Biochimica et Biophysica Acta 1841 (2014) 581–587

Contents lists available at ScienceDirect

Biochimica et Biophysica Acta journal homepage: www.elsevier.com/locate/bbalip

Kinetic and structural characterization of triacylglycerol lipases possessing phospholipase A1 activity Ahmed Aloulou a,⁎, Fakher Frikha a, Alexandre Noiriel b, Madiha Bou Ali a, Abdelkarim Abousalham b,⁎⁎ a

University of Sfax, ENIS, Laboratory of Biochemistry and Enzymatic Engineering of Lipases, 3038 Sfax, Tunisia CNRS Université Claude Bernard Lyon 1, ICBMS, Organization and Dynamics of Biological Membranes, UMR 5246, Bâtiment Raulin, 43, Boulevard du 11 Novembre 1918, 69622 Villeurbanne, France b

a r t i c l e

i n f o

Article history: Received 4 September 2013 Received in revised form 2 December 2013 Accepted 14 December 2013 Available online 22 December 2013 Keywords: Human pancreatic lipase Human pancreatic lipase related protein 2 Porcine pancreatic lipase Phospholipase A1 Activity Regioselectivity

a b s t r a c t The pancreatic lipase gene family displays various substrate selectivities for triglycerides and phospholipids. The structural basis for this difference in substrate specificity has not been definitively established. Based on a kinetic comparative study between various pancreatic lipase family members, we showed here that porcine pancreatic lipase (PPL), which was so far classified as “classical lipase”, was able to hydrolyze phosphatidylcholine (PC). Amino acid sequence alignments revealed that Val260 residue in PPL lid could be critical for the interaction with lipid substrate. Molecular dynamics was applied to investigate PC binding modes within the catalytic cavity of PPL and human pancreatic lipase (HPL), aiming to explain the difference of specificity of these enzymes towards phospholipids. Results showed that with HPL, the oxyanion hole was not able to accommodate the PC molecule, suggesting that no activity could be obtained. With PPL, the formation of a large pocket involving Val260 allowed the PC molecule to come near the catalytic residues, suggesting that it could be hydrolyzed. One more interesting finding is that human pancreatic lipase related protein 2 could hydrolyze phospholipids through its PLA1 and PLA2 activities. Overall, our study shed the light on new structural features of the phospholipase activity of pancreatic lipase family members. © 2013 Elsevier B.V. All rights reserved.

1. Introduction Triacylglycerol lipases (EC 3.1.1.3) are carboxylester hydrolases that catalyze the hydrolysis of long-chain acylglycerols at oil/water interfaces [1–3]. They are ubiquitous enzymes and have been found in most of the organisms belonging to the microbial, plant, and animal kingdoms. Most cells in the body hydrolyze triacyglycerols (TAGs) through similar pathways, using lipases, generally with a common purpose to provide fatty acids for energy demands [4,5]. Some of these lipases were assumed to be attractive targets for the treatment of dyslipidemias, viral infection and atherosclerosis [4,5]. Classical pancreatic lipases are the main enzymes involved in the digestion of dietary TAGs in the small intestine. Classical pancreatic lipases from human (HPL) [6,7], horse (HoPL) [8], and porcine (PPL) [9] have

Abbreviations: [ 14 C]PAPC, 1-palmitoyl-2-[1- 14 C]arachidonyl-sn-glycero-3phosphocholine; PPCEH, porcine pancreatic cholesterol ester hydrolase; HoPL, horse pancreaticlipase; HPL, human pancreatic lipase; HPTLC, high performance thin-layer chromatography; PC, phosphatidylcholine; PLA1, phospholipase A1; PLA2, phospholipase A2; PPPLA2, porcine pancreatic PLA2; PPL, porcine pancreatic lipase; NaTDC, sodium taurodeoxycholate; TAG, triacylglycerol; HPLRP2, human pancreatic lipase related protein 2 ⁎ Corresponding author. Tel.: +216 97 533 160; fax: +216 74 675 055. ⁎⁎ Corresponding author. Tel.: +33 4 72 44 81 02; fax: +33 4 72 44 79 70. E-mail addresses: [email protected] (A. Aloulou), [email protected] (A. Abousalham). 1388-1981/$ – see front matter © 2013 Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.bbalip.2013.12.009

been characterized extensively at the biochemical and the structural levels. Their three dimensional (3D) structure is composed of two domains (N- and C-terminal domains). The large N-terminal domain belongs to the α/β-hydrolase fold [10] and contains the active site with a catalytic triad formed by Ser, Asp and His. The β-sandwich C-terminal domain is important for colipase binding, a specific lipaseanchoring protein, present in the exocrine secretion of pancreas, that facilitates its adsorption at bile salt-covered lipid/water interfaces [11–13]. Structure–function studies on various lipases have shed light on the interfacial recognition sites present in the molecular structure of these enzymes and the conformational changes occurring in the presence of lipids and amphiphiles [14–16]. In many lipases, access to the active site is controlled by a so-called lid formed by a surface loop. This lid was found to undergo a conformational change in the presence of lipase inhibitors, making the active site accessible to solvent in the 3D structures of several lipases [17–19]. The discovery of novel pancreatic lipase-related proteins (PLRPs) has increased the complexity of structure–function relationships within this family of enzymes. Pancreatic lipase-related proteins 1 and 2 (PLRP1 and PLRP2) belong to the pancreatic lipase gene family, and they share 65–68% amino acid identity with the classical pancreatic lipase [20]. Their structural components such as the catalytic triad (Ser, Asp, His) are highly conserved, and overall, the 3D structures obtained so far are superimposable [7,21,22]. A deletion within the lid domain was however observed in the PLRP2 from guinea pig

