Colipase: structure and interaction with pancreatic lipase

Colipase is a small protein cofactor needed by pancreatic lipase for the efficient dietary lipid hydrolysis. It binds to the C-terminal, non-catalytic domain of lipase, thereby stabilising an active conformation and considerably increasing the overall hydrophobic binding site. Structural studies of the complex and of colipase alone have clearly revealed the functionality of its architecture. Interestingly, a structural analogy has recently been discovered between colipase and a domain in a developmental protein (Dickkopf), based on sequence analogy and homology modeling. Whether this structural analogy implies a common function (lipid interaction) remains to be clarified. Structural analogies have also been recognised between the pancreatic lipase C-terminal domain, the N-terminal domains of lipoxygenases and the C-terminal domain of alpha-toxin. These non-catalytic domains in the latter enzymes are important for interaction with membranes. It has not been established if these domains are also involved in eventual protein cofactor binding as is the case for pancreatic lipase.

Human pancreatic lipase: colipase dependence and interfacial binding of lid domain mutants

Five key amino acid residues from human pancreatic lipase (HPL) are mutated in some pancreatic lipase-related proteins 2 (PLRP2) that are not reactivated by colipase in the presence of bile salts. One of these residues (Y403) is involved in a direct interaction between the HPL C-terminal domain and colipase. The other four residues (R256, D257, Y267, and K268) are involved in the interactions stabilizing the open conformation of the lid domain, which also interacts with colipase. Here we produced and characterized three HPL mutants: HPL Y403N, an HPL four-site mutant (R256G, D257G, Y267F, and K268E), and an HPL five-site mutant (R256G, D257G, Y267F, K268E, and Y403N), in which the HPL amino acids were replaced by those present in human PLRP2. Colipase reactivated both the HPL Y403N mutant and HPL, and Y403 is therefore not essential for lipase-colipase interactions. Both the HPL four-site and five-site mutants showed low activity on trioctanoin, were inhibited by bile salts (sodium taurodeoxycholate, NaTDC) and were not reactivated by colipase. The interfacial binding of the HPL four-site mutant to a trioctanoin emulsion was suppressed in the presence of 4 mM NaTDC and was not restored by addition of colipase. Protein blotting/protein overlay immunoassay revealed that the HPL four-site mutant-colipase interactions are not abolished, and therefore, the absence of reactivation of the HPL four-site mutant is probably due to a lid domain conformation that prevents the interfacial binding of the lipase-colipase complex. The effects of colipase were also studied with HPL(-lid), an HPL mutant showing an 18-residue deletion within the lid domain, which therefore has only one colipase interaction site. HPL(-lid) showed a low activity on trioctanoin, was inhibited by bile salts, and recovered its lipase activity in the presence of colipase. Reactivation of HPL(-lid) by colipase was associated with a strong interfacial binding of the mutant to a trioctanoin emulsion. The lid domain is therefore not essential for either the interfacial binding of HPL or the lipase-colipase interactions.

Structural basis for the substrate selectivity of pancreatic lipases and some related proteins

The classical human pancreatic lipase (HPL), the guinea pig pancreatic lipase-related protein 2 (GPLRP2) and the phospholipase A1 from hornet venom (DolmI PLA1) illustrate three interesting steps in the molecular evolution of the pancreatic lipase gene family towards different substrate selectivities. Based on the known 3D structures of HPL and a GPLRP2 chimera, as well as the modeling of DolmI PLA1, we review here the structural features and the kinetic properties of these three enzymes for a better understanding of their structure-function relationships. HPL displays significant activity only on triglycerides, whereas GPLRP2 displays high phospholipase and galactolipase activities, together with a comparable lipase activity. GPLRP2 shows high structural homology with HPL with the exception of the lid domain which is made of five amino acid residues (mini-lid) instead of 23 in HPL. The lid domain deletion in GPLRP2 allows the free access to the active site and reduces the steric hindrance towards large substrates, such as galactolipids. The role of the lid domain in substrate selectivity has been investigated by site-directed mutagenesis and the substitution of HPL and GPLRP2 lid domains. The addition of a large-size lid domain in GPLRP2 increases the substrate selectivity for triglycerides by depressing the phospholipase activity. The phospholipase activity is, however, not induced in the case of the HPL mutant with GPLRP2 mini-lid. Therefore, the presence of a full-length lid domain is not the unique structural feature explaining the absence of phospholipase activity in HPL. The 3D structure of the GPLRP2 chimera and the model of DolmI PLA1 reveal a higher hydrophilic/lipophilic balance (HLB) of the surface loops (beta5 loop, beta9 loop, lid domain) surrounding the active site, as compared to the homologous loops in HPL. This observation provides a potential explanation for the ability of GPLRP2 and DolmI PLA1 to hydrolyze polar lipids, such as phospholipids. In conclusion, the beta5 loop, the beta9 loop, and the lid domain play an essential role in substrate selectivity towards triglycerides, phospholipids and galactolipids.

