[22]

TIME-DEPENDENT

PHOSPHOLIPASE A 2 ACTIVATION

[22] A c t i v a t i o n o f P h o s p h o l i p a s e

249

A 2 on Lipid Bilayers

B y JOHN D. BELL and RODNEY L. BILTONEN

Introduction The activation o f phospholipase A 2 is a complex process that depends on the conformation of the e n z y m e ~-5 and the structure and dynamics of the lipid bilayer with which it interacts. 5-9 The physical nature of the lipid bilayer is of particular importance. Thus, one must choose carefully the experimental system to be used for studies of the mechanism of the activation process as well as studies of the mechanisms of action of putative activators or inhibitors of the enzyme. Ideally, the system would be simple, well-characterized physically, and amenable to investigations of timedependent processes involved in e n z y m e activation. Substrates commonly used, such as sonicated vesicles of hen egg phosphatidylcholine or of saturated phosphatidylcholines, are frequently undesirable because at least some o f the phospholipases A 2 a r e not very active toward vesicles in the liquid crystalline state, and hydrolysis time courses with sonicated vesicles in the gel state do not always exhibit a time-dependent activation. Also, sonicated vesicles of phosphatidylcholine are not stable in the gel state 1° and aggregate or fuse into structures that are not well-defined, thus complicating interpretation of the data. One experimental system that offers several advantages for the study of the activation o f soluble phospholipases A 2 is large unilamellar vesicles (LUV) of dipalmitoylphosphatidylcholine (DPPC): (1) E n z y m e activation 1M. F. Roberts, R. A. Deems, and E. A. Dennis, Proc. Natl. Acad. Sci. U.S.A. 74, 1950 (1977). 2 D. O. Tinker and J. Wei, Can. J. Biochem. 57, 97 (1979). 3 G. Romero, K. Thompson, and R. L. Biltonen, J. Biol. Chem. 263, 13476 (1988). 4 W. Cho, A. G. Tomasselli, R. L. Heinrikson, and F. J. K6zdy, J. Biol. Chem. 263, 11237 (1988). 5 j. D. Bell and R. L. Biltonen, J. Biol. Chem. 264, 12194(1989). 6 H. M. Verheij, A. J. Slotboom, and G. H. de Haas, Rev. Physiol. Biochem. Pharmacol. 91, 91 (1981). 7 M. Menashe, G. Romero, R. L. Biltonen, and D. Lichtenberg, J. Biol. Chem. 261, 5328 (1986). s N. Gheriani-Gruszka, S. Almog, R. L. Biltonen, and D. Lichtenberg, J. Biol. Chem. 263, 11808 (1988). 9 M. K. Jain, B.-Z. Yu, and A. Kozubek, Biochim. Biophys. Acta 908, 23 (1989). ~0M. Wong, F. H. Anthony, T. W. Tillack, and T. E. Thompson, Biochemistry 21, 4126 (1982).

METHODS IN ENZYMOLOGY, VOL. 197

Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

250

PHOSPHOLIPASE STRUCTURE--FUNCTION TECHNIQUES

[22]