582

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587

(GPLRP2), which is able to accommodate more hydrophilic substrates than classical pancreatic lipase, such as phospholipids and galactolipids with large polar heads [23]. Besides this previously determined structural difference, PLRPs also differ from classical pancreatic lipases by their biochemical and physiological properties. PLRP2 showed lipase, phospholipase and galactolipase activities, whereas classical pancreatic lipase only showed lipase activity [23], and PLRP1 is an inactive lipase against all known substrates [22]. PLRP2 is also produced at a high level in species lacking pancreatic phospholipase A2 (PLA2), and it was suggested that it might play a significant role in phospholipid digestion [24,25]. The occurrence of pancreatic lipases in the pancreas of various herbivorous, carnivorous, omnivorous and avian species was investigated [24]. Classical pancreatic lipases, PLRP1 and PLRP2 were detected in the pancreatic enzymatic equipment of humans and rats [24,26,27]. However, all attempts to identify porcine PLRP2 in both pure pancreatic juice and a protein fraction isolated from zymogen granules were unsuccessful [24]. Likewise, only classical pancreatic lipase was detected, too, in the case of bird (ostrich and turkey) pancreases [24,28,29]. The regioselectivity of PLRP2s towards phospholipids was not proved yet experimentally, even though amino acid sequence comparisons have shown that PLRP2 might act as a phospholipase A1 (PLA1) [30,31]. In contrast to other phospholipases, such as PLA2, phospholipase C and phospholipase D isozymes, the physiological function(s) of PLA1 enzymes remain largely unknown. PLA1 enzymes may have specific roles in producing bioactive lysophospholipids, such as lysophosphatidylserine and lysophosphatidic acid [30]. All mammalian extracellular PLA1 enzymes belong to the pancreatic lipase gene family which is conserved in a wide range of organisms from insects to mammals [23]. While some extracellular PLA1s have a broad substrate specificity and hydrolyzing both phospholipids, TAGs and galactolipids, various other PLA1s such as phosphatidylserine (PS)-specific PLA1 (PS-PLA1), membrane-associated phosphatidic acid (PA)-selective PLA1α (mPA-PLA1α) and mPA-PLA1β show a strict substrate specificity and hydrolyze exclusively PS and PA, respectively [30]. The 3D structures of various lipases revealed that the lid domain covering the active site in the closed form is stabilized via van der Waals interactions formed between the lid itself and two other loops called β5 and β9 [7,13]. The lid domain, β5 and β9 loops were hypothesized to undergo a special reorganization in order to facilitate the entrance of the substrate in the active site. The present study was thus undertaken in order to ascertain whether human PLRP2 (HPLRP2) does hydrolyze ester bonds of phospholipids at the sn-1 and/or the sn-2 position. Interestingly, a comparative kinetic study of various pancreatic lipases from different origins has revealed that PPL, besides its high TAG hydrolase activity, does possess PLA1 activity. We used molecular dynamics calculations of lipase–phospholipid transition-state complexes to provide information about molecular interactions which are important for phospholipid breakdown by lipases. Our data suggest that avian pancreatic lipases from ostrich and turkey may act, too, as PLA1 enzymes. It is therefore suggested that classical pancreatic lipase may fulfill in some cases new biological functions as a PLA1 enzyme, compensating PLRP2 deficiency in the digestive tract. 2. Materials and methods 2.1. Reagents Purified egg yolk L-α-phosphatidylcholine (PC), tributyrin, BSA and sodium taurodeoxycholate (NaTDC) were purchased from SigmaAldrich-Fluka Chemie (St-Quentin-Fallavier, France). High-performance thin-layer chromatography (HPTLC) plates pre-coated with silica gel 60 were from Merck (Fontenay Sous Bois, France). Molecular mass markers were from Amersham Biosciences (Uppsala, Sweden). 1-Palmitoyl2-[1- 14 C]arachidonyl-sn-glycero-3-phosphocholine ([ 14 C]PAPC) was from PerkinElmer Life Sciences (Waltham, Massachusetts). All

