The role of phospholipid transfer protein in the development of atherosclerosis

Abstract

Purpose of Review:

Phospholipid transfer protein (PLTP), a member of lipid transfer protein family, is an important protein involved in lipid metabolism in the circulation. This article reviews recent PLTP research progresses, involving lipoprotein metabolism and atherogenesis.

Recent Findings:

PLTP activity influence atherogenic and anti-atherogenic lipoprotein levels. Human serum PLTP activity is a risk factor for human cardiovascular disease and is an independent predictor of all-cause mortality. PLTP deficiency reduces VLDL and LDL levels and attenuates atherosclerosis in mouse models, while PLTP overexpression exerts an opposite effect. Both PLTP deficiency and overexpression result in reduction of HDL which has different size, inflammatory index, and lipid composition. Moreover, although both PLTP deficiency and overexpression reduce cholesterol efflux capacity, but this effect has no impact in macrophage reverse cholesterol transport in mice. Furthermore, PLTP activity is related with metabolic syndrome, thrombosis, and inflammation.

Summary:

PLTP could be target for the treatment of dyslipidemia and atherosclerosis, although some potential off-target effects should be noted.

Introduction

Phospholipid transfer protein (PLTP) belongs to a family of lipid transfer proteins (1). Members of this family include bactericidal/permeability-increasing protein (BPI) (2), cholesterol ester transfer protein (CETP) (3), and lipopolysaccharide-binding protein (LPS binding protein, LBP) (4). In general, each of these transfer proteins have a boomerang shape, with the N-terminal and C-terminal barrel structures linked together to form a hydrophobic channel for the transport of lipid molecules. The protein structures of the other three members are known, but the structure of PLTP has not yet been elucidated.

Human PLTP cDNA is 1,750 base pairs in length, and includes an open reading frame of 1,518 nucleotides and a 3’-untranslated region (UTR) of 184 nucleotides(5) (Table 1). Mature PLTP contains 476 amino acids and 6 N-glycosylation sites, which allows it to change its molecular weight (55 or 81 kDa) after different degrees of glycosylation modification (5). Mature mouse PLTP contains the same number of amino acids as human PLTP, with 83% homology in the whole sequence and greater difference in 3’-UTR (6) (Table 1). Mouse PLTP contains 7 N-glycosylation sites. About 42% of the amino acid residues of human and mouse PLTP are hydrophobic. PLTP mediates non-specific transport activity of amphiphilic molecules, such as phospholipids, free cholesterol, alpha-tocopherol, diacylglycerol, lipopolysaccharide, and sphingosine-1-phosphate (7, 8).

Table 1.

The loci of human and mouse PLTP gene

Species	Human	Mouse
Location	Complement (45,898,620..45,912,364)	Complement (164,839,518..164,857,708)
Length	13,745 nt	18,191 nt
position	NC_000020.1145,905,452	NC_000068.7 164,848,728
Gene position	6,913	8,981
Exon count	17	16
Gene ID	5360	18830

The information was from NCBI Gene.

Almost all tested tissues express PLTP (5, 6). Importantly, PLTP is highly expressed in human atherosclerotic lesions (9, 10). There are two forms, low active one and high active one, of PLTP in human plasma, the former is complexed with apoA-I and the latter is complexed with apoE (11–13). However, no similar complex has been reported in human cerebrospinal fluid, rabbit plasma, and mouse plasma.

Under physiological conditions, plasma PLTP is mainly released from tissues or cells, such as the liver, adipose tissue, and macrophages. In global PLTP-deficient mice, hepatic PLTP contributes about 20% of plasma PLTP activity (14). In hematopoietic cells-specific PLTP-deficient mice, plasma PLTP activity reduces to 40%-50% of that of wild type mice, indicating that hematopoietic cells are one of the major sources of plasma PLTP activity (15). In adipocyte-specific Pltp knockout mice, plasma PLTP activity decreases by approximately 15% (16). However, based on a very recent report, PLTP activity in brown adipose tissue (BAT)-less mice was reduced by 58.1%, and thus, BAT is a significant source of circulating PLTP (17). This needs to be confirmed.

In the lung, the expression level of PLTP in mice is very high. After knocking out Pltp in the tissue, plasma PLTP activity decreased by nearly 15%, suggesting that the lung is also another important source of plasma PLTP activity (18). Certain pathophysiological changes, such as hypoxia, can induce high expression of PLTP in the lung, and it remains to be proven whether induction of PLTP is related with surfactant transportation (18). In multiple sclerosis, while plasma PLTP activity is significantly increased, cerebrospinal fluid PLTP activity is significantly decreased (19, 20), suggesting that PLTP-related metabolism may be involved in multiple sclerosis.

