Lipid Transfer Inhibitor Protein Activity Deficiency in Normolipidemic Uremic Patients on Continuous Ambulatory Peritoneal Dialysis
Abstract We previously demonstrated that lipid transfer inhibitor protein (LTIP) is a potent modifier of lipid transfer protein (LTP) function in vitro. Based on these studies, we proposed that LTIP activity is an important determinant of lipoprotein size and composition, which leads to a stimulation of reverse cholesterol transport. To further evaluate this hypothesis, we have studied a normolipidemic, uremic patient population undergoing continuous ambulatory peritoneal dialysis (CAPD) that is deficient in LTIP activity (<18% of control). LDL from CAPD plasma was triglyceride enriched; the diameters of both CAPD LDL and HDL were increased and CAPD HDL was dominated by the largest subfraction, HDL2b. In CAPD patients, the plasma cholesterol esterification rate was only 61% of control; this decrease was due mainly to the poor reactivity of CAPD lipoproteins. CAPD lipoprotein-deficient plasma promoted twofold greater transfer of radiolabeled cholesteryl ester (CE) between standard lipoproteins than control, although LTP itself was increased only 39%. This twofold increase was not equally expressed among individual lipoprotein classes; CE transfers involving LDL were increased 2.4-fold, whereas those not involving LDL were increased only 50%. In whole plasma, CE net mass transfer to VLDL was slightly increased in CAPD plasma; relative to their CE content, control HDL contributed twofold more CE mass to VLDL than control LDL, but in CAPD plasma this preferential transfer of CE from HDL was absent. Collectively, the aberrations in CAPD lipoprotein composition and metabolism are consistent with the hypothesized role of LTIP. The data further support the role of LTIP in modulating the participation of HDL in CE mass transfers to VLDL. This is the first report of LTIP activity deficiency in humans.
- lipid transfer inhibitor protein
- lipid transfer protein
- continuous ambulatory peritoneal dialysis
- lecithin:cholesterol acyltransferase
- Received September 9, 1996.
- Accepted January 8, 1997.
Lipid transfer protein (LTP) in human plasma catalyzes the transfer of cholesteryl ester (CE) and triglyceride (TG) between lipoproteins.1–4 LTP facilitates CE and TG transfer by an exchange mechanism whereby lipid transfer between two lipoproteins involves either the exchange of CE for CE or TG for TG, or the heteroexchange of CE for TG.5 LTP action leads to the net transfer of CE from HDL and LDL toward VLDL, in association with a reciprocal transfer of TG from VLDL to LDL and HDL.5–7 Through this mechanism, LTP plays an important role in determining the balance of cholesterol between VLDL, LDL, and HDL, and in the reverse transport of cholesterol from extrahepatic tissues to the liver.8,9
Increased lipid transfer activity is often considered pro-atherogenic because elevated activity can augment the redistribution of CE from anti-atherogenic lipoproteins (HDL) to atherogenic lipoprotein classes (VLDL and LDL).6 LTP activity in human plasma is often increased in pathological states that are associated with lipoprotein abnormalities and characterized by high risk for the development of atherosclerosis, such as dyslipidemia,10 hypertriglyceridemia,11 hypercholesterolemia,12 and in both insulin-dependent13 and non-insulin dependent14,15 diabetes mellitus. However, the role of LTP in atherogenesis is unclear, because it also promotes reverse cholesterol transport5 and has recently been shown to be antiatherogenic in hypertriglyceridemic LTP-transgenic mice.16 Similarly, genetic LTP deficiency in humans has been shown in one study to be an independent risk factor for coronary heart disease.17 Together, these results suggest that other blood-borne factors may be important in defining the atherogenic potential of LTP activity.
Human plasma contains a specific protein that inhibits LTP activity in vitro.18–20 This protein, termed lipid transfer inhibitor protein (LTIP), is characterized as a unique acidic glycoprotein with a molecular weight ranging from 29 000 to 35 000.18,19 Although the mechanism of LTIP is not well understood, binding studies have demonstrated a direct correlation between the suppression of LTP activity by LTIP and the disruption of LTP-lipoprotein binding by this protein,21 suggesting that LTIP inhibits LTP by competing with the transfer protein for the interaction with lipoprotein surface. LTIP preferentially suppresses lipid transfers reactions involving LDL, and thus can alter the pattern of lipid fluxes mediated by LTP and promote the involvement of HDL in transfer reactions with VLDL.20,22 Therefore the effectiveness of lipid transfer activity in plasma is determined by both LTP and LTIP activities.
Based on our previous work,20,22 we have hypothesized that LTIP plays an important role in tailoring the pattern of lipid transfer reactions mediated by LTP. This remodeling of transfer reactions is proposed to influence the distribution of cholesterol between LDL and HDL, alter lipoprotein size, and modify the substrate properties of lipoproteins for cholesterol esterification reactions. The objective of this study was to gain further evidence for our hypothesis that previously has been based on reconstitution studies and on in vitro alterations of the LTIP levels of normal plasma. We have identified a uremic patient population undergoing continuous ambulatory peritoneal dialysis (CAPD) that has aberrant LTIP activity and have characterized the lipid transfer processes in the plasma of these individuals.
