Crystal Structure of Human Gastric Lipase and Model of LysosomalAcid Lipase, Two Lipolytic Enzymes of Medical Interest*

(Received for publication, December 1, 1998, and in revised form, January 29, 1999)

Alain Roussel‡§, Stephane Canaan§¶, Marie-Pierre Egloff‡, Mireille Riviere¶, Liliane Dupuis¶,Robert Verger¶, and Christian Cambillau‡i

From the ‡Architecture et Fonction des Macromolecules Biologiques, CNRS-IFR1 UPR 9039 and ¶Laboratoire de LipolyseEnzymatique, CNRS-IFR1 UPR 9025, 31 chemin Joseph Aiguier, 13402 Marseille cedex 20, France

Fat digestion in humans requires not only the classi-cal pancreatic lipase but also gastric lipase, which isstable and active despite the highly acidic stomach en-vironment. We report here the structure of recombinanthuman gastric lipase at 3.0-Å resolution, the first struc-ture to be described within the mammalian acid lipasefamily. This globular enzyme (379 residues) consists of acore domain belonging to the a/b hydrolase-fold familyand a “cap” domain, which is analogous to that presentin serine carboxypeptidases. It possesses a classical cat-alytic triad (Ser-153, His-353, Asp-324) and an oxyanionhole (NH groups of Gln-154 and Leu-67). Four N-glycosy-lation sites were identified on the electron densitymaps. The catalytic serine is deeply buried under a seg-ment consisting of 30 residues, which can be defined asa lid and belonging to the cap domain. The displacementof the lid is necessary for the substrates to have access toSer-153. A phosphonate inhibitor was positioned in theactive site that clearly suggests the location of the hy-drophobic substrate binding site. The lysosomal acidlipase was modeled by homology, and possible explana-tions for some previously reported mutations leading tothe cholesterol ester storage disease are given based onthe present model.

Since 1990, when the first three-dimensional structures of afungal (Rhizomucor miehei lipase) and a mammalian lipase(human pancreatic lipase (HPL))1 were published, growing in-terest in lipolysis has led to the structural determination ofseveral lipases of various origins, including those present inbacteria, fungi, and mammals. All the lipases investigated sofar vary considerably in size and in their amino acid sequences.However, they are all serine esterases belonging to the a/bhydrolase superfamily (1) in which the nucleophilic serine, partof a Ser-His-(Asp/Glu) triad, is located in an extremely sharpturn (nucleophilic elbow). Another feature that is common to allthe members of the a/b hydrolase superfamily as well as toproteases is the occurrence of an oxyanion hole, which stabi-

lizes the transition state. Some organophosphorous compoundsinhibit lipases in a similar way to what occurs in the case ofserine proteases (2).

The three-dimensional structures of several complexes con-sisting of lipases bound to covalent inhibitors have been solved:R. miehei lipase (3, 4) and Candida antartica B lipases boundto a C6-alkyl phosphonate (5), Candida rugosa lipase bound tolong chain alkyl sulfonyl (6), Pseudomonas cepacia lipase (7) aswell as cutinase (8) bound to a dialkylcarbamoylglycerophos-phonate, and human pancreatic lipase-colipase complex boundto C11-alkyl phosphonate (9). It has been established that thecovalently inhibited lipases are in the so called “open” confor-mation, i.e. that the lid has moved away to give free access tothe active-site serine. It has been suggested that this mecha-nism may be instrumental in the binding of lipases to thewater-lipid interface and that the presence of a lid in thestructure of the enzyme may be involved in the interfacialactivation process (4, 10).

Among the mammalian lipases, the acid lipases belong to afamily of enzymes that have the ability to withstand acidicconditions. This family that includes the preduodenal lipasesand human lysosomal lipase shows no sequence homology withany other known lipase families (11). The preduodenal lipasesform a group of closely related enzymes originating either fromthe stomach, the tongue, or the pharynx (12). They all have alow pH optimum, and none of them require any specific proteincofactor. Human gastric lipase (HGL, EC 3.1.1.3) is secreted bythe chief cells located in the fundic part of the stomach (13),where it initiates the digestion of triacylglycerols (14, 15). Themaximum specific activities of HGL are 1160 units/mg on TC4(pH 6.0), 1110 units/mg on TC8 (pH 6.0), and 600 units/mg onIntralipidTM (pH 5.0) (16). Native HGL has an apparent mo-lecular mass of 50 kDa and is a highly glycosylated moleculewith 4 potential N-glycosylation sites (17). The glycan moietywas estimated to account for around 15% of its total proteinmass (18).

This enzyme plays a crucial role in newborns, because pan-creatic lipase is not yet fully developed at this age (15). Thephysiological importance of gastric lipase has been suspectedfor some time, based on pathological situations involving pan-creatic exocrine insufficiency, such as the late stage of chronicpancreatitis or cystic fibrosis. In these cases, even in the com-plete absence of pancreatic lipase, the patients still absorb ahigh percentage of their ingested dietary fat (19, 20). In sub-stitutive enzymatic therapy, the use of acidic-resistant lipasesshould in principle yield more satisfactory results than thepancreatic preparations currently in use. The co-administra-tion of acidic lipases, which hydrolyze dietary lipids underacidic conditions, should help to treat patients with variousforms of pancreatic deficiency. Physiological studies have

* This research was carried out in the framework of the EU Biotech-lipase project (BIO2-CT94-3041). The costs of publication of this articlewere defrayed in part by the payment of page charges. This article musttherefore be hereby marked “advertisement” in accordance with 18U.S.C. Section 1734 solely to indicate this fact.