is slow and can be monitored by following the time course of vesicle hydrolysis) (2) With certain phospholipases A 2, such as the monomer aspartate-49 enzyme from Agkistrodon piscivorus piscivorus (AppD49), changes in the intrinsic tryptophan fluorescence of the enzyme can be measured simultaneously with the hydrolysis reaction to correlate changes in the state of the enzyme during activation. 5 (3) Changes in the structure of the vesicles can be concurrently monitored using fluorescent probes to correlate the role of vesicle structure and dynamics in the activation process. (4) The vesicles can be well-characterized structurally and thermodynamically.l°'ll Although the methods described in this chapter focus on DPPC L U V and certain snake venom phosphlipases A2, they are certainly adaptable to the study of other lipid hydrolases and substrates. Preparation of Vesicles The preparation of fused L U V of DPPC is based on procedures and principles previously described. 3'1°This method is applicable to DPPC and its ether analog but is not necessarily applicable to other lipids. The phospholipid, suspended in chloroform, is dried by evaporation. Final traces of the chloroform are removed with a lyophilizer. A solution of 50 mM KC1 and at least 1 mM NaN3 to prevent bacterial growth is added to the tube of dried lipid at a temperature of 450-55 °, a temperature greater than the gel-liquid crystalline transition temperature. The lipid sample should be maintained at this temperature for at least 1 hr and vortexed vigorously for several seconds about every 10 rain. This procedure yields a suspension of multilamellar vesicles. 12 The multilamellar vesicles are then dispersed by sonication at 450-55 °. 13 Sonication of the sample 3 or 4 times for 3 min each is generally sufficient to yield a dispersion of small unilamellar vesicles (diameter about 25 nm). The sample should appear transparent at the end of the sonication. Titanium grains from the sonicator probe and nondispersed lipid are removed from the sample by either ultracentrifugation 14or a 5-min centrifugation at 14,000 rpm in a microfuge using 1.5-ml tubes. If ultracentrifugation is used, the sample must be maintained at no less than 45 ° during the separation. The upper two-thirds of the resulting supernatant is carefully removed and transferred to a glass tube and sealed with Teflon tape. The sample is allowed to stand at room temperature overnight and then II j. Suurkuusk, B. R. Lentz, Y. Barenholz, R. L. Biltonen, and T. E. Thompson, Biochemistry 15, 1393 (1976). 12 A. D. Bangham, J. De Gier, and G. D. Greville, Chem. Phys. Lipids 1, 225 (1967). 13 C. Huang, Biochemistry 8, 344 (1969). 14 C. Huang and T. E. Thompson, this series, Vol. 32, p. 485.

[22]

TIME-DEPENDENT PHOSPHOLIPASE A 2 ACTIVATION

251

transferred to storage at 4 ° for 3 weeks. The storage temperature is critical for reproducible vesicle fusion. Below 2 °, the vesicles aggregate irreversibly. It should also be noted that the vesicles do not fuse properly in the presence of added CaC12. At the end of the 3-week storage, the sample contains mostly large unilamellar vesicles with a diameter of about 90 to 100 nm. ~0The appearance of the sample should resemble dilute milk. Contaminating multilamellar vesicles that have formed during the fusion process must be removed to obtain a homogeneous sample. Two methods are useful for the separation. The sample may be applied to a column of Sepharose CI-2B, 3 or it may be centrifuged for 5 min at 14,000 rpm in a microfuge as described above. The supernatant contains the LUV, and the vesicles appear to be stable for several months if stored at room temperature in the presence of NAN3. ~0The purity of the L U V can be monitored by thin-layer chromotography 3 and by differential scanning calorimetry.l°,H The calorimetric scan of DPPC L U V should show a single nearly symmetrical peak at 41.4 ° which is broader than that of multilamellar vesicles. The apparent Tm and width of the calorimetric scan will depend on the scan rate and calorimeter used. H Most importantly, a preparation of L U V containing n o small unilamellar vesicles is characterized by the absence of a gel-liquid transition at about 37 °. Combined Enzyme Assay and Fluorescence Spectroscopy Assay of the time course of vesicle hydrolysis by phospholipase A2 using a pH stat has been described previously. ~5In order to make temporal correlations of changes in the intrinsic fluorescence intensity of the enzyme with changes in the catalytic activity during the time course, the pH stat instrument (Radiometer) has been used with a spectrofluorometer (SLM 8000C)) This is accomplished by inserting a small combination pH electrode (5 mm diameter) and the glass pipette tip from the pH stat buret in a two-hole rubber stopper as shown in Fig. I. A hole is drilled in a lid for the fluorometer sample chamber to fit the stopper and to center the electrode and pipette over the sample cuvette. A stainless steel cannula is inserted through the rubber stopper for sample injection. The height of the rubber stopper is adjusted so that the electrode, pipette, and cannula tips are all submerged in the sample but do not interfere with the fluorometer light path. This is possible in the SLM 8000C fluorometer using a standard 1 cm a fluorometer cuvette with a stirring bar at the bottom and 2.5 ml of sample. Adjustments may be needed depending on the position of the 15 W. Nieuwenhuizen, H. Kunze, and G. H. de Haas, this series, Vol. 32, p. 147.

