A kinetic mechanism for modulation of the activity of microsomal UDP-glucuronyltransferase by phospholipids. Effects of lysophosphatidylcholines.

The affinity of delipidated microsomal UDP-glucuronyltransferase (EC 2.3.1.17) for UDP is greater than that for UDP-glucuronic acid. Measurement of KIglucuronic acid reveals that glucuronic acid binds to the enzyme. Hence, the difference in affinity of the enzyme for UDP versus UDP-glucuronic acid indicates that inherent binding energy for interactions between enzyme and this substrate is used for purposes other than enhancing binding. A reasonable interpretation of these data is that the binding of UDP-glucuronic acid to enzyme requires distortion of the substrate and/or the enzyme. Inherent binding energy due to interactions between enzyme and UDP and glucuronic acid is utilized to effect such distortions. This type of mechanism can cause significant rate enhancement. Phospholipid activators of UDP-glucuronyltransferase activate by amplifying this basic mechanism. Thus, addition of various species of lysophosphatidylcholine to the delipidated enzyme increase the activity at Vmax and enhance the affinity for UDP, glucuronic acid, and UDP-glucuronic acid. However, activators enhance the affinity of the enzyme for UDP-glucuronic acid to a significantly smaller extent than they enhance affinity for the UDP and glucuronic acid portions of the substrate. Calculations of the amount of binding energy for interactions between enzyme and UDP-glucuronic acid that can be used for stimulating activities at Vmax yield values in agreement with the observed enhancement of activities at Vmax for enzyme reconstituted with various types of lysophosphatidylcholine.

The affinity of delipidated microsomal UDP-glucuronyltransferase (EC 2.3.1.17) for UDP is greater than that for UDP-glucuronic acid. Measurement of gIglucunacid reveals that glucuronic acid binds to the enzyme. Hence, the difference in affinity of the enzyme for UDP versus UDP-glucuronic acid indicates that inherent binding energy for interactions between enzyme and this substrate is used for purposes other than enhancing binding. A reasonable interpretation of these data is that the binding of UDP-glucuronic acid to enzyme requires distortion of the substrate and/or the enzyme. Inherent binding energy due to interactions between enzyme and UDP and glucuronic acid is utilized to effect such distortions. This type of mechanism can cause significant rate enhancement. Phospholipid activators of UDP-glucuronyltransferase activate by amplifying this basic mechanism. Thus, addition ofvarious species of lysophosphatidylcholine to the delipidated enzyme increase the activity at Vm,, and enhance the affinity for UDP, glucuronic acid, and UDP-glucuronic acid. However, activators enhance the affinity of the enzyme for UDP-glucuronic acid to a significantly smaller extent than they enhance affinity for the UDP and glucuronic acid portions of the substrate. Calculations of the amount of binding energy for interactions between enzyme and UDP-glucuronic acid that can be used for stimulating activities at Vm., yield values in agreement with the observed enhancement of activities at Vm,, for enzyme reconstituted with various types of lysophosphatidylcholine.
UDP-glucuronyltransferase (EC 2.4.1.17) is an integral component of the microsomal membrane. We have proposed previously that the kinetic properties of this enzyme are regulated via interactions between it and the phospholipids of the membrane (1)(2)(3). The experimental basis for this idea is that altering the composition and/or structure of the lipid portion of the microsomal membrane changes the kinetic properties of UDP-glucuronyltransferase (1-7). Moreover, delipidation of a partially purified preparation of enzyme leads to a marked diminution of activity, which can be restored by adding selected types of phospholipids (8,9). Reconstitution of the delipidated form of the enzyme appears to require a phospholipid containing a phosphorylcholine moiety (8). In addition, the specific activity of reconstituted enzyme is influenced by the length and unsaturation of the acyl groups of added phospholipids. UDP-glucuronyltransferase thus is one of several membrane-bound enzymes regulated by lipid-protein interactions. The problem in defining further how lipids affect the properties of UDP-glucuronyltransferase is 2-fold. It is essential to understand how lipid-protein interactions effect the observed kinetic changes and to determine the chemistry of the interactions between each of these components. We report, in this paper, observations on the kinetic mechanism for activation of delipidated UDP-glucuronyltransferase by a series of lysophosphatidylcholines. The data also reveal how the chain length of the acyl groups of the lysophosphatidylcholine determines the activity at Vmax of reconstituted enzyme.