Cholesteryl [1α,2α(N)-3H]oleate (50 Ci/mmol) and [4-14C]cholesterol (52 mCi/mmol) were purchased from Amersham Corp. Cholesterol and cholesteryl oleate were purchased from Nu-Chek Prep. Lipid solutions were prepared in chloroform containing 10 μg/mL butylated hydroxytoluene and stored under N2 at −20°C. Reagents for chromatofocusing were from Pharmacia Fine Chemicals. All other reagents were obtained from Sigma Chemical Co.
Plasma, obtained from residual clinical samples of patients undergoing CAPD treatment, was provided by Dr Martin Schreiber from our institution. Samples were obtained randomly from a patient population that was 40% male, 60% female with an average age of ∼60 years. This population was composed of 70% African American and 30% Caucasian individuals. All patients had been on CAPD treatment for at least 36 months and received 8 to 12 L of Dianeal dialysate per day (Baxter Travenol). Patients typically received medication for hypertension (primarily calcium channel blockers) and were treated with erythropoietin, phosphorus binders, plus iron and vitamin D supplements. Control plasma (n=22) was obtained from randomly selected donors to the Cleveland Clinic Foundation Blood Bank or from volunteers within our research facilities.
The plasma lipid analysis in the total group of CAPD patients (n=84) revealed that hypertriglyceridemia was diagnosed in CAPD plasma in 52.4% of the cases, and hypercholesterolemia in 63.2%. To eliminate the effect of elevated plasma lipids on the results obtained, a subset of normolipidemic CAPD patients (n=40) was selected. The average TG levels in the selected CAPD (130.9±6.7 mg/dL) and control (127.8±10.3 mg/dL) groups, and the total cholesterol levels (174.3±3.3 and 166.7±5.4 mg/dL, respectively) did not differ. This study was approved by the Institutional Review Board, and informed consent was obtained when appropriate.
Plasma Collection and Processing
Blood samples were collected by venous puncture into sodium EDTA-containing tubes, and samples were immediately centrifuged at 2000g for 15 minutes to obtain plasma. In some instances, plasma samples were pooled before further analysis as indicated in the figure legends. Lipoprotein-deficient plasma was prepared by using a modification20 of the dextran sulfate precipitation method of Burstein et al.23 Lipoprotein-deficient plasma (<2 mL) was dialyzed twice (24 hours each) at 4°C against 2 L of 50 mmol/L Tris-HCl, 150 mmol/L NaCl, 0.02% NaN3, 0.01% EDTA, pH 7.4, containing 1 g/L Chelex 100 resin. Samples, either with or without subsequent heat treatment (described below), were filtered (0.45 μm, Millipore) before use in transfer assays. LTP and LTIP activities in lipoprotein-deficient plasma were determined immediately after dialysis.
Isolation and Radiolabeling of Lipoproteins
Human lipoproteins were labeled with [3H]cholesteryl oleate by the lipid dispersion technique of Morton and Zilversmit.18 Labeled and unlabeled lipoproteins in plasma were isolated at 4°C by sequential ultracentrifugation24 at solvent densities of 1.019, 1.063, and 1.21 g/mL to yield VLDL, LDL, and HDL, respectively. Under these labeling conditions, cholesteryl ester specific activities were ≥2000 dpm/μg. Lipoproteins were extensively dialyzed against a solution of 0.9% NaCl, 0.01% EDTA, 0.02% NaN3, pH 8.5, and stored at 4°C.
Isolation of Lipid Transfer Protein
Partially purified LTP was isolated from human lipoprotein-deficient plasma by hydrophobic and CM-cellulose chromatography and stored as previously described.20 Partially purified LTP did not contain detectable LCAT activity25 and was devoid of phospholipid-specific transfer activity.26
Assay of Lipid Transfer Activity
Lipid transfer assays were carried out as previously described.25,26 Typically, radiolabeled donor lipoprotein and unlabeled acceptor lipoprotein (at the concentrations indicated) were incubated in the presence of a lipid transfer activity source (partially purified LTP, plasma, or lipoprotein-deficient plasma) at 37°C for 1.5 to 6 hours. Lipid transfer activity was terminated by selectively precipitating one of the two lipoproteins by addition of PO43- and Mn2+,18 or alternatively, lipoproteins were separated by ultracentrifugation. The extent of transfer was assessed either by determining the radiolabel content of the fraction of interest, or by chemical measurements to determine the extent of mass transfer. The fraction of radiolabeled donor lipid transferred (kt) was calculated as described before25 and is reported as percent lipid transfer (kt × 100), or as μg lipid transfer, which was calculated by multiplying the kt value times the mass of the lipid in the donor particle. Radiolabeled lipid “transfer” in the absence of LTP was subtracted before these calculations.