In the brain, PLTP expression is also high, while its function is still not quite understood (21, 22). PLTP activity seems to play an important role in ensuring neuronal structural integrity and completing signal transduction pathways. The concentration of vitamin E in the brain of Pltp knockout mice was significantly lower than that of wild-type mice (23). PLTP expression levels in patients with Alzheimer’s disease are significantly downregulated (21, 22). PLTP deficiency can aggravate β-amyloid-related memory deficits (23). The expression of PLTP mRNA in cerebrovascular cells is 6.8 times higher than the average level of the whole brain (24). PLTP seems to have a function for maintaining the blood brain barrier (25).

PLTP Regulation

PLTP can be regulated by transcription factors. Both fenofibrate and chenodeoxycholic acid can promote the expression of PLTP, suggesting that peroxisome proliferator-activated receptor (PPAR) and farnesoid X-activated receptor (FXR) are involved in its regulation (26). Human and mouse Pltp promoters contain Liver X receptor (LXR) response elements and can be regulated by LXR (27, 28).

Many physiological factors and pathological stimuli can also regulate the expression of PLTP activity. Hypoxia, glucose, and bile acid can stimulate the upregulation of PLTP expression, while lipopolysaccharide (LPS) and insulin can reduce PLTP expression (29, 30). Diacylglycerol can also regulate PLTP activity (31).

Recently, we found that profurin (prodomain of furin) overexpression in Ldl receptor KO mice, an atherogenic mouse model, significantly reduces atherosclerosis and LDL levels in the circulation (32). This phenomenon is associated with profurin-mediated PLTP degradation in hepatocytes(33*).

In alveolar epithelial cells, oxidized LDL can stimulate PLTP expression through TGF-β1/smad3/Ras/ERk pathway(34). Cigarette smoke extract can induce PLTP expression through TGF-β1/CyclinD1/CDK4 pathway(35).

PLTP and Cardiovascular Diseases

We reported that serum PLTP activity is a risk factor for human cardiovascular disease and suggested that PLTP could be a therapeutic target for atherosclerosis (36). Many human studies agreed with our observation (37–40). Moreover, it has been reported that Pltp gene variation, two Pltp tagging single nucleotide polymorphisms, leads to decreased risk of cardiovascular disease among 5 cohorts (41). A relatively recent Framingham Heart Study indicated that higher plasma PLTP activity predicted a first cardiovascular event, defined as fatal or non-fatal coronary heart disease and stroke, among men (42). Moreover, human left ventricular systolic dysfunction is also associated with higher PLTP activity (43, 44). Importantly, we found that plasma PLTP activity levels are a strong and independent predictor of all-cause mortality in 5 years and higher PLTP activity had higher mortality (45). Contradictorily, PLTP concentration was reduced in a small group of cardiovascular disease patients (46), although it seems to be accepted that plasma PLTP concentration does not equal to PLTP activity (47, 48), since there are two forms of PLTP in the circulation, inactive one and active one (11–13). Also, the effect of PLTP activity on peripheral artery disease are both inconsistent and limited (49, 50).

In mouse models, global PLTP deficiency attenuates the development of atherosclerosis (51), whereas PLTP overexpression has an opposite effect (52–54). Global PLTP deficiency is also associated with a decrease in thrombotic response (55) and decreases the incidence of abdominal aortic aneurysms (56). In a rabbit model, overexpression PLTP promotes the occurrence of atherosclerosis (57). However, a very recent study made the situation becoming more complicated. Scavenger receptor B1 (Sr-bl) knockout mice develop atherosclerotic lesions after 20 weeks on Western-type diet (58), while Sr-bl/Pltp double knockout mice reduce the lesion, despite the induction of proatherogenic metabolic complications(59). This seems to be related with HDL normalization(59). This observation needs to be confirmed. In general, the majority studies indicated that PLTP is a risk factor for atherogenesis. Then, what is the mechanism? It could be related with the production of very-low density lipoprotein (VLDL).