MATERIALS AND METHODS
Phospholipase A2, from Naja naja venom, was purified by the method of Cremona and Kearney (10). Egg lysolecithin was prepared by treating purified lecithin from eggs (11) with phospholipase A2 (12,13). Lysolecithin used in the reconstitution experiments was purchased from Applied Science Laboratories, State College, PA, or P-L Biochemicals, Milwaukee, WI. They were used without further purification. Phospholipid phosphorus was determined after digestion of CHC13/CH30H extracts of different forms of UDP-glucuroyltransferase (8). Protein was measured by the biuret method (14) except when glycerol was present. In this instance protein was determined with the method of Lowry et al. after precipitation of the protein with trichloracetic acid (15).
Egg lysolecithin was added to the DEAE-Cl-6B dialyzate to a fmal concentration of 2 mg/rnl. The mixture was sonicated for 200 s, as described above, on ice. The solution was applied to a DE52 gel (300-400 ml prewashed with 10 mM Tris, pH 8.0 (0 'C), in a Buchner funnel. The gel was washed with 10 mM Tris, pH 8.0 (0 'C), and fractions of 100 ml were collected. The fractions with high enzyme activity were pooled (-700ml). The DE52 eluate was sonicated, as previously described, for 80 s on ice. The solution was poured through 60 ml of CM-C25 gel (prewashed with 20 mM PIPES,' pH 7.0 (0 'C), in a scintered glass funnel). The CM-C25 gel was washed with 100 ml of 40 mm NaCl in 20 mM PIPES, pH 7.0 (0 'C), and the enzyme was eluted with 100 ml of 250 mM NaCl in 20 mM PIPES, pH 7.0 (0 'C). Fractions of 20 ml were collected. The fractions with high activity (60-80 ml) were pooled and dialyzed against 35 volumes of 20 mM Tris, pH 8.0 (0 'C), overnight. The enzyme solution was frozen at -80 'C and could be stored for several months without appreciable loss of activity.
Fifty ml of the CM-C25 dialysate were concentrated 10-fold on an Amicon XM-50 membrane (total protein, 8-10 mg). Tris (pH 8, 0 'C), dithioerythritol, EDTA, sodium cholate, and glycerol were added to the concentrated dialysate after chromatography on CM-C25. Final concentrations were 50 mM Tris, 0.1 mM dithioerythritol, 2 mM EDTA, 0.5% (w/v) cholate, and 10% (v/v) glycerol. The enzyme was applied to a hydroxylapatite column, prewashed with 50 ml of the same buffer. The enzyme was eluted from the column with a 0-0.5 M K2HPO3 gradient (100 x 100 ml) in the same buffer. Fractions of 3.5 ml were collected. The fractions with maximal activity were pooled and dialyzed twice against 30 volumes of 0.1 mM dithioerythritol, 2 Based on activities at Vm., and after correction for detergentinduced activations, the purification procedure described above yields an enzyme preparation that is enriched 140-fold as compared with microsomes. The yield of microsomal protein in the final preparation is about 0.1%. Enzyme eluted from the hydroxylapatite column contained no detectable phospholipids.
Enzyme Assays-All enzyme activities were determined at 30 'C, in 50 mM Tris, pH 7.5, by following the change in optical density ofpnitrophenol at 400 nm (8). For determination of kinetic constants, activities were measured at eight different concentrations of UDPglucuronic acid and six different fixed concentrations ofp-nitrophenol. Best straight lines were fitted to the data and to the secondary plots, according to the scheme for kinetic analysis of a bisubstrate enzyme with a random, rapid equilibrium kinetic mechanism (17). Initially, all experiments were carried out in duplicate. We found, however, that replicate determinations of kinetic constants gave values differing by no more than approximately 15%. Analysis of inhibition studies and calculation of values for K, were carried out according to the kinetic scheme outlined in Segal (18). Initial rates of formation of pnitrophenylglucuronide were difficult to measure with highly active enzyme because of product inhibition by UDP. Addition of 1 mM Mg2+ to assays prevented this product inhibition. Careful estimates of initial rates of activity at several different concentrations of substrates in the absence and presence of Mg2`indicated, however, that Mg2`had no effect on the rate of glucuronidation. This is true because the tendency of UDP for formation of chelates with divalent metal ions is much greater than for chelates between UDP-glucuronic acid and metal ions (19). Hence, 1 mM Mg2`was added to assays except in studies with UDP added as inhibitor. When glucuronic acid was added to assays, the total concentration of Tris plus glucuronic acid, preadjusted to pH 7.5, was maintained at 50 mm. This was done because evidence from separate experiments suggested that salt concentrations above 100 mm inhibited the activity of reconstituted enzyme.
