The Influence of Apolipoproteins on the Hepatic Lipase-mediated Hydrolysis of High Density Lipoprotein Phospholipid and Triacylglycerol*

This study describes the influence of apolipoproteins on the hepatic lipase (HL)-mediated hydrolysis of phospholipids and triacylglycerol in high density lipoproteins (HDL). HL-mediated hydrolysis was assessed in well characterized, homogeneous preparations of spherical reconstituted high density lipoproteins (rHDL). The rHDL were comparable in size and lipid composition and contained either apoA-I ((A-I)rHDL) or apoA-II ((A-II)rHDL) as their sole apolipoprotein constituent. Preparations of rHDL containing only cholesteryl esters (CE) in their core, (A-I/CE)rHDL and (A-II/CE)rHDL, were used to assess phospholipid hydrolysis. Preparations of rHDL that contained triacylglycerol as their predominant core lipid, (A-I/TG)rHDL and (A-II/TG)rHDL, were used to assess both triacylglycerol and phospholipid hydrolysis. The rHDL contained trace amounts of either radiolabeled phospholipid or radiolabeled triacylglycerol. Hydrolysis was measured as the release of radiolabeled nonesterified fatty acids (NEFA) from the rHDL. Kinetic analysis showed that HL had a greater affinity for the phospholipids in (A-II/CE)rHDL (K m (app) = 0.2 mm) than in (A-I/CE)rHDL (K m (app) = 3.1 mm). This was also evident when hydrolysis was measured directly by quantitating NEFA mass. HL also had a greater affinity for the phospholipids and triacylglycerol in (A-II/TG)rHDL than in (A-I/TG)rHDL. TheV max for phospholipid hydrolysis was, by contrast, greater for (A-I/CE)rHDL than for (A-II/CE)rHDL: 309.3versus 49.1 nmol of NEFA formed/ml of HL/h. ComparableV max values were obtained for the hydrolysis of the phospholipids in (A-II/TG)rHDL and (A-I/TG)rHDL. In the case of triacylglycerol hydrolysis, the respective V maxvalues for (A-I/TG)rHDL and (A-II/TG)rHDL were 1154.8 and 240.2 nmol of NEFA formed/ml of HL/h. These results show that apolipoproteins have a major influence on the kinetics of HL-mediated phospholipid and triacylglycerol hydrolysis in rHDL.

Hepatic lipase (HL) 1 is a 476-amino acid glycoprotein of molecular weight 64,000 -69,000 (1) that is bound to liver sinusoidal endothelial cells (2). HL hydrolyzes acyl ester bonds of triacylglycerol and the sn-1 acyl ester bond of phospholipids. The main plasma substrates for HL are very low density lipoproteins and high density lipoproteins (HDL). The role of HL in HDL metabolism is of considerable importance, as shown by strong negative associations between HL activity and plasma HDL 2 levels (3)(4)(5) and the dramatic reduction in the HDL levels of rabbits that have been made transgenic for human HL (6).
Unlike lipoprotein lipase (LPL), which requires apolipoprotein C-II (apoC-II) for maximal activity, there is no known protein cofactor for HL. However, there is some conflicting evidence to suggest that the apoA-II in HDL may influence the HL-mediated hydrolysis of triacylglycerol in HDL (7)(8)(9)(10). Some investigators have reported that apoA-II enhances (7,8), while others have concluded that it inhibits, the HL-mediated hydrolysis of triacylglycerol in HDL (9,10).
The present study was carried out in order to determine whether there are significant differences in the HL-mediated hydrolysis of phospholipids and triacylglycerol in HDL that differ in their apolipoprotein composition. This has been achieved by using well defined, homogeneous preparations of spherical reconstituted HDL (rHDL) as substrates for HL. The rHDL were comparable in size and lipid composition and contained either apoA-I or apoA-II as their sole apolipoprotein constituent. The results show that apolipoproteins not only have a major influence on the HL-mediated hydrolysis of the triacylglycerol and phospholipids in rHDL but also regulate the affinity of HL for the rHDL surface.