252

PHOSPHOLIPASE STRUCTURE--FUNCTION TECHNIQUES i

/---F

C

,-

[22]

imG E

D

A

liB

~--

FIG. 1. Detail of the experimental setup for simultaneous measurement of enzyme fluorescence and hydrolysis. A, Standard fluorometer cuvette; B, magnetic stirring bar; C, combination pH electrode (from pH stat); D, cannula for enzyme injection; E, pipette tip (from pH stat buret); F, rubber stopper; G, fluorometer sample chamber lid.

optical windows in the sample chamber of other fluorometers. The rubber stopper assembly can be conveniently placed and removed as a unit between experiments as samples are exchanged. It is necessary for these experiments that the sample chamber of the fluorometer be temperaturecontrolled and equipped with a magnetic stirrer. Temperature is critical in studies involving DPPC (see below), and adequate stirring must be maintained for maximum resolution and sensitivity in monitoring the hydrolysis time course and to maintain sample homogeneity. Prior to the experiment, the vesicles must be equilibrated in the appropriate solution; I0 mM CaC12, 35 mM KCI, and 1 mM NaN3 has been commonly used. 5After adjusting the solute concentration, the vesicles are slowly and repeatedly heated and cooled through the phase transition (41.5°) over a period of several hours to achieve equilibrium of all components across the vesicle bilayer. All solutions must be degassed to eliminate CO2, which interferes with the pH stat assay by buffering the sample pH. Finally, the vesicles are temperature-equilibrated in the fluorometer sample chamber with the rubber stopper assembly in place and the magnetic stirrer turned on to ensure proper equilibration. The pH is then adjusted to 8.0 (or as desired) with the pH stat, and the experiment is initiated by injection of phospholipase A2. Figure 2 shows typical simultaneous fluorescence and hydrolysis time courses using the AppD49 phospholipase A2. Note that the fluorescence intensity of the enzyme rapidly increases at the onset of rapid vesicle hydrolysis. The decrease in the intensity during the initial portion of the time course results from adsorption of the enzyme onto the quartz cuvette.5 A number of methods to prevent this adsorption have been examined including treatment with silane, dextran, or polylysine. To date, only

[9-2]

TIME-DEPENDENT PHOSPHOLIPASEA 2 ACTIVATION

i

253

1000 g

~- 500

4-J

~ ,7

o

100 200 300 400 Time (seconds)

500

FIG. 2. Time courses of DPPC L U V hydrolysis and AppD49 phospho]ipase A 2 fluorescence monitored with the system described in Fig. 1. Vesicles (0.4 mM) in 2.5 ml of 35 mM

KC1, 10 mM CaC12, and approximately l mM NaN3 were equilibrated at 39° and adjusted to pH 8.0 using 10 mM NaOH as the titrant. Enzyme (20/~g/ml) was added at t = 30 sec, and fluorescence (excitation 280 nm, emission 340 nm) and vesicle hydrolysis (number of protons titrated with NaOH by the pH stat) were simultaneously recorded as described.

polylysine has been found to be effective in preventing the binding of e n z y m e to the cuvette. Unfortunately, polylysine appears to interfere with the hydrolysis reaction. More acidic phospholipases, such as the dimer enzymes from Crotalus atrox and A. piscivorus piscivorus, do not adsorb appreciably to the cuvette walls. Changes in the physical properties of the vesicles during the hydrolysis time course can also be simultaneously monitored. Changes in vesicle light scattering suggest changes in vesicle size or refractive index during the time course and provide the data needed for correction of certain optical artifacts (see below). Light scattering can be monitored simultaneously by using a second emission m o n o c h r o m a t o r mounted 90 ° from the excitation m o n o c h r o m a t o r and 180° from the other emission monochromator. The wavelength of the second emission m o n o c h r o m a t o r is set near the excitation wavelength. F o r example, the excitation wavelength for protein fluorescence is generally 280 nm, and one emission monochromator is set at 340 nm for the fluorescence and the second monochromator is set at 290 nm for measuring light scattering rather than at the excitation wavelength to avoid excessive intensity of the scattered light. A fluorescent probe of membrane structure, trimethylammonium diphenylhexatriene (TMA-DPH), has been used to monitor other properties of the bilayer during the hydrolysis time c o u r s e ) This probe is sensitive

254

PHOSPHOLIPASE STRUCTURE-FUNCTION TECHNIQUES

[22]

== =,- 30

2c

H a0

+~

0

,

,

,

,

i

,

500

,

-

-

,

.

