WO2000030669A2 - Use of plasma phospholipid transfer proteins in treating cardiovascular dieases - Google Patents

Abstract

The present invention relates to the field of cardiovascular diseases, in particular to the field of diseases associated with elevated plasma levels of cholesterol and/or triglycerides in mammals, particularly in man. The present inventors have found beneficial pharmaceutical effects of plasma phospholipid transfer protein (PLTP) and have shown that it is feasible to make pharmaceutical compositions for the treatment of cardiovarcular disorders based on PLTP.

Description

Title: Use of plasma phospholipid transfer proteins in treating cardiovascular diseases.

The present invention relates to the field of cardiovascular diseases, in particular to the field of diseases associated with elevated plasma levels of cholesterol and/or triglycerides in mammals, particularly in man. Cardiovascular diseases are a leading cause of death, particularly in the western world and especially for men over 45 and women over 65 years of age. World-wide mortality for cardiovascular diseases exceeds 7 million people per annum. Many factors play a role in the aetiology of cardiovascular diseases, such as coronary artery disease. Important ones are hyperlipidaemia, hypercholesterolaemia, hypertension and smoking.

Hyperlipidaemia and hypercholesterolaemia are strongly implicated in the development of atherosclerotic plaques, the growth of which leads to impaired blood flow or occlusion of arteries. Until recently most therapies available were dietary lowering of atherogenic lipids or lipoproteins, possibly applied together with lipid-lowering drugs, such as statins, in high risk patients. For some time now, therapies aim at reducing LDL (low density lipoproteins) cholesterol and triglyceride levels as well as at increasing HDL (high density lipoproteins) cholesterol. Statins are drugs which are capable of lowering LDL cholesterol values by more than 15%, which effect is accompanied by significant reductions in total plasma cholesterol and triglycerides. They also seem to have a moderate increasing effect on HDL cholesterol levels (for a review see Farnier et al. Am. J. Cardiol 1998 ; 82 : 3J-10J) . The present inventors are interested in HDL as an anti- atherogenic agent for the following reasons. Plasma levels of HDL are among the best indicators for the risk of atherosclerosis in epidemiological studies (7,8). HDL plasma levels are inversely related to atherosclerosis. HDL are thought to be anti-atherogenic, for example because they mediate efflux of cholesterol from peripheral cells and transport cholesterol to the liver for degradation into bile acids and excretion. This process is known as reverse cholesterol transport (8-10) . Preβ-HDL, a subclass of HDL, are very efficient acceptors of cellular cholesterol in vitro (13-14). Phospholipid transfer protein (PLTP) is a protein which may be involved in pathways towards the generation of preβ-HDL in plasma (15) . However, the role of PLTP until the present invention has been far from clear. /Among others PLTP is thought to be capable of promoting the net transfer of phospholipids between plasma lipoproteins (1-4) and to mediate conversion of HDL (5,6) . Recently a human cDNA for human PLTP has been cloned (16) after which transgenic mice were made encoding human PLTP (17, 18). Unfortunately, these mice expressed only very low levels of the transgene, and no physiological effects of PLTP could be measured to any reliable extent. Particularly because in one of the studies, the transgene was introduced in a compound transgenic background with apolipoprotein A-I, which makes it difficult, if not impossible to relate any effect to the PLTP transgene. However, in order to be able to develop new and efficient medical treatments for cardiovascular diseases and pharmaceutical compositions for such diseases a good and reliable model system is required.

The present invention provides such a model system in that transgenic mice are provided which overexpress human PLTP 2.5 to 4.5-fold. Thereby the present inventors have found beneficial pharmaceutical effects of huPLTP for the first time and have shown that it is feasible to make pharmaceutical compositions for the treatment of cardiovascular disorders in vivo based on huPLTP. The effects seen in the transgenic mice according to the invention include decreased total plasma cholesterol levels, increased rates of formation of preβ-HDL and a high plasma capability to prevent cholesterol accumulation and/or improving cholesterol excretion, for example in macrophages . It is thus clear through this invention that huPLTP is a protein which can be used as a pharmaceutical. Thus the invention provides, in one aspect, a plasma phospholipid transfer protein or a functional derivative or fragment thereof for use as a pharmaceutical. It is of course preferred to apply huPLTP for human applications, in order to avoid an immunological response to heterologous proteins, as it is preferred to apply PLTP of any given species in the treatment of said same species. PLTP herein is defined as any mammalian protein or proteinaceous substance having at least one activity, but preferably all activities of PLTP., which is adopted from Day et al (16). Having the same activity or activities in this respect means that the activities are the same in kind, not necessarily in amount. PLTP, or functional fragments or derivatives thereof, can be produced recombinantly, synthetically, in cell free systems, or isolated from a suitable source. It is however preferred to produce it recombinantly, particularly in a eukaryotic host cell, more preferably in a higher eukaryote, such as a mammalian or insect cell. Transgenic animals (such as the mice provided herein) can also be a suitable source for PLTP. Typically, the person skilled in the art is able to choose a suitable expression system for PLTP based on the teachings of e.g. Sambrook, Fritsch and Maniatis, Molecular Cloning, Cold Spring Harbor Laboratory Press, 1989. Functional fragments and/or derivatives of PLTP are defined as such derivatives or fragments which have sequence homology with PLTP over at least a part of the molecule and which share at least the anti-atherogenic properties of PLTP. It is always difficult to say at which level of homology such activity will be present, but the person skilled in the art is capable with present day techniques to identify the regions in PLTP which is necessary for activity, which typically will be the area of the protein in which derivatives need to have a high homology with the original PLTP's for retaining activity. Such a region is of course also the region of interest to derive functional fragments from and to do site-directed mutations in, in order to enhance activities wanted and decrease unwanted activities by recombinant techniques. As a rule of thumb general homology of a derivative and/or a fragment (over the parts that are present) should be recognisable at the amino and/or nucleic acid level by standard homology search programmes and/or by hybridisation.

