Reciprocal Effect of Apolipoprotein C-I1 on the Lipoprotein Lipase-catalyzed Hydrolysis of p-Nitrophenyl Butyrate and Trioleoylglycerol*

Interaction of purified bovine milk lipoprotein lipase (LpL) with sonicated vesicles of dipalmitoyl phospha- tidylcholine in the gel phase is associated with an increase in the rate of the LpL-catalyzed hydrolysis offp- nitrophenyl butyrate. There is a 6-fold increase in Vmm. Apolipoprotein C-II, the activator protein for LpL, in- hibits the LpL-catalyzed hydrolysis of p-nitrophenyl butyrate. With 0.5 mol 5% tri['4C]~leoylglycerol present in the dipalmitoyl phosphatidylcholine vesicles and in the presence of 20 m~ Ca2+, the rate of p-nitrophenyl butyrate hydrolysis is decreased reciprocally compared to trioleoylglycerol hydrolysis and is dependent on apolipoprotein C-II. These results suggest that apolipoprotein C-I1 enhances the activity of LpL by increas- ing the affinity of the active site of LpL for triacylglycerol. Lipoprotein lipase (EC 3.1.1.34) catalyzes the hydrolysis of plasma lipoprotein triacylglycerol, phosphatidylcholine, and

catalyze the hydrolysis of a long chain triacylglycerol in the presence of apoC-I1 or the water-soluble substrate which does not require the activator protein.

MATERIALS AND METHODS
DPPC was purchased from Applied Science Laboratories. PNPB and heparin (porcine intestinal mucosa, 169.9 units/mg) were purchased from Sigma. Tri['4C]oleoglycero1 (50 mCi/mmol) was obtained from New England Nuclear.
LpL was purified to homogeneity from bovine skimmed milk by chromatography on heparin-Sepharose as described by Kinnunen (9). The purified enzyme had a specific activity of 30 mmol of released oleic acid/mg of protein/h using tri['4C]oleoylglycerol emulsified with Triton X-loc) (10). ApoC-11 was isolated from triglyceride-rich lipoproteins by gel filtration of the delipidated proteins on Sephadex G-75 followed by ion exchange chromatography on DEAE-Sephacel at 4 "C in 6 M urea (11). To prepare DPPC vesicles, phospholipid (10 mg) was dissolved in chloroform, evaporated under a stream of ultrapure N,, and lyophilized for 15 min. Phospholipid dispersions were prepared by adding 5 ml of 0.9% NaCI, 0.01 M Tris-HC1, pH 7.2. The lipid dispersion was sonicated at 42 "C for 15 min using a cell disrupter (Heat Systems Ultrasonics, Inc. model W-225R). The sonicated vesicles were subjected to ultracentrifugation at 15 "C in a Beckman type 50 rotor for 1 h at 50,000 rpm. After ultracentrifugation, the top 1.0 ml was removed by aspiration and discarded. The middle 4.0 ml was then removed and used in the experiments described below. DPPC vesicles containing 0.5 mol % tri['4C]oleoylglycerol were prepared exactly as described above; lipid vesicles were prepared daily and used immediately. Phospholipid phosphorus was determined by the method of Bartlett (12). The LpL-catalyzed hydrolysis of PNPB was determined by monitoring the increase in absorbance at 400 nm or by extraction of p-nitrophenol as described previously (8). ['4C]Oleic acid was extracted from the reaction mixture by the method by Belfrage and Vaughan (13). The hydrolysis of PNPB and tri['4C]oleoylglycerol was linear for the indicated incubation times.

RESULTS
The time course of the LpL-catalyzed hydrolysis of PNPB is shown in Fig. 1. The addition of DPPC vesicles to the incubation was associated with an immediate 4-fold increase in the rate of PNPB hydrolysis by LpL and is consistent with previous findings (8). The addition of apoC-I1 (Fig. 1, curue  B ) to give an apoC-I1 concentration of 4 p g / d caused an 80% inhibition in the rate of hydrolysis.