Reconstitution-Lysophosphatides were added in 10-fold excess (on a weight basis) to delipidated enzyme. This amount yielded maximal activities. Activation was instantaneous on the time scale of the experiments. A mixture of lysophosphatides and enzyme was stable at 0 'C for at least 8 h. pH optima for activities at Vm,,, and 'The abbreviations used are: PIPES, 1,4-piperazinediethanesulfonic acid; UDP-GA, UDP-glucuronic acid; pNP, p-nitrophenol. KUDPGA were determined, and were close to 7.5 for all forms of UDPglucuronyltransferase studied.

RESULTS AND DISCUSSION
Effects of Lysophosphatidylcholine on the Activity at Vm,, of Delipidated UDP-glucuronyltransferase-The partially purified, delipidated preparation of UDP-glucuronyltransferase has measurable activity. The presence of residual activity in the delipidated preparation of enzyme stimulated careful reanalysis for contamination with small amounts of lipid phosphorus. None was found. The activity of the delipidated preparations also could not be attributed to small amounts of cholate because complete removal of trace amounts of cholate did not reduce activity. We believe, therefore, that the low residual activity of delipidated UDP-glucuronyltransferase in our preparations reflects self-association between molecules of this enzyme as well as between it and other integral components of the microsomal membranes. These associations must satisfy the requirements of the enzyme for interaction with amphipathic materials, which accounts for the complete dispersal in water of the partially purified preparation.
Addition of lysophosphatidylcholine enhances the activity at Vm,, of delipidated UDP-glucuronyltransferase (Table I).
The importance of chain length and unsaturation in stimulating catalysis is evident in the 25-fold variation between enzyme reconstituted with stearoyl lysophosphatidylcholine as compared with oleoyl lysophosphatidylcholine. Effects ofLysophosphatidylcholine on Kinetically Derived Binding Constants for Interactions of UDP-glucuronyltransferase with Its Substrates-The kinetic mechanism of UDPglucuronyltransferase is rapid equilibrium, random order (20). Thus, there are two binding constants for each substrate, one for binding of substrate to free enzyme and the other for binding to enzyme already saturated with the other substrate. The constants KUDP-GA and KPNP are for the binding of each substrate to free enzyme. The constants K'UDP-GA and K'PNP are for the binding of each substrate to enzyme already saturated with the other substrate. All the lysophosphatidylcholines tested enhance the affinity of free, delipidated UDPglucuronyltransferase for UDP-glucuronic acid and p-nitrophenol (Table II). The extent to which this occurs appears to be independent of the nature of the acyl chain of the phospholipid, except for the somewhat lower KUDP-GA for enzyme treated with stearoyl lysolecithin. Comparison of the constants KUDP-GA versus K'UDP-GA and KPNP versus K'pNP indicates that there are significant substrate-induced effects on the properties of the enzyme. Prior binding of UDP-glucuronic acid alters the affinity of the E.S complex, enzyme. UDPglucuronic acid, forp-nitrophenol as compared with the affinity of free enzyme for p -nitrophenol. The reverse also occurs. The presence and type of phospholipid appears to determine whether binding of the first substrate enhances or diminishes affinity of an E. S complex for the binding of the second  Effect of selected lysophosphatidylcholines on the binding constants ofpartially purified, delipidated UDPglucuronyltransferase Enzyme was prepared and assayed at 30 'C as under "Materials and Methods." Binding constants were determined from kinetic data for a random order, rapid equilibrium kinetic mechanism, according to Cleland (17). Units are millimolar. The terms KUDP-GA and KPNP are for binding of substrate to enzyme in the absence of the second substrate. The terms K'UDP-GA and K'PNP are for binding of substrate to enzyme already saturated with the alternate substrate. substrate. For example, prior binding of UDP-glucuronic aicd to delipidated UDP-glucuronyltransferase enhances subsequent binding ofp-nitrophenol as compared with the binding of p-nitrophenol to the enzyme in the absence of UDP-glucuronic acid. By contrast, prior binding of UDP-glucuronic acid to enzyme reconstituted with myristoyl, palmitoyl, and oleoyl lysophosphatidylcholines diminishes the affinity of the enzyme for p-nitrophenol. Since the binding of UDP-glucuronic acid alters the active site of UDP-glucuronyltransferase, as reflected by differences between KPNP and K'PNP, the inherent binding energies for the interaction E + UDP-glucuronic acid E. UDP-glucuronic acid must be greater than the observed binding energies calculated from the values of KUDP-GA in Table II by using the equation AG = RT In KS. This must be so because the energy required to produce the observed substrate-induced effects on the properties of UDPglucuronyltransferase (the difference between KPNP AND K'PNP) comes at the expense of the inherent binding energy of the enzyme-substrate interactions. The observed binding energy of UDP-glucuronic acid will be reduced as compared with its inherent binding energy by an amount equal to the energy used to alter the E.S complex. The data in Table II indicate that the amount of inherent binding energy of UDPglucuronic acid utilized to alter the affinity of the E. S complex for p-nitrophenol (KPNP versus K'PNp) depends on the species of lysophospatidylcholine added to the enzyme. The nature of the acyl chain of the lysophosphatidylcholine added to UDPglucuronyltransferase thus has a significant effect on the inherent binding energies of substrates, but this is not obvious from the values of KUDP-GA and KPNP for different forms of the enzyme.