Pancreatic lipase-related protein 1 (PLRP1) is present in the pancreatic juice of several species

Pancreatic lipase-related protein 1 (PLRP1) was purified from human, canine, porcine and rat pancreatic juices. The four PLRP1s were identified using microsequencing methods after performing gel filtration on Ultrogel AcA-54 followed by chromatography on Heparin-Sepharose cation-exchanger. Polyclonal antibodies specific to human PLRP1 (HPLRP1) were raised in the rabbit using a synthetic decapeptide from HPLRP1. The results of Western blotting analysis showed that these antibodies recognized native HPLRP1 and recombinant HPLRP1 produced by insect cells, and cross-reacted only with rat PLRP1 (RPLRP1). No significant lipolytic activity was observed with native canine PLRP1 and recombinant HPLRP1 on various glycerides, phospholipid and vitamin esters, or on cholesterol esters. It was established for the first time that this protein is secreted in variable amounts by the adult exocrine pancreas of several species.

Reactivation of the totally inactive pancreatic lipase RP1 by structure-predicted point mutations

Both classical pancreatic lipase (DPL) and pancreatic lipase-related protein 1 (DPLRP1) have been found to be secreted by dog exocrine pancreas. These two proteins were purified to homogeneity from canine pancreatic juice and no significant catalytic activity was observed with dog PLRP1 on any of the substrates tested: di- and tri-glycerides, phospholipids, etc. DPLRP1 was crystallized and its structure solved by molecular replacement and refined at a resolution of 2.10 A. Its structure is similar to that of the classical PL structures in the absence of any inhibitors or micelles. The lid domain that controls the access to the active site was found to have a closed conformation. An amino-acid substitution (Ala 178 Val) in the DPLRP1 may result in a steric clash with one of the acyl chains observed in the structures of a C11 alkyl phosphonate inhibitor, a transition state analogue, bound to the classical PL. This substitution was suspected of being responsible for the absence of DPLRP1 activity. The presence of Val and Ala residues in positions 178 and 180, respectively, are characteristic of all the known PLRP1, whereas Ala and Pro residues are always present in the same positions in all the other members of the PL gene family. Introducing the double mutation Val 178 Ala and Ala 180 Pro into the human pancreatic RP1 (HPLRP1) gene yielded a well expressed and folded enzyme in insect cells. This enzyme is kinetically active on triglycerides. Our findings on DPLRP1 and HPLRP1 are therefore likely to apply to all the RP1 lipases.

Human pancreatic lipase: an exposed hydrophobic loop from the C-terminal domain may contribute to interfacial binding