The atomic coordinates and structure factors (code 1HLG) have beendeposited in the Protein Data Bank, Brookhaven National Laboratory,Upton, NY.

§ These authors have made equal contributions to the present study.i To whom correspondence should be addressed. Fax: 33-491-16-45-

36; E-mail: cambillau@afmb.cnrs-mrs.fr.1 The abbreviations used are: HPL, human pancreatic lipase; HGL,

human gastric lipase; rHGL, recombinant HGL; HLAL, human lysoso-mal acid lipase; SIRAS, single isomorphous replacement with anoma-lous scattering; HPP, human protective protein.

THE JOURNAL OF BIOLOGICAL CHEMISTRY Vol. 274, No. 24, Issue of June 11, pp. 16995–17002, 1999© 1999 by The American Society for Biochemistry and Molecular Biology, Inc. Printed in U.S.A.

This paper is available on line at http://www.jbc.org 16995

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

shown that preduodenal lipases are capable of acting not onlyin the stomach but also in the duodenum in synergy with apancreatic lipase (14). Various clinical studies have been con-ducted on both animals and humans to assess the efficacy ofenzymatic replacement therapies using acid-resistant lipasesto treat exocrine pancreatic insufficiency (21). This treatmentsignificantly increased the weight and reduced the steatorrheain dogs.

Despite the close amino acid sequence similarities (59% ofthe amino acids are identical) between HGL (17) and humanlysosomal acid lipase (HLAL, EC 3.1.1.3) (22, 23), HGL lacksthe cholesteryl ester hydrolase activity reported in HLAL. Thelatter enzyme hydrolyzes not only the triglycerides that aredelivered to the lysosomes by low density lipoprotein receptor-mediated endocytosis but also cholesteryl esters (24). The cho-lesterol released by this reaction plays an important regulatoryrole in cellular sterol metabolism. Defective HLAL activity hasbeen found to be associated with two rare autosomal recessivetraits, Wolman disease and cholesteryl ester storage disease. InWolman disease (25), a lack of HLAL activity results in apronounced accumulation of cholesteryl esters and triacylglyc-erols in the lysosomes in most of the body tissues. The patientsusually succumb to hepatic and adrenal failure within the firstyear of life. Cholesteryl ester storage disease, the other clini-cally recognized phenotypic form of HLAL deficiency, follows amore benign clinical course (26), and a residual HLAL activityhas been detected. Since the cloning of the cDNA and determi-nation of the genomic organization of the gene (LIPA) locatedon chromosome 10, which encodes HLAL (22, 23, 27), somedeleterious LIPA gene mutations have been identified (28–34).Most of these mutations affect either the mRNA splicing or theamino acid sequence of HLAL. The exact correlations betweenthese mutations and the biochemical and clinical phenotypesstill remains to be elucidated, however. In this study, thecrystal structure of recombinant HGL at 3.0-Å resolution isdetermined, and a model of HLAL is discussed and used topossibly explain the previously reported cholesteryl ester stor-age disease mutations.

MATERIALS AND METHODS

Recombinant HGL (rHGL) Expression and Purification—rHGL wasexpressed in the baculovirus/insect cell system (48). The active enzymewas produced on a large scale (5–13 mg/liter) from recombinant bacu-lovirus-infected insect cells using a bioreactor and its specific activity(mmolezmin21zmg21) was around 700 units/mg (49). The amino acidsequence (KLHPG) of rHGL in the purified protein starts at residue 4.

rHGL Crystallization and Enzymatic Activity in the Crystal—Crys-tallization experiments were performed using the hanging-drop vapordiffusion method. Crystals of the rHGL were obtained by mixing 2 ml ofa well solution (2 M ammonium sulfate, 1.4% tert-butanol, at pH 5.0)with 2 ml of a protein solution at 5–6 mg/ml. The crystals are cubic,space group I 21 3 with cell dimensions a 5 b 5 c 5 244.0 Å3. Theprotein mass, 47,673 Da, was determined from collected crystals bymatrix-assisted laser desorption ionization time-of-flight spectroscopy.There are two molecules in the asymmetric unit (see below), and the Vm

was estimated to be 6.35 Å3/Da (81% water content). For the mercuryderivative preparation, the crystals were transferred in a syntheticliquor corresponding to the well solution containing 23 mM mercuricacetate.

To assess the catalytic activity of the crystallized enzyme, tests wereperformed on both dissolved and intact crystals. Ten crystals werewashed three times in the crystallization buffer and were subsequentlydissolved in 50 ml of water, and the protein concentration was estimatedby absorbance at 230 nm. The lipase activity was measured titrimetri-cally at 37 °C using a pH stat (metrohm) at pH 5.7 with a tributyrinemulsion as the substrate: 0.5 ml of tributyrin added to 14.5 ml of 150mM NaCl, 2 mM taurodeoxycholate, and 2 mM bovine serum albumin(16).

The specific activity (mmolezmin21zmg21) was calculated and found tobe around 580 units/mg. To test the catalytic activity in situ, a singlerHGL crystal was incubated in the crystallization solution containing

0.1 mM Nitroblue tetrazolium and 0.75 mM 5 bromo-chloro-3-indoylbutyrate as the substrate. After 24 h of incubation, the crystal wasintensely colored in blue/gray.

