15°C, which was due to the formation of non-specific primer dimers that characteristically have substantially lower melting temperatures than the specific PCR product. The presence of the specific 196 bp product following real-time PCR, was confirmed by subjecting the reactions to 2% agarose gel electrophoresis and visualization of the PCR products by ethidium bromide staining (data not shown).
|
Human apoE gene transcription in the liver
Total RNA was isolated from the livers of animals injected intravenously with the adenovirus vector Ad-CMV/apoE to evaluate the level of human apoE transcription over the course of the study. ApoE-specific cDNAs were readily detectable in the two animals injected intravenously with 2 x 1010 vp and killed 8 and 16 days post-injection (Fig. 13). However, a decline in amplified product was observed between these time points, a trend that was confirmed by the analysis of total RNAs from animals killed 70 days after receiving low (1 x 1010 vp, n = 4) and high (6 x 1010 vp, n = 4) dose intravenous injections. In the case of animals receiving the low dose, no apoE cDNAs were detected, whereas the high dose injections resulted in the presence of low levels of apoE transcripts. These results using conventional PCR amplification of apoE cDNA target sequences were confirmed by real-time PCR, where the amplified product was only detected in those animals killed at 8 and 16 days and those animals given the high dose injections (data not shown). The decline in apoE transcripts over the course of the study was validated by the amplification of the housekeeping gene for glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (Fig. 13). In addition, real-time amplification gave similar levels of GAPDH transcripts as the cDNAs produced a mean cycle threshold of 18.08 ± 0.33 with a mean specific melt temperature of 81.74 ± 0.035, giving further confirmation of normalization of the cDNAs and showing the specificity of the real-time amplification, respectively.
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Previous studies have demonstrated correction of hypercholesterolaemia and protection against atherosclerosis by the intravenous injection of adenovirus vectors expressing apoE (16–20). Conventional first generation adenovirus vectors contain deletions in the E1 and/or E3 regions of the vector genome (31), and are prone to low levels of viral gene expression; this leads to immune clearance of adenovirus-transduced cells and hence rapid loss of transgene expression (23,24). In this study, we utilized a new improved second generation adenovirus vector with the aim of achieving persistence of the vector genome in vivo and consequently, sustained expression of apoE in the treated apoE–/– mice. The improved [E1–, E3–, polymerase–] adenovirus vectors include the deletion of the adenovirus DNA polymerase (E2b) gene, which results in a significantly reduced potential for viral late gene expression and therefore diminishes the risk of an immune response directed against the virus (25–28).
Previously, we assessed the functionality of the transgene cassette containing the full CMV enhancer/promoter driving expression of the human apoE cDNA, by transfection of cultured mouse C2C12 cells with a plasmid expression vector containing the transgene cassette (21). The secreted proteins became incorporated into spherical lipoprotein particles similar to those involved in cholesterol efflux (3–5), thereby providing evidence for the functional activity of the expressed apoE.
In this study, the integrity of the transgene cassette within the [E1–, E3–, polymerase–] adenovirus vector Ad-CMV/apoE was confirmed by detection of apoE in culture supernatants of HepG2 (in addition to endogenously expressed apoE) and C2C12 myoblast cells, following infection with the adenovirus vector. In addition, the infections demonstrated a high efficiency of transduction in liver-derived HepG2 cells compared to the mouse myoblast C2C12 cell line. This reflects the propensity of adenovirus virus vectors to target the liver following intravenous administration in rodents (32–35).
The appearance of apoE in the circulation following the intravenous injection of Ad-CMV/apoE into apoE–/– mice coincided with dramatic reductions in total plasma cholesterol below wild-type levels [total plasma cholesterol levels in normal mice: 80–150 mg/dl (11,36)], with complete normalization of lipoprotein distribution. At 70 days after the injections, the proportion of HDL particles to total lipoproteins was still raised compared to pre-treatment levels. In contrast, following the intramuscular injections, the reduction in total plasma cholesterol observed with the high dose injections corresponded to peak plasma apoE levels at 7 days, after which apoE was lost from the circulation. However, a mild increase in HDL particles compared to the VLDL/IDL particle population was still evident. Muscle tissue expresses 100-fold lower levels of the coxsackie and adenovirus receptor (CAR) than liver (37). This may explain the low efficiency of adenovirus-mediated apoE gene transfer to muscle compared to the high level of transduction observed in the liver following the intravenous injections, as CAR is the receptor for virus attachment (38).