EXPERIMENTAL PROCEDURES
Purification of ApoA-I and ApoA-II-ApoA-I and apoA-II were prepared from pooled human plasma donated by the Transfusion Service, Royal Adelaide Hospital. HDL were isolated from the plasma by sequential ultracentrifugation in the 1.07 Ͻ d Ͻ 1.21 g/ml density range (11). The isolated HDL were delipidated (12), and the resulting apoHDL was subjected to anion exchange chromatography on Q Sepharose Fast Flow (Amersham Pharmacia Biotech, Uppsala, Sweden) (13). The purified apoA-I and apoA-II appeared as single bands following electrophoresis on a homogeneous 20% SDS-polyacrylamide PhastGel (Amersham Pharmacia Biotech) and Coomassie staining.
Purification of Lecithin:Cholesterol Acyltransferase (LCAT)-LCAT was purified from pooled human plasma (Transfusion Service, Royal Adelaide Hospital) as described previously (14). The purified LCAT appeared as a single band following electrophoresis on a homogeneous 20% SDS-gel and silver staining. LCAT activity was assessed as described by Piran  Purification of Cholesteryl Ester Transfer Protein (CETP)-CETP was isolated from pooled human plasma (Transfusion Service, Royal Adelaide Hospital) as described previously (16,17). Transfer activity was quantitated as the transfer of [ 3 H]CE from [ 3 H]CE-HDL 3 to low density lipoproteins (LDL) (18,19). The assay was linear when less than 30% of the total counts were transferred from HDL 3 to LDL during a 3-h incubation at 37°C. The CETP preparation used in this study had 12.6 units of activity/ml, where 1 unit is the transfer activity of 1 ml of pooled, lipoprotein-deficient human plasma.
Purification of Phospholipid Transfer Protein (PLTP)-PLTP was purified from pooled human plasma (Transfusion Service, Royal Adelaide Hospital) as described elsewhere (20). PLTP activity was quantitated as the transfer of L-3-1,2-di[1-14 C]palmitoylphosphatidylcholine ([ 14 C]DPPC) (112 mCi/mmol) (Amersham Pharmacia Biotech) from [ 14 C]DPPC-labeled phospholipid vesicles to ultracentrifugally isolated HDL during a 2-h incubation at 37°C (21). The PLTP preparation used in this study transferred 2700 nmol of phospholipid/ml of PLTP/h. Preparation of Spherical rHDL (Fig. 1)-Three pairs of substrates (a and b, c and d, and e and f) were used to study the HL-mediated hydrolysis of phospholipids and triacylglycerol in rHDL (Fig. 1). Each pair of substrates was comparable in size and lipid composition and contained either apoA-I or apoA-II as its sole apolipoprotein.
For most of the experiments, the substrates were labeled either with [ 14 C]DPPC, for assessing phospholipid hydrolysis, or [9, (22). The discs were incubated with LDL and LCAT as described previously (23) to generate spherical rHDL with CE in their core and apoA-I as the sole apolipoprotein constituent, (A-I/CE)rHDL. The (A-I/CE)rHDL were dialyzed extensively against 0.01 M Tris-buffered saline (TBS) (pH 7.4) containing 0.15 M NaCl, 0.005% (w/v) EDTA-Na 2 , and 0.006% (w/v) NaN 3 before use.