Data Collection and Heavy Metal Derivative Search—All data setswere collected at 100 K using a cryo-stream cooler from Oxford Cryo-systems. A first native and a derivative data set were collected in-houseusing a 300-mm MAR Research imaging plate detector mounted on aRU200 rotating anode generator (Rigaku, Tokyo, Japan). The generatorwas operated at 3.2 kW with a focal spot size of 0.3 3 0.3 mm2. A secondnative data set was collected at LURE (Orsay, France) on DW32 beam-line at 0.963 Å wavelength using a 345-mm MAR Research imagingplate. All data were collected in frames of 1.0 degree and processed withDENZO. The scaling was performed with SCALA (CCP4 (50)), and thederivative was merged with FHSCAL. Data collection statistics aregiven in Table I. The fact that most of the crystals diffracted only verypoorly made the heavy metal derivative search laborious. In the end, itled to one mercury derivative diffracting to 3.6-Å resolution, isomor-phous to the native crystal.

Phase Determination—The position of the heavy metal was deter-mined using Patterson methods. The MLPHARE software program (51)was then used to refine the heavy atom parameters and calculatephases to 4.0 Å, taking into account both isomorphous and anomalousdifferences. Resulting single isomorphous replacement with anomalousscattering (SIRAS) phases were considerably improved after flattening80% of the solvent using density modification. At this stage, the mapindicated clearly that the handedness of the helices was wrong. Repeat-ing the previous steps with the heavy atom position giving the negativecoordinates yielded a suitable map for determining an envelope for eachof the two molecules in the asymmetric unit. The noncrystallographicsymmetry operators were determined using both the GLRF programand the relation between the two molecular replacement solutions(automated molecular replacement density modification (52)) obtainedfrom a search using a bones model (MAPMAN (53)) calculated on thebasis of one molecule. The density modification program was then usedagain to improve the initial phases (from MLPHARE) by performingsimultaneously solvent flattening, histogram matching, 2-fold noncrys-tallographic symmetry averaging, and phase extension. The best mapwas obtained when a suitable mask around one of the molecules wasadded to the program. The resolution of the derivative data set wasactually better than that of the native one, and the map was thereforecalculated using the derivative structure factor amplitudes with thephases extended to 3.6 Å.

Model Building, Refinement, and Analysis—The model was builtwith the program TURBO-FRODO (54) using the recently developedab-initio building tools, which make it possible to build a model fromplanar pseudo-residues in a very short time. When the connectivity andthe direction of the polypeptide chain have been determined, the pseu-do-residues are automatically replaced by the actual residues in thesequence. Side-chain fitting is then performed manually. When most ofthe model had been built, a 3.0-Å native data set was collected onsynchrotron radiation at the LURE. These better data were then usedto calculate a new map, with which the model was completed (370residues of 376 in the recombinant protein). The four sugar glycosyla-tion sites were already clearly identified in this experimental electrondensity map. Refinement was carried out in X-PLOR (55) against the

TABLE IData collection, phasing and refinement statistics

Native 1 Hg derivative Native 2

Resolution (Å) 15–4.1 15–3.6 15–3.0Completeness (%) 97.9 (99.4) 98.9 (95.6) 99.6 (100)R-merge (highest shell) (%) 26.9 (47.5) 16.7 (34.7) 6.8 (39.9)I/s I 2.7 (1.6) 3.6 (2.1) 10.9 (2.0)Redundancy 5.3 (5.2) 3.1 (3.1) 4.0 (4.1)

Riso 28.4 (15–4.1 Å)Rcullis 0.85 acentric/ 0.76 centricFigure of merit 0.268 (0.255 acentric/ 0.456 centric)Phasing power 1.14 acentric/ 0.9 centricRefinement statistics

Resolution 15.0–3.0R / Rfree (%) 22.5 / 25.1Nr reflections 45522 (2381 free)Nr atoms 6194s bond (Å) / s angle (°) 0.007/ 1.392Ramachandran Procheck

(61) zones 1–3 (%)85.5 / 13.3 / 1.2

Crystal Structure of Recombinant Human Gastric Lipase16996

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

3.0-Å data. Five percent of the data were set aside for calculating andmonitoring of the free R-factor (Rfree). Refinement involved cycles ofsimulated annealing using the slow-cool procedure interspersed withmanual rebuilding. Noncrystallographic symmetry restraints were ap-plied during the whole refinement procedure. Individual B-factor re-finement was subsequently performed. The final model consisted of6194 atoms; the final R-factor was 22.5%, and the R-free factor was25.1%. The complete refinement statistics, given in Table I, the qualityof the Ramachandran plot, and the electron density indicate that themodel is better than expected for a structure at 3.0-Å resolution. Mapscalculated from the derivative structure factor amplitudes showed aclear electron density for the mercury derivative bound to cysteine 244but no significant movements of any part of the model. Fig. 1 was drawnup with Molscript (56) and Raster 3D (57), Fig. 2 with Clustal W (58)and Alscript (59), Figs. 3, 4, and 6 with Turbo-Frodo (54), and Fig. 5with GRASP (60). The coordinates and structure factor amplitudeshave been deposited in the Protein Data Bank with entry code 1HLG.

RESULTS AND DISCUSSION

After numerous trials using weakly diffracting crystal formsobtained from the native gastric lipases purified over the last

10 years from humans, rabbit, and dog (18), useful cubic crys-tals were finally obtained from rHGL expressed in the insectcell/baculovirus expression system. The structure of rHGL hasbeen solved by a combination of SIRAS, solvent flattening, and2-fold averaging at 4.5 Å resolution. Phase extension proce-dures have made it possible to extend the resolution to 3.6 Å. Apreliminary model was constructed at this resolution. The finalstructure was refined to 3.0-Å resolution using restrained non-crystallographic symmetry between the two molecules. Thecrystallographic R-factor based on all the data between 15.0and 3.0 Å is 22.6%, and the corresponding R-free is 25.1% witha model containing residues 9 to 53 and 57 to 379, 6 sugarresidues located on the 4 potential N-glycosylation sites, and 46water molecules/monomer.