The improved clearance of these remnant lipoprotein particles and the consequent reduction in plasma cholesterol observed in these investigations, is due to the incorporation of circulating apoE into these apoB-containing lipoprotein particles, allowing their clearance via the liver through the binding of apoE with the low density lipoprotein receptor (LDL-R) and the LDL-R-related protein (LRP) (1,39), and also cell-surface heparan sulfate proteoglycans (HSPG) (40). The observed increase in HDL particles in the presence of circulating human apoE, can be explained by the availability of more apoA1 for HDL formation, since apoA1 becomes a structural component in VLDL and IDL particles in the absence of apoE (11,13).
In the case of low levels of circulating human apoE in apoE–/– mice, the apolipoprotein can still affect the plasma lipoprotein distribution. ApoB48 is a truncated version of the apolipoprotein apoB100, which is missing the LDL-R binding domain and therefore VLDL and IDL particles containing apoB48 accumulate in the plasma. ApoE–/– mice are characterized by the accumulation of VLDL/IDL in the plasma, which is reported to be due to 20-fold elevated levels of apoB48 compared to wild-type (11). Therefore, apoE–/– mice are dependent on apoE for the efficient clearance of these particles from the circulation, and the introduction of circulating apoE by adenovirus-mediated gene transfer may result in the rapid clearance of these particles and hence lead to low or even undetectable levels of apoE (21,22). A disproportionate increase in circulating transgene product was observed following the intravenous injections, where a 6-fold dose increase produced a >80-fold increase in plasma apoE. This may reflect that saturation of LDL-R, LRP and HSPG by apoE-containing remnant lipoprotein particles in the space of Disse of the liver, has been reached by the high dose. In the normal physiological range plasma apoE is present in excess (5–8 mg/dl) (30), and is involved in the recruitment of remnant lipoprotein particles for receptor-mediated uptake in the space of Disse of the liver (40).
Intravenous injection of adenovirus vectors results in the majority of the virus targeting the liver (32–35), followed by the clearance of >90% of the viral genomes within the first 2 days following the injections, due to direct toxicity of the virus infection (23,41). In the case of first generation adenovirus vectors, virus late gene expression leads to a cell-mediated immune response against the adenovirus-transduced liver cells due to MHC class I presentation of virus antigens (23,24,35). This leads to the elimination of the remainder of transduced cells by 3 weeks after the injections, resulting in the loss of transgene expression (23,24,35,41). In fact, CD4+ T-cell depletion of immunocompetent C57BL/6 mice before intravenous injection of a first generation adenovirus vector carrying the LacZ reporter gene, resulted in prolonged expression of ß-galactosidase in the livers of these animals from a baseline of less than 2–7 weeks (42). The loss of transgene expression by 9 weeks coincided with the emergence of cytotoxic T lymphocyte activity against the adenovirus-transduced cells.
With regard to the improved [E1–, E3–, polymerase–] adenovirus vectors, the deletion of the E2b region profoundly diminishes the risk of virus late gene expression (25–27). Since previous studies with early generation vectors have demonstrated the clearance of residual adenovirus-transduced cells by cell-mediated immune responses (23,24,35), the persistence of residual vector genome in this study, suggests that removal of virus late gene expression may avoid this effect in the liver. Moreover, the absence of an immune response against the transgene product after liver-directed gene transfer supports this conjecture.
Following intravenous administration of adenovirus vector into apoE–/– mice, secretion of human apoE into the circulation was evident throughout the course of the study (70 days). In a previous study, human apoE gene transfer to apoE–/– mice, utilizing a first generation adenovirus vector containing the CMV promoter driving expression of human apoE, resulted in transient apoE expression for 1 month after the injections, with peak expression levels of circulating apoE being attained at 4 days, ranging from 1.5 to 650 mg/dl (16). This transient expression is likely to be due to a combination of immune clearance of adenovirus-transduced cells as a result of using a first generation adenovirus vector (23,24,35,41), an immune response against the transgene product (24), and CMV promoter shutdown in the liver (43–46). In this study, RT–PCR analysis has demonstrated a decline in apoE transcripts but not vector genome over the course of the study, which is strongly suggestive of CMV promoter shutdown. Verification that CMV promoter shutdown is a major factor in the loss of plasma apoE, would require reactivation of the CMV promoter to be demonstrated experimentally (45). Adenovirus infection of liver results in activation of NF
B, a transcription factor that stimulates expression of the transgene by direct interaction with the CMV promoter (45). The loss in CMV promoter-driven gene expression is explained by the fall in activated NFB as the liver recovers from the infection and the generation of an immune response against the virus, resulting in the production of the cytokine -interferon, which can down-regulate the activity of the CMV promoter (46). In comparison, the prolonged expression observed with the [E1–, E3–, polymerase–] adenovirus vector suggests the removal of an immune response directed against the vector due to a lack of viral late gene expression that is characteristic of first generation adenovirus vectors.