other chemicals and solvents were of reagent or better quality and were obtained from local suppliers. 2.2. Proteins Recombinant HPL was expressed and purified from insect cells as described by Thirstrup et al. [32]. Recombinant HPLRP2 was produced in the yeast Pichia pastoris and purified as described previously [25,33]. Horse pancreatic lipase (HoPL) was purified at the laboratory as described previously [34]. PPL and porcine colipase devoid of phospholipase contamination were purified using methods previously described [35,36]. Porcine pancreatic PLA2 (PPPLA2) and porcine pancreatic cholesterol ester hydrolase (PPCEH) were purchased from SigmaAldrich-Fluka Chemie. 2.3. Lipolytic activity measurements The release of fatty acids was continuously assayed potentiometrically with a pH-stat apparatus (Metrohm 718 Stat Titrino, Zofingen, Switzerland) under mechanical stirring in a 15-mL reaction vessel at 37 °C, adding 0.1 N NaOH and using either tributyrin or egg yolk PC [37] emulsions as substrates. The pH was adjusted to fixed end point pH 8.0 value and the (phospho)lipase solutions were added at zero time after recording the background level for 5 min. When olive oil was used as substrate, gum arabic was used as an emulsifier as previously described [38]. Tributyrin or olive oil assay: 0.5 mL of tributyrin or 5 mL of olive oil prepared in gum arabic [38] was added and mechanically emulsified in 14.5 mL or 10 mL, respectively, of 2.5 mM Tris–HCl, 150 mM NaCl, 5 mM CaCl2 and 4 mM NaTDC. Egg PC assay: the assay was performed as previously described [37] with slight modifications. Four grams of egg yolk PC was homogenized by probe sonication in 100 mL of water and filtered through cheesecloth. Five milliliters of the above substrate was added to 9.5 mL of 20 mM NaTDC and 0.5 mL of 0.2 M CaCl2 and mechanically emulsified. HPL, PPL and HoPL activities were measured in the presence of porcine colipase at a colipase/lipase molar excess of 5. Under the above assay conditions, one international (phospho)lipase unit (U) corresponds to the release of 1 μmol of fatty acid released per minute. The specific activity was expressed in U·mg −1 of pure enzyme. For determination of regioselectivity, the lipase was incubated (100 μL final volume) under continuous stirring for 15 min at room temperature in 25 m M Tris–HCl (pH 8.0), 8 mM CaCl2 , 0.4% (w/v) egg yolk PC, 6 mM sodium deoxycholate and [ 14 C]PAPC (3.6 kBq, 4.2 GBq·mmol − 1 ). Lipids were extracted immediately after sampling according to Folch's procedure [39] and then separated by HPTLC on silica gel 60. The sample migration was performed with chloroform/methanol/water (65/35/5, v/v/v). The plate was dried and exposed overnight for Phosphor-Imager (PerkinElmer, Waltham, Massachusetts) analysis. 2.4. Protein quantitation and gel electrophoresis Protein concentrations were determined routinely by using Bradford's procedure [40], with BSA as standard. Alternatively, protein concentrations were determined by amino acid analysis. The homogeneity of the various enzymes was routinely assessed by performing SDS-PAGE on 12% gels using Laemmli's procedure [41]. Electrophoresis in the presence of SDS was carried out on a Mini-Protean II dual vertical slab gel electrophoresis cell (Bio-Rad, Marnes-la-Coquette, France). 2.5. Structural alignment, modeling and docking The sequence alignment was generated with Clustal Omega [42]. The 3D coordinates of open forms of HPL (1LPB) and PPL (1ETH) were

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587

extracted from the protein data bank (PDB) (http://www.rcsb.org/pdb). The structures of PPL and HPL in complex with the covalently bound PC molecule were manually modeled on the basis of the open structures of PPL [9] and HPL [43]. The PPL–PC and HPL–PC complexes were subjected to molecular mechanistic optimization using HyperChem Professional version 7.52 for Windows Molecular Modelling System. The molecular mechanistic optimization using the CHARMM27 force field (energy minimization) was performed until a gradient of 0.01 kcal/(Å·mol). Simulation was accomplished using CHARMM27 force field with the distance-dependent dielectric constant and with the one to four electrostatic interactions scaled by 0.5. All non-bonded interactions were calculated along with simulation (no cut-off). The simulated molecule was first equilibrated for 5 ps, followed by a production run, whose data were collected for analysis. The molecular dynamics simulations were run, at a 330 K, for 10 ps with a 0.001 ps step size. The simulations unroll in vacuo with a constant temperature and a bath relaxation time equal to 0.001 ps. 3. Results and discussion 3.1. Assessment of enzyme purity The purity of all lipases and PLA2 used in this study was checked by performing SDS-PAGE followed by Coomassie blue staining (Fig. 1). Highly purified PPL, HPLRP2, HPL, and HoPL, devoid of any detectable contaminants, showed protein bands corresponding to molecular masses of around 48 kDa for PPL, HPL and HoPL, and around 50 kDa for HPLRP2 (Fig. 1). It is worth noting that HPLRP2 and HPL migrated in the form of a doublet. It was shown previously that mass spectrometry analysis of HPL and HPLRP2 yielded two different average masses consistent with a heterogeneous glycosylation in these enzymes [25]. Commercial enzymes PPCEH and PPPLA2 showed protein bands corresponding to molecular masses of 85 and 14 kDa, respectively (Fig. 1). 3.2. Phospholipase activity of TAG lipases from the pancreatic lipase family Till now, the phospholipase activity of PPL has never been described. In order to investigate the ability of PPL and other lipases to hydrolyze phospholipids, the specific activities were measured, with the pH-stat method, on egg yolk PC as substrate using the steady state rate of the reaction. The hydrolysis of egg yolk PC by PPL showed a triphasic kinetic pattern (Fig. 2): a slow accelerating phase (the lag phase) followed by a second phase reaching a steady state with a high hydrolytic activity, and finally, a slow phase with a low hydrolytic activity. The lag time was defined as the point where the extrapolated steady-state curve intersected with the time axis (Fig. 2). The natural substrates of lipolytic enzymes, such as TAGs and phospholipids, are usually long-chain fatty