.

1000

.

.

,

.

.

.

.

t500

Time (seconds)

FIG. 3. Correlation of the time courses of phospholipase A2 fluorescence, vesicle light scattering, and TMA-DPH fluorescence during the hydrolysis of DPPC LUV. Vesicles (0.4 mM DPPC) were equilibrated with 0.4/zM TMA-DPH in 2.5 ml of 35 mM KCI, 10 mM CaC12, 10 mM sodium borate (pH 8), and 1 mM NAN3. AppD49 phospholipase A2 (8/zg/ml) was added at t = 90 sec. Curve a, Enzyme fluorescence (excitation 280 nm, emission 340 nm); curve b, light scattering (excitation 280 nm, emission 290 nm); curve c, TMA-DPH fluorescence (excitation 360 nm, emission 430 nm).

to the fluidity of the phospholipid molecules within the bilayer 16 and to quenching by water. The TMA-DPH (2 mM in dimethylformamide) is diluted 5000-fold into the vesicle solution and equilibrated at least 1 hr at 45 ° in the dark. The concentrations of probe and vesicle are selected so that the ratio of lipid to probe in the bilayer will be at least 1000: 1. Figure 3 shows a typical time course of simultaneous measurement of enzyme fluorescence (curve a), vesicle light scattering (curve b), and TMA-DPH fluorescence (curve c) during lipid hydrolysis. These three optical properties were obtained simultaneously by arranging the monochromators in the T format described above and by electronically switching the excitation wavelength between 280 nm (for the protein fluorescence and the light scattering) and 360 nm (for the TMA-DPH fluorescence). Likewise, one emission monochromator was switched between 340 nm (protein fluorescence) and 290 nm (light scattering). The other emission monochromator monitored the fluorescence at 430 nm (TMA-DPH). The switching of wavelengths is accomplished rapidly under computer control with the SLM 8000C instrument. The time required to complete the cycle is 9 sec. As shown in Fig. 3, apparent changes in vesicle physical properties occur during the time course of hydrolysis. The light scattering decreases substantially after the increase in enzyme fluorescence and may reflect t6 F. G. Prendergast, R. P. Haugland, and P. J. Callahan, Biochemistry 20, 7333 (1981).

[22]

T I M E - D E P E N D E N T PHOSPHOLIPASE A 2 ACTIVATION

255

macroscopic changes in vesicle structure during the rapid phase of hydrolysis. 5 The TMA-DPH fluorescence decreases slowly during the slow phase of hydrolysis, begins to decrease more rapidly, increases transiently concurrent with the increase in enzyme fluorescence, and then decreases dramatically. The transient increase in TMA-DPH fluorescence (encircled in Fig. 3) is highly reproducible, though small, and is not an optical artifact. This change in TMA-DPH fluorescence concurrent with the increase in enzyme fluorescence has been interpreted to reflect a transition in bilayer structure arising from the accumulation of hydrolysis products that may be responsible for the rapid and sudden activation of the enzyme, 5 but it is not yet understood. Fluorescent probes such as TMA-DPH may adsorb very tightly to the electrode and pipette tip from the pH stat. Consequently, it may not always be desirable to use the pH stat when a lipid probe is included in the reaction mixture. If the pH stat is not used, one must include a buffer in the reaction solution. Phosphate buffer precipitates with calcium and therefore cannot be used. Amine-containing buffers or sulfonic acid derivatives may be used, but they are highly sensitive to temperature. Borate buffer is not very temperature-sensitive and has been used successfullyJ Finally, proper experimental design with this system requires at least a phenomenological understanding of the temperature dependence of LUV hydrolysis by phospholipase A 2. The reaction is highly sensitive to the thermotropic phase transition of the lipid bilayer (41.4 ° for DPPC). Significant rates of hydrolysis are generally seen only when the temperature of the experiment is within 3° or 4° of the transition temperature. Significantly below the transition temperature, rates of hydrolysis are very slow and vesicle hydrolysis may require many hours. Significantly above the thermotropic phase transition, the maximum hydrolysis rates are high, but the lag times preceding rapid hydrolysis may become so long that sample evaporation interferes with the hydrolysis time course. Data Analysis The light scattered by the vesicles in solution poses two artifactual problems for quantitative analysis of the fluorescence data. This is especially true since the light scattering changes with time (Fig. 3). First, a small amount of light scattered by the vesicles creates a background at the emission wavelength. The intensity of this scattering is proportional to the concentration of vesicles and to the intensity of scattered light measured at a wavelength near the excitation wavelength. Figure 4A shows the relationship between scattered light measured at 290 nm and that obtained at 340 nm (excitation wavelength 280 nm in both cases). The calibration