Assay of Lipid Transfer Inhibitor Protein Activity
LTIP activity in lipoprotein-deficient plasma was determined by two methods. The inhibition of exogenously added LTP by lipoprotein-deficient plasma was determined by measuring transfer activity mediated by partially purified LTP in the absence and the presence of lipoprotein-deficient plasma. Lipid transfer activity of exogenous LTP in the presence of lipoprotein-deficient plasma minus the transfer activity of lipoprotein-deficient plasma itself was divided by the activity of the exogenously added LTP alone to determine the extent of inhibition induced by lipoprotein-deficient plasma. Alternatively, LTIP activity was determined by a heat inactivation method,20 in which LTIP activity was quantitated by the change in endogenous LTP activity in lipoprotein-deficient plasma after inactivation of LTIP by heat.
Measurement of Lecithin:Cholesterol Acyltransferase Activity
Cholesterol esterification in plasma mediated by lecithin:cholesterol acyltransferase (LCAT) was measured by a proteoliposome method using the lecithin-[4-14C]cholesterol substrates containing apolipoprotein AI as cofactor.27 Apolipoprotein AI was isolated from fresh human plasma by combination of ultracentrifugation, HDL delipidation, and chromatofocusing of human apolipoprotein HDL.28
Lipoprotein Particle Size Determination
Freshly isolated plasma was adjusted to a solvent density of 1.21 g/mL with NaBr and centrifuged at 50 000 rpm in a Beckman 50.3 Ti rotor for 22 hours at 4°C. The top fraction was removed and analyzed by nondenaturing polyacrylamide gel electrophoresis on 2.5% to 16% or 4% to 30% gradient gels (Isolab, Inc.) as previously described.29 Gels were stained with colloidal Coomassie Blue G-250 (Gradipore LTD), and the absorbance of lipoprotein peaks was quantified by computer-assisted image analysis using a high resolution CCD camera (Sierra Scientific). Thyroglobulin was added as an internal standard to each sample. The lipoprotein particle size was determined as previously reported29 with high molecular weight standards (Pharmacia LKB Biotech) and 38-nm latex particles (Duke Scientific Corporation) as size standards. HDL were assigned to subfractions based on the criteria reported by Nichols et al.29
Protein was quantitated by the method of Lowry et al as modified by Peterson,30 with bovine serum albumin as a standard. Total and free cholesterol were determined by colorimetric, enzymatic assays using a Cholesterol 100 reagent kit (Sigma Chemical Co.) and Free Cholesterol C kit (Wako Pure Chemical Industries), respectively. CE was determined as the difference between total cholesterol and free cholesterol levels multiplied by 1.69 to correct for its fatty acid content. TG content was measured by the glycerol phosphate oxidase-Trinder enzymatic method (Sigma Chemical Co.). Lipid phosphorus was assayed by the method of Bartlett,31 and a conversion factor of 25 was used to determine phospholipid mass.
A nonparametric Mann-Whitney unpaired test was used to evaluate the difference between groups. Reported P values are based on two-tailed calculations. Data are presented as mean±SE values. Results were considered significant if P<.05.
Lipid Transfer Inhibitor Protein Activity in Lipoprotein-Deficient Plasma
Several subject populations were screened in efforts to identify a group of individuals in which LTIP activity was altered. Based on this survey, we selected uremic patients treated with CAPD for further characterization as they demonstrated markedly reduced LTIP activity. Because hyperlipidemia is well known to alter lipid transfer processes, a subgroup of normolipidemic CAPD individuals was selected to exclude this confounding variable on our studies. LTIP activity was quantitated in normolipidemic CAPD and control subjects by two independent assays. In the first, the determination of LTIP activity relies on the capacity of lipoprotein-deficient plasma to directly suppress exogenous LTP. CAPD lipid deficient plasma displayed much lower inhibition of LTP-mediated transfer of [3H]CE from LDL to HDL than control (Fig 1A⇓). Based on the average of four experiments, CAPD lipoprotein-deficient plasma (30 μL) inhibited LTP-mediated transfer by 22.3±2.9% in contrast to 52.3±3.9% in control lipoprotein-deficient plasma. When corrected for the nonlinearity of the LTIP dose response in this assay,22 CAPD lipoprotein-deficient plasma was determined to contained only 21.8% of the LTIP activity in control samples (31.5±4.1 versus 144.8±10.8 U of LTIP activity/mL, CAPD versus control, respectively; 1 U=10% inhibition). LTIP activities calculated relative to plasma protein levels rather than volume were not different from those reported above.