583

Fig. 2. Kinetic recording of the hydrolysis of egg yolk PC by the lipolytic carboxylester hydrolases. PPPLA2 (12.9 μg, curve 1), HPLRP2 (56.7 μg, curve 2), PPL (5.92 μg, curve 3) and PPCEH (52.5 μg, curve 4). The experiments were carried out at 37 °C, using the pH-stat technique, in a final assay volume of 15 mL containing 1.3% (w/v) egg yolk PC, 12.6 mM NaTDC, and 6.6 mM CaCl2 (see Materials and methods). Enzymes were added at zero time after recording the background of hydrolysis level for 5 min. The kinetic recordings shown here are typical of those obtained in three independent experiments.

acids esters. Upon hydrolysis, they generate poorly water-soluble lipolysis products, which remain transiently at the lipid–water interface. The progressive interfacial accumulation of these products could induce a change in the physicochemical properties of the interface, associated with either a decrease or an enhancement of the reaction rate, as characterized by lag periods [44,45]. PPL, in the presence of colipase excess, hydrolyzed the egg yolk PC with a steady state specific activity of around 1192 U·mg−1, as compared with 9017 U·mg− 1 on tributyrin (Table 1). HPL and HoPL showed no hydrolytic activity on egg yolk PC under the same experimental conditions, whereas their maximal specific activity was shown to be around 6481 and 6250 U·mg− 1 on tributyrin, respectively (Table 1). As control, PPPLA2 hydrolyzed egg yolk PC with a specific activity of 937 U·mg−1 (Fig. 2 and Table 1). It was reported [24] that other classical lipases, like turkey pancreatic lipase (TPL), rabbit pancreatic lipase (RPL) and ostrich pancreatic lipase (OPL), were able to hydrolyze egg yolk PC, under the same experimental conditions, with specific activities of 28 ± 5 U·mg−1, 3 ± 1 U·mg−1 and 4 U·mg−1, respectively [24]. In contrast to the classical pancreatic lipases which act only on mono-, di-, and triacylglycerols, PLRP2s show a broader range of substrate specificity by hydrolyzing TAGs, phospholipids, and galactolipids [33,46,47]. As shown in Fig. 2 and Table 1, the maximal specific activity of HPLRP2 on tributyrin and egg yolk PC was shown to be around 423 and 19 U·mg−1, respectively. The maximum catalytic turnover of HPLRP2 on triolein and egg yolk PC was found to be around 675 [48] and 50 U·mg −1 [25], respectively. Furthermore, it is now assumed that PLRP2 is the main enzyme involved in the gastrointestinal digestion of dietary galactolipids. The hydrolysis rate of natural long-chain

Table 1 Maximum specific activities of various lipolytic carboxylester hydrolases on tributyrin and egg yolk PC. Specific activity (U·mg −1) Tributyrin

Fig. 1. SDS-PAGE (12% acrylamide) of the lipolytic carboxylester hydrolases (~5 μg each) used in this study. The gel was stained by Coomassie blue to reveal the proteins. Line 1, Molecular mass markers; line 2, PPL; line 3, HPLRP2; line 4, HPL; line 5, HoPL; line 6, PPCEH; line 7, PPPLA2.

PPPLA2 PPL HPLRP2 PPCEH HPL HoPL

0 9017 ± 29 423 ± 8 2048 ± 107 6481 ± 314 6250 ± 396

Egg PC 937 ± 31 1192 ± 33 19 ± 1 38 ± 4 0 0

Phospholipase to TAG–lipase activity ratio – 0.13 0.04 0.02 0 0

584

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587

galactolipids by HPLRP2 was found to be around 2441 U·mg −1 [49]. On the basis of the high catalytic activity on (glycero)galactolipids, TAGs, and (glycero)phospholipids, PLRP2s were suggested to be renamed as (glycero)-lipases [1]. Classical lipases, such as PPL, HPL and HoPL, are known to be as efficient catalysts at lipid/water interfaces hydrolyzing long-chain TAGs with a specific activity of about 4000 U·mg − 1 [50,51]. It is worth noting that the phospholipase to TAG-lipase activity ratio was shown to be 0.013, 0.04 and 0.02 for PPL, HPLRP2 and PPCEH, respectively (Table 1). We suggested recently that the name of triacylglycerol lipases (EC 3.1.1.3) seems to be more appropriate to define functionally this enzyme family [1].

3.3. Regioselectivity of the phospholipase activity of PPL and HPLRP2 The regiospecificity of PPL and HPLRP2 phospholipase activities was further investigated by analyzing the [14C]PAPC lipolysis products, i.e. [14C]-arachidonic acid and [14C]-lysophosphatidic acid, using a Phosphor-Imager system following TLC separation (Fig. 3). The hydrolysis profile of PPL was typical of a PLA1 enzyme generating [14C]-lysophosphatidic acid, but not hydrolyzing the ester bond at the sn-2 position, since [14C]-arachidonic acid did not accumulate in the reaction mixture (Fig. 3). However, HPLRP2 was found to hydrolyze, under the same conditions, both the ester bonds at the sn-1 and sn-2 positions of [14C]PAPC, thus releasing [14C]-arachidonic acid and [14C]-lysophosphatidic. Interestingly, PLA1 activity seems to be almost 2-times higher than PLA2 activity of HPLRP2, as revealed by the intensity of the lipolysis product bands (see line 2, Fig. 3). As control experiments, HPL did not show any hydrolytic activity on [14C]PAPC, whereas PPPLA2 was found to hydrolyze strictly the ester bond at the sn-2 position, thus releasing [ 14C]-arachidonic acid (Fig. 3). It is worth noticing that only indirect evidence, based on amino acid sequence comparisons, had been reported previously for the existence of PLA1 activity of PLRP2s [30,31]. In this study, we provided for the first time the regioselectivity of HPLRP2 towards phospholipids by proving experimentally that it is a dual PLA1/PLA2 enzyme. This finding might explain the high production level of PLRP2 in species lacking