Overall Structure—rHGL consists of one globular domain(Fig. 1) and belongs to the a/b hydrolase-fold family (1). Thecore domain, which is located between residues 9–183 and309–379 (Figs. 1 and 2) contains a central b sheet composed of

FIG. 1. Stereo ribbon representation of human gastric lipase. A, view with the central b-sheet parallel to the paper plane. The cap domainon top of the a/b hydrolase core domain is colored magenta and the putative lid, green. For the a/b hydrolase fold, the color code is helices (red),strands (blue), turns and random coil (yellow). The catalytic triad, Ser-153, His-353, and Asp-324, the disulfide bridge, and the Asn-attached sugarsare shown in ball and stick form. The disordered loop is indicated by an arrow. The secondary structures were calculated with Dictionary ofSecondary Structure of Proteins. B, view rotated about 90° from the previous one. Cys-244 is visible above the catalytic triad. C, schematicrepresentation of the rHGL catalytic domain topology with the secondary structure elements identified (H, helix; B, strand) as well as the catalytictriad.

Crystal Structure of Recombinant Human Gastric Lipase 16997

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

8 strands, 7 of which are parallel and 1 antiparallel (strand 2)with 1(-2)435678 connectivity, and 6 helices, 3 on each side ofthe b-sheet (Fig. 1C). Two segments are missing in the electrondensity maps. The first of these was from residues 4 to 9,because the electron density starts abruptly at residue 9. In theregion of the second lacking segment (residues 54 to 56), somefaint electron density was observed, suggesting that the loopmay be intact but disordered (Fig. 1). The accessibility of thisloop is consistent with the previously reported preferentialtrypsin cleavage site (Arg-55) of rabbit gastric lipase (35). Incomparison with the canonical a/b hydrolase fold, an extrahelix (a1) is present at the N terminus. Helix aA is shorter andhas moved to the bottom of the structure, helix aB is replacedby two helices (aB1 and aB2), and helix aD is replaced by anextra domain (residues 184 to 308, Figs. 1 and 2) locatedbetween strands 6 and 7. Protrusions have been observed inother lipases, generally constituting the device covering theactive site and called the lid. A “cap” domain occurs at the samelocation in wheat serine carboxypeptidase II (residues 181 to311, WCSII) (36) and in human protective protein (residues 181to 347, HPP) (37), two protease members of the a/b hydrolase-fold family. In HPL, the lid (237 to 261) lies between strands 8and 9, and a b-sandwich domain extends the enzyme at the Cterminus (336 to 449). In R. miehei lipase (4), a lid-containingextra domain formed by an a-helix and two short strands(82–109) can be observed between b4 and aB. The situation is

more complex in C. rugosa lipase (6), where three protrusionshave been observed that together cover the central core of theenzyme: between strands 1 and 2 (residues 30 to 98), betweenb6 and aD (residues 241 to 331), and between b7 and aE(residues 339 to 415). In the latter enzyme, however, the lidcovering the active site is composed of two helices belonging tothe first protrusion and comprising residues 66 to 92.

The amino acid sequence of the enzyme includes four con-sensus N-glycosylation sites at Asn-15, -80, -252, and -308 (Fig.2). Electron density patches attributable to the first sugarattached to Asn have been observed at all four sites (Fig. 1).GlcNAc residues were therefore introduced that nicely sus-tained the refinement. No electron density was observed in thecase of the fucose residues attached at GlcNAc 1, in line withthe fact that the expression of the enzyme was carried out in aweakly glycosylating cell type. The residual electron densitymade it possible to introduce a second GlcNAc residue at sites80 and 252. Sites 15, 80, and 252 are located on one side of themolecule, whereas site 308 is on the other. Site 80 is located onthe core of the protein, whereas site 252, which belongs to thecap domain, lies between helices ae5 and ae6. It is worth notingthat the glycan chains on the above-mentioned sites are in closecontact, which enhances the interactions between the core andthe cap.

The amino acid sequence of HGL contains three cysteineresidues, one of which is free, whereas the other two are in-

FIG. 2. Sequence alignment of acid preduodenal lipases and human lysosomal acid lipase. Note that there are no insertions and onlyone deletion in the alignment. Identities are displayed with yellow shading, and the values with respect to HGL are: HLAL, 59%; rabbit gastriclipase, 85%; dog gastric lipase, 85%; rat lingual lipase, 76%; calf pregastric esterase, 75%. The catalytic triad, Ser-153, His-353, and Asp-324 is ingreen, blue, and red, respectively. The oxyanion hole residues are boxed. The secondary structures have been calculated with DSSP.

Crystal Structure of Recombinant Human Gastric Lipase16998

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

volved in a disulfide bridge (35, 38–40). Based on the presentthree-dimensional structure, the free cysteine can be unambig-uously assigned to residue 244 (Fig. 3A), and the disulfidebridge, to residues 227–236 (Fig. 1), as previously suggested byLohse et al. (40) using site-directed mutagenesis.