The The resulting (A-I/TG)rHDL were isolated by sequential ultracentrifugation in the density range 1.063 Ͻ d Ͻ 1.21 g/ml using a TLA-100.4 rotor (Beckman Instruments) as described previously (17). . This pair of substrates was used to study HL-mediated phospholipid hydrolysis in the absence of triacylglycerol. The second pair of radiolabeled substrates, c and d, contained triacylglycerol as the predominant core lipid and either apoA-I or apoA-II as the sole apolipoprotein, and these substrates were labeled with [ 14 C]DPPC. They were used to study phospholipid hydrolysis in the presence of triacylglycerol. The final two radiolabeled substrates, e and f, were identical to c and d, except they contained [ 3 H]triolein (ૺTG) in their core. They were used to study triacylglycerol hydrolysis. tion ϭ 4 mmol/liter) was then incubated for 1.5 h at 37°C with (A-I/ CE)rHDL (final CE concentration ϭ 0.1 mmol/liter) and CETP (final concentration ϭ 2.7 units/ml) in a final volume of 44.8 ml. The resulting [ 3 H]triolein-labeled (A-I/TG)rHDL were isolated by sequential ultracentrifugation in the density range 1.063 Ͻ d Ͻ 1.21 g/ml using a TLA-100.4 rotor (Beckman Instruments) as described elsewhere (17).
Preparation of Native (A-I)HDL 2 and (A-II)HDL 2 Labeled with [ 14 C]DPPC-HDL 2 were isolated from fresh human plasma by sequential ultracentrifugation in the 1.07 Ͻ d Ͻ 1.12 g/ml density range with two 24-h spins (50,000 rpm) at the lower density followed by a single 40-h spin (50,000 rpm) and one 16-h spin (100,000 rpm) at the higher density. The 50,000 rpm spins were carried out at 4°C using a Ti-50 rotor in a Beckman L8 -70M ultracentrifuge (Beckman Instruments). The 100,000 rpm spin was carried out at 4°C using a TLA-100.4 rotor in a Beckman TL-100 tabletop ultracentrifuge. The isolated HDL 2 , which contained apoA-I as the predominant apolipoprotein, is designated (A-I)HDL 2 . Purification of HL-HL was purified from the blood of patients injected with a bolus of 25,000 IU of heparin prior to undergoing angioplasty (Cardiovascular Investigational Unit, Royal Adelaide Hospital). Postheparin plasma was isolated by centrifugation at 3,000 rpm for 10 min at 4°C and stored at Ϫ70°C. The pooled postheparin plasma was thawed and added to an equal volume of 0.005 M sodium barbitone buffer, 0.45 M NaCl (pH 7.4). The postheparin plasma was applied to an HR 10/30 column containing Heparin Sepharose Fast Flow (Amersham Pharmacia Biotech) preequilibrated with 0.005 M sodium barbitone, 0.15 M NaCl (pH 7.4). HL was eluted from the column at a flow rate of 5 ml/min with a linear 0.8 -1.3 M NaCl gradient. Fifty IU of heparin was added to the eluted fractions (8 ml) before dialysis against TBS containing 6.3 IU/ml heparin. Active fractions were pooled and concentrated approximately 20-fold in a Centriprep-10 concentrator (Amicon Inc., Beverly, MA). Heparin was added to the pooled fractions to give a final concentration of 500 IU/ml. The purified HL appeared as a single band following SDS-polyacrylamide gel electrophoresis on a 20% homogeneous PhastGel (Amersham Pharmacia Biotech) and staining with Coomassie Blue. The HL was stored at Ϫ70°C.
HL activity was assessed as the nmol of triacylglycerol hydrolyzed/ml of HL/h using (A-I/TG)rHDL as a substrate. Purified HL (24 l) was incubated at 37°C for 1 h with (A-I/TG)rHDL (final triacylglycerol concentration ϭ 0.15 mmol/liter) and heparin (final concentration ϭ 500 IU/ml) in the presence or absence of 1 M NaCl. The final incubation volume was 120 l. Triacylglycerol hydrolysis was measured as the decrease in triacylglycerol mass compared with a control incubation that did not contain HL. Since the amount of triacylglycerol hydrolysis was identical in the presence and absence of 1 M NaCl, the HL was judged to be free of LPL activity. It should also be noted that the rHDL preparations were deficient in apoC-II, the cofactor for LPL-mediated triacylglycerol hydrolysis. Triacylglycerol mass was measured by enzymatic assay (24) using Boehringer Mannheim standards (Mannheim, Germany). The amount of HL activity in individual experiments is presented in the figure legends.