Fig. 3. Hydrolysis of [14C]PAPC vesicles by the lipolytic carboxylester hydrolases. HPLRP2 (120 μg, lane 2), HPL (112 μg, lane 3), PLA2 (10 μg, lane 4) and PPL (44 μg, lane 5). After 15 min incubation at room temperature at pH 8.0, lipids were extracted and separated by HPTLC. The signal from 14C-labeled species (free fatty acids, lyso-PC and PC) was visualized and quantified by a Phosphor-Imager.

pancreatic PLA2 [24,25]. PLRP2 might therefore contribute to phospholipid digestion through its PLA1 and PLA2 activities. Moreover, this study highlighted for the first time the PLA1 activity of classical lipases, such as PPL. It is worth noticing that PLRP2 has never been detected so far in the porcine pancreas [24]. Porcine pancreas could produce, however, high amounts of pancreatic PLA2 [52]. One can assume that the PLA1 activity of PPL might therefore compensate the lack of PLRP2 during the digestion of phospholipids.

3.4. Substrate modeling and structural basis for phospholipid binding To understand why PPL is able to hydrolyze phospholipid substrates, we compared its amino acid sequence to those of homologous TAG lipases, from the pancreatic lipase gene family, which are able to hydrolyze phospholipids (Fig. 4). Sequence comparison showed that residue Val260 of the PPL lid domain was highly conserved among the lipase family members exhibiting phospholipase activity, including lipoprotein lipase and endothelial lipase [53]. Strikingly, the equivalent of Val260 in PPL is an Ala residue in HPL and HoPL which have no phospholipase activity (see Table 1). To investigate if the presence of Val260 (in PPL) or Ala259 (in HPL) is correlated to differences in protein dynamics affecting phospholipid binding, we have carried out molecular dynamics during 10 ps for the open forms of PPL and HPL in complex with phosphatidylcholine (Fig. 5). The molecular dynamics of the PPL–PC complex showed that the distance between Val260 and Phe216, which belongs to the β9 loop, got increased over time, forming thus a large pocket which might accommodate the PC polar group (Fig. 5A). PPL was able to accommodate a PC molecule in its active site with the PC polar head stabilized by interactions with aromatic residues (Trp86, Tyr268). Similar results were obtained along the molecular dynamics simulations of the HPLRP2-PC complex (data not shown). This would explain the activity of PPL and HPLRP2 on PC. In the case of HPL, however, Ala259 of the lid domain was positioned in such a way that it remained close to Phe215 of the β9 loop, hindering thus the formation of a pocket allowing the accommodation of the PC polar group (Fig. 5B). As a control experiment, we have performed simulations of PPL Val260Ala and HPL Ala259Val mutants in complex with phosphatidylcholine (data not shown). Molecular dynamics of PPL Val260Ala and HPL Ala259Val mutants have shown, as expected, similar results with HPL and PPL, respectively, which indicates that the observed conformational change in PPL is most likely due to the presence of the Val260 residue. Since many years, several works have attempted to provide molecular and structural explanation for the phospholipase activity of pancreatic lipase family members. From crystal structure observations and substrate modeling within the active site, it was proposed that the

Fig. 4. Sequence alignment of the lid domain of pancreatic lipase family members. Conserved Val260 residue (PPL numbering) was shown in red, while conserved Ala259 residue (HPL numbering) was shown in blue. BPLRP2, bovine pancreatic lipase related protein 2; HEL, human endothelial lipase; MEL, mouse endothelial lipase; HLPL, human lipoprotein lipase; BLPL, bovine lipoprotein lipase.

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587

585

Fig. 5. Molecular dynamics of the open forms of PPL–PC (A) and HPL–PC (B) complexes. The figure shows a GRASP representation of each 2 ps simulation step of PPL and HPL. The Val260 is colored in yellow and the other residues in white. The PC molecule is shown as sticks. The pocket that may accommodate the phosphate group of PC is shown in box. The figures were generated using PyMOL (http://www.pymol.org).

phospholipase activity of GPLRP2 resulted from the deletion of the lid domain, thereby generating a larger and more polar active site in comparison with that of the TAG activity encoded by HPL [31]. Exchange of HPL lid and GPLRP2 “mini-lid” by site-directed mutagenesis reduced drastically the phospholipase activity for the GPLRP2 lid domain mutant. However, the phospholipase activity was not introduced in the HPL lid deletion mutant [54]. It was therefore concluded that the lid domain was a necessary, but not the only, structural element contributing to the substrate specificity of pancreatic lipases [23].