Epitope mapping was performed using four anti-HPL monoclonal antibodies (mAb's 81-23, 146-40, 315-25, and 320-24) directed against human pancreatic lipase (HPL). Three HPL mutants produced in insect cells were tested for this purpose: (i) N-HPL, which consists of only the N-terminal domain of HPL, (ii) HPL(-lid), in which a short loop consisting of 5 amino acid residues replaces the full-length 23-residue lid domain present in HPL, and (iii) N-GPLRP2/C-HPL chimera, a chimeric mutant consisting of the N-terminal domain of the guinea pig pancreatic lipase related protein 2 (GPLRP2) fused to the C-terminal domain of HPL. The C-terminal domain of HPL (C-HPL) was prepared in a pure form after performing chymotryptic digestion of HPL. The mAb 146-40 recognizes HPL, HPL(-lid), and N-HPL but not GPLRP2, N-GPLRP2/C-HPL chimera, or the C-HPL. The antibody mAb 146-40 therefore specifically recognizes the N-terminal domain of HPL, and the epitope recognized does not include the amphiphilic lid. On the other hand, mAb's 81-23, 315-25, and 320-24 react specifically to the C-terminal domain of HPL, since they recognize HPL, HPL(-lid), the N-GPLRP2/C-HPL chimera, and the C-HPL but not N-HPL or GPLRP2. It was further established that these three mAb's recognize the same conformational epitope, the structure of which is stabilized by the N-terminal domain in the presence of SDS at concentrations greater than its critical micellar concentration. This conformational epitope was found to be located in the vicinity of Met 397 and Arg 414. These two residues delineate a highly exposed peptide stretch extending from the HPL C-terminal domain, which includes a hydrophobic surface loop (beta5'). Kinetic studies on the HPL/mAb's complexes showed that the lipase activity was much lower in these complexes than in HPL. The results of the present study suggest for the first time that the beta5' loop from the C-terminal domain may be involved in the interaction of HPL with a lipid/water interface.

An inactive pancreatic lipase-related protein is activated into a triglyceride-lipase by mutagenesis based on the 3-D structure

Both classical dog pancreatic lipase (DPL) and dog pancreatic lipase-related protein 1 (DPLRP1) have been found to be secreted by the exocrine pancreas. These two proteins were purified to homogeneity from canine pancreatic juice and no significant catalytic activity was observed with DPLRP1 on any of the substrates tested: di- and tri-glycerides; phospholipids (PC); etc. DPLRP1 was crystallized and its structure solved by molecular replacement and refined at a resolution of 2.10 A. Its structure is similar to that of the classical pancreatic lipase (PL) structures determined in the absence of any inhibitors or micelles. The lid domain that controls the access to the active site was found to have a closed conformation. An amino-acid substitution (Ala 178 Val) in the DPLRP1 was suspected of being responsible for the absence of enzymatic activity by inducing a steric clash with one of the acyl chain observed in the structures of chiral C11 alkyl phosphonate inhibitors, bound to the classical PL. The presence of Val and Ala residues in positions 178 and 180, respectively, are characteristic of the three known pancreatic lipase-related protein 1 (PLRP1), whereas Ala and Pro residues are always present at the same positions in all the other members of the PL gene family. Introducing the double mutation Val 178 Ala and Ala 180 Pro into the human pancreatic-related protein 1 (HPLRP1) gene yielded a well expressed and folded enzyme in insect cells. This enzyme is kinetically active on tributyrin (1800 U/mg) as well as trioctanoin (2250 U/mg) and its activity is low in the presence of taurodeoxycholate and stimulated in the presence of colipase. Our findings on DPLRP1 and HPLRP1 are therefore likely to apply to all the PLRP1 lipases.

In vivo and in vitro studies on the stereoselective hydrolysis of tri- and diglycerides by gastric and pancreatic lipases

The stereoselectivity of dog gastric and dog pancreatic lipases was investigated both in vitro, under simulated physiological conditions, and in vivo, during the digestion of a liquid test meal. In vitro it was observed that although both lipases had a stereopreference for the sn-3 position in triglycerides, it was about three times higher in the case of the gastric lipase. On the other hand, both lipases clearly showed a comparable enantioselectivity for the sn-1 position when a racemic diolein was used as the substrate. In the case of pancreatic lipase, the enantiomeric excess of 1,2-sn-diolein generated in vitro by the hydrolysis of triolein was found to decrease significantly, and even to be slightly reversed, at high rates of hydrolysis (above 50%) due to the further stereoselective hydrolysis of diglycerides into monoglycerides. This finding may explain the low enantiomeric excess of the diglycerides observed in vivo during the early phase of intraduodenal digestion when pancreatic lipase plays a predominant role and the rate of triolein hydrolysis is already high. On the other hand, a large enantiomeric excess of 1,2-sn-diolein generated from triolein was always the fingerprint of the gastric lipase in vitro even at high hydrolysis rates. This fingerprinting of gastric lipase was observed during both the intragastric phase and the late intestinal phase of lipolysis. This feature was therefore taken as an index to determine the respective roles of gastric and pancreatic lipases during in vivo lipolysis. To the best of our knowledge, this is the first time that stereoselectivity has been used as a tool to discriminate between the activities of two enzymes hydrolyzing the same substrate in vivo.