Binding energy derived from interaction of enzyme and substrates can facilitate the catalytic steps of an enzymecatalyzed reaction, if the energy is used to lower the activation energy of the transition state (18). The differences between KPNP and K'PNP in Table II indicate that this is the case for enzyme activated with myristoyl, palmitoyl, and oleoyl lysophosphatidylcholines. In contrast, binding energy due to interaction between delipidated UDP-glucuronyltransferase and substrates does not facilitate catalysis because binding of the first substrate enhances affinity for binding of the second substrate. This increases the activation energy of the enzymecatalyzed reaction. The phospholipids interacting with UDPglucuronyltransferase, therefore, not only affect the inherent binding energy of substrates but also can have a significant effect on how inherent binding energy is used.
Binding of UDP and Glucuronic Acid to UDP-glucuron-yltransferase Although the data in Table II indicate that there are substrate-induced effects on the properties of UDPglucuronyltransferase and that the inherent binding energy for interactions between enzyme and substrate is used to effect such changes, they do not provide an estimate of how much of the inherent energy is used for purposes other than binding. This is so because inherent binding energy can be used directly to facilitate the catalytic steps as well as to alter affinity for ligands. In order to estimate the amount of inherent energy used for purposes other than enhancing affinity for ligands, one must know the difference between the inherent binding energies and the observed binding energies. In order to make some estimates of the former quantities, we determined the binding constants for binding of UDP and glucuronic acid to UDP-glucuronyltransferase. These binding constants were determined via inhibition studies. The applicable scheme for inhibition of a bisubstrate, random-order, reaction is given in Fig. 1 in which A is UDP-glucuronic acid and B is p-nitrophenol (18). The terms aKA and aKBs, respectively, correspond to the designations K'UDP-GA and K'PNP. The value of f8 in Fig. 1 reflects changes in the enzyme induced by binding of inhibitor.
Addition of lysophosphatidylcholines to delipidated UDPglucuronyltransferase enhances the binding of UDP to the enzyme (Table III). The most interesting aspect of these data, however, is that the affinity of enzyme for UDP is considerably greater than that for UDP-glucuronic acid for all forms of UDP-glucuronyltransferase. Moreover, the differences in affinities for UDP versus UDP-glucuronic acid are larger than the amount of binding energy utilized to alter the affinity of UDP-glucuronyltransferase for p-nitrophenol. For example, on calculating these energies (from the expression AG = RT In K.) the difference in the binding of UDP and UDP-glucuronic acid to delipidated enzyme is about 2900 calories/mol. The difference between KPNP and K'PNP for this form of UDPglucuronyltransferase is about 1500 calories. Corresponding values for enzyme reconstituted with oleoyl lysophosphatidylcholine are 2100 calories/mol (difference between KUDP and KUDP-GA) and 1500 calories/mol (difference between KPNP and K'PNP). Moreover, binding of UDP alone is sufficient to alter the properties of the active site of UDP-glucuronyltransferase as reflected by differences between KPNP and f8 KPNP in zyme and this ligand. These findings suggest one of two possibilities. The binding of the glucuronic acid moiety of UDP-glucuronic acid has a large positive free energy. Alternatively, the inherent binding energy of this part of the substrate is utilized for purposes other than enhancing binding of UDP-glucuronic acid. If the latter is a valid explanation for the differences between KUDP-GA and K1UDP, binding of the glucuronic acid portion of UDP-glucuronic acid also must lead to the "disappearance" of inherent binding energy due to the interaction between enzyme and the UDP portion of the substrate. These possible explanations for the differences between KUDP-GA and K,UDP were investigated by measuring K? lucuronic acid The data in Table IV indicate that glucuronic acid is an effective inhibitor of UDP-glucuronyltransferase. The differences in the affinity of UDP-glucuronyltransferase for UDP as compared with UDP-glucuronic acid cannot be attributed, therefore, simply to a positive free energy for the binding of the glucuronic acid moiety of the latter compound. They suggest, instead, that the inherent binding energy of the glucuronic acid part of the substrate is utilized for purposes other than enhancing binding. Also, binding of the glucuronic acid occurs in such a way as to decrease the binding energy of the UDP portion of UDP-glucuronic acid.