The present invention discloses a first pharmaceutical or medicinal use for PLTP. Of course this first use lies in the field of cardiovascular diseases, especially in the field of preventing or treating atherosclerosis and the results thereof. The invention especially provides uses under conditions of high plasma cholesterol and/or triglyceride levels, were these conditions need to be alleviated. Thus in a further aspect the invention provides the use of a plasma phospholipid transfer protein or a functional derivative or fragment thereof in the preparation of a medicament for the treatment or prevention of medical conditions associated with relatively high plasma levels of cholesterol and/or triglycerides. Generally, high cholesterol and/or triglyceride levels are associated with atherosclerosis and its detrimental results, such as stroke and coronary heart disease. However other cholesterol and/or triglyceride dependent conditions may also be treated or prevented with PLTP or a fragment or derivative thereof. Moreover, since PLTP seems to act at least in part through enhancing the production of preβ-HDL, any condition benefiting from higher levels of preβ-HDL can also be treated with PLTP or preβ-HDL. In a further aspect the invention provides a pharmaceutical composition for the treatment or prevention of medical conditions associated with relatively high cholesterol levels, comprising plasma phospholipid transfer protein or a functional derivative or fragment thereof and a suitable diluent or other pharmaceutical excipient. Typically, with PLTP being a proteinaceous substance, it is necessary to prevent breakdown of the active compound in the gastrointestinal tract of a patient to be treated. Especially the stomach's actions must be avoided. Preparations are known which protect proteinaceous substances from the stomach's environment. Another possibility is of course to choose routes of administration avoiding the stomach, such as other enteral or parenteral routes, such as intravenous compositions and the like. Excipients for all routes are by now well known and can be found in any pharmaceutical compendium. Levels of PLTP to be reached in plasma will depend on the condition to be treated and its severity. The upper limit of levels to be reached is of course determined by the patients tolerance or reactions to PLTP. However in general levels in plasma to be reached should be upward from physiological levels, e.g. 1- to 3-fold, or even up to 5- to 10-fold, expressed as units of PLTP. It will of course be different for derivatives and fragments of PLTP having other activity levels. The levels that can be reached are also dependent the route of administration. It is therefor hard to give amounts to be given per dosage unit for every possible route of administration with every kind of excipient. Eventually, the only relevant parameter is plasma levels. However, as a rule of thumb a daily dose of PLTP will allow 2- to 4- fold increase of PLTP plasma levels. It becomes even more difficult to give necessary doses in combined preparations. Of course PLTP can be combined in preparations or treatments with conventional or other cardiovascular drugs, such as statins or fibrates. Thus in a further aspect the invention provides a composition according to the invention, which further comprises other agents for the treatment or prevention of medical conditions associated with relatively high plasma levels of cholesterol and/or triglycerides. Typically such agents would be present in these compositions in their usual amounts or lower.

The invention further provides a composition for preventing the accumulation and/or improving excretion of cholesterol in a mammal comprising an effective amount of plasma phospholipid transfer protein or a functional derivative or fragment thereof.

In subjects which are genetically predisposed for cardiovascular conditions, especially those with genetically higher plasma levels of cholesterol and/or triglycerides or low levels of total HDL or preβ-HDL, or absence of genetic information encoding sufficient functional PLTP, it is of course preferred to provide such subjects with a permanent solution for said problem, in stead of having to repeat treatments or a life-time dietary regime. Thus in a further aspect the invention provides a gene delivery vehicle for delivering a recombinant nucleic acid molecule encoding comprising plasma phospholipid transfer protein or a functional derivative or fragment thereof to a host cell, whereby said recombinant nucleic acid molecule is integrated into the genome of said host cell. Transgenic mice are of course only a means to study possible roles and effects of expressed transgenes. It is of course highly unlikely that any medical treatment for cardiovascular diseases will ever include anything like transgenesis of humans. However, by providing the genome of a host cell, for example one or more liver cells, of the subject with integrated genetic information encoding PLTP activity, the subject obtains a higher capacity of preventing cholesterol accumulation and/or improving cholesterol excretion in the system. This can be done through gene therapy with a gene delivery vehicle.

Vehicles for gene delivery and integration into the host cell genome are well known. They include viral delivery vehicles such as vehicles based on adeno- and retroviruses or adeno associated viruses, liposome-like delivery vehicles and polymer based vehicles. The person skilled in the art is aware of these possibilities. The invention further provides a transgenic host cell obtainable by transfection with a gene delivery vehicle according to the invention. Said transgenic host cell, for example a hepatocyte, can be grown and cultivated in vi tro, and when needed, can be used to provide a patient with said cell having increased capacity of preventing cholesterol accumulation and/or improving cholesterol excretion in the system. As stated herein before it is preferred to apply this invention to humans, it is therefore also preferred that the compositions according to the invention comprise human plasma phospholipid transfer protein, preferably human plasma phospholipid transfer protein having sequence encoded by a cDNA of figure 5. This of course is also true for the uses according to the invention.

The invention will now be explained in more detail in the following experimental part without limiting the invention. Experimental part.

Materials and Methods

Generation of HuPLTP transgenic mice

A human cosmid library was constructed from high molecular weight DNA isolated from blood from a healthy volunteer. This library was screened for cosmids containing the PLTP gene, using human PLTP cDNA (kindly donated by Drs . A-Y. Tu and J.J. Albers) as a probe. Isolated cosmids were mapped using restriction fragments from the cDNA, and a cosmid with approximately 15 kb 5' to the first exon and approximately 3.5 kb 3' to the last exon (19) was selected. Vector sequences were removed by restriction endonuclease digestion and DNA was dissolved in micro-injection buffer (10 mM Tris- HC1, pH 7.5; 0.1 mM EDTA) at a concentration of 1-2 mg/ml . The DNA was micro-injected into fertilized oocytes from FVB mice. These oocytes were transferred into the oviducts of pseudopregnant foster females.

DNA analysis

Genomic DNA was isolated from tail clips of 10 days old mice, and analyzed for the presence of the HuPLTP transgene by PCR analysis: sense primer: 5 ' -GCCACAGCAGGAGCTGATGC-3 ' ; anti- sense primer: 5 ' -GCGGATGGACACACCCTCAGC-3 ' ; 25-30 cycles (94 ° C, 1 min; 65 °C, 1 min; 72 °C, 1 min) .

Breeding and treatment of transgenic mice

Transgenic founder mice were bred with FVB mice to obtain transgenic mice. FVB transgenic HuPLTP mice were backcrossed with C57B1/6 mice for four generations. These mice were intercrossed to obtain wild type, hemizygous and homozygous HuPLTP transgenic mice. Animals were kept on regular chow and fasted overnight prior to collection of blood from the orbital plexus.