An alternative, and more specific assay for LTIP activity in lipoprotein-deficient plasma exploits the differential heat stability of LTIP and LTP. The difference in endogenous LTP activity in heat-treated (LTIP inactive) and native lipoprotein-deficient plasma (LTIP active) reflects the extent to which endogenous LTP is suppressed by LTIP in vitro.20 Heat inactivation of LTIP in control lipoprotein-deficient plasma led to a marked 2.7-fold induction in CE transfer compared to a modest 1.2-fold increase seen in CAPD samples (Fig 1B⇑). This is equivalent to a 62.5% inhibition of endogenous LTP by LTIP in control plasma compared to 17.1% in CAPD plasma under these assay conditions. Correcting for the nonlinearity of the LTIP dose response,20 CAPD lipoprotein-deficient plasma contained only 13.7% of the LTIP activity in normal plasma (89.9- versus 654.7-μL equivalents, respectively). This large difference in LTIP activities was also evident in that before heating (LTIP active) CE transfer activity was about fourfold higher in CAPD than control subjects, but after heating (LTIP inactive) CE transfer activity was only about twofold different. Collectively, these assays indicate that normolipidemic CAPD subjects have <18% of normal LTIP activity.
Lipid Transfer Activity in Lipoprotein-Deficient Plasma
The effect of this LTIP activity deficiency on lipid fluxes between individual lipoprotein classes was readily noted. The lipoprotein-deficient fraction of CAPD plasma promoted [3H]CE transfer between a mixture of control VLDL, LDL, and HDL to a much greater extent than the lipoprotein-deficient fraction of control plasma. In this standardized assay, CAPD lipoprotein-deficient plasma (100 μL) facilitated a total of 6.45 μg of CE transfer among all lipoprotein classes (t=4.5 hours), whereas control transfer was only 3.08 μg. However, the effectiveness of [3H]CE transfer was variable between the different lipoprotein pairs. The transfers between VLDL and HDL did not increase proportionally to the total increase in CE transfer activity. Whereas the VLDL-LDL and HDL-LDL pathways were greatly stimulated (2.4-fold on average) in CAPD lipoprotein-deficient plasma versus control, HDL-VLDL transfers were increased only 1.5-fold (Table 1⇓).
The shift in relative transfer rates noted above is a hallmark of altered LTIP activity, however, a twofold increase in total CE transfer activity was also noted. To determine what portion of this rise was due to the LTIP activity deficiency versus changes in LTP levels, LTP activity in whole plasma, measured as the transfer of [3H]CE from LDL to HDL, was determined in the presence of excess exogenous donor and acceptor lipoproteins in as assay analogous to that described by Tall et al10 Under these assay conditions, LTIP has negligible activity (Serdyuk AP, Morton RE, unpublished observations) due to the very high concentration of lipoproteins in the assay, thus LTP activity correlates well with LTP mass.32 CAPD plasma transfer activity was 39% higher than control (4.21 versus 3.03% kt/5 μL plasma/6 hours), indicating an increase in LTP mass in the dialysis patients. Therefore an increase in LTP and a decrease in LTIP activity contribute to the higher total CE transfer activity in CAPD lipoprotein-deficient plasma.
The concentrations of VLDL, LDL, and HDL in control (22.8±2.2, 98.5±2.8, and 45.4±3.4 mg/dL, respectively (n=8)) and in normolipidemic CAPD subjects (29.6±3.4, 97.2±5.7, and 47.5±5.2 mg/dL, respectively (n=9)) were similar. Despite similar total plasma cholesterol and triglyceride concentrations in CAPD and control groups (see Methods), the chemical composition of CAPD lipoproteins was different from control (Table 2⇓). LDL TG and TG/CE were significantly raised in uremic patients on CAPD; the TG/CE ratio in CAPD HDL tended to be higher but did not reach statistical significance. Conversely, the TG level and TG/CE in CAPD VLDL were decreased. The free cholesterol/phospholipid ratio was decreased in CAPD VLDL, unchanged in CAPD LDL and increased slightly in CAPD HDL relative to the respective control lipoprotein fraction.
In addition to the chemical changes in CAPD lipoproteins, the size of the lipoprotein fractions was also different. The average particle diameters of CAPD LDL (25.29±0.28 nm (n=16) versus 23.73±0.30 nm (n=8), P<.001) and the HDL2b subfraction of HDL (11.12±0.05 nm (n=9) versus 10.61±0.09 nm (n=4), P<.01) were significantly increased compared to control. This increase in particle size resulted in a 21% increase in the molecular weight of CAPD LDL and 15% for CAPD HDL2b. Other HDL subfractions did not differ in mean diameter from control. However, the distribution of particles among the HDL subfractions was markedly altered in CAPD subjects (Fig 2⇓). The HDL spectrum in CAPD patients was shifted toward the largest, most lipid rich HDL2b particles whereas the content of the smaller HDL2a and HDL3 particles was decreased. The ratio of HDL2/HDL3 particles was 1.80 in CAPD versus 0.87 in control. Collectively, the altered composition and size of the CAPD lipoproteins, in the face of normal TG and cholesterol levels, suggest alterations in plasma lipoprotein metabolism in this patient group.
