Advertisement
JBC

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

Originally published In Press as doi:10.1074/jbc.M401016200 on March 2, 2004

J. Biol. Chem., Vol. 279, Issue 19, 20490-20500, May 7, 2004
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/19/20490    most recent
M401016200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, X.
Right arrow Articles by Roberts, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, X.
Right arrow Articles by Roberts, M. F.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

Cross-linking Phosphatidylinositol-specific Phospholipase C Traps Two Activating Phosphatidylcholine Molecules on the Enzyme*

Xin Zhang, Hania Wehbi, and Mary F. Roberts{ddagger}

From the Merkert Chemistry Center, Boston College, Chestnut Hill, Massachusetts 02467

Received for publication, January 29, 2004 , and in revised form, February 27, 2004.


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bacillus thuringiensis phosphatidylinositol-specific phospholipase C (PI-PLC), a bacterial model for the catalytic domain of mammalian PI-PLC enzymes, was cross-linked by 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride to probe for the aggregation and/or conformational changes of PI-PLC when bound to activating phosphatidylcholine (PC) interfaces. Dimers and higher order multimers (up to 31% of the total protein when cross-linked at pH 7) were observed when the enzyme was cross-linked in the presence of PC vesicles. Aggregates were also detected with PI-PLC bound to diheptanoyl-PC (diC7PC) micelles, although the fraction of cross-linked multimers (19% at pH 7) was lower than when the enzyme was cross-linked in the presence of vesicles. PI-PLC cross-linked in the presence of a diC7PC interface exhibited an enhanced specific activity for PI cleavage. The extent of this cross-linking-enhanced activation was reduced in PI-PLC mutants lacking either tryptophan in the rim (W47A and W242A) of this ({beta}{alpha})8-barrel protein. The higher activity of the native protein cross-linked in the presence of diC7PC correlated with an increased affinity of the protein for two diC7PC molecules as detected by matrix-assisted laser desorption-ionization time-of-flight mass spectrometry. In contrast to wild type protein, W47A and W242A had only a single diC7PC tightly associated when cross-linked in the presence of that activator molecule. These results indicate that (i) each rim tryptophan residue is involved in binding a PC molecule at interfaces, (ii) the affinity of the enzyme for an activating PC molecule is enhanced when the protein is bound to a surface, and (iii) this conformation of the enzyme with at least two PC bound that is stabilized by chemical cross-linking interacts more effectively with activating interfaces, leading to higher observed specific activities for the phosphotransferase reaction.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bacterial phosphatidylinositol-specific phospholipase C (PIPLC),1 a ({beta}{alpha})8-barrel, has a similar structure to the catalytic domain of mammalian PLC{delta}1 (1-3). Although the Bacillus thuringiensis PI-PLC is Ca2+-independent, whereas the mammalian enzyme requires Ca2+ for activity, both enzymes can be activated by nonsubstrate interfaces (4-7). B. thuringiensis PI-PLC-catalyzed cleavage of interfacial substrate PI (which is an anionic lipid) to water-soluble cIP and subsequent hydrolysis of cIP are dramatically enhanced by the enzyme binding to phosphatidylcholine (PC) interfaces (4-6). For the soluble cIP reaction, kcat was increased, and Km decreased in the presence of PC interfaces (4, 8). That work suggested that PC binds to an activator site on the bacterial enzyme that is distinct from the enzyme active site, a model confirmed recently using artificial substrates (9). However, the details of the conformational change upon PC binding that leads to improved enzyme efficiency are not known. Growing evidence suggests that a region at the top of the barrel rim containing two tryptophans exposed to solvent (Trp-47 in helix B and Trp-242 in the flexible loop 232-244) is crucial for this protein to bind to PC interfaces and for kinetic activation of the enzyme (10-12). Interfacial binding could enhance aggregation of the protein as well as alter the conformation of an individual protein molecule to generate a more active state. Recent work by Jain and co-workers (13) suggests that aggregation of the protein does occur with binding of an activator such as diC7PC. Micellar PC has been shown to increase kcat for PLC{delta}1 cleavage of cIP; the pleckstrin homology domain was critical for that kinetic activation (7). However, the interplay of the pleckstrin homology domain with the catalytic domain in PLC{delta}1 is unknown, and hydrophobic groups poised at the rim of PLC{delta}1 could mediate similar changes for interfacial activation of that enzyme as well.

Chemical cross-linking of proteins can be used to detect their aggregation state in the absence and presence of lipid surfaces (14, 15). In this study, we used 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride (EDC), a reagent that forms an amide bond between an amino and a carboxyl group of side chains that are spatially close, to probe for the aggregation and/or conformational changes of PI-PLC when bound to lipid interfaces. The sequence of the B. thuringiensis enzyme has 23 lysine, 20 glutamate, and 18 aspartate residues, most of which are on the surface of the protein. Previous attempts to cross-link the bacterial PI-PLC using dimethyl suberimidate dihydrochloride, a reagent that targets surface amino groups, in the absence or presence of Triton X-100, diC7PC, or cIP and diC7PC yielded only monomers.2 Therefore, it was thought that amide formation might be more effective at trapping PI-PLC aggregates. The relatively high fraction of acidic and basic residues on the surface of B. thuringiensis PI-PLC should enhance the probability of forming intermolecular cross-linked subunits. However, in the crystal structure of the related Bacillus cereus enzyme (1), there are nine lysine-acidic amino acid pairs within 4 Å that could be intramolecularly cross-linked by EDC as well (Fig. 1, Table I). None of these pairs are very close to the key tryptophan residues in the rim of the ({beta}{alpha})8-barrel that are critical for kinetic activation by PC interfaces (10, 11).



View larger version (65K):
[in this window]
[in a new window]
  		FIG. 1.
Model of B. cereus PI-PLC (Protein Data Bank file 1PTB [PDB] ) in SETOR showing the two rim tryptophan residues (in green) and the nine pairs of lysine (blue)/aspartate or glutamate (red) within 4 Å. Note the Lys-279/Glu-52 pair as the nearest intramolecular cross-link to the mouth of the barrel; an intramolecular cross-link of these two residues will also link N- and C-terminal portions of the molecule. The next closest pair, Lys-38/Glu-93, is also shown.

 


View this table:
[in this window]
[in a new window]
  		TABLE I
Lysine / glutamate (or aspartate) pairs within 4 Å in the B. cereus PI-PLC structure

The same paris should exist in the B. thuringiensis PI-PLC structure, which only differs by eight amino acids.

 
The results reported in this work show that PI-PLC does indeed form aggregates when bound to surfaces, since dimers and higher order aggregates (up to 31% of the total protein at pH 7, where enzyme activity is optimal) were observed when the enzyme was cross-linked by EDC in the presence of PC vesicles. Cross-linking in the presence of diC7PC micelles yielded at most 19% dimer and larger aggregates at pH 7. An intramolecularly cross-linked species (typically 40-50% of the total intensity on SDS-PAGE) was also detected as a major product of chemical cross-linking. More interestingly, a kinetic activation of PI cleavage was observed if the cross-linking occurred in the presence of diC7PC micelles. This kinetic activation does not correlate with aggregation of the protein but rather correlates with trapping the enzyme in a conformation where it has a higher affinity for individual diC7PC molecules. Cross-linking previously characterized rim tryptophan mutants of PI-PLC (W47A and W242A) also shows that each of these tryptophan residues appears to be part of a binding site for a PC molecule. This interaction is used to discuss how PC binding activates PI-PLC.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Chemicals—POPC, DOPMe, PI from bovine liver, and the short chain lipids diC6PC and diC7PC were purchased from Avanti Polar Lipids and used without further purification. diC6PMe was synthesized from diC6PC using Streptomyces sp. phospholipase D (see Ref. 16 for transphosphatidylation protocol). iPrOH and D2O were purchased from Sigma. EDC was purchased from Pierce. cIP was generated enzymatically from crude PI using PI-PLC as described previously (1). All other chemicals were reagent grade.