Determination of HL-mediated Hydrolysis in Radiolabeled Substrates-All incubations were carried out in stoppered plastic tubes in a shaking water bath at 37°C. Details of individual incubations are described in the legends to the figures. Chloroform/methanol (1 ml, 2:1 (v/v)), was added to stop the hydrolysis reactions. The lipids were extracted by the method of Folch et al. (25). NEFA were separated from the other rHDL (or native HDL 2 ) lipids by thin layer chromatography on 20 ϫ 20-cm Silica gel 60 plastic sheets (Merck, Darmstadt, Germany). The sheets were developed in chloroform/methanol/water (65:25:4, v/v/v) until the solvent front was 8 cm from the origin. The sheets were dried and then run in hexane/diethyl ether/acetic acid (70:30:1, v/v/v) until the solvent front was 17 cm from the origin. A mixture of triolein (Sigma), POPC, and sodium oleate (Sigma) (0.4 mg/ml of each dissolved in chloroform/methanol (2:1, v/v)) was used as a standard. The NEFA and the other lipids were visualized with I 2 . The spots corresponding to phosphatidylcholine, triacylglycerol, and NEFA were cut from the sheets and placed directly into 10 ml of Ready Safe™ liquid scintillation mixture (Beckman Instruments). Radioactivity was determined using a Beckman LS 6000TA liquid scintillation counter with automatic quenching correction (Beckman Instruments). The silica gel had a negligible effect on the counting.

Determination of HL-mediated Hydrolysis in Unlabeled (A-I/CE)r-HDL and (A-II/CE)rHDL-
These incubations were carried out exactly as described above for the radiolabeled substrates. Phospholipid hydrolysis was determined directly by assaying the mass of NEFA released from the rHDL. At the end of the incubation period, the tubes containing the incubation mixtures were placed on ice prior to assaying for NEFA. The NEFA concentration was determined using an enzymatic colorimetric assay kit (Wako Pure Chemical Industries, Osaka, Japan).
Calculations-HL-mediated hydrolysis in the radiolabeled substrates was determined as the amount of radiolabel in the NEFA relative to the total radiolabel in the substrate. HL-mediated hydrolysis in the unlabeled (A-I/CE)rHDL and (A-II/CE)rHDL was determined by direct mass assay of the NEFA formed. The kinetic parameters K m (app) and V max were estimated from the line of best fit by linear regression analysis of a Lineweaver-Burk double-reciprocal plot of the rate of hydrolysis versus the concentration of substrate. In all cases, the regression coefficients (r) were Ͼ0.98. V max was determined as the reciprocal of the intercept on the y axis. The K m (app) was calculated as the product of the slope and V max .
Other Techniques-All chemical analyses were carried out on a Cobas Fara centrifugal analyzer (Roche Diagnostics, Zurich, Switzerland). Boehringer Mannheim kits were used for phospholipid, UC, and total cholesterol assays. CE concentrations were calculated as the difference between the total and UC concentrations. The concentrations of apoA-I and apoA-II were determined by an immunoturbidometric assay (26). The size of the rHDL and native HDL 2 was determined by electrophoresis on 3-35% nondenaturing polyacrylamide gradient gels (Gradipore, Sydney, Australia) (27).
Statistical Methods-The one-tailed, Student's t test for two samples with equal variance was used to determine whether differences between values were significant. Table I) Table I are expressed relative to the number of apolipoprotein molecules. The composition of the native HDL 2 is expressed as percentage of mass. The data in Table I  Kinetics of the HL-mediated Hydrolysis of Phospholipids and Triacylglycerol in rHDL-The aim of these studies was to determine how apolipoproteins influence the HL-mediated hydrolysis of phospholipids and triacylglycerol in rHDL. Preliminary experiments established that the hydrolysis of phospholipids and triacylglycerol in both (A-I)rHDL and (A-II)rHDL was linear up to 30% (results not shown). Consequently, the kinetic studies described below were all conducted under conditions that gave less than 30% phospholipid or triacylglycerol hydrolysis. Table II)-(A-I/CE)rHDL and (A-II/CE)rHDL labeled with [ 14 C]DPPC were used to monitor the kinetics of the HL-mediated hydrolysis of phospholipids in the absence of triacylglycerol (Fig. 2).