Cholesteryl Ester Mass Transfer in Whole Plasma
The marked decrease in LTIP activity in CAPD lipoprotein-deficient plasma would be expected to alter the participation of LDL and HDL in net CE transfers to VLDL because LTIP selectively suppresses the participation of LDL in this event.22 Native plasma was incubated for 18 hours and the change in lipoprotein CE content relative to unincubated samples was determined. Net CE accumulation in VLDL was slightly higher in CAPD plasma (315.5±5.2 μg/mL) in contrast to control plasma (284.4±2.5 μg/mL) (Fig 3⇓). This occurred although the TG/CE gradient between VLDL and the LDL-HDL fraction was less in CAPD plasma than control (Table 2⇑); this gradient drives the heteroexchange of CE for TG that facilitates net lipid transfer.2 In CAPD plasma, LDL was the major donor of CE to VLDL whereas in controls HDL was the major contributor to VLDL CE enrichment (Fig 3⇓). A comparison of the fractional participation of LDL and HDL in donating CE mass to VLDL versus their fractional contribution to the transferable CE pool (LDL + HDL CE) permits the calculation of a preference ratio that reflects the transfer of CE from a given lipoprotein relative to its CE content (Table 3⇓). In controls, the preference ratio for HDL was 1.82, indicating that HDL contributed nearly twice as much CE to VLDL as would be expected based on its contribution to the plasma CE pool. Conversely, LDL in controls participated much less than expected, with a preference ratio of 0.71. In CAPD plasma, on the other hand, the preference ratios for LDL and HDL were similar and neared unity, indicating that in CAPD plasma LDL and HDL participated in CE donation to VLDL at rates roughly equivalent to their plasma concentrations.
Lipoprotein Substrate Capacity
The altered lipid flux noted in whole plasma above is consistent with the hypothesized role of LTIP, however, an alternative source of this aberrant flux may be the CAPD lipoproteins themselves because their chemical and physical properties are different. Disproportionate reactivity of LTP with LDL versus HDL in CAPD plasma as compared to control could cause the shift in CE flux noted above. To assess this, the capacity of individual lipoprotein fractions isolated from CAPD and control plasma to support LTP activity was measured in standardized LTP assays by adding radiolabeled CAPD or control lipoproteins to incubation mixtures containing LTP and a standard acceptor lipoprotein (Fig 4A⇓). Both CAPD LDL and HDL tended to be slightly poorer donors of CE relative to the corresponding control lipoprotein (CAPD/control=0.94 and 0.89, respectively). When CAPD lipoproteins were the acceptor in the transfer reaction from a standard donor lipoprotein (Fig 4B⇓), again CAPD HDL was slightly less effective as an acceptor of CE compared to control HDL (CAPD/control=0.93), however, CAPD LDL was a better acceptor of radiolabeled CE than control (CAPD/control=1.25). In both instances, ie, as an acceptor and donor, the response with CAPD VLDL compared to control VLDL was identical to that noted with LDL. These two assays taken together suggest that, on average, CAPD LDL mediates ∼10% greater CE transfer activity than control LDL, whereas CAPD HDL was not statistically different from control HDL. However, this value is likely to be an overestimation, because CAPD LDL contains less CE (Table 2⇑) than does the control. When the LDL transfer activities (Fig 4⇓, A and B) were adjusted for this difference in CE content, CAPD, and control LDL appear to have similar capacities to participate in CE transfer reactions.
Free fatty acids are known to stimulate LTP activity.33,34 The presence of albumin in the lipid transfer assays, which would negate the influence of these compounds on LTP activity measurements, may mask real differences in the lipoprotein reactivity that are due to abnormal free fatty acid levels. However, free fatty acid concentrations in CAPD plasma were not elevated compared to control subjects (0.245±0.066 mEq/L (n=8) versus 0.413±0.061 mEq/L (n=5), respectively). Therefore, elevated free fatty acid levels do not account for the greater LTP activity noted in CAPD plasma.
Cholesterol Esterification Rate in Plasma
We have previously demonstrated that LTIP, because of its selective effects on lipid transfer reactions involving LDL, stimulates CE efflux from HDL and augments LCAT activity.22 The low LTIP activity in CAPD plasma suggests that the cholesterol esterification pathways in CAPD plasma may be altered. To test this aspect of the LTIP hypothesis, cholesterol esterification rates were measured in CAPD and control plasma by assessing the decrease in free cholesterol over time. The cholesterol esterification rate in CAPD plasma was 61% of control plasma (Fig 5A⇓), averaging 4.79 versus 7.86 μg free cholesterol consumed/mL/h for CAPD versus control, respectively. A portion of this decline appears to be due to slightly lower LCAT activity in CAPD plasma. Utilizing proteoliposomes as substrate to assess LCAT activity, which correlates highly with LCAT mass,27 CE synthesis rates were 5.16±0.28 and 6.82±0.39% free cholesterol esterified/h for CAPD (n=11, P<.005) versus control (n=5) plasma, respectively.