By analogy with data for several enzymes (cf. review by Jencks (21)), but especially lysozyme (22), one can consider the UDP-glucuronic acid binding site of UDP-glucuronyltransferase to be composed of several subsites. Binding energies for UDP and glucuronic acid would be greater than that for UDP-glucuronic acid if there were some hindrance to binding at the subsite interacting with the region of UDPglucuronic acid that takes part in the catalytic step. For example, binding of UDP-glucuronic acid to the enzyme could require distortion of the bond linking the pyrophosphate of UDP to the glucuronic acid. Another possibility is that binding of UDP-glucuronic acid distorts the enzyme, but binding by UDP and glucuronic acid do not. The inherent binding energy of the UDP and glucuronic acid moieties would be needed to provide energy for these distortions, which would lead to apparent disappearance of inherent binding energy and account thereby for the observed differences between KUDP-GA and K,UDP. The energy used to distort the E. 8 complex in this manner could enhance catalysis.
Estimates of the Maximal (Inherent) Binding Energy Due to Interactions between UDP-glucuronyltransferase and UDP-glucuronic Acid-The maximal potential binding energy, or inherent binding energy, for the interaction between UDP-glucuronyltransferase and UDP-glucuronic acid is not a simple sum of the maximal potential binding energies for  the UDP and glucuronic acid portions of the molecule. The principal reason for this discrepancy is that the loss of entropy on the binding of free glucuronic acid to enzyme will be greater as compared to the loss of entropy on binding to the enzyme of glucuronic acid already attached covalently to UDP (19). The same reasoning applies to the entropy loss associated with the binding of free UDP to enzyme. Since there is no way to measure these quantities, it is not possible to determine exactly the enthalpy changes associated with the binding of these ligands. On the other hand, the maximal potential enthalpy change for the binding of glucuronic acid to enzyme will be greater for the glucuronic acid portion of UDP-glucuronic acid as compared with free glucuronic acid. The same is true for the enthalpy change associated with the binding of UDP. It is possible, at least, to make minimal estimates of the inherent binding energy available for enhancing catalysis. Moreover, we think it is reasonable to expect that the entropy changes associated with the binding of UDP and glucuronic acid to UDP-glucuronyltransferase will be similar for delipidated and phospholipid-activated forms. Comparing the differences in the amounts of inherent binding energy that disappear on activation with various lysophosphatidylcholines with the energy required to account for the observed activations at Vmax is especially useful, therefore, for evaluating whether inherent binding energy of UDP-glucuronic acid is used to facilitate catalysis. The inherent or maximal potential binding energy for UDP and glucuronic acid can be estimated from the values of K'UDP, K,glucuronicacid, and /3 in Tables III and IV. The latter are important because the inherent binding energy is different from RT In K, by the amount of energy used to change KPNP to 8KPNP, which equals RT ln ft. The maximal potential binding energies for UDP and glucuronic acid, calculated in the above manner, are tabulated in Table V. The differences between these values and the observed binding energies of UDP-glucuronic acid are tabulated in Table VI. The second column in Table VI is the inherent binding energy that is used for other purposes. The data in Table V indicate that there is positive correlation between the inherent binding energy for the binding of glucuronic acid to various forms of UDPglucuronyltransferase and the activity at Vmax of the different forms of the enzyme. Additionally, there is a positive correlation between the amount of binding energy that "disappears" (Table VI) and activities at V,,ax of UDP-glucuronyltransferase activated with different lysophosphatidylcholines.