Gene expression analysis by RT-PCR Total RNA was isolated from various tissues obtained freshly from either wild type, hemizygous or homozygous HuPLTP- transgenic mice. cDNA was obtained by reverse transcription primed by oligo(dT) . This was used in PCR reactions in the presence of [³²P]-ATP (20 cycles; 94 °C, 1 min; 62 °C, 1 min; 72 °C, 1 min), Primers used were: HuPLTP; sense: 5'-

CCTGCTGAGCCCAGCAGTG-3 ' ; anti-sense : 5 ' -CTGGACCTCAGGCTGGTCTG- 3'; MuPLTP; sense: 5 ' -TTGACTCTGCCATGGAGAGC-3 ' ; anti-sense: 5'-GCTCCACTTCGGGCAACATG-3' ; HPRT; sense: 5'- CGAAGTGTTGGATACAGGCC-3 ' ; anti-sense : 5 ' -GGCAACATCAACAGGACTCC- 3 ' . PCR products were run on polyacrylamide gels and visualized using a Phosphor Imager.

Assay of plasma PLTP activity

Plasma PLTP activity was assayed using a phospholipid vesicles-HDL system (1, 20) . EDTA-plasma samples (25 ml of 1:75 diluted plasma) were incubated with

[³H] dipalmitoylphosphatidylcholine-labelled (Amersham, U. K. ) phosphatidylcholine vesicles and excess pooled normal HDL for 45 min at 37 °C. Following incubation the vesicles were precipitated as described (20) and the radioactivity transferred to HDL was counted in the supernatant. Standard curves using dilutions of human plasma were included in each run. The measured activity is linear with time for 1 h. All samples were analysed in duplo and blanks without plasma were subtracted. Duplicates of pooled human plasma, stored at -70 °C, were also measured in each series (reference plasma) . The between assay coefficient of variation of reference plasma was 4.1%. Activities are expressed as mmoles phosphatidylcholine transferred/ml plasma/h (mmol/ml/h) . An ti -mouse apo A-I polyclonal antibodies

Apo A-I was purified from mouse plasma HDL (density range 1.063-1.21 g/ml) essentially as described (21) and used to immunize rabbits by subcutaneous injection with 100 mg of mouse apo A-I using standard procedures.

Quantitation of plasma lipids and apo A-I

Total cholesterol was enzymatically determined with the F- Choi kit from Boehringer Mannheim (Mannheim, Germany) after hydrolysis of cholesterylesters with cholesterol esterase from C. cylindraeca (Boehringer Mannheim) . Phospholipids were measured enzymatically using the PAP150 kit from BioMerieux ■ (Lyon, France) . Mouse apo A-I was quantitated by a sandwich ELISA, using a polyclonal rabbit anti-mouse apo A-I IgG, performed in 96- well plates coated with this antibody. Purified mouse apo A-I was used as a primary standard. Plasma samples were diluted in PBS-Tween 20 (0.1%)-BSA (0.5%). Bound apo A-I was detected by addition of polyclonal rabbit anti-mouse apo A-I, conjugated to horseradish peroxidase. The assay is linear in the range of 6.5 to 420 ng/ml .

Quantitation of preβ-HDL by crossed immuno-electrophoresis Freshly isolated plasma samples from mice were either directly frozen or incubated at 37 °C in order to measure the formation of preβ-HDL in vi tro . Iodo-acetate (1 mM) was added for complete inhibition of lecithin: cholesterol acyl transferase (LCAT) . This prevents maturation of the formed preβ-HDL into α-HDL by LCAT activity.

The crossed immuno-electrophoresis consisted of agarose electrophoresis in the first dimension for separation of lipoproteins with preβ- and -mobility. Electrophoresis in the second dimension, i.e. antigen migration from the first gel into an anti-apo A-I-containing gel, was used to quantitatively precipitate apo A-I. Lipoprotein electrophoresis was carried out in 1% (w/v) agarose gels in barbital buffer (50 mM, pH 8.6) and run in an LKB 2117 system (4 °C, 2 h, 250 V) . Five ml of plasma were applied per well. The track of the first agarose gel was excised and annealed with melted agarose to a gel containing 7.5% (v/v) rabbit anti-mouse apo A-I antiserum that was cast on GelBond film (Pharmacia) . The plate was run in an LKB 2117 system (4 °C, 20 h, 50 V) in barbital buffer. Unreacted antibody was removed by extensive washing in PBS. The gel was stained with Coomassie Brilliant Blue R250 and subsequently dried. Areas under the preβ-HDL and -HDL peaks were calculated by multiplication of peak height and width at half height. The preβ-HDL area is expressed as a percentage of the sum of a- HDL and preβ-HDL areas. Preβ-HDL concentrations are also given in absolute amounts (mg apo A-I present in preβ-HDL/ml plasma) . These values were calculated from the percentage of apo A-I present in preβ-HDL and the total plasma apo A-I concentrations .

Cholesterol flux experiments using mouse peritoneal ma crophages The ability of mouse heparin-plasma to interfere with the intracellular formation by ACAT of labeled cholesteryl oleate from [³H] oleate and AcLDL-derived cholesterol was tested using the assay developed by Brown et al . (22). The formation of labelled cholesteryl esters from [³H] oleate by ACAT is an estimate of intracellular cholesterol concentration.

Bovine serum albumin (fraction V grade) was delipidated by extraction of free fatty acids with activated carbon (23) . [9, 10 (n)³H]oleic acid (10.0 Ci/mmol, Amersham, U.K.) was complexed to BSA after evaporating 0.1 mmol of oleic acid (48 Ci/mol) to dryness under a stream of nitrogen. Subsequently, 10 ml of 12 % (w/v) fatty acid free BSA in DMEM at 56 °C was added to the dried oleate, and sterilized by passage through a 0.22 mm filter.

AcLDL was prepared from LDL (density range 1.019-1.063 g/ml) , isolated from human plasma by differential centrifugation and subsequently acetylated by repeated additions of acetic anhydride (24). Increased electrophoretic mobility of the acLDL was confirmed by agarose electrophoresis at pH 8.6 (12).