The above data suggest that other factors, such as the lipoprotein substrates themselves, may account for a significant part of the decline in whole plasma cholesterol esterification rates in CAPD subjects. To address this point, CAPD and control plasmas were fractionated by ultracentrifugation into lipoprotein (d<1.21 g/mL) and lipoprotein-deficient (d>1.21 g/mL) components. The role of each of these fractions in the reduced CAPD cholesterol esterification rate was assesses in a crossover design experiment (Fig 5B⇑). Recombination of the lipoprotein and lipoprotein-deficient fractions derived from the same plasma source reconstituted the deficient esterification activity noted above in native plasma. However, the cholesterol esterification activity in CAPD lipoprotein-deficient plasma could be increased to near control levels by the addition of control lipoproteins. Similarly, the control esterification activity was decreased markedly by the combination of control lipoprotein-deficient plasma with CAPD lipoproteins. There was no difference in the free cholesterol content of CAPD and control lipoprotein fractions. Overall, 65% of the decreased cholesterol esterification activity in CAPD plasma could be attributed to its lipoprotein fraction and 35% to differences in LCAT activity in the lipoprotein-deficient plasma fraction.
Our previous studies of LTIP in whole plasma and in reconstituted systems have led us to hypothesize that this protein plays an important role in tailoring lipid transfer reactions in vivo.18,20,22 LTIP inhibits the overall pattern of lipid transfers in plasma by uniquely affecting the individual transfer reactions, preferentially diminishing the transfer events involving LDL. The hypothesized consequences of this regulation by LTIP, which are partially supported by experimental observations, are to promote the involvement of HDL in transfer events with VLDL, enhance the substrate quality of the HDL fraction for LCAT, and to promote, in concert with lipolytic activity, a reduction of HDL size while LDL particle size increases.20,22 The greatly reduced LTIP activity in CAPD plasma affords a unique opportunity to test this hypothesis further by assessing the peculiarities of lipoprotein metabolism and composition that result when the activity of LTP’s inhibitor is pathologically low.
LTIP activity in normolipidemic CAPD subjects was <18% of control plasma. This lower activity was observed in two independent assays. Although the direct measurement of inhibitory activity in lipoprotein-deficient plasma is nonspecific, the heat inactivation assay measures inhibitory activity primarily attributable to LTIP.20 The agreement between these two assays suggests that LTIP is the only component in lipoprotein-deficient plasma that inhibits LTP activity under these assay conditions. Thus, variation in other apoprotein components35 in CAPD patients, some of which would also be reflected in lipoprotein-deficient plasma, do not appear to influence LTP activity in these assays.
LTP activity with lipoprotein-deficient plasma and exogenous lipoproteins indicated that CAPD lipoprotein-deficient plasma promotes greater total CE transfer, but it also mediates an overall derangement in the relative rates of lipid transfer between individual lipoprotein classes compared to control. In particular, CAPD lipoprotein-deficient plasma facilitated a two- to fourfold higher transfer of CE from and to LDL, but only a 40% to 50% increase in the transfer between HDL and VLDL compared to control lipoprotein-deficient plasma. Overall these data are consistent with our studies with isolated LTIP,22 and further support the conclusion that LTIP activity deficiency is the major determinant of the altered transfer activities mediated by CAPD lipoprotein-deficient plasma.
CE mass transfer experiments revealed significant differences in the participation of LDL and HDL in donating CE to VLDL. In control plasma, relative to its contribution to the plasma CE pool, HDL was 2.6-fold more active in CE transfer to VLDL than LDL. In CAPD plasma, the pool-size corrected contribution of HDL was only 1.3-fold greater than LDL, indicating that LDL and HDL were nearly equally effective contributors of CE to VLDL, relative to their CE content. Even though the total transfer of CE from LDL-HDL to VLDL was greater in CAPD plasma, CAPD HDL still contributed significantly less CE mass to VLDL than control HDL. In control plasma, HDL participates in CE mass transfer to VLDL at a rate that is twice that expected based on its plasma abundance. This preferential transfer of HDL CE is largely absent in CAPD plasma. It appears that only a small portion of the shift in lipoprotein preference can be attributed to changes in the substrate properties of the lipoproteins themselves. We speculate that the absence of this preferential transfer is due to the very low LTIP activity, which normally functions to enhance the participation of HDL in transfer reactions by hindering transfers with LDL.22 These studies suggest for the first time that LTIP may account for the disproportionately high participation of HDL in CE mass transfers in normolipidemic control subjects.