PI-PLC and Mutants—The recombinant B. thuringiensis PI-PLC (17) and tryptophan mutants W47A and W242A (10) were expressed and purified as detailed previously. The concentration of each protein was measured by absorbance at 280 nm (10) and by Lowry (18) and Bradford (19) assays.

Chemical Cross-linking of PI-PLC using EDC—Cross-linking of PIPLC in the presence of different interfaces with EDC as the cross-linking reagent used 100 µg/ml PI-PLC dissolved in 100 mM MES (pH adjusted to 5.0, 6.0, or 7.0). EDC (1 M in water) was added to a final concentration of 50 mM in the reaction mixture. As a control, PI-PLC was incubated with EDC (50 mM) in the absence of lipids. Cross-linking of PI-PLC was tested separately in the presence of micellar diC7PC (2 mM), monomeric diC6PC or diC6PMe (4 mM), or SUVs of POPC or DOPMe (2 mM). These mixtures were incubated for 2-3 h at room temperature. After this time, aliquots were removed for enzymatic assays, whereas the bulk of each reaction mixture was dialyzed against 10 mM Tris, pH 7.5, at 4 °C to remove excess EDC, lyophilized, rehydrated with 50 µl of water, and then analyzed by SDS-PAGE. An alternative to dialysis was concentration of the treated protein using a 10-kDa cut-off microconcentrator and washing the protein repeatedly with 20 mM Tris-HCl, pH 7.5.

Characterization of Cross-linked PI-PLC—The effect of cross-linking on secondary structure of PI-PLC was assessed by circular dichroism as described previously (10, 12). Separation of cross-linked dimer from intramolecularly cross-linked PI-PLC was attempted with gel filtration using a 10 x 110-mm column of Sephadex G-50 equilibrated with 10 mM Tris-HCl, pH 7.5. Two-dimensional PAGE (20, 21) was used to assess heterogeneity of cross-linked protein preparations. The isoelectric focusing dimension was done with the lyophilized sample rehydrated either with 190 µl of denaturant (9 M urea) along with 4% CHAPS, 15 mM dithiothreitol, and 0.2% Biolytes or with 190 µl of distilled water. Each sample was loaded onto the middle of a 11-cm immobilized pH gradient strip (pH gradient 3-10) that was then covered with 1 ml of mineral oil. The strip was placed in the isoelectric focusing cell and focused at 20 °C. The strip was then equilibrated with buffer containing 6 M urea, 2% SDS, 0.375 M Tris-HCl, pH 8.8, and dithiothreitol or {beta}-mercaptoethanol and then placed on a precast two-dimensional gel. The SDS-PAGE dimension was run essentially the same as for conventional one-dimensional gels.

Turbidity Assay for PI Cleavage—DAG produced from PI cleavage is poorly soluble in a water plus 30% iPrOH cosolvent system previously used to monitor PI-PLC activity (22). When more than 15% DAG is generated, the DAG begins to form microemulsions and partially phase-separates. This dramatically increases the optical density of the solution, here measured at 460 nm. Although absolute specific activities cannot be obtained, this assay is quick and very useful for screening PI-PLC activity for changes induced by cross-linking. Assay conditions included 1.5 mM PI, 30% iPrOH, 50 mM Tris HCl, pH 7.0, and 0.1 µg of PI-PLC. Typically 2-10 µl of the enzyme mixture from the cross-linking reaction was added to 1 ml of the assay mixture. The iPrOH cosolvent assay system solubilizes the small amount of added vesicles or micelles and dilutes any PC or PMe species present in the preincubation mixture to below 0.02 mM. This amount of diC7PC (or dC6PMe) has previously been shown to have no effect on PI cleavage by PI-PLC (8).

31P NMR Assays—31P (202.3 MHz) NMR spectra were acquired for two different assays: (i) an activity assay monitoring PI cleavage (phosphotransferase reaction) and cIP hydrolysis (cyclic phosphodiesterase reaction) as described previously (4, 22) and (ii) a binding assay where the 31P line width of the ligand monitors PI-PLC interactions (5). A Varian INOVA 500 spectrometer was used for both types of experiments. Two different PI assay mixtures were used to monitor PI cleavage: (i) 4 mM PI, 30% iPrOH, 50 mM Tris-HCl, pH 7.5, and (ii) 8 mM PI, 32 mM diC7PC, 0.05 M MES, pH 7.5. For cIP hydrolysis, 8 mM cIP was mixed with 5 mM diC7PC in the Tris buffer. Amounts of enzyme ranged from 0.1 to 10 µg, depending on the reaction monitored. The amount of phospholipids introduced (<20 µM) when adding the enzyme to each assay mix was well below the threshold needed to activate the enzyme in the case of the cosolvent system and minimal compared with the 32 mM diC7PC used in the second PI cleavage assay system. diC7PC was included in the reaction mixture with cIP, since this activator optimally enhances enzyme activity (4, 8) and since it would mask any effect of the small amount of residual PC from the cross-linking mixtures.

Measurements of the line width of the diC7PC 31P resonance in the presence of PI-PLC were carried out with samples containing 3 mg/ml (0.085 mM) enzyme in 0.05 M HEPES with 1 mM EDTA, pH 7 (5). diC7PC was titrated into the enzyme solution in the NMR tube, and the line width was measured at 28 °C; the number of transients varied (from 300 to 3000), depending on the concentration of diC7PC. A detailed analysis extracting binding constants is problematic, given the probability that (i) the enzyme lowers the critical micellar concentration of the short-chain PC and/or (ii) the exchange rate can vary from an intermediate to fast regime, depending on the number of ligand sites on the enzyme and relative affinities (e.g. the PC could form different types of aggregates with the enzyme depending on the ratio of PC to PI-PLC, and the ligand could have different off rates). Nonetheless, the large increase in line width can be used to compare PI-PLC mutants or, in this case, cross-linked enzyme to see how bound PC dynamics are affected.

Mass Spectrometry—MALDI-TOF mass spectrometry was performed on a Micromass TOFSpec-2E using a sinapinic acid matrix in linear positive mode. Most protein samples analyzed by mass spectrometry were extensively dialyzed against 10 mM Tris HCl, pH 7.5, and then concentrated to 20-25 µM using Microcon microconcentrators (10-kDa mass cut-off). The protein samples were analyzed by MALDI-TOF MS using a Micromass TofSpec-2E mass spectrometer operating in positive mode and using a sinapinic acid matrix. Spectra were calibrated using standard solutions of cytochrome c (12,360 Da), trypsinogen (23,980 Da), and/or bovine serum albumin (66,430 Da). A few protein samples were dialyzed at a much more basic pH (9.0) to remove any tightly bound phospholipids, since at this pH there is little partitioning of PI-PLC to PC or PMe vesicles (12). For this size protein, the resolution obtained by the particular MALDI-TOF mass spectrometer used was ±150 Da.