Pancreatic lipase structure-function relationships by domain exchange

We designed chimeric mutants by exchanging the lid domains of the classical human pancreatic lipase (HPL) and the guinea pig pancreatic lipase related protein 2 (GPLRP2). This latter enzyme possesses naturally a large deletion within the lid domain and is not activated by lipid/water interfaces. Furthermore, GPLRP2 exhibits phospholipase A1 and lipase activities in the same order of magnitude, whereas HPL has no significant phospholipase activity and displays a clear interfacial activation. An HPL mutant [HPL(-lid)] with GPLRP2 mini-lid domain does not display interfacial activation. Its specific activity toward triglycerides is, however, dramatically reduced. A GPLRP2 mutant [GPLRP2(+lid)] with HPL full-length lid domain is not interfacially activated, and its lid domain probably exists under a permanent open conformation. Therefore, the phenomenon of interfacial activation in HPL is not only due to the presence of a full-length lid domain but also to other structural elements which probably allow the existence of stabilized closed and open conformations of the lid. GPLRP2(+lid) phospholipase activity is significantly reduced as compared to GPLRP2, whereas its lipase activity remains at the same level. Therefore, the lid domain plays a major role in substrate selectivity and can be considered as part of the active site. However, the presence of a full-length lid domain is not sufficient to explain the absence of phospholipase activity in HPL since HPL(-lid) does not display any phospholipase activity. We also produced a chimeric GPLRP2 mutant in which the C-terminal domain was substituted by the HPL C-terminal domain. The colipase effects, i.e., anchoring and stabilization of the lipase at the interface, are clearly observed with the chimera, whereas GPLRP2 is insensitive to colipase. The kinetic characterization of this chimera reveals for the first time that the interfacial stability of pancreatic lipases depends on the structure of the C-terminal domain.

A pancreatic lipase with a phospholipase A1 activity: crystal structure of a chimeric pancreatic lipase-related protein 2 from guinea pig

BACKGROUND

The guinea pig pancreatic lipase-related protein 2 (GPLRP2) differs from classical pancreatic lipases in that it displays both lipase and phospholipase A1 activities; classical pancreatic lipases have no phospholipase activity. The sequence of GPLRP2 is 63 % identical to that of human pancreatic lipase (HPL), but the so-called lid domain, is much reduced in GPLRP2. A phospholipase A1 from hornet venom (Dolml PLA1) is very similar to HPL and GPLRP2 but is devoid of lipase activity; Dolml PLA1 also contains a reduced lid domain and lacks a region termed the beta9 loop, which is located in the vicinity of the HPL and GPLRP2 active sites. The structure determination of a chimera of GPLRP2 and HPL and domain building of Dolml PLA1 were undertaken to gain a better understanding of the structural parameters responsible for the differences in lipase versus phospholipase activity among these structurally related enzymes.

RESULTS

The crystal structure of a chimeric mutant of GPLRP2, consisting of the catalytic domain of GPLRP2 and the C-terminal domain of HPL, has been solved and refined to 2.1 A resolution. This enzyme belongs to the alpha/beta hydrolase fold family and shows high structural homology with classical pancreatic lipases. The active site is closely related to those of serine esterases, except for an unusual geometry of the catalytic triad. Due to the reduced size of the lid domain, the catalytic serine is fully accessible to solvent. Part of the beta9 loop, which stabilizes the lid domain in the closed conformation of the classical HPL, is totally exposed to the solvent and is not visible in the electron-density map.

CONCLUSIONS

The structures of the related enzymes, GPLRP2 and HPL and the model of Dolml PLA1, provide insights into the role played by the loops located above the active site in controlling substrate selectivity towards triglycerides or phospholipids. In GPLRP2, the lid domain is reduced in size compared to HPL, and hydrophilic residues are exposed to solvent. GPLRP2 is thus able to accommodate the polar head of phospholipids. The beta9 loop is still present in GPLRP2, making it possible for this enzyme to still accommodate triglycerides. In Dolml PLA1, the beta9 loop is absent, and this enzyme is unable to process triglycerides retaining only the phospholipase A1 activity.