The Catalytic Machinery—In lipases, as well as in serineproteases, the catalytic machinery consists of a triad and anoxyanion hole (2, 4). The present three-dimensional structureof rHGL is also equipped with this machinery (Fig. 3B). Thenucleophilic serine belonging to the usual consensus sequenceGX1SX2G (X1 and X2 being His and Gln, respectively, see Fig.2) is located at position 153, between b5 and aC. This Ser-153has an e conformation (41), which is a characteristic feature ofall the enzymes within the a/b hydrolase-fold family (1). His-353 (between b8 and aF) and Asp-324 (between b7 and aE) arethe other two residues forming the catalytic triad, as previouslyidentified by site-directed mutagenesis (42). The rHGL triad is

almost superimposable on all the other lipase catalytic triads.The nucleophilic serine is covered by the cap (184–308) and istherefore not freely accessible to lipidic substrate molecules(Fig. 1). The environment of the catalytic triad seems not todisplay any special features. However, one might expect His-353 to show local pKa decrease to be able to explain the acidicpH optimum of rHGL. There are no charged residues within a10-Å sphere centered on the Ser-153 Og atom. This absencemakes it difficult to suggest the mechanism possibly responsi-ble for the local pKa modulation.

Cys-244 was found to be very close to Ser-153 and His-353,and Og and Sg are at a distance of only 5.2 Å. This cysteine,which is completely buried, has nevertheless permitted thebinding of the mercury acetate derivative. In the presence ofspecific cysteine reagents (C12-docecyldithio-5-2-nitrobenzoicacid or 5,59-dithiobis(2-nitrobenzoic acid or 4,49-dithiopyri-dine), it has been established that a single cysteine can react

FIG. 3. The active site. A, stereo view of the electron density map around the active site residues Ser-153, His-353, and the single cysteine 244(The density has been contoured at 1 s level). B, representation of the HGL catalytic triad and of the oxyanion hole (Gln-154, Leu-67) superposedto the same part of HPL (green, in Ca tracing).

Crystal Structure of Recombinant Human Gastric Lipase 16999

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

stoichiometrically with a concomitant loss of enzymatic activity(38). Another member of the acid lipase family (calf pregastricesterose) has a threonine at this position (Fig. 2). Furthermore,rHGL Cys-244 Thr mutant has been constructed, and no loss ofactivity was observed in the purified enzyme.2 In view of all theabove findings, the possibility that cysteine 244 may partici-pate directly in the catalytic mechanism can be ruled out.Based on the three-dimensional structure, however, the spe-cific inhibition observed with sulfhydryl reagents was attrib-uted to a steric hindrance occurring at the level of the activesite (see Fig. 3A).

The oxyanion hole is a device that stabilizes the oxyaniontransition state via hydrogen bonds with two main-chain nitro-gens (2). It is often evidenced by using organophosphate ororganophosphonate complexes in which the two main chain NHgroups establish short hydrogen bonds with an oxygen belong-ing to the phosphonate group. In rHGL, the oxyanion hole hasbeen identified on the basis of comparisons with other esterasesor lipases. Within the a/b hydrolase-fold family, one mandatorycomponent of the oxyanion hole is the NH group of the residuefollowing the nucleophile Ser (Gln-154 in rHGL). The secondNH group occupies a different position along the sequence butalways originates from the same spatial region. In somelipases, the oxyanion hole is not preformed in the closed con-formation but results from the concerted movements duringthe opening process of the lid or of a small loop bearing thenitrogen atom of the main chain. Comparisons between rHGLand the open form of the C11-phosphonate HPL-colipase ter-nary complex indicates that the second oxyanion NH groupbelongs to Leu 67 (Fig. 3B). The Ca tracing of the loop bearingLeu-67 is fully superimposable on that of the open HPL (Fig.3B). It can therefore be concluded that the oxyanion hole ofrHGL is preformed as in esterases, cutinase (43), and C. rugosalipase (6). The second oxyanion hole residue can be of verydiverse nature in lipolytic enzymes. R. miehei lipase (4) orcutinase (43) possess a polar noncharged residue, Ser-82 andSer-42, respectively. Furthermore, the Ser-42 side chain incutinase has been found to be a third essential component ofthe oxyanion hole (44). It seems, however, that the Leu-67 sidechain in rHGL may play a similar role to that of Phe-77 in theopen form of HPL, leading to an exquisite interaction with thealkyl group of the phosphonate inhibitor (see below).

The Cap and the Putative Lid—The cap domain fold (resi-dues 184–308) is an intricate mixture of 8 helices, turns, andrandom coils (Figs. 1 and 2). When this domain is removed fromrHGL on a display, the active site becomes accessible to sol-vent. An apparently sufficient degree of accessibility can be

achieved, however, by removing a shorter stretch of residuesbetween residues 215 and 244 (Fig. 4). The flanking residues ofthis stretch are Pro-214 and Gly-245. It has been suggestedthat the trans-cis-proline isomerization may be one of the mainfactors involved in the C. rugosa lipase lid conformationalchanges (45). This, together with the flexibility of the glycineresidue, suggests that these residues might act as hinges in theopening of the lid. It seems unlikely, however, that this liddisplacement might occur as a rigid body movement, as de-scribed previously in R. miehei lipase (4). Given the complexfolding of this lid, the opening of rHGL is probably a morecomplex process than those described previously. Cysteine 244would therefore be the last residue in the lid, before the hinge.The resulting displacement of the chemically modified Cys-244,which occurs upon the lid opening, might therefore be toolimited to abolish the steric hindrance during the formation ofthe lipase-substrate complex.