The involvement of an immune response against the virus vector in combination with a promoter predisposed to shutdown in the liver, was also demonstrated in apoE–/– mice by the use of a first generation adenovirus vector containing the Rous sarcoma virus promoter driving expression of the human apoE cDNA (17). ApoE gene transfer resulted in low levels of transgene expression due to the lower strength of the RSV promoter (plasma apoE: 0.12 mg/dl) and was detected up to 1 month after the injections. The first report of the use of a second generation adenovirus vector with CMV-driven expression of human apoE, which contained a temperature-sensitive mutation in the adenovirus E2a gene, resulted in prolonged transgene expression to 3 months after the injections (
0.1 mg/dl) (18). In a recent study, apoE–/– nude mice were treated with a first generation adenovirus containing human apoE cDNA and complete regression of atherosclerosis was demonstrated 6 months post-injection (20). As these animals were immunodeficient, no immune response was mounted against the adenovirus vector, and recognition of human apoE as a neoantigen was minimal, which resulted in prolonged transgene expression.
Mice are highly resistant to the development of atherosclerosis, whereas apoE–/– animals are predisposed to the development of atherosclerotic lesions similar to those found in humans (11,12,47). As early as 2 months of age, apoE–/– mice develop fatty streak lesions that contain foam cell-rich depositions in their proximal aorta, with progression to more complex atherosclerotic lesions by 5 months (11,12,16,36). These advanced lesions consist of proliferating smooth muscle cells, a fibrous cap and a lipid-rich necrotic core that result in the occlusion of the affected artery. In this study, a significant retardation of early lesion development was seen, and acute regression of established atherosclerotic lesions in animals 10.5 months of age was observed 1 month after Ad-CMV/apoE vector administration. The older animals contained advanced lipid-laden atherosclerotic lesions at the time of vector administration and, under these conditions, the presence of circulating human apoE caused regression to significantly less than pre-treatment levels. Furthermore, plasma apoE associated with VLDL resulted in the increased clearance of this lipoprotein from the circulation, reducing the risk of these particles forming pro-atherogenic lipoprotein particles. Indeed, there is an apoE isoform-dependent relationship between the residence time of VLDL in the circulation and atherosclerotic lesion size, where extended VLDL residence times are associated with increased lesion size (48). Liver-directed administration of adenovirus vectors expressing apoE into apoE–/– mice have reported the localization of plasma-derived apoE in pre-existing atherosclerotic lesions with marked regression within 7 months after vector administration (16,19,20). Plasma apoE may gain access and induce cholesterol efflux from the lesions via LpE (3–5) and pre-ß-LpE (49), immature HDL particles that are major acceptors of cell-derived cholesterol.
In contrast to liver-derived apoE, macrophage-derived apoE following transplantation of apoE-expressing bone marrow cells in apoE–/– mice, protects against the development of atherosclerosis (13,14), but does not regress pre-existing lesions. Reduction of atherosclerotic lesions was reported only in young mice (from 5 to 13 weeks), and not in older animals (from 10 to 26 weeks), following retrovirus-mediated ex vivo transduction and re-implantation of apoE–/– bone marrow stem cells (15). Therefore, apoE expressed from macrophages appears to be effective at protecting against the early stages of atherogenesis but has no effect on advanced lesions. Furthermore, transplantation of apoE+/+ bone marrow cells into 8-week-old apoE–/– mice did not induce regression, but prevented lesion progression 20 weeks post-transplantation (50). The ability of adenovirus-mediated gene transfer to protect against early lesion progression as well as to induce dramatic regression of pre-existing advanced lesions, compared to the milder therapeutic effect of macrophage-derived apoE through bone marrow transplantation, could be related to the higher systemic levels of apoE achieved with the adenovirus vectors. The threshold level of plasma apoE that results in the lowering of total plasma cholesterol is 0.04 mg/dl, and below this level apoE is unable to clear remnant lipoprotein particles significantly from the circulation (30). Moreover, a dose-dependent effect has also been observed, where reduced lesion size in apoE–/– mice correlated with higher levels of plasma apoE (20). Normal levels of plasma apoE in mice and humans is in the range 5–8 mg/dl (30), therefore in this study, liver-directed apoE gene transfer utilizing an [E1–, E3–, polymerase–] adenovirus vector has achieved not only therapeutic, but physiological levels of circulating apoE.