To determine the effects of apoC-11 on the kinetic parameters for the LpL-catalyzed hydrolysis of PNPB, rates were determined at various concentrations of PNPB. In these experiments, the incubation mixtures contained a constant amount of DPPC (50 p g ) and LpL (5 p g ) and variable amounts, as indicated, of PNPB. As shown in Fig. 2, apo C-I1 caused an increase in the apparent K,,, with no effect on the maximal velocity (VmaX); the calculated V,,, was 8.33 pmol of product released/min/mg of LpL. In the absence of apoC-11, the apparent K,,, was 0.58 mM. With 0.2, 1.0, or 5.0 p g / d of apoC-11, the apparent K , values were 0.90, 1.37, or 3.17 m, respec-tively. Thus, apoC-I1 caused a 6.4-fold increase in the apparent K,.
In the next experiment, 0.5 mol 56 tri['4C]oleoylglycerol was incorporated into the DPPC vesicles and the effect of apoC-11 on the rate of hydrolysis of PNPB and tri['4C]oleoylglycerol was determined. To complex the released oleic acid, 20 mM CaC12 was included in the incubation mixture as described by Bengtsson and Olivecrona (14). As shown in Fig. 3, apoC-I1 had a reciprocal effect on the rate of hydrolysis of PNPB and tri['4C]oleoylglycerol. With increasing concentrations of apoC-11, the rate of hydrolysis of PNPB decreased and there was a corresponding increase in the rate of triacylglycerol hydrolysis. After incubation for 20 min at 30 "C, 3.25 ml of methanol: chloroform:heptane (1.00.30.7, v/v) and 1.0 ml of potassium borate, pH 10.5, were added to each incubation mixture. The mixtures were shaken vigorously for 10 s and centrifuged at 1500 X g for 5 min. After warming at 42 "C for 3 min, the supernatant fractions were removed. One-ml of the supernatant fraction was taken to dryness in a scintillation vial with a stream of nitrogen and oleic acid radioactivity was determined. p-Nitrophenol was determined by absorbance at 400 nm. The molar extinction coefficient of released p-nitrophenol in the extracted upper phase was 12,000, with a recovery of 95%.The results are expressed as the mean % S.E.  Fig. 3 for the LpL-catalyzed hydrolysis of PNPB versus apoC-I1 concentration, as described in Equation 1 of the text. The displayed linear fit was generated by linear least squares analysis. primary data.
Inset, replot of the slopesof the Lineweaver-Burk plots of Fig. 21

DISCUSSION
As shown previously (8), interfacial actidation of LpL by DPPC in the gel phase is associated with an increase in the V,,, of PNPB hydrolysis, with no change in K,. The present study shows that apoC-I1 inhibits the DPPC-stimulated LpLcatalyzed hydrolysis of PNPB. Furthermore,when the DPPC vesicles contain 0.5 mol % trioleoylglycerol, the addition of apoC-I1 results in a Corresponding enhancement of the LpLcatalyzed hydrolysis of trioleoylglycerol.
The kinetics pattern for DPPC-stimulated LpL-catalyzed hydrolysis of PNPB at various apoC-I1 concentrations is consistent with competitive inhibition by apoC-I1 (Fig. 2). The simpleFt mechanism for the observed inhibition by apoC-I1 is: where K,,, = ( k 1 + KCat)/kl and V,,, = Kcat Et(EI = total LpL concentration). According to this equation, the slopes of the Lineweaver-Burk plots for LpL-catalyzed hydrolysis of PNPB depend on the concentration of apoC-11, and therefore the plot constructed from the slopes and shown in the inset to Fig. 4 allows one to calculate Kcrr, the dissociation constant of the LpL:apoC-I1 complex formed at the vesicle surface. This analysis gives KcIl = 0.13 +-0.01 PM. Furthermore, the linearity of the plot supports the two mechanistic assignments made earlier (17). (a) The stoichiometry of the LpL:apoC-I1 complex is 1:l. If it were 1:2 (LpLapoC-II), for example, the velocity equation would contain a [CIII2 term and the plot would be nonlinear and quadratic in apoC-I1 concentration. (b) The LpL:apoC-I1 complex formed at the vesicle surface is catalytically inactive toward PNPB. If this complex retains PNPB hydrolysis activity, though diminished, the plot would be nonlinear and hyperbolic in apoC-I1 concentration.