Cap domains of this kind have been found to have the sametopology in WSCII (36), and in the related HPP, a serine car-boxypeptidase zymogen occurring in lysosomes (37). In HPP,the protruding domain is located, as in rHGL, between strands6 and 7, comprising residues 181–347. The putative excisionpeptide is thought to be situated between residues 285 and 298.When this excision peptide is not removed, the active-site ser-ine remains covered (Fig. 4A), in the same way as the putativelid covers the rHGL serine (Fig. 4B). In both cases, the rest ofthe cap domain forms a ring around the active site, ensuringthe substrate specificity in the case of HPP and the lipid affin-ity in the case of rHGL. In the framework of the above hypoth-esis, the lid remains covalently attached to the body of rHGL2 Canaan, S. (1999) Biochem. Biophys. Res. Commun., in press.

FIG. 4. Representations of humanprotective protein (A) and rHGL (B)in the same orientation with the cat-alytic triad in red. A, Corey-Pauling-Koltun view of the core domain of HPP(yellow) and of its cap domain (blue). Theputative excision peptide is given by a Catrace (green). B, Corey-Pauling-Koltunview of the core domain of rHGL (yellow)and its cap domain (blue). The putative lidis given by a Ca trace (green).

FIG. 5. Molecular surface representations of rHGL. A, the com-plete enzyme; B, same as in A but with the putative lid depleted and theC11-phosphonate enantiomers (red) of HPL represented in the activesite (9). Hydrophobic residues are colored blue.

Crystal Structure of Recombinant Human Gastric Lipase17000

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

after opening, in contrast to HPP.Active-site Accessibility and Model of a Phosphonate Com-

plex—On the water-accessible surface of rHGL, a large numberof cavities that are accessible to water can be observed in thestructure (not shown). These are predominantly located at theinterface between the core and the cap domains, which sug-gests that their packing may be imperfect. Of particular inter-est is the small cavity (23 Å3) located above the active-siteserine. Another cavity of 113 Å3, lined by residues 274 to 278and 284 to 287, makes it possible for the solvent to gain accessto Ser-153, provided that side-chain displacements can occur.

The question therefore arises as to whether or not the abovechannels might make it possible for substrates or inhibitors toreach to the active site. There exist several arguments againstthis hypothesis. It is in fact impossible for a bulky long chaintriglyceride to travel through these channels to be accommo-dated in the closed active site. When the putative lid domainwas removed on the display, from residues 215 to 244, a largehydrophobic surface appears around the active site and can actas a lipid binding site (Fig. 5). As observed in the three-dimen-sional structures of open lipases, this hydrophobic surface mayface the lipid interface after the lid opening. Accordingly, thecomplementary surface part of the lid domain is also composedof hydrophobic residues facing the putative lipid binding site inthe closed form of rHGL.

When positioning the two enantiomers of the C11-phospho-nate inhibitor, keeping their spatial position as observed in thethree-dimensional structure of the HPL-C11-phosphonate com-plex, it was noted that both molecules fit well against thebottom of the active-site cleft (Figs. 5B and 6). The C11 enan-tiomer corresponding to the hydrolyzable triglyceride acylchain fit nicely into the putative acyl binding site and inter-acted only with hydrophobic residues (Fig. 6). The other phos-phonate enantiomer is surrounded by a more polar environ-ment, as in the HPL-C11 phosphonate complex.

The Lysosomal Acid Lipase—HLAL displays 59% identityand 75% homology with HGL, without any insertion or deletion(Fig. 2). A straightforward homology model of HLAL was built,and this model was examined in the light of the mutations,resulting in a clinical phenotype in humans. Besides mutantsbearing large deletions, few single mutations associated withcholesteryl ester storage disease have been described. The de-letion of fragment 205 to 253 and fragment 254 to 277 affect thecap domain (29, 31, 32, 46). It is possible that the residual

activity reported in the latter study (5–10%) might be becauseof the fact that the a/b hydrolase fold of the catalytic domainhas remained intact. The mutation Leu-273 3 Ser creates anew potential glycosylation site (NMS), and this mutant ex-pressed in HeLa cells has been found to have a higher molec-ular mass than that of the wild type HLAL expressed in thesame system (29, 30). This residue is water-exposed, and thebulky glycan moiety may prevent either the substrate bindingor the movement of the lid. The mutation Leu-336 3 Pro islocated just inside helix aE and probably destabilizes its struc-ture, leading to an inactive enzyme as initially proposed bySeedorf et al. (32). The mutation Gly-66 3 Val is easily inter-pretable in structural terms, because the valine side chainclashes with the active-site serine 153 and also partly blocksthe putative triglyceride binding site, on similar lines to whathas been found to occur with pancreatic lipase RP1 (47).

Acknowledgments—We are grateful to Frederic Carriere for helpfuldiscussions and for his constant interest in this topic. The contributionof C. Wicker-Planquart is particularly acknowledged for setting up theexpression of rHGL in a baculovirus/insect cell system. R. Verger ac-knowledges the constant help of the Institut de Recherche Jouveinal/Park Davis since 1989. We thank C. Abergel and C. Jelsch for perform-ing the preliminary crystallization trials, J. Bonicel for performingmass spectroscopy on the rHGL, and S. Diotallevi for technical assist-ance. We thank J. Perez and A. Bentley for their assistance during thedata collection at the LURE. We also thank Dr. Jessica Blanc forcorrecting the English.