The absence of antibodies against apoE in animals that received intravenous injections was striking. The feature of [E1–, E3–, polymerase–] adenovirus vectors that leads to their persistence in vivo is the deletion of the adenovirus E2b gene that diminishes the risk of viral late gene expression (27). In effect, this removes the primary trigger that leads to an immune response against adenovirus vector and therefore may prevent an immune response being directed against the neoantigenic transgene product. Although a humoral immune response against virus capsid proteins was evident 1 week post-injection, this was directed against input virus particles in the circulation. Previous studies have failed to elicit an immune response against human blood coagulation factor IX (hFIX), following intravenous injection of C57BL/6 mice with adenovirus vector expressing hFIX, as opposed to the generation of an immune response in other immunocompetent mouse strains (51,52). Therefore, lack of a humoral immune response against apoE following intravenous injection of the [E1–, E3–, polymerase–] adenovirus vector, may not necessarily be extrapolated to other animal models and consequently to human gene therapy trials. However, this would not be an issue in the case of apoE deficiency, where apoE transgene expression would restore physiological levels of circulating apoE in these patients, who characteristically have plasma apoE levels at less than 1% of the normal physiological range (29,53,54).
The rapid loss in circulating plasma apoE following the intramuscular injections can be explained by the ability of the adenovirus vector to transduce and activate professional antigen-presenting cells that are resident in the muscle tissue (55). In addition, direct intramuscular injection results in localized trauma at the injection site, which may contribute to the strong inflammatory response, resulting in a mixed infiltrate of immune cells that include activated lymphocytes and macrophage/dendritic cells (55). Following intramuscular injection of immunocompetent mice with a first generation adenovirus vector expressing factor IX, CD8+ lymphocytes were detected within 3 days and anti-factor IX antibodies 13 days after the injections, with factor IX expression being lost within days of the injections (56). Furthermore, intramuscular injection of plasmid expression vectors results in strong humoral and cell-mediated immune responses against the expressed protein, suggesting a role for muscle in vaccination (57). The lack of anti-apoE antibodies following intravenous injection of Ad-CMV/apoE, may reflect the non-invasive nature of liver transduction by this route, minimizing the risk of inducing an inflammatory response. In the liver, kupffer cells derived from the monocyte–macrophage lineage can be infected by adenovirus, and are implicated in the clearance of adenovirus-transduced cells in the first hours after infection (41), but their subsequent role in the immune response against the adenovirus vector, adenovirus-transduced cell and transgene product is unknown.