Equation 1 shows that, when initial velocities are determined at a single PNPB concentration but varying apoC-I1 concentration, VL" is a linear function of apoC-II concentration. Fig. 4 shows such a plot, a Dixon plot (17), constructed from the data of Fig. 3 for the apoC-I1 inhibidon of LpLcatalyzed hydrolysis of PNPB. The linearity of the Dixon plot in Fig. 4 lends further support to the mechanism of Scheme 1. In addition, Fig. 3 shows that the apoC-I1 inhibition of the LpL-catalyzed hydrolysis of PNPB in the presence of DPPC vesicles containing 0.5 mol '% trioleoylglycerol is accompanied by a corresponding increase in the LpL-catalyzed hydrolysis of trioleoylglycerol. This reciprocal effect of apoC-I1 on LpLcatalyzed PNPB and trioleoylglycerol hydrolysis strongly suggests that the same LpL-apoC-I1 interaction is responsible for apoC-I1 inhibition of PNPB hydrolysis and activation of trioleoylglycerol hycirolysis. Furthermore, the data displayed in Figs. 1 and 2 demonstrate that triacylglycerol need not be present for apoC-I1 inhibition of DPPC-stimulated LpL-catalyzed PNPB hydrolysis to be expressed. It thus appears that apoC-I1 inhibits DPPC-stimulated LpL-catalyzed hydrolysis of PNPB by forming a 1:l complex with the mzyme in which the conformation of the active site of LpL is changed. In this altered conformation, LpL is no longer a catalyst for PNPB hydrolysis. One way that apoC-I1 inhibition of LpL-catalyzed hydrolysis of PNPB may be expressed is that in its altered conformation the LpL active site has a much higher affinity for DPPC monomers that are contained in the vesicle surface. This suggestion is supported by the fact that DPPC is a substrate, though poor, of LPL.~ However, apoC-I1 inhibits the LpL-catalyzed hydrolysis of water-soluble p-nitrophenyl esters even in the absence of phospholipid.' This inhibition also appears to involve 1:l LpL:apoC-I1 complexes, with dissociation constants in the range of 0.26 to 0.83 p~. An alternate possibility, that apoC-I1 sterically blocks access of PNPB to the active site, is precluded by the fact that apoC-I1 simultaneously inhibits LpL-catalyzed PNPB hydrolysis and stimulates LpL-catalyzed trioleoylglycerol hydroly~is.~ It is therefore reasonable to suggest that the LpL conformational change that causes inhibition of PNPB hydrolysis when the enzyme interacts with apoC-I1 is the molecular dynamic event responsible for apoC-I1 stimulation of trioleoylglycerol hydrolysis.
The analysis of kinetic data presented in this paper suggests that apoC-IT inhibits the DPPC-stimulated LpL-catalyzed hydrolysis of PNPB (and that apoC-I1 reciprocally stimulates the LpL-catalyzed hydrolysis of trioleoylglycerol contained in DPPC vesicles) through the formation of 1:l LpL.apoC-I1 complexes. This stoichiometry has previously been suggested by Chung and Scanu (18) for apoC-I1 stimulation of the rat heart LpL-catalyzed hydrolysis of trioctanoylglycerol monolayers, and by Fielding and Fielding (19) for apoC-I1 enhancement of the rat postheparin plasma LpL-catalyzed hydrolysis of trioleoylglycerol emulsified with dioleoylphosphatidylcholine. Moreover, the dissociation constant determined herein of the LpL:apoC-11 complex of 0.13 pM is in reasonable agreement with the corresponding dissociation constants estimated by Bengtsson and Olivecrona (20) for the bovine milk LpLcatalyzed hydrolysis of emulsified trioleoylglycerol (0.038 p~) and monooleoylglycerol (0.34 p~) , and with the dissociation constants determined by Smith et al. (21) for interaction in the absence of substrate of bovine milk LpL and dansylated apoC-I1 fragments, which are in the range of 0.22 to 4.0 PM.