C57B1/6 mice were elicited by intraperitoneal injection of 0.8 ml of eliciting agent, prepared from Baker's thioglycollate (Difco) according to manufacturer's instructions. After 4 days, macrophages were obtained as described (25) . The macrophage monolayers were washed with DMEM and incubated with 500 ml aliquots of DMEM that contained 3 mg/ml AcLDL, 0.1 mM BSA- [³H] oleate (48 Ci/mol) and 12.5 times diluted plasma from fasted mice. After 18 h, medium was removed and cells were washed twice with PBS.

Cholesteryl esters were extracted from the intact monolayers with 1 ml hexane-isopropanol (3:2, v/v) and purified by TLC as previously described (26) . Labeled cholesteryl ester bands were excised from the silica and radioactivity was determined. Protein was extracted from the cell remnants with 0.1 M NaOH and quantified by the method of Lowry et al . (27) using BSA as a standard. Duplicate assays were performed for each plasma sample.

Statistics

Data are given as mean ± SD. Differences between groups were analyzed by one way analysis of variance (ANOVA) . The Bonferroni correction was used for multiple pairwise comparisons when the ANOVA indicated a significant effect. Results

Generation of trangenic mice A cosmid clone containing the complete human PLTP gene was isolated and analyzed by Southern blotting. It included the 3' end but not the 5' end of the lysosomal protective protein gene. The purified 35 kb cosmid was micro-injected in fertilized oocytes. This resulted in 27 newborn mice, two of which harboured the transgene, as determined by PCR analysis and Southern blotting (data not shown) . Both founder mice were bred into two independent lines of HuPLTP transgenic mice.

Expression of the transgene

PLTP activity was measured in plasma samples of mice from two HuPLTP transgenic lines. In line #1 PLTP activity in plasma was increased by 281 % compared with the activity in plasma from wild type mice (Table 1), while in line #2 an increase by 262 % was measured (not shown) . Subsequent analyses were performed with line #1 only. The use of either human HDL or mouse HDL as acceptor in the PLTP activity assay showed the same differences between wild type, hemizygous and homozygous transgenic animals, demonstrating that human PLTP interacts both with human HDL and mouse HDL (not shown) .

The expression of the transgene was confirmed by Western blotting (Fig. 1). The mice homozygous for the transgene showed a higher plasma protein level of HuPLTP when compared with the hemizygous transgenic mice. As expected, no immunoreactive PLTP was detected in wild type mice.

Recombinant PLTP was used as a positive control and gave one single immuno-reactive band at a relatively low MW, due to a lower extent of glycosylation in the baculovirus expression system. The transgene was found to be expressed in all tissues analyzed (Fig. 2), with relatively high mRNA levels in adrenal, testis, and lung, and moderate mRNA levels in liver, kidney, intestine, brain and spleen. The tissue pattern of expression of the endogenous PLTP gene was similar to that of the transgene, again with the highest mRNA levels in adrenal, testis, and lung. The expression of endogenous PLTP was not affected by HuPLTP expression in any of the tissues tested (Fig. 2) .

Effects of HuPLTP overexpress ion on plasma lipids and lipopro teins

The overexpression of PLTP resulted in a decrease in plasma cholesterol levels in the HuPLTP hemizygous transgenic mice and a further decrease in the homozygous transgenic mice

(Table 1) . Plasma levels of cholesterol, phospholipids and apo A-I were decreased by about the same extent, indicating that the decrease reflects a lowering of HDL, which are the major lipoprotein in mouse plasma. Separation of plasma lipoproteins by gelfiltration confirmed that the decrease in plasma lipids is confined to the HDL fraction (not shown) .

In order to investigate changes in HDL subclass distribution, mouse plasma was analyzed by crossed- immunoelectrophoresis (Fig. 3) . Plasma samples from wild type, hemizygous and homozygous HuPLTP transgenic mice were collected and incubated in the presence of an inhibitor of lecithin: cholesterol acyl transferase (LCAT) in order to prevent maturation of the formed preβ-HDL into oi-HDL (14). The formation of preβ-HDL particles is clearly increased in plasma from transgenic mice when compared with wild type mice. It is also evident that the formation of preβ-HDL is at the expense of -HDL, demonstrating the origin of preβ-HDL (Fig. 3). Table 2 gives the preβ-HDL values, both before and after incubation in the presence of LCAT inhibitor. No significant differences are found in freshly frozen plasma between the different genotypes. Before incubation the percentage of preβ-HDL tends to be highest in the homozygous HuPLTP transgenic animals, but the differences between genotypes are not significant. However, in incubated samples clear differences arise with highest relative and absolute concentrations of preβ-HDL in the transgenic animals.

Effect of wild type and transgenic plasma on the accumulation of AcLDL-derived cholesterol in mouse peritoneal macrophages

We investigated whether differences in the HDL-subfraction pattern affect the ability of wild type and transgenic plasma to interfere with the esterification of intracellular cholesterol by acylCoA: cholesterol acyltransferase (ACAT) . ACAT activity determined this way is a measure of the intracellular cholesterol concentration. Mouse peritoneal macrophages were incubated in the presence of [³H] oleate, AcLDL and diluted mouse plasma containing the acceptor HDL particles (see Methods) . Fig. 4 shows that the formation of labeled cholesteryl oleate by ACAT was 25.7+9.7% lower in the presence of hemizygous transgenic plasma as compared to wild type plasma, indicating less accumulation of cellular cholesterol in the presence of plasma from transgenic animals. Thus, in spite of a lower HDL-cholesterol concentration, transgenic mouse plasma has the ability to prevent cholesterol accumulation to a greater extent. With plasma from homozygous HuPLTP transgenic mice a 42.4113.2% decrease in ACAT activity (compared with wild type plasma, see Fig. 4) is measured, showing even less accumulation. Discussion

Plasma phospholipid transfer protein (PLTP) is able to promote the net transfer of phospholipids between plasma lipoproteins (1-4) and mediate conversion of high density lipoproteins (HDL) (5, 6) . Since plasma levels of HDL are among the best indicators for the risk of atherosclerosis in epidemiological studies (7, 8), PLTP could play a role in the prevention of the development of coronary heart disease and stroke via its effect on HDL.

HDL are anti-atherogenic because they mediate efflux of cholesterol from peripheral cells and transport cholesterol to the liver, for excretion and degradation to bile acids. This process is known as reverse cholesterol transport (8- 10). It has been postulated that the anti-atherogenic effect of HDL can be attributed mainly to a quantitatively minor subclass of HDL, called preβ-HDL (11, 12). This assumption is based on in vi tro studies showing that preβ-HDL is a very efficient acceptor of cellular cholesterol (13, 14). The origin of preβ-HDL is not well understood, but the available evidence suggests that PLTP participates in its generation, at least in vi tro (15).