It is notable that the CE mass transfer studies show a slight increase in total CE mass transferred to VLDL in CAPD plasma, but not the twofold increase in lipid transfer activity seen in transfer assay measuring the transfer of radiolabeled CE. This is consistent with the studies of Mann et al11 which show that mass CE transfer in normolipidemic subjects is determined by the plasma TG concentration and not by LTP mass. Thus, although LTP activity, measured by radiolabeled lipid transfer, is increased markedly in CAPD subjects, this is not expressed in whole plasma because of the limited TG available for mass transfer.
Accompanying the reduced LTIP levels and aberrant participation of LDL and HDL in CE mass transfers was a significant reduction in plasma cholesterol esterification potential. Our data show that the reduced esterification in CAPD plasma is due to both lower absolute LCAT activity and, to a greater extent, the poor reactivity of the endogenous CAPD lipoproteins. This latter effect may be mediated in part by LTIP deficiency because we have shown that elevated LTIP activity in plasma enhances the substrate potential of lipoproteins in cholesterol esterification by LCAT.22 It is likely that the lower substrate potential of CAPD HDL is related to the shift in HDL subfraction levels and the increase in HDL size noted in these subjects.
The higher TG content of CAPD LDL, as evidenced by their elevated TG/CE ratios may be explained by the increased lipid transfer activity in CAPD plasma because LTP promotes the net mass transfer of CE from LDL and HDL towards VLDL with a reciprocal transfer of TG from VLDL towards LDL and HDL.5–7 TG-enriched LDL and HDL2b particles were larger than the control lipoprotein fractions, and the spectrum of HDL particles was shifted toward the largest, most lipid-rich subfraction in CAPD patients. The LTIP activity deficiency in CAPD subjects may contribute to this shift in HDL size by restricting the participation of HDL particles in lipid transfer events which normally function to reduce HDL size in conjunction with hepatic lipase.36 The mechanism underlying the increase in LDL size is unclear, but it may relate to the higher content of TG in CAPD LDL because this lipid has a greater molecular volume than CE.
The mechanism underlying the LTIP activity deficiency in CAPD lipoprotein-deficient plasma is unknown, but may reflect loss of LTIP protein during dialysis. CAPD therapy results in the loss of plasma proteins37,38; the rate of loss for a particular protein is directly related to its molecular mass.38 The greater molecular mass of LTP (∼74 000 Da)39 compared to LTIP (∼29 000 to 35 000 Da)18,19 theoretically may lead to a greater loss of LTIP in these subjects. However, because most LTP and LTIP are believed to be lipoprotein bound in vivo,19,21,22,40 such losses would be minimized due to the larger size of these complexes. Alternatively, LTIP protein levels may be unaltered but LTIP activity may be suppressed by an unknown mechanism. Distinction of these two possibilities will require the development of appropriate methods to quantitate LTIP mass.
Clearly, much remains to be elucidated about the factors that contribute to the decreased LTIP activity level in CAPD subjects. However, the LTIP activity deficiency in CAPD subjects does not appear to be related to this specific mode of dialysis. In preliminary studies, we have found that LTIP activity levels in hemodialysis patients are also markedly reduced (∼30% of control). Dialysis itself may contribute to the decrease in LTIP activity, because we have also noted that end-stage renal disease patients before dialysis appear to have near-normal LTIP activity (∼93% of control). This conclusion is consistent with our observation that LTIP activity was uniformly low in CAPD subjects despite the fact that this patient population is pathologically heterogeneous, although the use of pooled plasma samples may have masked some variation.
In summary, we describe here significant alterations in the reverse cholesterol transport pathways in CAPD subjects. CAPD plasma is characterized by reduced cholesterol esterification capacity, aberrant LTP-mediated CE fluxes between lipoprotein that are due to increased LTP activity and greatly diminished LTIP activity, and an overall shift of HDL particles toward larger, lipid-rich subfractions. The observed changes in these parameters are consistent with the hypothesized role of LTIP in lipoprotein metabolism and underscore the importance of LTIP in controlling the participation of LDL and HDL in CE flux to TG-rich lipoproteins. These data contribute to a growing body of evidence which suggests that, whereas LTP promotes lipid transfer between lipoproteins, the relative participation of lipoproteins in the transfer process is defined in large part by LTIP. The contribution of LTIP activity deficiency to the high risk for atherosclerotic disease in CAPD patients41 remains to be determined.
Selected Abbreviations and Acronyms
|CAPD||=||continuous ambulatory peritoneal dialysis|
|kt||=||fraction of donor lipid transfer|
|LTP||=||lipid transfer protein|
|LTIP||=||lipid transfer inhibitor protein|
This research was supported by grant HL29582 from the National Heart, Lung and Blood Institute, National Institutes of Health. This work was also supported by an Established Investigatorship (Dr Morton) from the American Heart Association and with funds contributed in part by the Iva D. Savage Award of the AHA Maryland Affiliate, Inc. The technical assistance of Diane Greene, Jeff Steinbrunner, and Tracy Evans is gratefully acknowledged.