One possible explanation for these results is that in the presence of vesicles of DPPC LpL may preferentially hydrolyze PNPB that is bound to the vesicles, and that the ratedetermining step may be the binding of PNPB to the vesicles. If such were true, apoC-I1 could inhibit the DPPC-stimulated LpL-catalyzed hydrolysis of PNPB by preventing PNPB binding to the vesicles. This explanation can be discounted for a number of reasons. (a) Shirai and Jackson (8) showed that stimulation of the LpL-catalyzed hydrolysis of PNPB by vesicles of DPPC was greater than that by vesicles of dimyristoyl phosphatidylcholine at 30 "C, although dimyristoyl phosphatidylcholine vesicles bound more PNPB than did DPPC vesicles. ( b ) Shirai and Jackson (8) found that DPPC vesicles stimulate the LpL-catalyzed hydrolysis of PNPB by increasing V, , , without affecting X,,,. Since Vmax is the maximum velocity of substrate turnover from the LpL-PNPB Michaelis complex, the putative binding step of PNPB to the DPPC vesicles must OCCUT before the rate-determining step measured by V,,,. (c) ApoC-I1 inhibits the DPPC-stimulated LpL-catalyzed hydrolysis of PNPB competitively; i.e. V,,, is unaffected but K,,, is increased. Two ways that apoC-I1 might increase K , that are consistent with preventing PNPB binding to DPPC vesicles are by directly binding to PNPB or by binding to the DPPC vesicle surface so that the on-step of PNPB to the vesicles is slowed. The analysis of apoC-I1 inhibition of the DPPC-stimulated LpL-catalyzed hydrolysis of PNPB determined in the present report yielded Kc11 = 0.13 p, which we assign to the dissociation constant of the LpLapoC-I1 complex. However, it is ( i n the more general sense) the concentration of apoC-I1 required for half-maximal inhibition when [PNPB] < K,. For example, when [PNPB] = 0.13 rn and [apoC-111 = KcII = 0.13 p~, the PNPB/apoC-I1 ratio is 10001, so that for 50% inhibition to occur each apoC-I1 molecule must bind 500 PNPB molecules, or -7 molecules of PNPB/amino acid. If such were the case, it is difficult to imagine that the resulting complex of apoC-I1 and PNPB could simultaneously stimulate trioleoylglycerol hydrolysis. Hence, direct binding of apoC-I1 and PNPB does not appear to explain the apoC-I1 inhibition of DPPC-stimulated LpL-catalyzed hydrolysis of PNPB. Moreover, Cardin et al. (22) determined a dissociation constant of 6.5 ( l~ for the interaction of apoC-I1 and DPPC vesicles. Since KcII = 0.13 PM is -50-fold smaller than the dissociation constant measured by Cardin et al. (22), it must reflect some molecular event other than interaction of apoC-I1 and the DPPC vesicle surface. We suggest that Kc11 reflects the LpL:apoC-I1 interaction. It thus seems that apoC-I1 binding to the vesicle surface such that PNPB binding is slowed cannot be the mechanism for apoC-11 inhibition of DPPC-stimulated LpLcatalyzed hydrolysis of PNPB. This argument is supported by the numerical similarity of Kc11 and the inhibition constants of 0.26-0.83 pM determined3 for apoC-I1 inhibition of the LpLcatalyzed hydrolysis of PNPB in the absence of lipids.