Following the cloning of a human PLTP cDNA (16) , two groups independently generated transgenic mice for human PLTP (17, 18) . Unfortunately, these mice showed low levels of expression of the transgene and, as a result, only small effects on plasma lipoproteins were observed. Changes in HDL levels and subtractions could only be demonstrated in a compound transgenic background with human apolipoprotein (apo) A-I (18) .

We now report the generation of transgenic mice which overexpress human PLTP (HuPLTP) 2.5- to 4.5-fold. This results in decreased plasma total HDL cholesterol levels, in increased formation of preβ-HDL, and in a high plasma capability to prevent cholesterol accumulation in macrophages .

We generated transgenic mice for human PLTP that have up to 4.5-fold elevation of plasma PLTP activity levels. The total plasma concentrations of cholesterol, phospholipids and apo A-I are decreased, due to a 30-40% reduction in total HDL. In contrast, preβ-HDL concentrations are not decreased in HuPLTP transgenic animals. Both the relative and absolute plasma levels of preβ-HDL are clearly increased in transgenic animals after incubation of plasma in the presence of an inhibitor of LCAT (Table 2) . LCAT is known to convert preβ- HDL into α-HDL (14) . Rapid preβ-HDL maturation, driven by LCAT, may obscure the effects of the HuPLTP gene on plasma preβ-HDL levels in the absence of an LCAT inhibitor. These data imply that HuPLTP transgenic plasma has a much greater ability to generate preβ-HDL than wild type plasma, in spite of only marginal differences in preβ-HDL levels.

HuPLTP transgenic mice have been described previously by two other groups, but these models did not show an appreciable overexpression of the transgene. The mice described by Albers et al . (17) showed little expression of the transgene and only small changes in plasma lipoproteins. Jiang et al . (18) reported a 29% increase in PLTP activity, but significant effects on total plasma lipids or lipoproteins were not observed. Only after their HuPLTP transgenic mice were crossbred with mice transgenic for human apoA-I, they detected small effects on plasma lipids and lipoproteins (including an increase in preβ-HDL levels) , together with a 47% elevation in plasma PLTP activity. It must be noted that human ApoA-I transgenic mice already have elevated levels of both total and preβ-HDL when compared to wild type mice (18). The HuPLTP transgenic mice described in the present paper have altered HDL metabolism, without the complication of the additional HuApoA-I gene, which by itself has substantial effects on HDL. Moreover, our mice are not transgenic for lysosomal protective protein. Jiang et al . (18) used a DNA construct containing both genes that have a 3' overlap on opposite DNA strands. Although unlikely, it is difficult to exclude the possibility that this condition interferes with lipoprotein metabolism.

HuPLTP has also been overexpressed in mice via adenovirus mediated transfer (30, 31) . These mice showed a 13- to 40- fold elevation of PLTP activity in plasma several days after treatment. This resulted in a dramatic decrease (by 91%) in HDL-levels, while preβ-HDL levels were substantially elevated (30). These data are in line with our present observations: high PLTP activity results in a decrease in total HDL, while preβ-HDL levels are increased. It is clear that the effects are transient and greatly exaggerated in the adenovirus treated mice, due to the extremely high plasma levels of PLTP. Four-fold elevated PLTP activity levels, measured at late time points after adenovirus transfection, did not result in altered plasma HDL cholesterol concentrations, while in our homozygous HuPLTP transgenic mice a 4-fold elevation in PLTP activity resulted in decreased HDL concentrations. Another notable difference is that in the adenovirus treated mice PLTP expression is restricted to the liver. The tissue pattern of expression of the HuPLTP gene in transgenic mice resembles the pattern of expression of the endogenous PLTP gene (Fig. 2; 32).

Plasma of HuPLTP transgenic mice was found to be much more efficient in partially preventing AcLDL-induced accumulation of intracellular cholesterol in cultured macrophages than plasma of wild type mice, in spite of lower levels of total HDL. The most likely explanation for this observation is the increased plasma concentration of preβ-HDL, which has been identified previously as a very efficient cholesterol acceptor (13, 14). It is well known that an operative cholesterylester cycle, as present in macrophages (33, 34), is important for cholesterol efflux. The present results imply that the distribution of HDL subclasses is of major importance for the efficacy of HDL-mediated reverse cholesterol transport, even more important than total plasma HDL levels.

Plasma levels of total HDL cholesterol are inversely correlated with the incidence of coronary artery disease in man (7, 8) . However, only a few studies have looked at the relationship between total HDL and preβ-HDL in human plasma. Recently, O'Connor et al . (14) analyzed the steady state levels of preβ-HDL in 136 normolipidemic individuals, using an isotope dilution technique. Their relative values for preβ-HDL in human plasma are quite comparable with the values measured in mouse plasma (see Table 2). The percentage preβ- HDL (percentage of total plasma apo A-I) was negatively correlated with total HDL cholesterol concentrations, in line with our observations in mice. Plasma PLTP activity was not measured in their study.

Reconstituted HDL particles enriched in triglycerides, which are model particles for HDL prevalent during alimentary lipemia, are more rapidly converted by PLTP to preβ-HDL than triglyceride-poor HDL (35) . This observation suggests that hypertriglyceridemia may be associated with increased generation of preβ-HDL by PLTP. In addition, Syvanne et al . (36) reported a positive correlation between PLTP activity and the capability of plasma from patients with diabetes mellitus and coronary heart disease to induce cholesterol efflux from Fu5AH rat hepatoma cells.

Taken together, the combination of increased PLTP activity, increased preβ-HDL formation and less accumulation of cellular cholesterol (as seen in our mouse model) is likely to exist also in the human situation. The rate of formation of preβ-HDL may very well be more important than its steady state concentration. The present work shows for the first time that PLTP is important for the ongoing generation of plasma preβ-HDL and that the ability of plasma to prevent cholesterol accumulation in macrophages is increased at high PLTP activity levels. Our findings show that PLTP is a potential anti-atherogenic factor, due to its ability to generate prebβ-HDL.