In conclusion, human apoE gene transfer to liver of apoE–/– mice using the improved second generation [E1–, E3–, polymerase–] adenovirus vector system, resulted in clear reductions in plasma cholesterol below wild-type levels, complete normalization of lipoprotein distribution and, more importantly, retardation of early fatty streak lesion progression and regression of advanced atherosclerosis to less than pre-treatment levels. In addition, the vector genome remained persistent in the livers of treated animals throughout the course of the study. Furthermore, no humoral immune response was generated against apoE following the intravenous adenovirus injections, although plasma apoE levels declined over the course of the study, suggesting CMV promoter shutdown. The lack of anti-apoE antibodies is a distinct advantage in developing a gene-based treatment where an antigenic therapeutic protein is expressed in the recipient. A liver-specific promoter driving expression of factor IX has been used in a rAAV-based vector and intravenous injection of the vector into mouse and canine models of haemophilia B resulted in sustained therapeutic levels of the transgene product for more than 5 months (58,59). Utilization of such a promoter driving expression of apoE in the context of an [E1–, E3–, polymerase–] adenovirus vector, may lead to continuous transgene expression, and hence regression of pre-existing lesions and protection against the development of atherosclerosis. The dramatic regression of pre-existing advanced atherosclerotic lesions just 1 month after a single administration of [E1–, E3–, polymerase–] adenovirus vector expressing apoE, together with normalization of the hyperlipidaemic phenotype and persistence of adenovirus vector genome in the liver, demonstrate the potential in the further modification of these second generation adenovirus vectors for the somatic gene therapy of atherosclerosis.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONRESULTSDISCUSSIONMATERIALS AND METHODSREFERENCES |
|---|
Recombinant adenovirus construction
Ad-CMV/apoE was constructed using the bacterial recombination procedure (60). ApoE3 cDNA was excised from pUC18-apoE3 (61) by digestion with HindIII and BamHI then ligated into BamHI–HindIII-digested pcDNA3. The apoE3 cDNA was then sub-cloned into pShuttle-CMV using HindIII and EcoRV. The plasmid pShuttle-CMV/apoE was linearized with PmeI and co-transformed with pAdEasy-
pol (26) into the BJ5183 strain of Escherichia coli. Positive recombinants were linearized with PacI and transfected into C7 cells (62). Recombinant Ad-CMV/apoE vector was isolated from an individual plaque of transfected C7 cells, plaque purified once and grown to high titre stock (1.8 x 1012 vp/ml) using routine procedures (63). The purified virus was stored at –80°C in viral storage buffer containing 10 mM Tris–HCL pH 8, 2 mM MgCl2 and 10% glycerol.
Culture and Ad-CMV/apoE infection of HepG2 and C2C12 cells
The human hepatic carcinoma cell line HepG2 (64) and the murine myoblast C2C12 cell line (65) were maintained in DMEM containing 10% fetal bovine serum (FBS), 2 mM L-glutamine, 100 U/ml penicillin, 100 µg/ml streptomycin and 250 ng/ml amphotericin B at 37°C and 8% CO2. HepG2 and C2C12 cells were seeded in six-well plates at 2 x 106 and 1.5 x 105 cells per well, respectively. The following day, the cells were infected with the same virus dilutions of Ad-CMV/apoE at the indicated MOIs (Fig. 2). Briefly, cell culture supernatants were replaced with 700 µl of infection mix containing the appropriate concentration of virus particles in phosphate buffered saline (PBS)/MgCl2/CaCl2 (PBS containing 49 mM MgCl2 and 68 mM CaCl2). After 90 min at 37°C/8% CO2 the infection mixes were replaced with fresh culture medium and the infected cells were incubated for 24 h, before the culture medium was replaced with medium containing 5% FBS. After a further 24 h culture, supernatant was harvested for the detection of secreted human apoE by western blotting analysis.
In vivo administration of Ad-CMV/apoE by intravenous and intramuscular injections
Female C57BL/6 apoE–/– mice provided by Glaxo Wellcome (Stevenage, UK), were generated by inactivation of the mouse apoE locus through homologous recombination as previously described by Piedrahita et al. (47). The animals were maintained on a normal chow diet and blood samples (
50 µl) were taken from the tail-vein following 4 h fasts. Before intramuscular injections of virus particles, animals were anaesthetized by intra-peritoneal administration of a 25% (v/v) solution of hypnorm and hypnovel (3 µl/g body weight). Animals to be treated by intravenous injection were placed on a heated pad to induce vasodilation. The Ad-CMV/apoE virus stock was diluted appropriately using diluent containing 10 mM Tris–HCL pH 8, 2 mM MgCl2 and 0.9% (w/v) NaCl.
To determine the effect of apoE gene transfer upon the hyperlipidaemic phenotype and early atherosclerotic lesions of apoE–/– mice, animals at 6–8 weeks of age were administered Ad-CMV/apoE by either intravenous or intramuscular injections. For intravenous injections, 1 x 1010 vp (low dose, n = 4) or 6 x 1010 vp (high dose, n = 4) were given in a total volume of 250 µl by tail-vein injection. In the case of the intramuscular injections the animals were injected with either 2 x 1010 vp (low dose, n = 4) or 9 x 1010 vp (high dose, n = 4), the dose being equally divided between both tibialis anterior muscles in 50 µl volumes. A group of untreated animals (n = 4) were used as the controls. The animals were killed 70 days after the injections. To assess the stability of the vector genome in vivo over time, two animals were each injected intravenously with 2 x 1010 vp and an animal was killed at 8 and 16 days post-injection.