Morton RE, Zilversmit DB. Purification and characterization of lipid transfer protein(s) from human lipoprotein-deficient plasma. J Lipid Res. 1982;23:1058–1067.
Morton RE, Zilversmit DB. Inter-relationship of lipids transferred by the lipid-transfer protein isolated from human lipoprotein-deficient plasma. J Biol Chem. 1983;258:11751–11757.
Albers JJ, Tollefson JH, Chen C-H, Steinmetz A. Isolation and characterization of human lipid transfer proteins. Arteriosclerosis. 1984;4:49–58.
Yen FT, Deckelbaum RJ, Mann CJ, Marcel YL, Milne RW, Tall AR. Inhibition of cholesteryl ester transfer protein activity by monoclonal antibody. Effects on cholesteryl ester formation and neutral lipid mass transfer in human plasma. J Clin Invest. 1989;83:2018–2024.
Glomset JA. The plasma lecithin:cholesterol acyltransferase reaction. J Lipid Res. 1968;9:155–167.
Tall A, Granot E, Brocia R, Tabas I, Hesler C, Williams K, Denke M. Accelerated transfer of cholesteryl esters in dyslipidemic plasma. J Clin Invest. 1987;79:1217–1225.
Mann CJ, Yen FT, Grant AM, Bihain BE. Mechanism of plasma cholesteryl ester transfer in hypertriglyceridemia. J Clin Invest. 1991;88:2059–2066.
Bagdade JD, Ritter MC, Subbaiah PV. Accelerated cholesteryl ester transfer in plasma of patients with hypercholesterolemia. J Clin Invest. 1991;87:1259–1265.
Hayek T, Masucci-Magoulas L, Jiang X, Walsh A, Rubin E, Breslow JL, Tall AR. Decreased early atherosclerotic lesions in hypertriglyceridemic mice expressing cholesteryl ester transfer protein transgene. J Clin Invest. 1995;96:2071–2074.
Morton RE, Zilversmit DB. A plasma inhibitor of triglyceride and cholesteryl ester transfer activities. J Biol Chem. 1981;256:11992–11995.
Nishide T, Tollefson JH, Albers JJ. Inhibition of lipid transfer by a unique high density lipoprotein subclass containing an inhibitor protein. J Lipid Res. 1989;30:149–158.
Morton RE, Steinbrunner JV. Determination of lipid transfer inhibitor protein activity in human lipoprotein-deficient plasma. Arterioscler Thromb. 1993;13:1843–1851.
Morton RE. Binding of plasma-derived lipid transfer protein to lipoprotein substrates: the role of binding in the lipid transfer process. J Biol Chem. 1985;260:12593–12599.
Morton RE, Greene DJ. Regulation of lipid transfer between lipoproteins by an endogenous plasma protein: selective inhibition among lipoprotein classes. J Lipid Res. 1994;35:836–847.
Burstein M, Scholnick HR, Morfin R. Rapid method for the isolation of lipoproteins from human serum by precipitation with polyanions. J Lipid Res. 1970;11:583–595.
Havel RJ, Eder HA, Bragdon JH. The distribution and chemical composition of ultracentrifugally separated lipoproteins in human serum. J Clin Invest. 1955;34:1345–1353.
Morton RE, Steinbrunner JV. Concentration of neutral lipids in the phospholipid surface of substrate particles determines lipid transfer protein activity. J Lipid Res. 1990;31:1559–1567.
Bartlett GR. Phosphorus assay in column chromatography. J Biol Chem. 1959;234:466–468.
McPherson R, Mann CJ, Tall AR, Hogue M, Martin L, Milne RW, Marcel YL. Plasma concentrations of cholesteryl ester transfer protein in hyperlipoproteinemia: relation to cholesteryl ester transfer protein activity and other lipoprotein variables. Arterioscler Thromb. 1991;11:797–804.
Tall A, Sammett D, Granot E. Mechanisms of enhanced cholesteryl ester transfer from high density lipoproteins to apolipoprotein B-containing lipoproteins during alimentary lipemia. J Clin Invest. 1986;77:1163–1172.
Deckelbaum RJ, Eisenberg S, Oschry Y, Granot E, Sharon I, Bengtsson-Olivecrona G. Conversion of human plasma high density lipoprotein-2 to high density lipoprotein-3: roles of neutral lipid exchange and triglyceride lipases. J Biol Chem. 1986;261:5201–5208.
Hesler CB, Tall AR, Swenson TL, Weech PK, Marcel YL, Milne RW. Monoclonal antibodies to the Mr 74,000 cholesteryl ester transfer protein neutralize all of the cholesteryl ester and triglyceride transfer activities in human plasma. J Biol Chem. 1988;263:5020–5023.
Cheung MC, Wolf AC, Lum KD, Tollefson JH, Albers JJ. Distribution and localization of lecithin:cholesterol acyltransferase and cholesteryl ester transfer activity in A-I-containing lipoproteins. J Lipid Res. 1986;27:1135–1144.