REFERENCES

1. Ollis, D. L., Cheah, E., Cygler, M., Dijkstra, B., Frolow, F., Franken, S. M.,Harel, M., Remington, S. J., Silman, I., Schrag, J., Sussman, J. L., Vers-chueren, K. H. G., and Goldman, A. (1992) Protein Eng. 5, 197–211

2. Kraut, J. (1977) Annu. Rev. Biochem. 46, 331–3583. Derewenda, Z. S., Derewenda, U., and Dodson, G. G. (1992) J. Mol. Biol. 227,

818–8394. Brzozowski, A. M., Derewenda, U., Derewenda, Z. S., Dodson, G. G., Lawson,

D. M., Turkenburg, J. P., Bjorkling, F., Huge-Jensen, B., Patkar, S. A., andThim, L. (1991) Nature 351, 491–494

5. Uppenberg, J., Ohrner, N., Norin, M., Hult, K., Kleywegt, G. J., Patkar, S.,Waagen, V., Anthonsen, T., and Jones, T. A. (1995) Biochemistry 34,16838–16851

6. Grochulski, P., Bouthillier, F., Kazlauskas, R. J., Serreqi, A. N., Schrag, J. D.,Ziomek, E., and Cygler, M. (1994) Biochemistry 33, 3494–3500

7. Lang, D., Mannesse, M., de Haas, G., Verheij, H., and Dijkstra, B. (1998) Eur.J. Biochem. 254, 333–340

8. Longhi, S., Mannesse, M., Verheij, H. M., de Haas, G. H., Egmond, M.,Knoops-Mouthuy, E., and Cambillau, C. (1997) Protein Sci. 6, 275–286

9. Egloff, M.-P., Marguet, F., Buono, G., Verger, R., Cambillau, C., and vanTilbeurgh, H. (1995) Biochemistry 34, 2751–2762

10. van Tilbeurgh, H., Egloff, M.-P., Martinez, C., Rugani, N., Verger, R., andCambillau, C. (1993) Nature 362, 814–820

11. Petersen, S. B., and Drabløs, F. (1994) in Lipases: Their Biochemistry, Struc-

FIG. 6. Stereo view of the interac-tions of the HPL C11-alkyl phospho-nate inhibitor (9) with rHGL. For theC11-phosphonate enantiomer the identi-fied hydrophobic cleft extends beyond theC11 carbon atom by about 7 additionalatoms (mimicking a stearoyl chain) andinvolves on each side: F203, V198, L202and F212, F207, I206, F205, respectively.

Crystal Structure of Recombinant Human Gastric Lipase 17001

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

ture, and Application (Woolley, P., and Petersen, S., eds) pp. 23–48, Cam-bridge University Press, Cambridge, England

12. Carriere, F., Gargouri, Y., Moreau, H., Ransac, S., Rogalska, E., and Verger, R.(1994) in Lipases: Their Structure, Biochemistry, and Application (Woolley,P., and Petersen, S. B., eds) pp. 181–205, Cambridge University Press,Cambridge, England

13. Moreau, H., Bernadac, A., Gargouri, Y., Benkouka, F., Laugier, R., and Verger,R. (1989) Histochemistry 91, 419–423

14. Carriere, F., Barrowman, J. A., Verger, R., and Laugier, R. (1993) Gastroen-terology 105, 876–888

15. Hamosh, M. (1990) Lingual and Gastric Lipases: Their Role in Fat Digestion(Hamosh, M., ed) CRC Press, Inc., Boston

16. Gargouri, Y., Pieroni, G., Riviere, C., Sauniere, J.-F., Lowe, P. A., Sarda, L.,and Verger, R. (1986) Gastroenterology 91, 919–925

17. Bodmer, M. W., Angal, S., Yarranton, G. T., Harris, T. J. R., Lyons, A., King,D. J., Pieroni, G., Riviere, C., Verger, R., and Lowe, P. A. (1987) Biochim.Biophys. Acta 909, 237–244

18. Moreau, H., Abergel, C., Carriere, F., Ferrato, F., Fontecilla-Camps, J. C.,Cambillau, C., and Verger, R. (1992) J. Mol. Biol. 225, 147–153

19. Muller, D. P. R., McCollum, J. P. K., Tompeter, R. S., and Harries, J. T. (1975)Gut 16, 838

20. Lapey, A., Kattwinkel, J., di Sant Agnese, P. A., and Laster, L. (1974) J. Pe-diatr. 84, 328–334

21. Suzuki, A., Mizumoto, A., Sarr, M. G., and Dimagno, E. P. (1997) Gastroen-terology 112, 2048–2055

22. Ameis, D., Merkel, M., Eckerskorn, C., and Greten, H. (1994) Eur. J. Biochem.219, 905–914

23. Anderson, R. A., and Sando, G. N. (1991) J. Biol. Chem. 266, 22479–2248424. Goldstein, J. L., and Brown, M. S. (1977) Annu. Rev. Biochem. 46, 897–93025. Patrick, A. D., and Lake, B. D. (1969) Nature 222, 1067–106826. Burke, J., and Schubert, W. (1972) Science 176, 309–31027. Aslanidis, C., Klima, H., Lackner, K. J., and Schmitz, G. (1994) Genomics 20,

329–33128. Aslanidis, C., Ries, S., Fehringer, P., Buchler, C., Klima, H., and Schmitz, G.