256

PHOSPHOLIPASE STRUCTURE-FUNCTION TECHNIQUES

[22]

4J

16 6

2

B

14 4

12

03

2 L" l0

0

4

8

12 0 4 Intensity at 290 nm

8

12

FIG. 4. Calibration curves for stray light and inner filter effects of light scattered by DPPC LUV. Various concentrations of LUV ranging from 0 to 0.4 mM were added to 2 ml of 35 mM KCI, 10 mM EDTA, 10 mM sodium borate (pH 8), and 1 mM NaN~ at 37°. The intensity of light at 340 and 290 nm (excitation 280 nm) was recorded, tryptophan (1.5 p.M final) was added, and the intensity at both wavelengths was again recorded. (A) Light intensity at 340 nm prior to the addition of tryptophan plotted as a function of the intensity at 290 nm. (B) Intensity of light at 340 nm prior to the addition of tryptophan subtracted from the intensity after addition of tryptophan, with the difference plotted as a function of the intensity at 290 nm. The calibration curves were fit by nonlinear regression to the function Y = A + B X + C X 2.

curve is calculated by nonlinear repression using the equation Y = A + B X + C X 2. One can then use this calibration curve and the measured light-scattering data obtained at 290 nm to calculate the background due to scattered light at 340 nm. This time-dependent background is subtracted from the fluorescence time course. The presence of vesicles in the solution causes a second optical artifact. The turbidity of the solution, which is proportional to vesicle concentration, reduces the intensity of excitation light reaching the fluorophore and also reduces the amount of emitted light reaching the monochromator. This artifact can be quantified by adding various concentrations of vesicles to a solution of tryptophan. The background (presumably due to light scattering) is measured at both 290 and 340 nm (excitation 280 nm) prior to the addition of the trypt6phan. Fluorescence emission at 340 nm is then measured after addition of the amino acid. The background at 340 nm is subtracted from the fluorescence signal, and the difference is plotted as a function of the intensity of scattered light at 290 nm (Fig. 4B). A calibration curve is again obtained by nonlinear regression, and the fraction of the

[22]

TIME-DEPENDENT PHOSPHOLIPASE A 2 ACTIVATION

257

true fluorescence that is actually transmitted to the monochromator for a given amount of light scattering is inferred from the calibration curve. One can then correct the measured enzyme fluorescence for inner filter effects during the time course by first calculating the fractional transmittance at each time point using the calibration curve (Fig. 4B) and the time course of light scattering at 290 nm. The correction is made by subtracting the background light (as described above) and then dividing the observed fluorescence intensity at each time point by the corresponding fractional transmittance. In general, the amount of light scattered is relatively small with LUV at lipid concentrations below 0.2 mM. These corrections are only needed for quantitative analysis of the fluorescence changes with LUV at lipid concentrations of at least 0.4 mM.