Intrinsic Fluorescence—All fluorescence measurements, obtained with a SHIMADZU RF5000U spectrofluorimeter, were carried out at 25 °C with ~2 µM protein in 50 mM HEPES, pH 7.5, with 1 mM EDTA, using an excitation wavelength of 290 nm and 5-nm excitation and emission slit widths (8, 10). The wavelength maximum for PI-PLC fluorescence, 337 nm, was unaltered for wild type and cross-linked proteins both in the absence and presence of ligands. There was no detectable light scattering for any of the protein samples with PC micelles added. Changes in the fluorescence intensity at 337 nm were expressed as (I - Io)/Io, where Io is the intensity of protein alone, and I is the intensity in the presence of an additive.

Vesicle Binding Assay—Binding of PI-PLC before and after cross-linking to POPC and DOPMe SUVs was carried out using a centrifugation/filtration assay described previously for this system (12). Phospholipid concentrations of 0.01, 0.02, 0.05, 0.1, and 0.2 mM were used in the binding assay with 35 µg of PI-PLC in a total volume of 2 ml.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
EDC Cross-linking of PI-PLC: Intra- and Intermolecular Species—EDC leads to amide formation between nearby amino and carboxyl groups on the surface of proteins and can be used to detect oligomerization of a protein. PI-PLC alone in solution (no phospholipid interface) appears mostly as a monomer as judged by gel filtration with moderate salt in the buffer (5). In keeping with this, EDC cross-linking of the protein at pH 7 trapped 15% dimers under these conditions. The major protein species (49%) generated by EDC cross-linking in the absence of any phospholipids was an intramolecularly cross-linked protein with greater mobility on SDS-PAGE (Fig. 2, lanes 2 and 3), reflecting its more compact nature as compared with uncross-linked, denatured PI-PLC. Similar anomalous migration has been seen with proteins linked by intramolecular disulfide binds (23, 24). The cross-linked protein was run on two-dimensional PAGE to try to detect how many different species were formed. There was a spread in the pI values for the intramolecularly cross-linked and normally migrating PI-PLC monomers when the isoelectric focusing was done in the presence of urea (the pI values for the major spots were between 6.9 and 9.6), indicating heterogeneity in the cross-linking reaction (Fig. 3A). Since EDC has no binding specificity for its targets but will only covalently link any spatially close residues, and since there are so many surface acidic and basic residue pairs (Table I), it is not unreasonable that there was a range of pI species. For comparison, the predicted pI for denatured B. thuringiensis PI-PLC is 5.9; this was verified by two-dimensional PAGE of uncross-linked material (data not shown). If the cross-linked protein was run in the first dimension without urea, a way of assessing the overall surface charge of the protein, one major spot was observed at both monomer molecular masses, consistent with an apparent pI of 6.9 ± 0.1 (Fig. 3B). This suggests that the either the surface charge is similar for all cross-linked species or that the protein now forms aggregates and focuses on the nondenaturing gel as a unit.



View larger version (69K):
[in this window]
[in a new window]
  		FIG. 2.
SDS-PAGE analysis of EDC cross-linked B. thuringiensis PI-PLC (100 µg/ml). Lane 1, molecular mass standards. Shown are PI-PLC cross-linked at pH 5.0 (lane 2) and at pH 7.0 (lane 3); PI-PLC cross-linked at pH 5.0 (lane 4) and at pH 7.0 (lane 5) in the presence of 0.5 mM POPC SUVs; and PI-PLC cross-linked in the presence of 4 mM diC7PC at pH 5.0 (lane 6), pH 6.0 (lane 7) and pH 8.0 (lane 8). The asterisk indicates the intramolecular cross-linked species with an altered mobility on the gel. S, protein monomer; S2, protein dimer. Higher order species can also be seen in the lanes with PC present for cross-linking.

 



View larger version (49K):
[in this window]
[in a new window]
  		FIG. 3.
Two-dimensional PAGE for PI-PLC samples cross-linked in the absence of diC7PC and run with urea in the first dimension (A), the same cross-linked sample run without urea in the isolectric focusing dimension (B), and PI-PLC cross-linked in the presence of 10 mM diC7PC and run without urea in the first dimension (C).

 
Cross-linking of PI-PLC in the presence of PC micelles (4 mM diC7PC; the critical micellar concentration of diC7PC is 1.5 mM (25)) or vesicles (0.5 mM POPC SUVs) still produced the intramolecular cross-linked PI-PLC (~40% of the total protein regardless of the pH of the cross-linking reaction); however, more protein aggregates were observed than in the EDC control (Table II). The pH of the mixture cross-linked had a significant effect on multimer formation with more dimers and higher order aggregates at pH 5 than at pH 7 or 8. This is consistent with previous work that showed that binding of PI-PLC to PC SUVs is pH-dependent, with a pKa of 8.4 (12). The formation of covalently linked PI-PLC multimers was greater for enzyme binding to PC vesicles (Fig. 2, lanes 4 and 5; 54% at pH 5 and 31% at pH 7 (Table II)) than to diC7PC micelles (36% at pH 5 and 19% at pH 8). To determine whether a PC surface was required for PI-PLC aggregation, cross-linking of the protein was carried out in the presence of 4 mM diC6PC. This concentration of diC6PC by itself would be monomeric, since its critical micellar concentration is 14 mM (25). Cross-linking of PI-PLC in the presence of diC6PC monomers (4 mM) led to ~15% covalently linked protein aggregates (all dimer and no higher order species), comparable with the control without any PC. Similarly, cross-linking of PI-PLC in the presence of diC6PMe, an inhibitory phospholipid that can bind to the active site (8, 25), at a concentration (4 mM) below its critical micellar concentration generated a gel pattern identical to the control with no increase in cross-linked species. Therefore, it is the presence of an interface that promotes the aggregation of PI-PLC that can be trapped covalently by EDC.


View this table:
[in this window]
[in a new window]
  		TABLE II
Distribution of cross-linked species (percentage of total protein on SDS-PAGE) for B. thuringiensis PI-PLC cross-linked by EDC in the absence and presence of PC interfaces

S* and S are monomers, with the former the intramolecularly cross-linked monomer with abnormal mobility. S2 is cross-linked dimer, and S3 + S4 +... is the fraction of trimer and higher order species.

 
Preparations of PI-PLC cross-linked at pH 7 in the absence and presence of diC7PC were examined by CD to see whether there was any substantial change in secondary structure. The far UV spectra were essentially superimposable, indicating that secondary structure was not significantly affected by cross-linking. However, PI-PLC cross-linked even in the absence of PC was more prone to aggregation. Gel filtration on a Sephadex G-50 column (in an attempt to separate monomer from covalently linked dimer) yielded ~60% of the protein in the void volume. SDS-PAGE analysis showed that this material had cross-linked dimer and monomer species, suggesting that the protein was aggregating nonspecifically. The mass of the material in the void volume was not >100 kDa, since centrifugation of the protein through a 100-kDa cut-off filter, as done in vesicle binding assays (12), caused >95% of all of the protein to elute in the filtrate. Therefore, whereas aggregates of the enzyme clearly formed in solution, most were not trapped by EDC.

Protein preparations cross-linked with diC7PC present and then incubated in urea prior to two-dimensional electrophoresis exhibited a similar range of pI values (hence similar heterogeneity) to protein cross-linked without the PC present. When the first dimension in these gels was run without the urea, the apparent pI values for the protein cross-linked with 2 mM diC7PC, a concentration close to its critical micellar concentration, were 6.8-7.0, similar to what was observed for protein cross-linked without PC (Fig. 3B). However, increasing the diC7PC in the cross-linking mixture (to 10 mM) caused a shift in the apparent pI determined without urea to 5.9-6.5 (Fig. 3C). This suggests that diC7PC is still associated with the protein during the focusing, and its binding alters the surface charge of the PI-PLC complexes.