To establish the location of apoE in the plasma of apoE–/– mice, following adenovirus vector administration, 6–8-week-old animals (n = 7) were injected intravenously with Ad-CMV/apoE at 2 x 1010 vp per animal, and bloods taken subsequently at 5, 14, 28, 56 and 132 days.
To determine the effect of apoE gene transfer on advanced atherosclerotic lesions, apoE–/– mice at 10.5 months of age were given 2 x 1010 vp (n = 4) by tail-vein injection and killed 4 weeks later. To control for variation in lesion development over the course of the 4 week experiment, a group of untreated animals were killed at the beginning (baseline control group, n = 5) and at the end (endpoint control group, n = 4) of the study.
Tail-vein bleeds were carried out where 50 µl of blood was anti-coagulated with sodium citrate. Plasma samples were stored at –80°C.
Detection of human apoE in culture supernatants and plasma from apoE–/– mice by western blotting analysis
Culture supernatant from transduced cells (17 µl) or plasma samples (4 µl), were denatured by the addition of SDS–PAGE sample buffer containing 2.5% (v/v) ß-mercaptoethanol and heating at 100°C for 5 min. Samples were then subjected to 4–12% SDS–NuPAGE electrophoresis (Invitrogen/Novex, Groningen, The Netherlands), and resolved proteins transferred to an ECL-nitrocellulose membrane (Amersham Pharmacia Biotech, Little Chalfont, UK). Nitrocellulose blots were incubated with goat anti-human apoE antibody (1:3000 dilution; Biogenesis Ltd, Poole, UK ) followed by an anti-goat-HRP secondary antibody (1:3000 dilution; Sigma, Poole, UK). Blots were developed using the ECL detection system (Amersham Pharmacia Biotech).
Enzyme-linked immunosorbent assay (ELISA) for quantification of human apoE
Human apoE in mouse plasma was detected by a two-antibody sandwich ELISA. Polyclonal goat anti-human apoE antibodies were used both for capture (DiaSorin Inc., Stillwater, MN) and, after biotinylation using a commercial kit (Amersham Pharmacia Biotech), for detection (Biogenesis Ltd). Purified human apoE (Technoclone Ltd, Dorking, UK) was used as a standard. Standard and mouse plasma samples were diluted in assay buffer, 150 mM NaCl, pH 7.4 containing 0.5% (w/v) bovine serum albumin, 0.05% (w/v) -globulin, 0.01% (v/v) Tween-40 and 50 mM Tris; the sensitivity for detection of apoE was 10 ng/ml.
Determination of plasma cholesterol levels
Plasma from tail-vein bleeds was diluted 1/10 in PBS and 10 µl was used to estimate total cholesterol levels by the infinity cholesterol reagent assay system (Sigma), as described by the manufacturer. Briefly, the assay was performed in microtitre 96-well plates and 90 µl of infinity cholesterol reagent per sample was used followed by an incubation at 37°C for 5 min and then measuring the absorbance at 510 nm.
Detection of anti-apoE and anti-adenovirus antibodies by western blotting analysis
For the detection of anti-human apoE antibodies in mouse plasma, human apoE was partially purified from CHO cells stably expressing the protein (CHO-apoE3) (66). CHO-apoE3 cells were cultured in CD-CHO medium (Life Technologies, Paisley, UK), which requires no addition of serum and contains 20 times less protein than conventional growth media. Human apoE3 was partially purified from the culture supernatant by concentration using a Vivaspin 30 kDa filter (Vivascience Ltd, Binbrook, UK), followed by dialysis against PBS. Human apoE protein (1 µg) was denatured by the addition of SDS–PAGE sample buffer containing 2.5% (v/v) ß-mercaptoethanol and heating to 100°C for 5 min. Subsequently the protein was loaded into a preparative well of a 4–12% SDS–NuPAGE (Invitrogen/Novex) and subjected to electrophoresis. For the detection of antibodies against the virus vector, 9 x 1010 vp of Ad-CMV/apoE was used in place of the partially purified human apoE. The electrophoresed proteins were transferred to a Hybond-P (PVDF) membrane (Amersham Pharmacia Biotech) and placed in a Mini-PROTEAN® II multiscreen apparatus (Bio-Rad, Hemel Hempstead, UK). Equal volumes of mouse plasma from each animal within each treatment group were pooled and 2.5 µl was diluted in 600 µl of incubation buffer [TBST containing 2.5% (w/v) milk powder, 0.05% (v/v) Tween-20 and 0.2% (w/v) 2-chloroacetamide] and this was used to screen for antibodies against human apoE. The same primary incubation buffer was used for the detection of antibodies against adenovirus capsid proteins. A goat anti-mouse-HRP (Jackson Laboratories, Bar Harbor, ME) was used as a secondary antibody, followed by detection using ECL western blotting detection reagents (Amersham Pharmacia Biotech).