Atherosclerosis in transgenic mice

We generated several lines of transgenic mice over-expressing human PLTP. The lines huPLTPl and huPLTP4 have been generated using a cosmid containing the entire gene plus its natural flanking sequences (fig. 6) and thus express PLTP driven by its native promoter (mice) . These mice have different levels of plasma activity of PLTP (see table 3) .

These mice as well as wild type mice were fed a high sucrose diet for 2 weeks (LFC; 40) and subsequently a high fat, high cholesterol diet (HFC/0.5%; 40 ; containing 1% cholesterol and 0.5% cholate; Hope Farms, Woerden, The Netherlands), which is a standard diet to study diet-induced atherosclerosis in mice (41) . Analysis after 4 weeks on this diet of the plasma lipoproteins by gel chromatography (figure 7) showed that cholesterol was lowered in all lipoprotein fractions when compared with plasma from wild type mice, including the atherogenic VLDL/IDL fraction. The lowest cholesterol levels were found in the PLTP4 mice, that have the highest PLTP activity. Therefore, PLTP protects against diet-induced increase of atherogenic VLDL/IDL. Atherosclerotic lesions in the mice were analyzed after 16 weeks of HPC/0.5% diet. The area of aortic atherosclerotic lesions was smaller in mice with higher levels of plasma PLTP activity.

In conclusion, PLTP protects against diet-induced increase in plasma cholesterol levels, and protects against the development of diet-induced atherosclerosis.

Preβ-HDL in transgenic mice

We studied transgenic mice over-expressing either human PLTP (huPLTP mice) , human cholesteryl ester transfer protein

(CETP; huCETP mice) or both (huPLTP/huCETP mice) . Both the activity of PLTP as well as formation of preβ-HDL in plasma were studied. It appeared that the activity of plasma PLTP was equal in wild type compared to huCETP mice. Also, in huPLTP compared to huPLTP/huCETP mice, plasma PLTP activity was equal. The formation of preβ-HDL in plasma (fig. 8) did not appear to be affected by the presence of the CETP transgene: in lanes W (wild type) and C (huCETP transgenic mice) preβ-HDL formation was very low whereas in lanes P (huPLTP mice) and CP (huPLTP/huCETP mice) a prominent preβ- HDL band with approximately the same intensity could be detected.

We conclude that PLTP determines the rate of formation of plasma preβ-HDL, independent of CETP levels.

We also studied transgenic mice overexpressing apolipoprotein A-l (huAITg; donated by Dr. A. Tall; 18) . HuAITg mice are less susceptible to diet-induced atherosclerosis in comparison with wild type mice (42) . Plasma PLTP activity is higher in the huAITg mice than in wild type mice (Fig. 9) . We also measured the ability of plasma from these mice to prevent cholesterol accumulation in mouse peritoneal macrophages in vitro (Fig. 10) . Plasma from huAITg mice was more effective in this respect.

We conclude that the elevated activity of plasma PLTP contributes to the low susceptibility for diet-induced atherosclerosis in mice that overexpress apolipoprotein A-l.

Legends to the Figures

Fig. 1. Western blot analysis of the plasma levels of HuPLTP in transgenic mice. SDS-PAGE was carried out in 12.5 % (w/v) gels, proteins were electrophoretically transferred to PVDF membranes and visualized with a rabbit polyclonal antibody raised against a synthetic peptide with identity to amino • acids 470-493 of HuPLTP. Goat anti-rabbit IgG conjugated to peroxidase was used as a secondary antibody. Antigen-antibody complexes were visualized by chemiluminiscence using the ECL system (Amersham) . Blots were exposed for 20 sec to a Kodak XAR-5 film. Re refers to recombinant PLTP from a baculovirus expression system. Plasma from wild type mice and hemizygous and homozygous HuPLTP mice is indicated by wt, he and ho, respectively.

Fig. 2. Expression of PLTP in various tissues. Total RNA was isolated from various tissues of wild type (wt) , hemizygous (he), or homozygous (ho) HuPLTP transgenic mice. RNA levels of either the transgene (HuPLTP) or the endogenous, murine gene (MuPLTP) were examined by gel electrophoresis of RT-PCR products. RT-PCR on HPRT was used as a loading control.

Fig. 3. Apolipoprotein A-I immunoprecipitation patterns of plasma from wild-type or HuPLTP transgenic mice obtained after crossed immunoelectrophoresis . The figure shows a representative display of the preβ- and -HDL bands obtained by analysis of mouse plasma, incubated for 3 h at 37 °C in the presence of iodoacetic acid (1 mM) , an inhibitor of LCAT. Samples were then analyzed by crossed-immunoelectrophoresis as described in Methods. In order to confirm mono-specificity of the anti apo A-I antibody towards α-HDL and preβ-HDL fractions, we performed immunoelectrophoresis as described (28). Two single precipitation arches appeared without the formation of spurs, indicating that the antibody recognizes immunologically identical determinants in both HDL subfractions (29). Left panel: plasma from a wild type mouse. Right panel: plasma from a homozygous HuPLTP transgenic mouse. The positions of preβ- and α-HDL migration are indicated.

Fig. 4. Esterification of cholesterol by ACAT in peritoneal macrophages incubated simultaneously with AcLDL, [³H] oleate and plasma of wild type (Wt) , hemizygous (He) or homozygous HuPLTP transgenic mice (Ho) . Plasma was obtained from age- matched wild type (Wt), hemizygous (He) and homozygous (Ho) HuPLTP transgenic mice. Peritoneal macrophages were incubated for 18 h in the presence of 3 mg/ml AcLDL, 0.1 mM [³H]oleate- BSA (48 Ci/mol) and diluted mouse plasma. Esterification by ACAT is expressed as dpm in cholesteryl oleate/mg cell protein. A decreased esterification indicates a decreased cellular cholesterol concentration. For further details see Materials and Methods. Values for individual plasma samples are shown (open circles), as well as mean values (closed circles) with standard deviations (vertical bars). In the absence of plasma, ACAT activity was 1770 ± 90 dpm/ g. ACAT activity was very low in the absence of AcLDL: 16.0 ± 0.4 dpm/mg with wild type plasma. Values for hemizygous and homozygous mice differed at the p=0.021 (n=8) and P=0.001 (n=6) level from those for wild type mice (n=7), respectively. The data presented show a typical example out of 4 independent experiments performed.