Cross-linked PI-PLC Specific Activity—The increased fraction of EDC-trapped PI-PLC aggregates at PC surfaces could result from concentrating the protein on the two-dimensional surface compared with its bulk concentration in bulk solution, or it might reflect a more specific aggregation promoted by PC surfaces only. If the latter effect describes the behavior of PI-PLC at lipid interfaces, then multimer formation might contribute to the kinetic activation of PI-PLC observed in the presence of PC surfaces.

A turbidity assay was used as an initial screen for effects of cross-linking on PI cleavage by PI-PLC (Fig. 4). DAG generated by PI-PLC has limited solubility in H2O/diC7PC and H2O/iPrOH mixtures and leads to microemulsion and partial phase separation that increases the solution optical density. With this turbidity assay, PI-PLC cross-linked in the presence of PC interfaces was always more active (i.e. more rapid increase in turbidity) than control PI-PLC (x in Fig. 4). No enhancement of enzymatic activity was observed if the enzyme was preincubated with 4 mM diC6PC or diC7PC (without the addition of EDC) and then assayed for PI cleavage by monitoring turbidity. Although PI-PLC cross-linked in the presence of POPC exhibited roughly 50% more aggregated protein than that cross-linked in the presence of diC7PC micelles or diC6PC monomers (Fig. 2, compare lanes 5 and 7), the enzymatic specific activity was not as high as for the enzyme that had been cross-linked in the presence of the two short-chain PC species as judged by the turbidity assay (Fig. 4, compare {blacktriangledown} and for cross-linking with POPC or diC7PC, respectively). Thus, the enhanced activity of PI-PLC cross-linked in the presence of a PC interface) does not directly correlate with the presence of more covalently linked PI-PLC aggregates.



View larger version (25K):
[in this window]
[in a new window]
  		FIG. 4.
Optical density at 460 nm as a function of time after the addition of 0.1 µg of PI-PLC to 1 ml of 1.5 mM PI, 30% iPrOH, 50 mM Tris HCl, pH 7.0. x, enzyme alone; {blacktriangleup}, EDC; {blacksquare}, EDC + diC6PC; , EDC + diC7PC; {blacktriangledown}, EDC + POPC SUVs; {diamond}, EDC + diC6PMe; {circ}, diC6PC; {square}, diC6PMe; {diamondsuit}, DOPMe SUVs.

 
In contrast to the results with PC, cross-linking of PI-PLC in the presence of PMe monomers and micelles generated enzyme that was significantly inhibited compared with uncross-linked PI-PLC (Fig. 4). If PI-PLC was incubated with diC6PMe (4 mM) or DOPMe (2 mM) (inhibitory lipids to which the enzyme binds tightly (5, 6, 12, 26)) in the absence of EDC, enzyme activity was also reduced but not as dramatically as that seen when the enzyme was also incubated with EDC (compare {square} and {diamond} for PI-PLC incubated with monomeric diC6PMe in the absence and presence of EDC).

More quantitative assays of PI-PLC activity, chemically cross-linked in the absence or presence of the best activator, diC7PC, toward PI solubilized in iPrOH (22) and in diC7PC (4), both conditions that exhibit optimal enzyme activity, were carried out using 31P NMR spectroscopy to monitor cIP production. PI-PLC cross-linked in the absence of diC7PC showed a higher specific activity compared with uncross-linked protein (1.4-1.5-fold increase). However, PI-PLC cross-linked in the presence of PC surfaces exhibited a specific activity that was ~1.8-fold higher toward PI than the specific activity of PI-PLC cross-linked without PC present. This represents a 2.4-2.8-fold higher activity than uncross-linked enzyme comparing activities observed in PI/iPrOH and PI/diC7PC assay systems (Table III). The enhanced activity of PI-PLC cross-linked in the presence of diC7PC was not due to preincubation of the protein with that lipid, since the phosphotransferase activity of PI-PLC incubated with diC7PC (4 mM) in the absence of EDC was comparable with that of uncross-linked PI-PLC (Table III). If cross-linking was carried out at pH 5, where B. thuringiensis PI-PLC has a higher affinity for PC surfaces but lower activity (12), the enzyme was trapped in a low activity form with only 4% of the activity of the protein observed when it was assayed at pH 7. Nonetheless, cross-linking of PI-PLC in the presence of diC7PC at this lower pH showed the same extent of activation (2-fold increase with both PI assay systems) as observed for protein cross-linked at pH 7.


View this table:
[in this window]
[in a new window]
  		TABLE III
Phosphotransferase (PI cleavage) and cyclic phosphodiesterase (cIP hydrolysis) activities of EDC-cross-linked PI-PLC

 
In contrast to the activation observed for cross-linked PIPLC cleavage of PI, cIP hydrolysis was significantly inhibited (30-50% of the uncross-linked control activity) when the enzyme was cross-linked in the presence of diC7PC (Table III).

Is the enhanced PC activation of PI cleavage caused by aggregation of the protein after cross-linking? To test this, we examined the specific activity of the cross-linked protein that was eluted in the void volume from chromatography on the Sephadex G-50 resin. The specific activity of the mixture in the void volume toward PI/diC7PC was 52% of the specific activity of the protein prior to gel filtration. Unfortunately, the protein retained on the gel filtration column was not eluted where expected but instead eluted over a wide range with molecular masses smaller than that of the protein, making an accurate measurement of its specific activity impossible. Such aberrant behavior on gel filtration has previously been seen for this PI-PLC (13) as well as for the enzyme from Listeria monocytogenes (27) and is thought to reflect interactions of the protein with the resin. The observation that the material in the void volume was not the more active fraction suggests that the aggregation alone does not activate the protein. Enhanced PI cleavage by cross-linked PI-PLC appears less the result of enzyme aggregation than of trapping the enzyme in an altered conformation of the protein that is stabilized for PI cleavage (and not cIP hydrolysis, since that activity is inhibited by cross-linking (Table III)).

Cross-linking of Interfacial Activation-impaired Mutants of PI-PLC—The kinetic activation of PI-PLC by PC surfaces (vesicles and micelles) requires both rim tryptophan residues (10, 11). Both W47A and W242A exhibit impaired binding to diC7PC micelles, although at least W47A still binds as judged by the increased intrinsic fluorescence of the protein in the presence of diC7PC micelles (10). These two mutants are not activated by PC interfaces to the same extent as wild type enzyme. We can evaluate how those mutants behave when cross-linked by EDC as a clue to understanding what conformation of the enzyme is stabilized by cross-linking. When W47A was cross-linked at pH 7, the dimer represented 8% of the total PI-PLC; this was only marginally changed (to 9%) if the cross-linking occurred with diC7PC present (Fig. 5). Cross-linking at pH 5 enhanced dimer formation (to 16%), but again the presence of diC7PC had only a small effect on dimer formation (increased to 19% of total protein). The specific activity of W47A increased 1.5-fold when it was cross-linked at pH 7 without the diC7PC present. The fractional increase in activity was comparable with what was observed with native protein cross-linked without diC7PC (Table IV). Cross-linking in the presence (2 and 10 mM) of diC7PC only increased the activity 1.7-fold compared with the uncross-linked enzyme when the PI/diC7PC assay system was used. In other words, the enhanced activation observed upon cross-linking wild type enzyme with diC7PC present (2.4-fold) was now significantly reduced. W47A specific activity toward PI solubilized in iPrOH was much lower than the activity detected with the PI/diC7PC system, consistent with that specific tryptophan involved in the conformational change promoted by iPrOH (10). When the cross-linking of W47A was carried out in the presence of diC7PC (2 mM), an amount around the apparent Kd for PC vesicle interfaces (10), and then the enzyme was assayed toward PI/iPrOH, the specific activity increased 2.1-2.2-fold, (Table IV) an activation close to the 2.8-fold activation for wild type PI-PLC cross-linked with diC7PC present and then assayed toward PI in 30% iPrOH. Thus, a larger activation is seen for assaying the protein toward PI plus 30% iPrOH than toward PI/diC7PC, consistent with iPrOH and diC7PC affecting enzyme activity by different pathways.