Plasma lipoprotein distribution analysis and lipoprotein-apoE immunoblotting
Lipoprotein profiles of pooled plasma (2 µl) were performed by electrophoresis on pre-cast alkaline buffered (pH 8.8) agarose gels (YSI Limited, Farnborough, UK) followed by staining with lipid-specific Sudan black according to the manufacturer’s instructions. The relative proportions of the Sudan black stained lipoprotein fractions were determined by scanning densitometry using an imaging densitometer Model GS-670 with Molecular Analyst Version 1.4 software (Bio-Rad).
To determine the location of apoE in the plasmas of treated animals, following electrophoresis the lipoprotein particles were transferred to a Hybond-P membrane (Amersham Pharmacia Biotech) for immunoblotting with a goat anti-human apoE antibody (1:3000 dilution; Biogenesis Ltd) and anti-goat-HRP secondary antibody (1:3000 dilution; Sigma), followed by detection using the ECL detection system (Amersham Pharmacia Biotech).
Dissection and examination of the aortic arch for atherosclerotic lesions
Having killed the animals, the hearts and thoracic aortas were taken and all adventitial fat was dissected away before cutting the aortas longitudinally and then pinning them out en face onto cork beds. The dissected aortas were stained with Oil-Red-O stain (Sigma) modified from previously described methods (19,21). Briefly the aortas were fixed in PBS/4% formaldehyde for 5 days, washed in PBS followed by staining with 1.8% Oil-Red-O in 60% isopropanol for 15 min at room temperature, then destained in 60% isopropanol for 5 min. The stained aortas were transferred to PBS for storage at 4°C. Images of the aortas were captured with a Nikon digital camera and analysis of aortic lesion area from the aortic root of the heart down to the diaphragm was achieved using the image analysis software Sigma Scan Pro5.
Quantification of vector genomes after liver-directed apoE gene transfer
To determine the levels of vector genome in the livers of those animals given an intravenous injection of Ad-CMV/apoE, real-time PCR was performed on isolated liver genomic DNA using Sybr Green I for detection in conjunction with the melting temperature of the amplified product to demonstrate the specificity of the analysis. Genomic DNA was isolated from liver samples using DNAzol reagent (Life Technologies) and purified further by Prepanol precipitation (DNAmp Ltd, Farnborough, UK) according to both manufacturer’s instructions. Real-time PCR was performed using the Sybr Green I 2x reaction system (Eurogentec, Seraing, Belgium) which utilizes Hot Goldstar DNA polymerase and uracil N-glycosylase (UNGase). Each 25 µl reaction contained 125 ng genomic DNA, Sybr Green I 2x reaction buffer, Sybr Green I (1/66 000 stock dilution), 5 mM MgCl2 and 0.3 µM of each primer spanning nucleotides 26541–26737 of Ad-CMV/apoE within the adenovirus fibre gene (Ad fibre primer 1, 5'-CCGCACCCACT ATCTTCATG-3'; Ad fibre primer 2, 5'-AACTAGAGGTTCGGATAGGC-3') yielding an amplified product of 196 bp. To quantify the vector genome concentration in the DNA samples, a standard curve of serial 10-fold dilutions of the vector plasmid pAd-CMV/apoE from 33.7 to 3.37 x 107 copies/reaction were prepared. Target sequences were amplified using a Cepheid Smart Cycler (Oswel, Southampton, UK) and the reaction conditions were 50°C for 2 min to allow UNGase digestion of any carryover amplified product; 95°C for 10 min to activate the Hot Goldstar DNA polymerase and denature the DNA; then 40 cycles of 95°C for 15 s, 60°C for 30 s and 79°C for 15 s, with optical fluorescence monitoring switched on for the latter extension temperature only. The Cepheid Smart Cycler software determined the cycle threshold for the standards and unknown DNA samples, which was directly related to the amount of target sequence present in the reactions. These values were used to create a standard curve in order to determine the amount of vector genomes in the unknown samples. Following completion of the real-time amplification, the reactions were subjected to a 0.2°C/s ramping from 60 to 95°C in order to demonstrate the specific melt temperature of the amplified product. To verify the size of the amplified product, a proportion of the reactions were electrophoresed on a 2% agarose gel containing ethidium bromide.