Kinetics of HL-mediated Phospholipid Hydrolysis in (A-I/ CE)rHDL and (A-II/CE)rHDL (Figs. 2 and 3 and
In this experiment, the substrate concentration was increased progressively in incubations that contained a constant amount of HL. The duration of the incubation was 3 h. Fig. 2A shows that the rate of phospholipid hydrolysis increased as the con-   To ensure that the hydrolysis of [ 14 C]DPPC in the radiolabeled rHDL reflected the hydrolysis of the rHDL bulk phospholipid, NEFA formation was measured directly by mass assay in unlabeled (A-I/CE)rHDL and (A-II/CE)rHDL (Fig. 3A). As the NEFA mass assay is much less sensitive than the measurement of radiolabel, the lowest concentration of substrate used in these experiments was 0.08 mM phospholipid (compared with 0.05 mM phospholipid for the radiolabeled substrate). These results were comparable with what was obtained using [ 14 C]DPPC-labeled (A-I/CE)rHDL and (A-II/CE)rHDL. When these results were transformed into a Lineweaver-Burk double reciprocal plot (Fig. 3B), it was apparent that, as with the radiolabeled substrate, HL has a greater affinity for the phospholipid in (A-II/CE)rHDL than in (A-I/CE)rHDL, with the V max being greater for (A-I/CE)rHDL than for (A-II/CE)rHDL. These results show that the hydrolysis of [ 14 C]DPPC, when present in trace amounts in rHDL, reflects the hydrolysis of the bulk phospholipid. 2 (Fig. 4)-To ensure that the above results are an accurate reflection of phospholipid hydrolysis in native HDL, an additional experiment was carried out with HDL 2 that had been isolated from human plasma and radiolabeled with [ 14 C]DPPC. ApoA-I constituted more than 90% of the apolipoproteins in this preparation (Table I). The HDL 2 also contained a minimal amount of triacylglycerol. Native HDL 2 , in which apoA-II comprised more than 90% of the apolipoproteins, was prepared as described under "Experimental Procedures." The relationship between HL-mediated phospho- mM, the rate of phospholipid hydrolysis was greater in (A-I)HDL 2 than in (A-II)HDL 2 (p Ͻ 0.01). Since transformation of this data into a Lineweaver-Burk double reciprocal plot did not yield a straight line, the kinetic parameters were not determined. The nonlinearity of the transformed data highlights the problems associated with using substrates that are heterogeneous in size and composition. (Fig. 5, Table II (Table II). The HL had a greater affinity for the phospholipids in (A-II/TG)rHDL (K m (app) ϭ 0.4 mM) than in (A-I/TG)rHDL (K m (app) ϭ 0.9 mM). The V max /K m (app) values for the two substrates were comparable.

TABLE II
Kinetic parameters of HL-mediated phospholipid and triacylglycerol hydrolysis in rHDL A range of concentrations of spherical rHDL were incubated with a constant amount of HL as described in the legends to Figs. 2-6. For the radiolabeled substrates, the resulting NEFA were separated from the other rHDL lipids by thin layer chromatography as described under "Experimental Procedures." Kinetic parameters were estimated from double-reciprocal plots of the rate of hydrolysis versus the concentration of rHDL phospholipid or triacylglycerol. All values are means of triplicate determinations. Since different preparations of HL were used for each study, the V max values cannot be compared for the different experiments.   (Table II). As with the phospholipid hydrolysis, HL had a greater affinity for the triacylglycerol in (A-II/ TG)rHDL (K m (app) ϭ 0.1 mM) than in (A-I/TG)rHDL (K m (app) ϭ 1.0 mM). The V max /K m (app) for triacylglycerol hydrolysis in (A-II/TG)rHDL was approximately double that for (A-I/TG)rHDL.