Summary So far, three phospholipases A 2 that display activation kinetics during the time course of hydrolysis of DPPC LUV have been found to undergo a fluorescence change coincident with the activation: the monomer (AppD49) and the dimer enzymes from A. piscivorus piscivorus and the dimer enzyme from C. atrox. The porcine pancreatic enzyme produces similar time courses of hydrolysis but does not display a concurrent fluorescence change. 17It is assumed that other phospholipases A2 will behave similarly in terms of the hydrolysis reaction. Which enzymes respond with a similar change in intrinsic fluorescence during the time course may well depend on the position of tryptophan residues and the amino acid sequence. Even though a given phospholipase A 2 may not change its fluorescent properties on activation, the simultaneous monitoring of the hydrolysis reaction and the fluorescence of probes of the bilayer structure can be done with any phospholipase A2. A variety of probes exist which are sensitive to slightly different membrane properties and could be used as described here for TMA-DPH. For example, 1,3-dipyrenylpropane is sensitive to the apparent microviscosity of the bilayer is terms of the ability of molecules to translationally diffuse in the membrane.18 6-Palmitoyl-2{[2-(trimethylammonio)ethyl]methylamino}naphthalene chloride is sensitive to the ability of a molecule to rotate in the bilayer and displays large changes in its steady-state fluorescence as the anisotropy of the bilayer 17 j. D. Bell and R. L. Biltonen, unpublished results (1989). 18 R. L. Melorick, H. C. Haspel, M. Goldenberg, L. M. Greenbaum, and S. Weinstein, Biophys. J. 34, 499 (1981).

258

PHOSPHOLIPASE STRUCTURE-FUNCTION TECHNIQUES

[23]

changes. ~9 6-Propionyl-2-(dimethylamino)naphthalene is sensitive to the polarity and degree of hydration of its environment. 2° Finally, a compound titled NK-529 has recently been introduced that apparently monitors the lateral phase separation of fatty acids in the bilayer. 9 The fact that activation of phospholipase A2 can be monitored during the time course of hydrolysis ofDPPC L U V makes this system an excellent choice for studying the mechanisms of activation and possible effects of various activators and inhibitors. The experimental system described here provides a way to determine whether such regulators exert their effects through alterations of the properties of the membrane and/or the enzyme. Importantly, this system allows one to seek temporal correlations of the various events in the process. Acknowledgments This work was made possible by funding from the National Institute of General Medical Science (Grants GM37658 and GMl1838) and from the Office of Naval Research (N0001488-K-0326).

19j. R. Lakowicz, D. R. Bevan, B. P. Maliwal, H. Cherek, and A. Balter, Biochemistry 22, 5714 (1983). 20 G. Weber and F. J. Farris, Biochemistry 18, 3075 (1979).

[23] Phospholipase Stereospecificity at Phosphorus

By KAROL S. BRUZIKand MING-DAW TSAI Introduction 1

Two types of stereochemical information can be obtained from studies of enzymatic reactions utilizing substrates which are stereogenically 2 (chirally) labeled at phosphorus. 3-t° (1) The steric course of the nucleophilic 1 This is paper 21 in the series "Phospholipids Chiral at Phosphorus." For Paper 20, see Ref. 16. 2 K. Mislow and J. Siegel, J. Am. Chem.Soc. 106, 3319 (1984). 3 F. Eckstein, Angew. Chem. Int. Ed. Engl. 22, 423 (1983); F. Eckstein, P. J. Romaniuk, and B. A. Connolly, this series, Vol. 87, p. 197. 4 p. A. Frey, J. P. Richard, H.-T. Ho, R. S. Brody, R. D. Sammons, and K.-F. Sheu, this series, Vol. 87, p. 213; P. A. Frey, Tetrahedron 33, 1541 (1984). 5 j. A. Gerlt, J. A. Coderre, and S. Mehdi, Adv. Enzymol. Relat. Areas Mol. Biol. 55, 291 (1983).

METHODS IN ENZYMOLOGY, VOL. 197

Copyright © 1991 by Academic Press, Inc. All rights of reproduction in any form reserved.

Activation of phospholipase A2 on lipid bilayers.

So far, three phospholipases A2 that display activation kinetics during the time course of hydrolysis of DPPC LUV have been found to undergo a fluores...
539KB Sizes 0 Downloads 0 Views