Fig. 5a. Nucleotide sequence and derived amino acid sequence of human PLTP. Data as reported by Day et al. (1994) (GenBank accession number L26232). The 30 C-terminal amino acids have been reported to be indispensable for phospholipid transfer activity (Huuskonen et al . , 1998). Fig. 5b. PLTP cDNA sequence and alignment with several proteins. The upper line contains the nucleotide sequence of human PLTP cDNA as reported by Day et al. (1994) (GenBank accession nubmer L26232) . PLTP belongs to a family of proteins that are structurally and functionally related (Hubacek et al., 1997). Homologous domains have been described by Albers et al. (1996). The homology of these proteins is indicated by shaded boxes. CETP: Human cholesteryl ester transfer protein (GenBank accession number M30185) ; LBP: Human lipopolysaccharide binding protein

(GenBank accession number M35533) ; BPI: Human bactericidal permeability increasing protein (GenBank accession number J04739) .

Fig. 6. Schematic drawing of the PLTP and lyosomal protective protein (LLP) genes, that are on opposite DNA strands. The sequences included in the cosmid used to generate transgenic mice, are indicated

Fig. 7. Gel filtration profiled of plasma from wild-type

(TCwt), huPLTPl (TCtgl) or huPLTP4 (TCtg4) mice after 2 weeks LFC followed by 4 weeks HFC/0.5% diet. Pooled plasma from 5-6 mice were subjected to gel filtration by FPLC. Cholesterol concentrations in the individual fractions are shown. Fraction 1-4: VLDL, fraction 5-8: IDL, fraction 9-12: LDL; fraction 13-18: HDL

Fig. 8 Preβ-HDL formation in plasma from wild-type (W) , huCETP mice (C) , huPLTP mice (P) or huPLTP/huCETP mice (CP) . The figure shows a representative display of the preβ- and α- HDL bands obtained by analysis of mouse plasma, incubated for 3 h at 37°C in the presence of iodoacetic acid (1 mmol/L) , an inhibitor of LCAT. The positions of preβ- and α-HDL migration are indicated.

Fig. 9 Plasma PLTP activity in wild type (wt) or huAITg mice. W=P<0.05

Fig. 10 Prevention of acLDL induced cholesterol accumulation in cultured mouse peritoneal macrophages by plasma from wild type (wt) or huAITg mice. Cholesterol accumulation is measured by the rate of intracellular cholesterol esterification due to ACAT activity, a = P < 0.05.

References

1. Damen, J. , Regts, J. & Scherphof, G. (1982) Biochim . Biophys . Acta 712, 444-452.

2. Tall, A.R., Abreu, E. & Shuman, J. (1983) J. Biol . Chem . 258, 2174-2180. 3. Albers, J.J., Tollefson, J.H., Chen, CH. & Steinmetz, A. (1984) Arteriosclerosis 4, 49-58.

4. Albers, J.J., Tu, A.Y., Wolfbauer, G., Cheung, M.C. & Marcovina, S.M. (1996) Curr. Opin . Lipidol . 7, 88-93.

5. Jauhiainen, M. , Metso, J. , Pahlman, R., Blomqvist, S., Van Tol, A. & Ehnholm, C. (1993) J. Biol . Chem . 268, 4032-4036.

6. Tu, A.Y., Nishida, H.I. & Nishida, T. (1993) J^". Biol . Chem . 268, 23098-23105.

7. Gordon, D.J. & Rifkind, B.M. (1989) N. Engl . J. Med. 321, 1311-1316. 8. Tall, A.R. (1990) J. Clin . Invest . 86, 379-384.

9. Fielding, C.J. & Fielding, P.E. (1995) J. Lipid Res . 36,

211-228.

10. Barter, P.J. & Rye, K.A. (1996) Curr. Opin . Lipidol . 7,

82-87. 11. Von Eckardstein, A., Huang, Y. & Assmann, G. (1994) Curr.

Opin . Lipidol . 5, 404-416.

12. Cheung, M.C, Wolfbauer, G. & Albers, J.J. (1996) Biochim .

Biophys . Acta 1303, 103-110.

13. Castro, G.R. & Fielding, C.J. (1988) Biochemistry 27, 25- 29.

14. O'Connor, P.M., Zysow, B.R., Schoenhaus, S.A., Ishida,

B.Y., Kunitake, S.T., Naya-Vigne, J.M., Duchateau, P.N.,

Redberg, R.F., Spencer, S.J., Mark, S., Mazur, M., Heilbron,

D.C, Jaffe, R.B., Malloy, M.J., Kane, J.P. (1998) J. Lipid Res . 39, 670-678. 15. Von Eckardstein, A., Jauhiainen, M., Huang, Y., Metso, J. ,

Langer, C, Pussinen, P., Wu, S., Ehnholm, C & Assmann, G.

(1996) Biochim . Biophys . Acta 1301, 255-262.

16. Day, J.R., Albers, J.J., Lofton-Day, C.E., Gilbert, T.L., Ching, A.F., Grant, F.J., O'Hara, PJ, Marcovina, S.M. &

Adolphson, J.L. (1994) J. Biol . Chem . 269, 9388-9391.

17. lbers, J.J., Tu, A.Y., Paigen, B., Chen, H. , Cheung, M.C

& Marcovina, S.M. (1996) Int . J. Clin . Lab. Res . 26, 262-267.

18. Jiang, X., Francone, O.L., Bruce, C, Milne, R. , Mar, J. , Walsh, A., Breslow, J.L. & Tall, A.R. (1996) J. Clin . Invest .

98, 2373-2380.

19.Shimmoto, M. , Nakahori, Y., Matsushita, I., Shinka, T.,

Kuroki, Y., Itoh, K. & Sakuraba, H. (1996) Biochem . Biophys .

Res . Commun . 220, 802-806. 20.Speijer, H., Groener, J.E., Van Ramshorst, E. & Van Tol,

A. (1991) Atherosclerosis 90, 159-168.

21.Edelstein, C, Lim, C.T. & Scanu, A.M. (1972) J. Biol .

Chem . 247, 5842-5849.

22. Brown, M.S., Ho, Y.K. & Goldstein, J.L. (1980) J. Biol . Chem . 255, 9344-9352.

23. Chen, R.F. (1967) J. Biol . Chem . 242, 173-181.

24.Fraenkel-Conrad, H. (1957) Methods Enzymol . 4, 247-269.