Lipid binding and activating properties of porcine pancreatic colipase split at the Ile79-Thr80 bond

Porcine colipase, the protein cofactor of pancreatic lipase, was isolated from pancreas freshly collected on animals and from a side fraction from the production of insulin (Novo Nordisk A/S). Samples of purified colipase were analyzed for homogeneity by polyacrylamide gel electrophoresis, reverse-phase high-performance liquid chromatography (RPLC), quantitative N-terminal sequence determination and mass spectrometry. The activating properties of colipase preparations were assayed against tributyrin, triolein or the commercial Intralipid emulsion, in presence of bile salt. Two fractions of colipase with the same specific activity were purified from fresh pancreas. The major fraction (85%) contained one single protein corresponding to fragment 1-93 of the 95-residue form of colipase (procolipase) previously characterized in porcine pancreatic juice. The other fraction (15%) corresponded to fragment 1-91 of procolipase. Also, two fractions of colipase were purified from the side fraction supplied by Novo. These fractions consisted of the 95-residue proform of colipase and of fragment 1-93, respectively, both specifically cleaved at the Ile79-Thr80 peptide bond with partial removal of isoleucine at position 79 and serine at position 78. Procolipase split at the 79-80 bond retained full activity on tributyrin and triolein and on the Intralipid emulsion but the kinetics of hydrolysis of triacylglycerol substrates showed much longer lag periods than those observed with native procolipase. Also, all forms of procolipase split at the 79-80 bond showed one peak in RPLC but their retention time was markedly decreased as compared to that of native procolipase which indicated a weaker hydrophobic binding capacity. The value of the retention time was of the same order of magnitude as that of inactive reduced procolipase. Treatment of native procolipase by pancreatic endopeptidases showed that elastase is likely responsible for specific cleavage at the 79-80 bond of procolipase purified from the Novo extract. Limited proteolysis by trypsin of the proforms of colipase split at the 79-80 bond reduced the lag period. Results presented in this communication provide the first direct evidence showing that the finger-shaped peptide segment between half-cystine residues at positions 69 and 87 is involved in colipase-lipid interaction as previously hypothesized from the three-dimensional structure of the protein.

Cloning and expression in insect cells of two pancreatic lipases and a procolipase from Myocastor coypus

The physiological role of pancreatic lipases has traditionally been assigned solely to triacylglyceride metabolism, while the digestion of phospholipids requires the presence of the pancreatic phospholipase A2, a 14-kDa enzyme unrelated to pancreatic lipases. However, in the guinea pig, it was observed that the pancreatic phospholipase A2 was absent and that a guinea pig pancreatic-lipase-related protein 2 (GPL-RP2) was responsible for phospholipase activity, in contrast to the situation observed in other mammalian species. As the guinea pig is a member of the hystricomorph rodents, it was of interest to investigate if other species within this evolutionary suborder display similar characteristics. The coypu (Myocastor coypus) also a member of the hystricomorph rodents, was chosen for further investigations. The cDNAs encoding two pancreatic lipases and a procolipase from the coypu were cloned, expressed and characterized. One lipase, CoPL-RP2, was identified as belonging to the RP2 subfamily, while the second, CoPL, was found to belong to the classical pancreatic lipase subfamily. Enzymic characterization and sequence data suggest a role for coypu colipase as a specific cofactor for CoPL, while this coypu colipase cannot be an important cofactor for CoPL-RP2 in vivo. Also, the new lipase cDNA sequences were used in a phylogentic analysis to reinvestigate the taxonomical position of the hystricomorph rodents (e.g. coypu and guinea pig) with respect to the myomorph rodents (e.g. rat and mouse).

Lipase structures at the interface between chemistry and biochemistry

In this chapter we review recent molecular knowledge on two structurally related mammalian triglyceride lipases which have evolved from a common ancestral gene. The common property of the lipase family members is that they interact with non-polar substances. Pancreatic lipase hydrolyzes triglycerides in the small intestine in the presence of many dietary components, other digestive enzymes and high concentrations of detergents (bile salts). Lipoprotein lipase acts at the vascular side of the blood vessels where it hydrolyses triglycerides and some phospholipids of the circulating plasma lipoproteins. A third member of the gene family, hepatic lipase, is found in the liver of mammals. Also, this lipase is involved in lipoprotein metabolism. The three lipases are distantly related to some non-catalytic yolk proteins from Drosophila (Persson et al., 1989; Kirchgessner et al., 1989; Hide et al., 1992) and to a phospholipase A1 from hornet venom (Soldatova et al., 1993).