View larger version (66K):
[in this window]
[in a new window]
  		FIG. 5.
SDS-PAGE analysis of B. thuringiensis PI-PLC mutants cross-linked by EDC in the absence and presence of 2 mM diC7PC at pH 7.0. Lane 1, molecular mass standards; lane 2, W242A + EDC; lane 3, W242A + EDC + diC7PC; lane 4, W47A + EDC; lane 5, W47A + EDC + diC7PC; lane 6, W47A (no cross-linking). Note that with the overloading of protein in order to see the dimer, the lower molecular mass intramolecularly cross-linked material is not as distinct as in Fig. 1.

 


View this table:
[in this window]
[in a new window]
  		TABLE IV
Specific activities of native and single tryptophan mutants of PI-PLC cross-linked by EDC in the absence and presence of diC7PC

 
W242A has an even weaker affinity for PC surfaces but shows higher activities than W47A for PI in a water-iPrOH cosolvent assay system (10). Cross-linking of the protein in the absence of diC7PC led to a 1.5-fold increase in activity, as observed for both the native and W47A proteins (Table IV). When W242A was cross-linked at pH 7 in the presence of 2 mM diC7PC, a band for the dimer was not detected (Fig. 5, lane 3), consistent with the dramatically increased Kd for PC for this mutant (10). Enzyme cross-linked with diC7PC present exhibited a 1.8-fold increased specific activity compared with uncross-linked PI-PLC assayed toward PI in iPrOH and only 1.6-fold when assayed toward PI solubilized in diC7PC mixed micelles (Table IV). Thus, the increased specific activity for protein cross-linked in the presence of diC7PC requires the enzyme to bind to micelles. If cross-linking was done at pH 5, a pH that should strengthen binding of PI-PLC to PC surfaces (12), then the enhancement of activity upon cross-linking with diC7PC was increased to 2-fold in the PI/iPrOH cosolvent system.

Clearly, EDC cross-linking of wild type PI-PLC in the presence of diC7PC, which formed mostly intramolecular cross-links (~40%) but some dimers (~20%) and higher order aggregates (<=10% and depending on pH) as well, stabilized a conformation where the enzyme was more active toward PI solubilized in diC7PC or iPrOH. The reduced rate enhancement for the two rim tryptophan single mutants cross-linked in the presence of diC7PC and then assayed toward PI/diC7PC confirms that these two tryptophans do play a role in PC activation of the enzyme and are critical for promoting the conformation of the enzyme that is stabilized by EDC cross-linking. For both mutants, the enhanced activity in the PI/iPrOH assay system after cross-linking with diC7PC suggests that cosolvent activation can compensate to some degree for the impaired interfacial activation of these tryptophan mutants.

Binding of Cross-linked PI-PLC to Interfaces—The results presented thus far are consistent with EDC cross-linking stabilizing a conformation of the enzyme that is more sensitive to kinetic activation by diC7PC if the cross-linking is carried out with diC7PC present. The enhanced kinetic activation could reflect altered interactions of the enzyme with interfaces. Several methods have been particularly useful in monitoring the interaction of this enzyme with interfaces: (i) an increase in the intrinsic fluorescence of the protein when it is bound to interfaces (5, 6, 26), (ii) partitioning of the protein to PC SUVs (10, 12), and (iii) enzyme-induced increases in the 31P line width of diC7PC (5).

The intrinsic fluorescence of the native protein cross-linked in the absence of an interface was examined after dialysis to remove EDC. The addition of micellar diC7PC did not enhance the intrinsic fluorescence of cross-linked PI-PLC (Fig. 6), although there was a small but reproducible increase around the critical micellar concentration of the short-chain PC. This suggests that in the conformation of the enzyme trapped by EDC, Trp-242, the major fluorophore responsible for the interface-induced fluorescence change (10), is no longer very sensitive to the addition of an interface.



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 6.
Intrinsic fluorescence of PI PLC (0.07 mg/ml) as a function of added diC7PC. , uncross-linked protein; {circ}, PI-PLC cross-linked by EDC.

 
Cross-linking of the protein in the absence of an interface also affected its ability to partition to both POPC and DOPMe SUVs. What becomes apparent in Fig. 7 is that only a fraction of both monomer and dimer forms of the cross-linked PI-PLC bind tightly to PC vesicles (using a Langmuir analysis (12), nKd = 9 and 4 µM for monomeric and dimeric PI-PLC forms). Very little (<17%) of the cross-linked monomeric protein partitioned to POPC SUVs, whereas closer to 50% of the small amount of dimer was bound to PC vesicles upon "saturation." In contrast, all of the protein partitioned to DOPMe SUVs, but with an nKd of 69 µM, a value considerably larger than what is observed for uncross-linked PI-PLC (nKd = 1-2 µM (12)). The perturbations in cross-linked PI-PLC binding to anionic vesicles probably reflect alterations in key charged residues, since a similar result was observed with R69D mutant (12). Although fairly inactive, R69D exhibited an increased affinity for PC and decreased affinity for DOPMe SUVs. The net result of these binding studies is that cross-linking generates a heterogeneous population of PI-PLC, only a fraction of which can bind tightly to PC vesicles. However, the cross-linked protein that does bind does so with higher affinity. Furthermore, cross-linking reduces the affinity of the enzyme for anionic surfaces. Can we determine what has changed upon cross-linking that activates PI-PLC?



View larger version (13K):
[in this window]
[in a new window]
  		FIG. 7.
Partitioning of EDC-cross-linked PI-PLC (0.1 mg/ml) onto POPC SUVs (, monomer; {blacksquare}, cross-linked dimer) and DOPMe (x, monomer) as a function of bulk phospholipid concentration. The binding assays were carried out at pH 7.5.

 
Previously, we saw that when an excess of diC7PC was incubated with PI-PLC, there was a large change in the 31P line width of the phospholipid consistent with intermediate exchange between PC in different environments (e.g. different PI-PLC/diC7PC aggregates as well as free diC7PC monomers and micelles) and/or generation of large particles (5). As shown in Fig. 8 using the recombinant PI-PLC and conditions slightly different from those used previously (5), the increase in diC7PC line width caused by recombinant PI-PLC was most pronounced with micellar diC7PC. If one assumes fast exchange for the free and enzyme-bound diC7PC at phospholipid concentrations above 2.5 mM, a bound line width of 900 ± 25 Hz is obtained. This is much too large for particles of <100 kDa being produced and strongly indicates that most of the line width is due to intermediate exchange conditions.



View larger version (14K):
[in this window]
[in a new window]
  		FIG. 8.
31P line width (Hz) of diC7PC (in 50 mM HEPES, pH 7) incubated with 0.086 mM PI-PLC (), or 0.086 PI-PLC cross-linked by EDC in the presence of 4 mM diC7PC at pH 7 and then dialyzed to remove excess EDC and diC7PC ({circ}). As a control, the line width for PC alone is shown as a function of lipid concentration (x). The arrow indicates the critical micellar concentration of pure diC7PC.