25.McCarron, R.M., Goroff, D.K., Luhr, J.E., Murphy, M.A. &

Herscowitz, H.B. (1984) Methods Enzymol . 108, 274-284. 26.Miyazaki, A., Rahi , A.T., Araki, S., Morino, Y. &

Horiuchi, S. (1991) Biochim . Biophys . Acta . 1082, 143-151.

27.Lowry, O.H., Rosebrough, N.J., Farr, A.L. & Randall, R.J.

(1951) J. Biol . Chem . 193, 265-275.

28. Mills, G.L., Lane, P.L. & Weech, P.K. (1984) In Labora tory Techniques in Biochemistry and Molecular Biology, Vol 14 : A

Guidebook to Lipoprotein Technique, eds . Burdon R.H. & Van

Knippenberg P.H. (Elsevier, Amsterdam, New York, Oxford), pp.

171-220. 29.Sidorowicz, W., Hsia, W.C, Maslej -Zownir, 0. & Behal,

F.J. (1980) Clin . Chim . Acta . 107, 245-256.

30.Fόger, B., Santamarina-Fojo, S., Shamburek, R.D., Parrot,

C.L., Talley, G.D. & Brewer, H.B., Jr. (1997) J. Biol . Chem . 272, 27393-27400.

31. Ehnholm S., Willems van Dijk K. , Van 't Hof B., Van der

Zee A., Olkkonen V.M., Jauhiainen M., Hofker, M., Havekes L.,

Ehnholm C (1998) J. Lipid Res . 39 1248-1253.

32. Albers, J.J., Wolfbauer, G., Cheung, M.C, Day, J.R., Ching, A.F., Lok, S. & Tu, A.Y. (1995) Biochim . Biophys . Acta

1258, 27-34.

33. Goldstein, J.L., Ho, Y.K., Basu, S.K. & Brown, M.S. (1979)

Proc . Na tl . Acad. Sei . USA 76, 333-337.

34.McGookey, D.J. & Anderson, R.G. (1983) J. Cell Biol . 97, 1156-1168.

35. Rye, K.A. , Jauhiainen, M., Barter, P.J. & Ehnholm, C

(1998) J. Lipid Res . 39, 613-622.

36.Syvanne, M., Castro, G., Dengremont, C, De Geitere, C,

Jauhiainen, M. , Ehnholm, C, Michelagnoli, S., Franceschini, C, Kahri, J. & Taskinen, M.R. (1996) Atherosclerosis 127,

245-253.

37. Ishibashi, S., Goldstein, J.L., Brown, M.S., Herz, J. &

Burns, D.K. (1994) J. Clin . Invest . 93, 1885-1893.

38. Hubacek, J.A., Bϋchler, C, Aslanidis, C, Schmitz, G. (1997) : The genomic organization of the genes for human lipopolysaccharide binding protein (LBP) and bactericidal permeability increasing protein (BPI) is highly conserved. Biochem Biophys Res Commun , 236, 427-430.

39. Huuskonen, J. , Jauhiainen, M., Ehnholm, C, and Olkkonen, V.M. (1998): Biosynthesis and secretion of human plasma phospholipid transfer protein. J. Lipid Res, 39, 2021-2030.

40. van Vlijmen B.J. et al., J. Clin Invest . 1994 ; 93 : 1403- 1410 41. Plump, A.S. et al. Cell, 1992;71:343-53

42. Rubin E.M. et al., Nature, 1991;353:265-267

Table 1

PLTP activity, apo A-I and lipid concentrations in plasma of HuPLTP transgenic mice

G

Values are means ± SD. Values between brackets are percentage of the wild type values. For further details see Methods.

^'Significantly different from wild type mice, P<0.05

^**Significantly different from wild type mice, P<0.001

^#Significantly different from homozygous mice, P<0.001

Significantly different from homozygous mice, P<0.05

Table 2

Preβ-HDL levels in plasma of HuPLTP transgenic mice before and after incubation at 37°C

G

Values are means + SD. For details see Methods.

"significantly different from wild type mice, P<0.001 ^*Significantly different from wild type mice, P<0.01 *Significantly different from homozygous mice, P<0.001

Table 3

Plasma PLTP activity in mice after 2 weeks LFC followed by 4 weeks HFC/0.5% diet

Wild type 190%

PLTP1 489%

PLTP4 628%

Plasma PLTP activities were measured in pooled plasma from 5-6 mice and expressed as percentage reference plasma

Claims

1. Plasma phospholipid transfer protein or a functional derivative or fragment thereof for use as a pharmaceutical.

2. Use of a plasma phospholipid transfer protein or a functional derivative or fragment thereof in the preparation of a medicament for the treatment or prevention of medical conditions associated with relatively high plasma levels of cholesterol and/or triglycerides.

3. Use according to claim 2 for the treatment of atherosclerosis .

4. A pharmaceutical composition for the treatment or prevention of medical conditions associated with relatively high plasma levels of cholesterol and/or triglycerides, comprising plasma phospholipid transfer protein or a functional derivative or fragment thereof and a suitable diluent or other pharmaceutical excipient.

5. A composition according to claim 4, which comprises diluents and/or excipients for intravenous administration.

6. A composition according to claim 4 or 5, which further comprises other agents for the treatment or prevention of medical conditions associated with relatively high plasma levels of cholesterol and/or triglycerides.

7. A composition for preventing cholesterol accumulation and/or improving cholesterol excretion in a mammal comprising an effective amount of plasma phospholipid transfer protein or a functional derivative or fragment thereof.

8. A gene delivery vehicle for delivering a recombinant nucleic acid molecule encoding plasma phospholipid transfer protein or a functional derivative or fragment thereof to a host cell, whereby said recombinant nucleic acid molecule is integrated into the genome of said host cell.

9. A transgenic host cell obtainable by transfection with a gene delivery vehicle according to claim 8.

10. A composition according to any one of claims 4-7, wherein plasma phospholipid transfer protein is human plasma phospholipid transfer protein.

11. A composition according to claim 10, wherein human plasma phospholipid transfer protein comprises a sequence encoded by a cDNA of figure 5.

12. A use according to claim 1-3 wherein plasma phospholipid transfer protein is human plasma phospholipid transfer protein.

13. A use according to claim 12 wherein plasma phospholipid transfer protein comprises a sequence encoded by a cDNA of figure 5.