(1996) Genomics 33, 85–9329. Pagani, F., Garcia, R., Pariyarath, R., Stuani, C., Gridelli, B., Paone, G., and

Baralle, F. E. (1996) Hum. Mol. Genet. 5, 1611–161730. Pariyarath, R., Pagani, F., Stuani, C., Garcia, R., and Baralle, F. E. (1996)

FEBS Lett. 397, 79–8231. Ameis, D., Brockmann, G., Knoblich, R., Merkel, M., Ostlund, R. E., Yang,

J. W., Coates, P. M., Cortner, J. A., Feinman, S. V., and Greten, H. (1995)J. Lipid Res. 36, 241–250

32. Seedorf, U., Wiebusch, H., Muntoni, S., Christensen, N. C., Skovby, F., Nickel,V., Roskos, M., Funke, H., Ose, L., and Assmann, G. (1995) Arterioscler.Thromb. Vasc. Biol. 15, 773–778

33. Anderson, R. A., Byrum, R. S., Coates, P. M., and Sando, G. N. (1994) Proc.Natl. Acad. Sci. U. S. A. 91, 2718–2722

34. Pagani, F., Zagato, L., Merati, G., Paone, G., Gridelli, B., and Maier, J. A.(1994) Hum. Mol. Genet. 3, 1605–1609

35. De Caro, J., Verger, R., and De Caro, A. (1998) Biochim. Biophys. Acta 1386,39–49

36. Liao, D. I., Breddam, K., Sweet, R. M., Bullock, T., and Remington, S. J. (1992)Biochemistry 31, 9796–9812

37. Rudenko, G., Bonten, E., d’Azzo, A., and Hol, W. (1995) Structure 3, 1249–125938. Gargouri, Y., Moreau, H., Pieroni, G., and Verger, R. (1988) J. Biol. Chem. 263,

2159–216239. Aoubala, M., Bonicel, J., Benicourt, C., Verger, R., and De Caro, A. (1994)

Biochim. Biophys. Acta 1213, 319–32440. Lohse, P., Lohse, P., Chahrokh-Zadeh, S., and Seidel, D. (1997) J. Lipid Res.

38, 1896–190541. Brenner, S. (1988) Nature 334, 528–53042. Lohse, P., Chahrokh-Zadeh, S., Lohse, P., and Seidel, D. (1997) J. Lipid Res.

38, 892–90343. Martinez, C., Nicolas, A., van Tilbeurgh, H., Egloff, M.-P., Cudrey, C., Verger,

R., and Cambillau, C. (1994) Biochemistry 33, 83–8944. Nicolas, A., Egmond, M., Verrips, C. T., De Vlieg, J., Longhi, S., Cambillau, C.,

and Martinez, C. (1996) Biochemistry 35, 398–41045. Grochulski, P., Li, Y., Schrag, J. D., and Cygler, M. (1994) Protein Sci. 3, 82–9146. Klima, H., Ullrich, K., Aslanidis, C., Fehringer, P., Lackner, K., and Schmitz,

G. (1993) Clin. Invest. 92, 2713–271847. Roussel, A., de Caro, J., Bezzine, S., Gastinel, L., de Caro, A., Carriere, F.,

Leydier, S., Verger, R., and Cambillau, C. (1998) Proteins 32, 523–53148. Wicker-Planquart, C., Canaan, S., Riviere, M., Dupuis, L., and Verger, R.

(1996) Protein Eng. 9, 1225–123249. Canaan, S., Dupuis, L., Riviere, M., Faessel, K., Romette, J. L., Verger, R., and

Wicker-Planquart, C. (1998) Protein Expression Purif. 14, 23–3050. SERC Daresbury Laboratory (1994) Collaborative Computing Project 4. A

Suite of Programs for Protein Crystallography, Acta. Crystallogr. Sec. D. 50,760

51. Otwinowski, Z. (1991) in (Evans, P. and Leslie, A., eds) pp. 80–85, Isomor-phous Replacement and Anomalous Scattering, SERC Daresbury Labora-tory, Warrington WA4, UK

52. Navaza, J. (1994) Acta Crystallogr. Sec. A 50, 157–16353. Kleywegt, G., and Jones, T. (1994) in From First Map to Final Model, pp.

59–66 EPSRC, Daresbury Warrington WA4, UK54. Roussel, A., and Cambillau, C. (1991) in Silicon Graphics Geometry Partners

Directory (Graphics, S., ed) pp. 86, Silicon Graphics, Mountain View, CA55. Brunger, A. T. (1992) X-PLOR, Version 3.1, Yale University Press, New Ha-

ven, CT56. Kraulis, P. (1991) J. Appl. Crystallogr. 24, 946–95057. Merrit, E. A., and Murphy, M. E. P. (1994) Acta Crystallogr. Sec. D 50,

869–87358. Thompson, J. D., Higgins, D. G., and Gibson, T. J. (1994) Nucleic Acids Res. 22,

4673–468059. Barton, G. J. (1993) Protein Eng. 6, 37–4060. Nicholls, A., Sharp, K. A., and Honig, B. (1991) Proteins 11, 281–28661. Laskowski, R. A., MacArthur, M. W., Moss, D. S., and Thorton, J. M. (1993)

J. Appl. Cryst. 26, 283–291

Crystal Structure of Recombinant Human Gastric Lipase17002

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from

Robert Verger and Christian CambillauAlain Roussel, Stéphane Canaan, Marie-Pierre Egloff, Mireille Rivière, Liliane Dupuis,

Two Lipolytic Enzymes of Medical InterestCrystal Structure of Human Gastric Lipase and Model of Lysosomal Acid Lipase,

doi: 10.1074/jbc.274.24.169951999, 274:16995-17002.J. Biol. Chem. 

  http://www.jbc.org/content/274/24/16995Access the most updated version of this article at

 Alerts:

  When a correction for this article is posted• 

When this article is cited• 

to choose from all of JBC's e-mail alertsClick here

  http://www.jbc.org/content/274/24/16995.full.html#ref-list-1

This article cites 54 references, 8 of which can be accessed free at

by guest on July 1, 2020http://w

ww

.jbc.org/D

ownloaded from