Structure-function relationships in naturally occurring mutants of pancreatic lipase

From primary structure comparison, the pancreatic lipase family is now divided into three subgroups: classical pancreatic lipases, pancreatic lipase-related proteins 1 (RPI) and pancreatic lipase-related proteins 2 (RP2). Among the RP2 subfamily, the guinea-pig and coypu enzymes share kinetic properties which differ from those of classical pancreatic lipases. Both enzymes display a high phospholipase activity and are not interfacially activated using a short chain triglyceride as substrate. Their activity towards insoluble triglycerides is inhibited by micellar concentrations of bile salts and is not restored by addition of colipase. These atypical kinetic properties are discussed in the light of amino acid sequence comparison between RP2 and classical pancreatic lipases, based on the closed and open conformations of the 3-D structure of human pancreatic lipase.

Evidence for a pancreatic lipase subfamily with new kinetic properties

Several new members of the pancreatic lipase family have been reported recently, and amino acid sequence comparison reveals that this family can now be divided into three subgroups: (1) "classical" pancreatic lipases, (2) related proteins 1 (RP1), and (3) related proteins 2 (RP2) (Giller, T., et al. (1992) J. Biol. Chem. 267(23), 16509-16516). Whereas "classical" pancreatic lipases are well characterized with respect to kinetic properties, i.e., interfacial activation and dependence on colipase in the presence of bile salts, the two latter subfamilies have been poorly investigated so far. The kinetic behavior of a lipase from guinea pig pancreas differs, however, from that of "classical" lipases (Hjorth, A., et al. (1993) Biochemistry 32, 4702-4707). This enzyme is highly homologous to RP2 lipases with the exception of a deletion in the so-called lid domain that regulates access to the active center of pancreatic lipases. We have now characterized a novel lipase from coypu (Myocastor coypus) pancreas. This enzyme, also belonging to the RP2 subfamily, possesses a full-length lid domain, but its kinetic properties are very similar to those of the guinea pig enzyme: (1) a high phospholipase activity, (2) the absence of interfacial activation, and (3) the absence of a colipase effect at high bile salt concentrations. Since both guinea pig and coypu pancreas produce a classical pancreatic lipase and no measurable phospholipase A2 activity, it is suggested that RP2 enzymes act as real phospholipases under physiological conditions. In fact, all RP2 lipases from other species might share phospholipase activity and fulfill new biological functions.

Cloning of the classical guinea pig pancreatic lipase and comparison with the lipase related protein 2

Starting from total pancreatic mRNAs, the classical guinea pig pancreatic lipase was cloned using rapid amplification of 3' and 5' cDNA ends. Internal oligonucleotide primers were designed from a partial cDNA clone including the region coding for the lid domain. Using this strategy, we did not amplify the cDNA corresponding to the pancreatic lipase related protein 2 in which the lid domain is deleted. Amino acid sequences of the classical guinea pig pancreatic lipase and the related protein 2 were compared based on the primary and tertiary structures of the classical human pancreatic lipase. Their distinct physiological roles are discussed in the light of functional amino acid differences.

One-step purification and characterization of human pancreatic lipase expressed in insect cells

A cDNA clone encoding the sequence of human pancreatic lipase (HPL) was subcloned into the baculovirus transfer vector pVL1392 and used in co-transfection of Spodoptera frugiperda (Sf9) insect cells with wild-type Autographa californica nuclear polyhedrosis virus (AcNPV) DNA. A single recombinant protein (50 kDa) secreted by Sf9 cells was detectable in the culture medium 24 h post-infection using both anti-HPL polyclonal antibodies and potentiometric measurements of lipolytic activity. The expression level reached 40 mg/l of enzyme at 6 days. A single cation-exchange chromatography was sufficient to obtain a highly pure recombinant HPL as demonstrated by N-terminal sequencing, amino acid composition and carbohydrate analysis, as well as by mass spectrometry. These analyses revealed the production of mature protein with the correct processing of signal peptide and an homogenous glycosylation pattern. The kinetic properties of recombinant and native HPL were compared. Both enzymes showed similar profiles of interfacial activation, inhibition by bile salts and re-activation by colipase.