DISCUSSION
In the present study, we have used well characterized, homogeneous, apolipoprotein-specific preparations of spherical rHDL and preparations of native HDL 2 to investigate the influence of apoA-I and apoA-II on the HL-mediated hydrolysis of HDL phospholipids and triacylglycerol. The (A-I)rHDL and (A-II)rHDL used in this study were comparable in size and lipid composition and differed only in their apolipoprotein con-tent. The results show unequivocally that, although HL has a higher affinity for the phospholipids and triacylglycerol in (A-II)rHDL than in (A-I)rHDL, the maximal rate of hydrolysis for both constituents is greater in (A-I)rHDL than in (A-II)rHDL.
The HDL in human plasma consist of two major apolipoprotein-specific subpopulations of particles: those containing apoA-I without apoA-II, (A-I)HDL, and those with both apoA-I and apoA-II, (A-I/A-II)HDL (28). There is also a minor HDL subpopulation that contains apoA-II without apoA-I, (A-II)HDL (29). The present studies conducted in vitro with rHDL and native HDL 2  There are several reasons why apolipoproteins may influence the interaction of HL with HDL. For instance, the different affinity of HL for (A-I)rHDL relative to (A-II)rHDL may reflect differences in particle charge. Rye and Barter (23) have reported that the surface of (A-II)rHDL is less negatively charged than that of (A-I)rHDL. Since HL has a net negative charge (30), it is likely to have a greater affinity for the surface of (A-II)rHDL compared with (A-I)rHDL. Consistent with this, Laboda et al. (30) showed that the HL-mediated hydrolysis of triolein is reduced when a negative charge is incorporated into a lipid monolayer. Another explanation for the different affinities of HL for (A-I)rHDL and (A-II)rHDL relates to lipid-water interfacial hydration. It has been shown in earlier work from this laboratory that the lipid-water interface of (A-I)rHDL is more hydrated than that of (A-II)rHDL (23). Since the catalytic sites of most lipases are hydrophobic (31,32), it is conceivable that such enzymes would associate preferentially with the less hydrated lipid-water interface of (A-II)rHDL.
The present data show that when the substrate concentration is not rate-limiting, HL hydrolyzes both phospholipids and triacylglycerol more rapidly in (A-I)rHDL than in (A-II)rHDL. This is consistent with the phospholipid and triacylglycerol acyl chains in (A-I)rHDL being more accessible to the active site of HL than those in (A-II)rHDL. One explanation for this observation relates to the less ordered phospholipid head group packing in (A-I)rHDL compared with (A-II)rHDL (16). Such a difference may lead to significant structural changes in the microenvironment of the rHDL interface, which could influence access of the enzyme to its substrates in such a way as to enhance the hydrolysis of both phospholipids and triacylglycerol in (A-I)rHDL (33).
Tansey et al. reported recently on the HL-mediated hydrolysis of phospholipids in discoidal and spherical rHDL that had been prepared with a range of phospholipids, including DPPC (34). In that study, minimal phospholipid hydrolysis was observed in either the discoidal or spherical rHDL that contained DPPC. This is an interesting finding given the results of the present study, where [ 14 C]DPPC was used as a tracer to monitor POPC hydrolysis. It should be noted that other investigators have also used [ 14 C]DPPC successfully as a tracer to monitor phospholipid hydrolysis (35,36). These discrepancies suggest that HL is sensitive to the phase state of substrate lipids. Indeed, Thuren et al. have found this to be the case in their monolayer studies (37). Thus, although DPPC is a poor substrate for HL when it is present as the bulk lipid in an interface, it is hydrolyzed readily by HL when it is present in trace amounts in a more physiological substrate. The current results show clearly that this is the case, since comparable results were obtained when phospholipid hydrolysis was determined using rHDL containing a trace amount of [ 14 C]DPPC (Fig. 2) as well as in experiments where NEFA formation was measured directly with a mass assay (Fig. 3).