More recently, Aoki et al. [30] have suggested that substrate specificity of pancreatic lipase family members is determined by the length of the two surface loops, the lid domain and the β9 loop. They have hypothesized that: (i) enzymes with a short lid domain and a short β9 loop exhibit only phospholipase activity (e.g. PS-PLA1 and PA-PLA1); (ii) enzymes with a short lid domain and a long β9 loop exhibit both phospholipase and TAG lipase activities (e.g. endothelial lipase and GPLRP2); and (iii) enzymes with a long lid domain and a long β9 loop exhibit only TAG lipase activity (e.g. classical pancreatic lipases).

586

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587

However, one exception is that some PLRP2s (such as HPLRP2), hepatic lipase and lipoprotein lipase exhibit both TAG lipase and phospholipase activities although they have a long lid and a long β9 loop [30]. In 2008, the lid domain of HPLRP2 was found to be in an open conformation in the crystal structure [21], without the requirement of detergents, lipids or substrate analogues, which suggested that HPLRP2 might adopt this conformation in solution. This hypothesis was supported by the preference of HPLRP2 for substrates forming small micellar aggregates in solution, such as monoolein and phospholipids. Once again, one exception is that rat PLRP2 adopted a close conformation in the crystal structure although it exhibited both TAG lipase and phospholipase activities [55]. Our finding might provide new insights on the structural basis of the phospholipase activity of the pancreatic lipase family members, through the involvement of the Val260 (PPL numbering) in the formation of a pocket accommodating a phospholipid substrate. With this in mind, we think it would be interesting to substitute the Ala259 in HPL by a Val residue in order to check either this punctual mutation would be able or not to induce alone phospholipase activity in the HPL Ala259Val mutant. Likewise, we expect that the HPLRP2 Val261Ala mutant might lose its phospholipase activity once the Val261 (in HPLRP2) is mutated by an Ala residue. In conclusion, there are here interesting prospects to better understand the structural features of the phospholipase activity of pancreatic lipase family members. References [1] Y. Ben Ali, R. Verger, A. Abousalham, Lipases or esterases: does it really matter? Toward a new bio-physico-chemical classification, Methods Mol. Biol. 861 (2012) 31–51. [2] H.L. Brockman, General Features of Lipolysis: Reaction Scheme, Interfacial Structure and Experimental Approaches, in: B. Borgstrom, H.L. Brockman (Eds.), Lipases, Elsevier, Amsterdam, 1984, pp. 3–46. [3] R. Verger, ‘Interfacial activation’ of lipases: facts and artifacts, Trends Biotechnol. 15 (1997) 32–38. [4] A.D. Quiroga, R. Lehner, Liver triacylglycerol lipases, Biochim. Biophys. Acta 1821 (2012) 762–769. [5] A.D. Quiroga, R. Lehner, Role of endoplasmic reticulum neutral lipid hydrolases, Trends Endocrinol. Metab. 22 (2011) 218–225. [6] H. van Tilbeurgh, L. Sarda, R. Verger, C. Cambillau, Structure of the pancreatic lipase– procolipase complex, Nature 359 (1992) 159–162. [7] F.K. Winkler, A. D'Arcy, W. Hunziker, Structure of human pancreatic lipase, Nature 343 (1990) 771–774. [8] Y. Bourne, C. Martinez, B. Kerfelec, D. Lombardo, C. Chapus, C. Cambillau, Horse pancreatic lipase. The crystal structure refined at 2.3 A resolution, J. Mol. Biol. 238 (1994) 709–732. [9] J. Hermoso, D. Pignol, B. Kerfelec, I. Crenon, C. Chapus, J.C. Fontecilla-Camps, Lipase activation by nonionic detergents. The crystal structure of the porcine lipase– colipase–tetraethylene glycol monooctyl ether complex, J. Biol. Chem. 271 (1996) 18007–18016. [10] D.L. Ollis, E. Cheah, M. Cygler, B. Dijkstra, F. Frolow, S.M. Franken, M. Harel, S.J. Remington, I. Silman, J. Schrag, et al., The alpha/beta hydrolase fold, Protein Eng. 5 (1992) 197–211. [11] H. van Tilbeurgh, S. Bezzine, C. Cambillau, R. Verger, F. Carriere, Colipase: structure and interaction with pancreatic lipase, Biochim. Biophys. Acta 1441 (1999) 173–184. [12] H. van Tilbeurgh, M.-P. Egloff, C. Martinez, N. Rugani, R. Verger, C. Cambillau, Interfacial activation of the lipase-procolipase complex by mixed micelles revealed by X-ray crystallography, Nature 362 (1993) 814–820. [13] H. van Tilbeurgh, Y. Gargouri, C. Dezan, M.P. Egloff, M.P. Nesa, N. Ruganie, L. Sarda, R. Verger, C. Cambillau, Crystallization of pancreatic procolipase and of its complex with pancreatic lipase, J. Mol. Biol. 229 (1993) 552–554. [14] A. Aloulou, J.A. Rodriguez, S. Fernandez, D. van Oosterhout, D. Puccinelli, F. Carriere, Exploring the specific features of interfacial enzymology based on lipase studies, Biochim. Biophys. Acta 1761 (2006) 995–1013. [15] V. Belle, A. Fournel, M. Woudstra, S. Ranaldi, F. Prieri, V. Thome, J. Currault, R. Verger, B. Guigliarelli, F. Carriere, Probing the opening of the pancreatic lipase lid using site-directed spin labeling and EPR spectroscopy, Biochemistry 46 (2007) 2205–2214. [16] S. Ranaldi, V. Belle, M. Woudstra, J. Rodriguez, B. Guigliarelli, J. Sturgis, F. Carriere, A. Fournel, Lid opening and unfolding in human pancreatic lipase at low pH revealed by site-directed spin labeling EPR and FTIR spectroscopy, Biochemistry 48 (2009) 630–638. [17] A.M. Brzozowski, H. Savage, C.S. Verma, J.P. Turkenburg, D.M. Lawson, A. Svendsen, S. Patkar, Structural origins of the interfacial activation in Thermomyces (Humicola) lanuginosa lipase, Biochemistry 39 (2000) 15071–15082.