PLTP and VLDL Production

Human genome-wide association studies (GWAS) have shown that plasma PLTP levels are positively associated with plasma apoB and triglyceride levels (38, 60). PLTP plays an important role in VLDL production. PLTP deficiency in mice attenuates VLDL secretion from the liver (14, 51), while PLTP overexpression promotes VLDL overproduction (61, 62). Induction of PLTP activity permits triglyceride incorporation into VLDL (63). Importantly, PLTP plays an important role in the initiation of VLDL assembly in mouse primary hepatocytes (64). We also found that the major function of hepatic PLTP is to drive VLDL production (Figure 1) (14, 65) and prodomain of furin-mediated PLTP proteasomal degradation can prevent VLDL production (33*). Human Pltp transgenic rabbits have a significant increase in plasma LDL (66) which could be related with more VLDL production. Unlike microsomal triglyceride transfer protein inhibition which results in fatty liver in mice and humans (67), PLTP-deficient mice have no lipid accumulation in the liver (51), potentially, PLTP inhibition in humans could results in reduction of VLDL and LDL with no consequence of fatty liver.

Figure 1. Effect of PLTP on the production of VLDL and HDL.

PLTP can mediate the fusion of primordial VLDL with apoB-free lipid droplet, thus, facilitating apoB lipidation and VLDL secretion. Profurin-mediated PLTP degradation can prevent VLDL secretion. PLTP can assist ABCA1 to efflux lipids onto free apoA-I for the formation of nascent HDL. PLTP can transfer lipids from the surface of VLDL or chylomicron (CM) onto nascent HDL for the formation of mature HDL.

PLTP and High-Density Lipoprotein (HDL)

There are four pathways for HDL biosynthesis (68) and PLTP plays a role in all these processes. PLTP can transfer lipids, phospholipids and free cholesterol, from the surface of VLDL or chylomicron (CM) onto nascent HDL for the formation of mature HDL (69)(Figure 1). The dramatic reduction of HDL in PLTP-deficient mice should be related with the reduction of VLDL secretion and the blockage lipid transfer, from VLDL to HDL (14, 69). Moreover, PLTP deficiency significantly increases HDL turnover rate (70, 71). It is well known that hepatocytes, enterocytes, and peripheral tissue cells (such macrophages) make contribution to HDL formation through ABCA1 (68, 72). Since PLTP can help ABCA1 to efflux lipids (73), PLTP activity should have an impact in HDL production in these three type cells (Figure 1). PLTP is also a putative fusion factor to make bigger HDL (74–76). Moreover, PLTP also exchanges phospholipids among the lipoproteins (77, 78).

Adenovirus and adenovirus-associated virus-mediated PLTP overexpression resulted in a dramatic induction of PLTP activity in the circulation of mice (52, 79). These mice had increased preβ-HDL levels but decreased α-HDL levels. In human PLTP transgenic mice, plasma total HDL level is significantly reduced and plasma preβ-HDL level is significantly induced (80).

Researches have indicated that HDL particles are very heterogeneous, and indeed increasing total HDL cholesterol does not reduce cardiovascular risk (81–86). Thus, it is important to characterize each HDL particle subclass in terms of its pro- or anti-atherogenicity. HDLs isolated from PLTP transgenic, knockout, and wild type mice were compared. The comparison indicated that PLTP activity plays an important role in determining plasma HDL size, inflammatory index and lipid composition (16). HDLs isolated from different mice have different sizes: Pltp transgenic mouse HDL > wild type mouse HDL > Pltp knockout mouse HDL (16). Different HDLs have different inflammatory index: Pltp transgenic moue HDL > wild type mouse HDL > Pltp knockout HDL (16). Different HDLs have different lipid contents(16). We also found that liver-specific PLTP deficiency significantly decreases HDL and apoA-I levels (65). A recent study indicated that both PLTP deficiency and overexpression in mice can reduce HDL mass and cholesterol efflux capacity but these effects have no impact in macrophage reverse cholesterol transport (87).

Recent research has focused on Sphingosine-1-phosphate (S1P) and HDL metabolism because S1P is an important constituent of HDL(88–91). HDL is the primary carrier of circulating S1P and S1P-HDL is important in numerous physiological and cellular processes(92). HDL associated apolipoprotein M (apoM) is a physiological chaperone of S1P (93). However, systemic apoM deficiency in mice can only decrease plasma S1P levels by 25-45% (93, 94). Another S1P carrier is albumin, however, Apom/albumin double deficiency can only reduce plasma S1P by 60%-65% (95). Thus, there should be other factors that contribute to S1P transport or carriers. ApoA4 is a one(95) and PLTP could be another one. We found that PLTP deficiency causes a dramatical reduction of plasma S1P levels (60%) (8). PLTP can transport S1P from red blood cells to HDL(8). Since red blood cells are a major source of plasma S1P, this result indicates that PLTP may be an important determinant of plasma S1P. Moreover, PLTP-deficiency does not influence plasma apoM content (8).