RT–PCR and real-time PCR analysis of apoE transcription
To assess the level of apoE gene expression in apoE–/– mice injected intravenously with the adenovirus vector Ad-CMV/apoE, total RNA was isolated from the livers of animals injected with high (n = 4) and low (n = 4) intravenous doses. To assess the stability of apoE expression over the course of the study, total RNAs were isolated from the livers of the two animals injected intravenously with 2 x 1010 vp, which were killed at 8 and 16 days post-injection. Total RNAs were isolated from livers that had been snap-frozen in liquid nitrogen then stored at –80°C. Frozen portions of tissue were thawed in 0.5 ml of RNAlater (Ambion, Abingdon, UK) and sliced into small pieces, followed by the isolation of total RNA with Trizol reagent (Life Technologies), according to the manufacturer’s instructions. DNaseI-treated RNAs (5 µg) were reverse transcribed in a 50 µl reaction containing 25 U RNaseH– MMuLV reverse transcriptase (GeneSys Ltd, Farnborough, UK), 15 ng/µl oligo dT(15)-Y-N primer, 0.08 U RNasin ribonuclease inhibitor (Promega, Madison, WI), 0.5 mM dNTPs and 5x RNaseH– MMuLV-RT reaction buffer. The reaction conditions were 42°C for 45 min, then 75°C for 10 min. The derived cDNAs were used to amplify a 155 bp product using a primer set specific for the human apoE sequence (Fig. 1). In addition, amplification of the housekeeping gene for glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was performed using a published primer set (67), to demonstrate the presence of equivalent copies of GAPDH transcripts between the RNA preparations. Each 50 µl reaction contained 45 µl Accurase PCR Master Mix (DNAmp Ltd) and the remaining 5 µl contained 1 µl cDNA and 20 pmol of each primer (apoE primer 1, 5'-CTGGGAACTGGCACTGGG-3'; apoE primer 2, 5'-CCGATTTGTAGGCCTTCAAC-3'). The reaction conditions were 92°C for 2 min, then 25 cycles of 92°C for 20 s, 62°C (for apoE) or 60°C (for GADPH) for 20 s, then 72°C for 20 s, followed by a final extension of 72°C for 10 min. The PCR reactions were subjected to 2% agarose gel electrophoresis and the products visualized by ethidium bromide staining.
To confirm the transcript levels determined by conventional PCR, real-time PCR was performed on the cDNAs using the Sybr Green I 2x reaction system, to quantify the level of apoE transcripts in the livers relative to GADPH transcript levels. Each 25 µl reaction contained 0.5 µl cDNA, Sybr Green I (1/66 000 stock dilution), 5 mM MgCl2 and 0.3 µM of the appropriate primer sets. Reaction conditions were 50°C for 2 min, 95°C for 10 min, then 40 cycles of 95°C for 15 s, 62°C for 30 s, 84°C for 15 s (for apoE) or 40 cycles of 95°C for 15 s, 60°C for 30 s (for GADPH), with the optical fluorescence monitoring switched on for the latter respective extension temperatures for determination of the cycle thresholds of the reactions. The melt temperatures of the amplified products were determined by running a ramping program from 60 to 95°C at 0.2°C/s.
|
ACKNOWLEDGEMENTS |
|---|
We gratefully thank J.L.Breslow for providing the human apoE3 cDNA. We thank A.Tagalakis for providing the partially purified recombinant apoE. Parts of this work were supported by grants from the British Heart Foundation, Sir Jules Thorn Charitable Trust, European Union and Wellcome Trust. A.A. was supported by NIH-grant R01-52925.