[18] M.P. Egloff, L. Sarda, R. Verger, C. Cambillau, H. van Tilbeurgh, Crystallographic study of the structure of colipase and of the interaction with pancreatic lipase, Protein Sci. 4 (1995) 44–57. [19] A. Roussel, N. Miled, L. Berti-Dupuis, M. Riviere, S. Spinelli, P. Berna, V. Gruber, R. Verger, C. Cambillau, Crystal structure of the open form of dog gastric lipase in complex with a phosphonate inhibitor, J. Biol. Chem. 277 (2002) 2266–2274. [20] T. Giller, P. Buchwald, D. Blum-Kaelin, W. Hunziker, Two novel human pancreatic lipase related proteins, hPLRP1 and hPLRP2. Differences in colipase dependence and in lipase activity, J. Biol. Chem. 267 (1992) 16509–16516. [21] C. Eydoux, S. Spinelli, T.L. Davis, J.R. Walker, A. Seitova, S. Dhe-Paganon, A. De Caro, C. Cambillau, F. Carriere, Structure of human pancreatic lipase-related protein 2 with the lid in an open conformation, Biochemistry 47 (2008) 9553–9564. [22] A. Roussel, J. de Caro, S. Bezzine, L. Gastinel, A. de Caro, F. Carrière, S. Leydier, R. Verger, C. Cambillau, Reactivation of the totally inactive pancreatic lipase RP1 by structure-predicted point mutations, Proteins 32 (1998) 523–531. [23] F. Carrière, C. Withers-Martinez, H. van Tilbeurgh, A. Roussel, C. Cambillau, R. Verger, Structural basis for the substrate selectivity of pancreatic lipases and some related proteins, Biochim. Biophys. Acta 1376 (1998) 417–432. [24] J. De Caro, C. Eydoux, S. Cherif, R. Lebrun, Y. Gargouri, F. Carriere, A. De Caro, Occurrence of pancreatic lipase-related protein-2 in various species and its relationship with herbivore diet, Comp. Biochem. Physiol. B Biochem. Mol. Biol. 150 (2008) 1–9. [25] C. Eydoux, J. De Caro, F. Ferrato, P. Boullanger, D. Lafont, R. Laugier, F. Carriere, A. De Caro, Further biochemical characterization of human pancreatic lipase-related protein 2 expressed in yeast cells, J. Lipid Res. 48 (2007) 1539–1549. [26] J. De Caro, F. Carriere, P. Barboni, T. Giller, R. Verger, A. De Caro, Pancreatic lipase-related protein 1 (PLRP1) is present in the pancreatic juice of several species, Biochim. Biophys. Acta 1387 (1998) 331–341. [27] M.J. Wishart, P.C. Andrews, R. Nichols, G.T. Blevins, C.D. Logsdon, J.A. Williams, Identification and cloning of GP-3 from rat pancreatic acinar zymogen granules as a glycosylated membrane-associated lipase, J. Biol. Chem. 268 (1993) 10303–10311. [28] A. Ben Bacha, Y. Gargouri, S. Bezzine, H. Mosbah, H. Mejdoub, Ostrich pancreatic phospholipase A2: purification and biochemical characterization, J. Chromatogr. B Analyt. Technol. Biomed. Life Sci. 857 (2007) 108–114. [29] A. Fendri, F. Frikha, H. Mosbah, N. Miled, N. Zouari, A.B. Bacha, A. Sayari, H. Mejdoub, Y. Gargouri, Biochemical characterization, cloning, and molecular modelling of chicken pancreatic lipase, Arch. Biochem. Biophys. 451 (2006) 149–159. [30] J. Aoki, A. Inoue, K. Makide, N. Saiki, H. Arai, Structure and function of extracellular phospholipase A1 belonging to the pancreatic lipase gene family, Biochimie 89 (2007) 197–204. [31] C. Withers-Martinez, F. Carriere, R. Verger, D. Bourgeois, C. Cambillau, A pancreatic lipase with a phospholipase A1 activity: crystal structure of a chimeric pancreatic lipase-related protein 2 from guinea pig, Structure 4 (1996) 1363–1374. [32] K. Thirstrup, F. Carriere, S. Hjorth, P.B. Rasmussen, H. Woldike, P.F. Nielsen, L. Thim, One-step purification and characterization of human pancreatic lipase expressed in insect cells, FEBS Lett. 327 (1993) 79–84. [33] B. Sias, F. Ferrato, P. Grandval, D. Lafont, P. Boullanger, A. De Caro, B. Leboeuf, R. Verger, F. Carriere, Human pancreatic lipase-related protein 2 is a galactolipase, Biochemistry 43 (2004) 10138–10148. [34] A. Abousalham, C. Chaillan, B. Kerfelec, E. Foglizzo, C. Chapus, Uncoupling of catalysis and colipase binding in pancreatic lipase by limited proteolysis, Protein Eng. 5 (1992) 105–111. [35] M.F. Maylié, M. Charles, M. Astier, P. Desnuelle, On porcine pancreatic colipase: large scale purification and some properties, Biochem. Biophys. Res. Commun. 52 (1973) 291–297. [36] R. Verger, G.H. de Haas, L. Sarda, Desnuelle, purification from porcine pancreas of two molecular species with lipase activity, Biochim. Biophys. Acta 188 (1969) 272–282. [37] A. Abousalham, R. Verger, Egg yolk lipoproteins as substrates for lipases, Biochim. Biophys. Acta 1485 (2000) 56–62. [38] A. Tiss, F. Carriere, R. Verger, Effects of gum arabic on lipase interfacial binding and activity, Anal. Biochem. 294 (2001) 36–43. [39] J. Folch, M. Lees, G.H. Sloane Stanley, A simple method for the isolation and purification of total lipides from animal tissues, J. Biol. Chem. 226 (1957) 497–509. [40] M.M. Bradford, A rapid and sensitive method for quantitation of microgram quantities of protein utilizing the principle of protein–dye binding, Anal. Biochem. 72 (1976) 248–254. [41] U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature 227 (1970) 680–685. [42] F. Sievers, A. Wilm, D. Dineen, T.J. Gibson, K. Karplus, W. Li, R. Lopez, H. McWilliam, M. Remmert, J. Soding, J.D. Thompson, D.G. Higgins, Fast, scalable generation of high-quality protein multiple sequence alignments using Clustal Omega, Mol. Syst. Biol. 7 (2011) 539. [43] M.P. Egloff, F. Marguet, G. Buono, R. Verger, C. Cambillau, H. van Tilbeurgh, The 2.46 A resolution structure of the pancreatic lipase-colipase complex inhibited by a C11 alkyl phosphonate, Biochemistry 34 (1995) 2751–2762. [44] T.H. Callisen, Y. Talmon, Direct imaging by cryo-TEM shows membrane break-up by phospholipase A2 enzymatic activity, Biochemistry 37 (1998) 10987–10993. [45] T. Wieloch, B. Borgström, G. Piéroni, F. Pattus, R. Verger, Product activation of pancreatic lipase. Lipolytic enzymes as probes for lipid/water interfaces, J. Biol. Chem. 257 (1982) 11523–11528. [46] L. Andersson, F. Carriere, M.E. Lowe, A. Nilsson, R. Verger, Pancreatic lipase-related protein 2 but not classical pancreatic lipase hydrolyzes galactolipids, Biochim. Biophys. Acta 1302 (1996) 236–240. [47] K. Thirstrup, R. Verger, F. Carriere, Evidence for a pancreatic lipase subfamily with new kinetic properties, Biochemistry 33 (1994) 2748–2756.