 
More importantly, when the same experiment was carried out with PI-PLC that had been cross-linked in the presence of diC7PC (5 mM) and then concentrated and washed to exchange out any PC not tightly bound to the protein, the large increase in line width due to binding to PI-PLC was no longer observed (Fig. 8). Indeed, except for points below 2 mM, the observed line width was only 1-1.5 Hz larger than the line width of diC7PC alone. Either (i) the PC exchange rate between cross-linked PI-PLC-bound and free environments has been decreased such that it is now in the slow exchange regime or (ii) the cross-linked enzyme does not bind tightly to diC7PC micelles. Since cross-linked enzyme is still activated by PC micelles (and shows a greater rate enhancement than normal recombinant enzyme), the PI-PLC must interact with diC7PC molecules. Furthermore, even if one assumes that only a fraction of the cross-linked protein binds to micelles, say 10% as in the monomer vesicle binding assays, a large line width change should still have been observed around the critical micellar concentration of diC7PC if the exchange rate was not appreciably affected. Therefore, EDC cross-linking of PI-PLC in the presence of diC7PC micelles enhances binding of that amphiphile to the enzyme such that the exchange rate between free diC7PC molecules (both monomers and micelles) and those bound to cross-linked PI-PLC has been reduced significantly.

MS Analysis of Cross-linked PI-PLC—Assuming that PIPLC protein exists as a equilibrium mixture of different conformations with varying activities, cross-linking by EDC in the presence of diC7PC micelles would appear to shift the equilibrium toward a species that is more active, possibly by binding the activator more tightly. EDC cannot covalently link the diC7PC to the protein. However, if it cross-links a state of the enzyme that has an increased affinity for PC, it may be possible to detect the presence of the bound diC7PC by MALDI-TOF mass spectrometry. Therefore, we used this technique to quantify any change in mass of different species of PI-PLC resulting from chemical cross-linking by EDC. This type of experiment is best done with diC7PC micelles rather than POPC SUVs, since molecules that are not tightly bound can be dialyzed away from the protein. Samples were incubated for 2-3 h in the absence or presence of EDC and diC7PC. After cross-linking but before MALDI-TOF analysis, the samples were dialyzed in 10 mM Tris, pH 7.5, and concentrated to 20-25 µM using Centricon filters (10-kDa mass cut-off). The spectrum of uncross-linked PI-PLC exhibited a major peak at 34.8 kDa (close to the expected mass of 34.6 kDa) and two minor peaks at 17.5 kDa, due to protein molecules with +2 charge, and at 69.4 kDa, representing a small amount of protein dimer (Fig. 9A). PI-PLC incubated with phospholipids in the absence of EDC and then dialyzed and analyzed by MS did not show evidence of larger aggregate formation, any perturbation in the ratio of species detected, or any changes in the masses of the species detected. The mass spectrum of PI-PLC incubated in the presence of EDC and the absence of any lipids exhibited an increase in multimers, notably peaks for trimer and a small amount of tetramer as well as monomer and dimer (Fig. 9B), indicating that the cross-linking did in fact trap protein aggregates. The distribution of peaks was reasonably consistent with the SDS-PAGE analysis of protein cross-linked by EDC. PI-PLC cross-linked in the presence of diC7PC (4 mM) formed aggregates in more or less the same ratios as for PI-PLC cross-linked without PC (again consistent with SDS-PAGE analysis). However, the molecular mass of the monomer peak was increased ~1 kDa compared with the PI-PLC cross-linked in the absence of PC (Table V). Similarly, the masses for the dimer, trimer, and tetramer increased 2.2, 2.9, and 4 kDa. The average increase in mass for a PI-PLC molecule was 1.0 kDa. This increase in mass, well outside the range of experimental errors (±0.15 kDa) in determining the mass of PI-PLC, is equivalent to 2.1 ± 0.1 mol of diC7PC (481.6 Da/PC) tightly bound on average per 1 mol of protein.



View larger version (12K):
[in this window]
[in a new window]
  		FIG. 9.
MALDI-TOF spectra of recombinant PI-PLC (A) and PI-PLC cross-linked with EDC (B).

 


View this table:
[in this window]
[in a new window]
  		TABLE V
Change in PI-PLC molecular mass and specific activity with EDC cross-linking

 
Previous 31P NMR (5) and fluorescence titration (13) studies of monomeric diC7PC binding to native (uncross-linked) PIPLC were consistent with two molecules of PC binding to the enzyme with Kd of ~0.2 mM. If the Kd for diC7PC binding to PI-PLC in this EDC-stabilized conformation were similar, the bound lipid molecules would have easily been removed by dialysis. However, if Kd is less than 1 µM, it would be difficult to remove the PC just by the dialysis conditions used. Extensive dialysis of PI-PLC cross-linked with diC7PC in 20 mM Tris, pH 9.0 (a pH where binding of PI-PLC to PC surfaces is weakened (12)) led to a PI-PLC monomer mass within 0.1 kDa of that for PI-PLC cross-linked without diC7PC. If the protein was first cross-linked with EDC and then incubated with diC7PC, the mass of the protein was also unaltered, indicating that under these conditions the diC7PC does not bind tightly to the previously cross-linked protein. Thus, EDC cross-linking of the protein in the presence of diC7PC trapped the enzyme in a conformation where it has a higher affinity for individual diC7PC molecules. This state with a higher affinity for a discrete number of PC molecules correlates with the enhanced kinetic activation.

MALDI-TOF MS analyses of cross-linked W47A and W242A were also carried out. In contrast to wild type PI-PLC protein, cross-linking of these mutants in the presence of diC7PC led to a smaller increase in protein mass (Table V). At pH 7, both Trp mutants had <0.7 diC7PC tightly associated with them; this increased to close to one dC7PC for cross-linking at pH 5. Thus, each of these two rim tryptophan mutants has lost a tight diC7PC binding site. Loss of either site reduced the enzymatic activity of the cross-linked material and masked kinetic activation of the enzyme in the PI/diC7PC but not the PI/iPrOH cosolvent assay system.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
B. thuringiensis and the closely related B. cereus PI-PLC are interfacially active enzymes, where the activity is enhanced when the substrate PI is present in an interface compared with PI monomers (28, 29). The second step of the reaction, hydrolysis of cIP to I-1-P, can also be affected by nonsubstrate interfaces (4, 5, 8). PC surfaces enhance cIP hydrolysis, whereas anionic surfaces inhibit enzyme activity (by acting as competitive inhibitors). Similar effects are also observed for the phosphotransferase activity of PI-PLC using long-chain and short-chain PI (6, 12) as well as monomeric nonlipid substrates (9). It has been documented that several peripheral membrane proteins oligomerize when bound to membranes (e.g. Streptomyces chromofuscus phospholipase D (14) and annexin V (15)), and this may be a means of controlling their biological activity.

In the case of PI-PLC, a possible mechanism for the PC interfacial activation is that the enzyme specifically oligomerizes at the lipid surface, and the aggregate has higher activity than monomeric PI-PLC. EDC covalently links spatially close amino and carboxyl groups of proteins. It did trap some PI-PLC dimers in the absence of PC surfaces and an increased proportion of dimers, trimers, and higher order species when the cross-linking reaction was carried out with PI-PLC bound to PC bilayer surfaces. However, intramolecularly cross-linked protein with a slightly altered mobility was also observed and was the major species trapped. Inspection of the B. cereus PI-PLC structure indicates that an intramolecular cross-link between the side chains of Glu-52 and Lys-279 would tether both ends of the molecule. In combination with other intramolecular cross-links (such as Glu-93 and Lys-38), these new amide bonds would give the protein a more compact structure with electro-phoretic mobility consistent with a lower molecular mass protein.