A structural domain (the lid) found in pancreatic lipases is absent in the guinea pig (phospho)lipase

Typically pancreatic lipases are characterized by the following properties: (1) they are activated by lipid/water interfaces (interfacial activation), (2) they are inhibited by bile salts but reactivated by colipase (a small activator protein), and (3) they do not hydrolyze significantly phospholipids. A cDNA clone encoding a guinea pig pancreatic (phospho)lipase (GPL) has been sequenced and expressed. The enzyme (recombinant as well as native) differs from other pancreatic lipases in that (1) it is not interfacially activated, (2) its activity is unaffected by the presence of bile salts and/or colipase using tributyrin as substrate, and (3) it exhibits equally phospholipase A1 and lipase activities. The amino acid sequence of GPL is highly homologous to that of other known pancreatic lipases, with the exception of a deletion in the so-called lid domain that regulates access to the active centers of other lipases. We propose that this deletion is directly responsible for the anomalous behavior of this enzyme. Thus GPL challenges the classical distinction between lipases, esterases, and phospholipases.

Gastric and pancreatic lipase levels during a test meal in dogs

The levels of gastric and pancreatic lipases in the duodenum and the jejunum were measured during the digestion of a test meal in dogs. Using both a specific enzymatic titration and an enzyme-linked immunosorbent assay, it is shown for the first time that gastric lipase remains active in the duodenal and jejunal contents. An experimental device was set up for measuring the secretions and the intestinal flows of lipases during the digestion of a liquid test meal. In a dog equipped with gastric and duodenal cannulae, the secretion of gastric lipase was stimulated by food ingestion, reaching 3.0 +/- 0.3 mg/h (three times the basal secretion rate) during the 1st h of digestion. The total secretory outputs of gastric and pancreatic lipases recorded over a 3-h period of digestion were 7.2 +/- 1.2 mg and 18.7 +/- 1.2 mg, respectively.

Isoform purification of gastric lipases. Towards crystallization

Several isoforms of rabbit and human gastric lipases have been purified. These isoforms have the same apparent molecular weight (Mr approximately 50,000), but very different isoelectric points. Some of these isoforms were purified: pI 7.2 and 6.5 in the case of rabbit gastric lipase; and pI 7.4 and 7.2 in that of human gastric lipase. All the purified isoforms were found to have the same specific lipase activity (around 1200 units per mg of protein, measured on tributyrin as substrate). The isoforms of dog gastric lipase are more closely related, and could not be separated. Partial enzymatic deglycosylation of human gastric lipase reduced the apparent molecular weight from Mr approximately 50,000 to Mr approximately 43,000 and induced a change in the isoelectrofocusing pattern and the emergence of a new isoform (pI 7.3). It is concluded that the charge heterogeneity of gastric lipases is at least partly due to the glycan moiety of the molecule, which amounts to approximately 14% of the total molecular weight. Several crystallization trials on purified native preparations of rabbit and human gastric lipases were unsuccessful, whereas crystals were obtained from native dog gastric lipase and all the purified isoforms of rabbit and human gastric lipases, some of which were crystallographically characterized.

Dog gastric lipase: stimulation of its secretion in vivo and cytolocalization in mucous pit cells

Dog gastric lipase (DGL) secretion is stimulated in vivo by urecholine, pentagastrin, histamine, 16,16-dimethyl prostaglandin E2, and secretin. Under fasting conditions, DGL is irreversibly inactivated by gastric acid below pH 1.5; consequently, DGL output can be underestimated. This problem has been resolved by buffering the acid or by using an antisecretory drug such as omeprazole during stimulation. There is a clear parallelism between the secretion of DGL and of gastric mucus. This observation led to the present investigation of the cellular localization of DGL using immunofluorescence techniques. Results showed that DGL is cytolocalized in mucous pit cells of gastric glands. Pepsinogen is found in chief cells. To the authors' knowledge, this is the first description of an enzyme (gastric lipase) secreted by mucous-type gastric cells. In contrast to other species, gastric lipase of the dog is located in cardiac, fundic, and antral mucosae.