Estimates of Amounts of Inherent Binding Energy That
Can Be Used to Facilitate Catalysis-A minimal estimate (for the reasons given above) of the inherent or maximal Oleoyl lysophosphatidylcholine -7800 potential binding energy for the interaction between delipidated enzyme and UDP-glucuronic acid is -10,800 calories. The observed binding energy of UDP-glucuronic acid to delipidated UDP-glucuronyltransferase is -4,600 calories/mol. Hence, a minimal estimate of the binding energy that can be used to facilitate catalysis by the delipidated form of UDPglucuronyltransferase is -6,100 calories/mol. We know from  Table VII. Also tabulated are the observed differences in catalytic rate, expressed as calories, between reconstituted and delipidated enzyme. There is excellent agreement between the calculated amounts of energy available for enhancing catalysis by phos-pholipid-activated UDP-glucuronyltransferase versus delipidated enzyme and the observed differences in rates of catalysis. Some of the disparity between calculated and observed energies could be due to differences in the energy level of enzyme-bound UDP-glucuronic acid. There are two interrelated reasons for concluding, however, that corrections due to this factor will be small. First, enhanced binding of UDPglucuronic acid to lipid-treated as compared with delipidated enzyme can be accounted for almost completely, in all cases, by enhanced affinity of the enzyme for the UDP portion of the substrate. Second, UDP does not take part in the catalytic steps.
Proposed Mechanism of Activation of UDP-glucuronyltransferase by Lysophosphatidylcholine-The data presented above provide evidence for the idea that some of the inherent binding energy for the binding of UDP-glucuronic acid to the delipidated form of UDP-glucuronyltransferase can be used to enhance the catalytic rate. The most reasonable explanation for this type of effect, based on the available data, is that binding of UDP-glucuronic acid distorts the region of the substrate in which UDP and glucuronic acid are linked and/or alters the active site of the enzyme in a manner that decreases the activation energy of the catalytic steps. Species of lysophosphatidylcholine appear to activate the delipidated form of UDP-glucuronyltransferase by amplifying this mechanism. Lysophosphatidylcholines induce a conformational change in the active site of the enzyme that enhances the inherent affinity of the enzyme for the UDP portion of UDPglucuronic acid (Table III). The best activators, e.g. myristoyl and oleoyl lysophosphatidylcholine, also increase the inherent affinity of UDP-glucuronyltransferase for the glucuronic acid moiety of the substrate (Table IV). The conformational changes that enhance inherent affinities for substrate seem to be associated with either an alteration that puts greater strain on the region of UDP-glucuronic acid linking the pyrophosphate of UDP with the sugar or that allows the binding energy of the substrate to distort the enzyme. The basis for this conclusion is that activators enhance the affinity of enzyme for UDP-glucuronic acid to a significantly smaller extent than they increase affinity for the UDP and glucuronic acid portions of the substrate. This mechanism explains, in an easily visualized form, how lysophosphatidylcholines can alter delipidated UDP-glucuronyltransferase in a way that allows more of the inherent binding energy to be used to facilitate catalysis. The kinetic studies do not allow for deciding whether binding energy is used to distort the substrate or the enzyme Binding energy utilized to enhance catalysis Amounts of potential binding energy utilized to facilitate catalysis were calculated as outlined in the text. The observed differences in activities at Vm., for the pair of enzyme forms compared were used to calculate the difference in free energy of activation for glucuronidation according to the equation AG = -RTIn (V,/ V2). Energies are calories/ mol at 30 'C. tidylcholine except for the fact that binding of one substrate changes the affinity of the enzyme for the second substrate. The nature of this distortion in the enzyme depends on the presence of lysophosphatides and on the chain length and unsaturation of the attached acyl groups. For, whereas some of the binding energy due to interactions between delipidated enzyme and substrate can be used to enhance affinity for binding of the second substrate and not for facilitation of catalysis, lysophosphatidylcholines that are good activators of UDP-glucuronyltransferase do not allow this to happen. In the presence of a good activator such as oleoyl lysophosphatidylcholine, all the binding energy that disappears is used to facilitate catalysis. The enzyme's lipid environment influences the inherent binding energy of ligands, and can determine how this energy is used, as for example, to facilitate binding of a second ligand or to facilitate catalysis. The property ofthe lipid environment that is important in determining the structure of UDP-glucuronyltransferase remains to be elucidated.