Although the cross-linking reaction yielded heterogeneous products in that a given PI-PLC preparation showed a range of pI values when examined with two-dimensional denaturing gels, all of the wild type PI-PLC samples cross-linked in the presence of diC7PC interfaces showed an enhanced ability to cleave PI in two very different assay systems: PI solubilized in diC7PC and PI dispersed in 30% iPrOH. Furthermore, the extent of the activation was similar for many different cross-linked protein samples, indicating that heterogeneity in cross-linking did not necessarily translate to altered enzymatic activity. This is probably because most of the acidic and basic residues are well removed from the active and interfacial binding sites. Three possible factors could account for this increased activity: (i) PI-PLC multimers are more active than monomers; (ii) the intramolecularly cross-linked PIPLC species is trapped in a more active conformation than uncross-linked PI-PLC; or (iii) a reduction of charged groups on the protein surface enhances interaction of the protein with activating interfaces. POPC vesicles were particularly effective at enhancing oligomerization of the enzyme compared with diC7PC micelles. However, the presence of a high proportion of cross-linked dimers, trimers, etc., did not correlate with enhanced PI-PLC activity, since enzyme activity was higher with enzyme cross-linked in the presence of diC7PC micelles. Furthermore, the two-dimensional PAGE profile using denaturing conditions in the first dimensions for PI-PLC cross-linked in the absence and presence of diC7PC looked essentially the same. These results suggest that an altered distribution of oligomers is not likely to be what is modulating enzyme activity. When the first dimension was run without denaturant, the pI of the major spot decreased significantly in the presence of micellar diC7PC. The shift in the pI suggests that micellar PC binding reduces the surface charge of the complex and that all cross-linked species exhibit similar behavior (hence the relatively constant extent of kinetic activation although there is cross-linking heterogeneity).

What can we infer about the conformation of the more active PI-PLC cross-linked with diC7PC? Changes in the mass of PI-PLC cross-linked in the presence of diC7PC indicate that on average two diC7PC molecules are tightly bound. These phospholipid molecules can be removed by dialysis under basic conditions but are tightly bound at pH values below 7.5. There is essentially no change in secondary structure upon cross-linking with or without diC7PC, so that the changes must involve positioning of side chain residues around the PC molecules in the activator site.

The same conformation of PI-PLC that exhibits enhanced phosphotransferase activity shows inhibited diC7PC activation of cIP hydrolysis. diC7PC clearly activates both steps of PI cleavage, since under these conditions with uncross-linked PIPLC, cIP hydrolysis would be decreased about 20-fold without the diC7PC present (4)). However, the EDC cross-linking reduced the magnitude of diC7PC activation for cIP. Why this occurs is not clear, but it might suggest a slightly different kinetic pathway/orientation of active site residues than just the reverse of the phosphotransferase reaction as has been suggested by others (2, 17, 30).

The two rim tryptophan mutants that are impaired in interfacial activation by diC7PC (10) show enhanced activity upon EDC cross-linking in the presence of diC7PC but not to the same extent as wild type protein. The lower extent of activation correlates with the detection of one rather than two diC7PC molecules tightly bound to the protein. The chemical cross-linking is only able to stabilize the binding of a single diC7PC interacting with these mutant PI-PLC enzymes, consistent with each rim tryptophan creating a tight and probably independent binding site for a single PC molecule. There is fairly extensive documentation of tryptophan residues inserting into bilayers and localizing in the head group/glycerol backbone region of the membrane (31). Perhaps the EDC cross-linking in the presence of diC7PC micelles traps most of the protein with the rim tryptophans in this orientation. The energetics for interaction of a specific PC molecule with each tryptophan could be {pi}-cation stacking as observed in peptide systems (32), but here the positively charged choline N(CH3)3 substitutes for the lysine or arginine cationic moiety. The tryptophan-diC7PC complex would then be poised for more extensive interactions with a zwitterionic surface.

Does this model for diC7PC activation of B. thuringiensis PI-PLC shed light on the mammalian PLC enzymes? PLC{delta}1 has a hydrophobic ridge with residues Leu-320, Tyr-358, and Phe-360 in a loop comparable with the 232-244 loop in the bacterial enzyme and Trp-555 in a helix that could be analogous to Trp-47 in the B. thuringiensis enzyme (2). W555A has significantly reduced catalytic activity (33) that could result from loss of interfacial interactions of the catalytic domain that are critical for optimal activity in a fashion similar to the bacterial PI-PLC. However, diC7PC micelle activation of PLC{delta}1, which increases kcat but has no effect on Km for the water-soluble substrate cIP, has been shown to depend on the presence of the N-terminal pleckstrin homology domain (7). Nonetheless, the picture of the bacterial PI-PLC binding tightly to specific membrane activator molecules via aromatic side chains in the barrel rim and these helping drive the protein to an optimized active configuration may well apply but with perhaps aromatic residues from the pleckstrin homology domain contributing to this interaction.


   FOOTNOTES
 
* This work was supported by National Institutes of Health Grant GM 60418 (to M. F. R.). The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked "advertisement" in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. Back

{ddagger} To whom correspondence should be addressed. Tel.: 617-552-3616; Fax: 617-552-2705; E-mail: mary.roberts{at}bc.edu.

1 The abbreviations used are: PI-PLC, phosphatidylinositol-specific phospholipase C; PLC{delta}1, phospholipase C{delta}1; PC, phosphatidylcholine; diC6PC, dihexanoyl-PC; diC7PC, diheptanoyl-PC; POPC, 1-palmitoyl-2-oleoyl-PC; PI, phosphatidylinositol; cIP, cyclo-1,2-phosphoinositol; PMe, phosphatidylmethanol; DOPMe, dioleoyl-PMe; diC6PMe, dihexanoyl-PMe; DAG, diacylglycerol; EDC, 1-ethyl-3-[3-dimethylaminopropyl]carbodiimide hydrochloride; iPrOH, isopropyl alcohol; MES, 2-(N-morpholino)ethanesulfonic acid; MALDI-TOF, matrix-assisted laser desorption/ionization time-of-flight; MS, mass spectrometry; SUV, small unilamellar vesicle; CHAPS, 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonic acid. Back