There are several conflicting reports as to the effects of apolipoproteins on the HL-mediated hydrolysis of phospholipids and triacylglycerol in HDL. On the one hand, it has been suggested that apoA-II inhibits (9, 10), while others have found that it enhances, HL-mediated hydrolysis of HDL triacylglycerol (7,8,38,39). Mowri et al. (38) reported that hydrolysis of phospholipids and triacylglycerol in HDL is greater in (A-I/A-II)HDL 2 compared with (A-I)HDL 2 . However, those observations cannot be compared directly with what was observed in the present study, since the HDL 2 used by Mowri et al. (38) were heterogeneous in size and composition. They also contained apolipoproteins other than apoA-I and apoA-II that may have influenced the hydrolysis (40). Finally, Mowri et al. (38) measured total fatty acid liberation (i.e. phospholipid and triacylglycerol hydrolysis combined) rather than the hydrolysis of individual lipids.
The results of Zhong et al. (9) are, to a large extent, consistent with the present findings. Those investigators found that the HL-mediated hydrolysis of triacylglycerol in a lipid emulsion was inhibited by the addition of HDL from mice that had been made transgenic for either human apoA-I or human apoA-II or for both human apoA-I and apoA-II. Since the HDL from the apoA-II and apoA-I/apoA-II transgenic mice mediated a greater reduction in the rate of HL-mediated triacylglycerol hydrolysis than the HDL from the apoA-I transgenic mice, it was concluded that apoA-II inhibits HL activity more than apoA-I. Given the findings in the present study, which show that HL has a greater affinity for the less reactive apoA-IIcontaining rHDL, it would be predicted that (A-II)HDL would be more effective than (A-I)HDL as inhibitors of triacylglycerol hydrolysis in lipid emulsions.
In another study, Jahn et al. (39) found that reconstituted particles containing apoA-II and HDL lipids enhanced the HLmediated hydrolysis of triacylglycerol in microemulsions. These investigators obtained similar results when the reconstituted particles were substituted with ultracentrifugally isolated HDL. Jahn et al. (39) explained their observations in terms of apoA-II partitioning from the reconstituted particles (or isolated HDL) to the triacylglycerol emulsion and thus increasing the affinity of HL for the triacylglycerol emulsion. However, these investigators also found that the triacylglycerol hydrolysis in the microemulsions decreased at high concentrations of reconstituted particles (or HDL). This may have been caused by the high concentrations of reconstituted particles or HDL competing with the triacylglycerol emulsion for HL.
The results of the present study may explain some of the conflicting conclusions that have been drawn about the relative effects of apoA-I and apoA-II on HL-mediated phospholipid and triacylglycerol hydrolysis (9,38). The lower K m (app) values for both the phospholipids and triacylglycerol in (A-II)rHDL relative to (A-I)rHDL suggest that HL has a greater affinity for HDL that contain apoA-II than for HDL that contain apoA-I. In other words, at a low concentration of substrate, the amount of HL interacting with HDL (and hydrolyzing HDL lipids) may be much greater in HDL that contain apoA-II than in (A-I)HDL. Therefore, when a study is conducted at a low substrate concentration, it may be concluded that apoA-II-containing HDL are superior to (A-I)HDL as substrates for HL. By contrast, if experiments are conducted at high substrate concentrations, under conditions where substrate availability is no longer a limiting factor, and the V max for phospholipid and triacylglycerol hydrolysis is greater in (A-I)rHDL than in (A-II)rHDL, it may be concluded that (A-I)HDL are superior substrates compared with HDL that contain apoA-II.
In conclusion, these studies provide the first description of the kinetics of the HL-mediated hydrolysis of phospholipids and triacylglycerol in spherical, apolipoprotein-specific rHDL. In addition, they show that apolipoproteins have a major impact on these processes. It remains to be determined whether the interaction of (A-I/A-II)HDL with HL resembles that of either the (A-I)rHDL or (A-II)rHDL or whether it is distinct from each.