A. Aloulou et al. / Biochimica et Biophysica Acta 1841 (2014) 581–587 [48] M.E. Lowe, Properties and function of pancreatic lipase related protein 2, Biochimie 82 (2000) 997–1004. [49] S. Amara, N. Barouh, J. Lecomte, D. Lafont, S. Robert, P. Villeneuve, A. De Caro, F. Carriere, Lipolysis of natural long chain and synthetic medium chain galactolipids by pancreatic lipase-related protein 2, Biochim. Biophys. Acta 1801 (2010) 508–516. [50] I. Crenon, S. Granon, C. Chapus, B. Kerfelec, Molecular cloning and expression of two horse pancreatic cDNA encoding colipase A and B, Biochim. Biophys. Acta 1213 (1994) 357–360. [51] A. Tiss, S. Ransac, H. Lengsfeld, P. Hadvary, A. Cagna, R. Verger, Surface behaviour of bile salts and tetrahydrolipstatin at air/water and oil/water interfaces, Chem. Phys. Lipids 111 (2001) 73–85.

587

[52] W. Nieuwenhuizen, H. Kunze, G.H. de Haas, Phospholipase A2 (phosphatide acylhydrolase, EC 3.1.1.4) from porcine pancreas, Methods Enzymol. 32 (1974) 147–154. [53] N. Griffon, E.C. Budreck, C.J. Long, U.C. Broedl, D.H. Marchadier, J.M. Glick, D.J. Rader, Substrate specificity of lipoprotein lipase and endothelial lipase: studies of lid chimeras, J. Lipid Res. 47 (2006) 1803–1811. [54] F. Carriere, K. Thirstrup, S. Hjorth, F. Ferrato, P.F. Nielsen, C. Withers-Martinez, C. Cambillau, E. Boel, L. Thim, R. Verger, Pancreatic lipase structure–function relationships by domain exchange, Biochemistry 36 (1997) 239–248. [55] A. Roussel, Y. Yang, F. Ferrato, R. Verger, C. Cambillau, M. Lowe, Structure and activity of rat pancreatic lipase-related protein 2, J. Biol. Chem. 273 (1998) 32121–32128.