PLTP and Metabolic Syndrome

Metabolic syndrome is a group of conditions that occur together, increasing risk of heart disease, stroke, insulin resistance, and type 2 diabetes. Onsetting of three of five following pathological conditions can be consider as metabolic syndrome: obesity, high triglyceride, high cholesterol, increased blood pressure, and high blood sugar (96). It is known for a quite long time that circulating PLTP levels are high in humans with obesity and insulin resistance(97–99). PLTP activity is also higher in obese and dietary induced obese mice (17, 100). It is also known that PLTP-deficient mice have significantly lower VLDL production rate (51, 101) and liver PLTP expression with a PLTP-null background induces VLDL secretion and increases plasma triglyceride and non-HDL-cholesterol levels(14). Recently, we evaluate the effect of PLTP deficiency on dietary induced obesity and insulin resistance. We found that the deficiency can significantly decrease obesity and insulin resistance (102).

Mechanistic studies showed that the deficiency significantly enhance insulin receptor and Akt phosphorylation levels in liver, adipose tissue, and muscle tissue after insulin stimulation (102). Moreover, the deficiency induces more GLUT4 protein on the membranes of adipocytes and muscle cells after insulin stimulation(102). Based on what we have observed, PLTP deficiency could prevent the development of metabolic syndrome. However, a very recent study had a different view. It reported that systemic or brown adipose tissue-specific overexpression of PLTP, mediated by adenovirus or adenovirus associated virus, can control energy expenditure and systemic glucose/lipid homeostasis(103). The authors claimed that the metabolic benefit of PLTP is mediated through a brown adipose tissue-liver interorgan communication(103). Thus, the role of PLTP in obesity and insulin resistance in mice is still not that clear and more work are needed.

PLTP and Thrombosis

PLTP-deficient mice have a longer blood clotting time than that of controls. This observation is associated with a reduction of phosphatidylserine externalization (104). Moreover, PLTP deficiency can reduce thrombotic response to acute intravascular oxidative stress(55). Thus, PLTP depletion seems to be related with hypocoagulation. On the other hand, certain studies suggested that PLTP has an anticoagulation effect (105, 106). Thus, it is still not quite clear what kind of role PLTP plays in the process of coagulation. Recently, we found that PLTP can stimulate phosphatidylserine externalization, as a flippase, on the plasma membrane of platelets and accelerate collagen-, and ADP-mediated platelet aggregation. Therefore, PLTP is associated with hypercoagulation (107*) and PLTP inhibition could be a novel approach for countering thrombosis.

PLTP and Inflammation

The relationship between PLTP and inflammation remains controversial. PLTP-deficient mice reduce interleukin-6 (IL-6) in the circulation (108, 109). A shift of T helper (Th) lymphocytes towards the anti-inflammatory subset Th2 was observed in PLTP-deficient mice (110). However, other studies (most of them are related with lipopolysaccharide treatment) showed that PLTP has an anti-inflammatory function (111–113). LPS treatment causes higher mortality in PLTP deficient mice (111). LPS and cigarette smoke exposition can decrease PLTP expression and activity (112). Since PLTP can bind and neutralize LPS, this can potentially explain its anti-inflammatory functions (111, 114). Moreover, PLTP might also have an anti-inflammatory properties in macrophages through an interaction with the ABCA1, then JAK2/STAT3 pathway(113, 115).

Conclusion

PLTP activity plays an important role in the development of cardiovascular disease, and it could be associated with dyslipidemia, metabolic syndrome, blood hypercoagulability, and inflammation. However, the intracellular functions of PLTP are still poorly understood. In addition, more epidemiological studies are needed to better understand the role of PLTP in human cardiovascular disease. We learned a lot from germline global PLTP gene knockout mice, in terms of its role in atherogenesis. However, it is known that germline deficiency does not necessarily represent the best technical approach for studying the function of a protein or an enzyme in vivo, particularly in adult mice. Thus, an inducible PLTP-deficient model is required. We are currently preparing this mouse model and hopefully, in the near future, we should be able to investigate the impact of inducible PLTP deficiency in atherogenesis. We have evidence to show that PLTP is target for the treatment hypercholesterolemia and atherosclerosis, although some other effects should also be considered. In our view, PLTP inhibition in human cardiovascular disease patients should have much more beneficial effects than unwanted effects.