2 C. Zhou and M. F. Roberts, unpublished results. Back



   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Heinz, D. W., Ryan, M., Bullock, T. L., and Griffith, O. H. (1995) EMBO J. 14, 3855-3863[Medline] [Order article via Infotrieve]
  2. Griffith, O. H., and Ryan, M. (1999) Biochim. Biophys. Acta 1441, 237-254[Medline] [Order article via Infotrieve]
  3. Essen, L. O., Perisic, O., Cheung, R., Katan, M., and Williams, R. L. (1996) Nature 380, 595-602[CrossRef][Medline] [Order article via Infotrieve]
  4. Zhou, C., Wu, Y., and Roberts, M. F. (1997) Biochemistry 36, 347-355[CrossRef][Medline] [Order article via Infotrieve]
  5. Zhou, C., Qian, X., and Roberts, M. F. (1997) Biochemistry 36, 10089-10097[CrossRef][Medline] [Order article via Infotrieve]
  6. Qian, X., Zhou, C., and Roberts, M. F. (1998) Biochemistry 37, 6513-6522[CrossRef][Medline] [Order article via Infotrieve]
  7. Wu, Y., Perisic, O., Williams, R. L., Katan, M., and Roberts, M. F. (1997) Biochemistry 36, 11223-11233[CrossRef][Medline] [Order article via Infotrieve]
  8. Zhou, C., and Roberts, M. F. (1998) Biochemistry 37, 16430-16439[CrossRef][Medline] [Order article via Infotrieve]
  9. Birrell, G. B., Zaikova, T. O., Rukavishnikov, A. V., Keana, J. F., and Griffith, O. H. (2003) Biophys. J. 84, 3264-3275[CrossRef][Medline] [Order article via Infotrieve]
  10. Feng, J., Webhi, H., and Roberts, M. F. (2002) J. Biol. Chem. 277, 19867-19875[Abstract/Free Full Text]
  11. Feng, J., Bradley, W., and Roberts, M. F. (2003) J. Biol. Chem. 278, 24651-24657[Abstract/Free Full Text]
  12. Wehbi, H., Feng, J., Kolbeck, J., Ananthanarayanan, B., Cho, W., and Roberts, M. F. (2003) Biochemistry 42, 9374-9382[CrossRef][Medline] [Order article via Infotrieve]
  13. Berg, O. G., Yu, B. Z., Apitz-Castro, R. J., and Jain, M. K. (2004) Biochemistry 43, 2080-2090[CrossRef][Medline] [Order article via Infotrieve]
  14. Stieglitz, K., Seaton, B., and Roberts, M. F. (1999) J. Biol. Chem. 274, 35367-35374[Abstract/Free Full Text]
  15. Concha, N. O., Head, J. F., Kaetzel, M. A., Dedman, J. R., and Seaton, B. A. (1992) FEBS Lett. 314, 159-162[CrossRef][Medline] [Order article via Infotrieve]
  16. Yang, H., and Roberts, M. F. (2003) Protein Sci. 12, 2087-2098[CrossRef][Medline] [Order article via Infotrieve]
  17. Hondal, R. J., Riddle, S. R., Kravchuk, A. V., Zhao, Z., Liao, H., Bruzik, K. S., and Tsai, M.-D. (1997) Biochemistry 36, 6633-6642[CrossRef][Medline] [Order article via Infotrieve]
  18. Lowry, O. H., Rosebrough, N. J., Farr, A. L., and Randall, R. J. (1951) J. Biol. Chem. 193, 265-275[Free Full Text]
  19. Bradford, M. M. (1976) Anal. Biochem. 72, 248-254[CrossRef][Medline] [Order article via Infotrieve]
  20. Deutscher, M. P. (1990) Methods Enzymol. 182, 1-818[CrossRef][Medline] [Order article via Infotrieve]
  21. Fichmann, J., and Estermeier, R. (1998) Methods Mol. Biol. 112, 1-8
  22. Wehbi, H., Feng, J., and Roberts, M. F. (2003) Biochim. Biophys. Acta 1613, 15-27[Medline] [Order article via Infotrieve]
  23. Sahoo, D., Weers, P. M. M., Ryan, R. O., and Narayanaswami, V. (2002) J. Mol. Biol. 321, 201-214[CrossRef][Medline] [Order article via Infotrieve]
  24. Mitchinson, C., and Wells, J. A. (1989) Biochemistry 28, 4807-4815[CrossRef][Medline] [Order article via Infotrieve]
  25. Bian, J., and Roberts, M. F. (1992) J. Coll. Int. Sci. 153, 420-428[CrossRef]
  26. Volwerk, J. J., Shashidhar, M. S., Luppe, A., and Griffith, O. H. (1990) Biochemistry 29, 8056-8062[CrossRef][Medline] [Order article via Infotrieve]
  27. Goldfine, H., and Kolb, C. (1992) Infect. Immun. 60, 4059-4067[Abstract/Free Full Text]
  28. Lewis, K. A., Garigapati, V. R., Zhou, C., and Roberts, M. F. (1993) Biochemistry 32, 8836-8841[CrossRef][Medline] [Order article via Infotrieve]
  29. Hendrickson, H. S., Hendrickson, E. K., Johnson, J. L., Khan, T. H., and Chial, H. J. (1992) Biochemistry 31, 12169-12172[CrossRef][Medline] [Order article via Infotrieve]
  30. Hondal, R. J., Zhao, Z., Kravchuk, A. V., Liao, H., Riddle, S. R., Yue, X., Bruzik, K. S., and Tsai, M. D. (1998) Biochemistry 37, 4568-4580[CrossRef][Medline] [Order article via Infotrieve]
  31. Wallace, B. A., and Janes, J. W. (1999) Adv. Exp. Med. Biol. 467, 789-799[Medline] [Order article via Infotrieve]
  32. Tsou, L. K., Tatko, C. D., and Waters, M. L. (2002) J. Am. Chem. Soc. 124, 14917-14921[CrossRef][Medline] [Order article via Infotrieve]
  33. Ellis, M. V., James, S. R., Perisic, O., Downes, C. P., Williams, R. L., and Katan, M. (1998) J. Biol. Chem. 273, 11650-11659[Abstract/Free Full Text]

Add to CiteULike CiteULike   Add to Complore Complore   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us   Add to Digg Digg   Add to Reddit Reddit   Add to Technorati Technorati    What's this?


This article has been cited by other articles:


Home page
 J. Biol. Chem.Home page
X. Shi, C. Shao, X. Zhang, C. Zambonelli, A. G. Redfield, J. F. Head, B. A. Seaton, and M. F. Roberts
Modulation of Bacillus thuringiensis Phosphatidylinositol-specific Phospholipase C Activity by Mutations in the Putative Dimerization Interface
J. Biol. Chem., June 5, 2009; 284(23): 15607 - 15618.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
A. Jerga, D. J. Miller, S. W. White, and C. O. Rock
Molecular Determinants for Interfacial Binding and Conformational Change in a Soluble Diacylglycerol Kinase
J. Biol. Chem., March 13, 2009; 284(11): 7246 - 7254.
[Abstract] [Full Text] [PDF]

Home page
 J. Biol. Chem.Home page
C. Shao, X. Shi, H. Wehbi, C. Zambonelli, J. F. Head, B. A. Seaton, and M. F. Roberts
Dimer Structure of an Interfacially Impaired Phosphatidylinositol-specific Phospholipase C
J. Biol. Chem., March 23, 2007; 282(12): 9228 - 9235.
[Abstract] [Full Text] [PDF]

Home page
 Mol. Cell. ProteomicsHome page
B. X. Huang and H.-Y. Kim
Interdomain Conformational Changes in Akt Activation Revealed by Chemical Cross-linking and Tandem Mass Spectrometry
Mol. Cell. Proteomics, June 1, 2006; 5(6): 1045 - 1053.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow All Versions of this Article:
279/19/20490    most recent
M401016200v1
Right arrow Submit a Letter to Editor
Right arrow Alert me when this article is cited
Right arrow Alert me when eLetters are posted
Right arrow Alert me if a correction is posted
Right arrow Citation Map
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Zhang, X.
Right arrow Articles by Roberts, M. F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Zhang, X.
Right arrow Articles by Roberts, M. F.
Social Bookmarking
Add to CiteULike   Add to Complore   Add to Connotea   Add to Del.icio.us   Add to Digg   Add to Reddit   Add to Technorati  
What's this?

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
 All ASBMB Journals   Molecular and Cellular Proteomics 
 Journal of Lipid Research   ASBMB Today 
Copyright © 2004 by the American Society for Biochemistry and Molecular Biology.
Advertisement
spacer
Advertisement
Advertisement