Search

PDF file - 60K, Lack of competition between bsFab and human IgG for FcγRIIIa binding.

PDF file - 70K

PDF file - 120K

Antibody-dependent cell-mediated cytotoxicity, one of the most prominent modes of action of antitumor antibodies, suffers from important limitations due to the need for optimal interactions with Fcγ receptors. In this work, we report the design of a new bispecific antibody format, compact and linker-free, based on the use of llama single-domain antibodies that are capable of circumventing most of these limitations. This bispecific antibody format was created by fusing single-domain antibodies directed against the carcinoembryonic antigen and the activating FcγRIIIa receptor to human Cκ and CH1 immunoglobulin G1 domains, acting as a natural dimerization motif. In vitro and in vivo characterization of these Fab-like bispecific molecules revealed favorable features for further development as a therapeutic molecule. They are easy to produce in Escherichia coli, very stable, and elicit potent lysis of tumor cells by human natural killer cells at picomolar concentrations. Unlike conventional antibodies, they do not engage inhibitory FcγRIIb receptor, do not compete with serum immunoglobulins G for receptor binding, and their cytotoxic activity is independent of Fc glycosylation and FcγRIIIa polymorphism. As opposed to anti-CD3 bispecific antitumor antibodies, they do not engage regulatory T cells as these latter cells do not express FcγRIII. Studies in nonobese diabetic/severe combined immunodeficient gamma mice xenografted with carcinoembryonic antigen–positive tumor cells showed that Fab-like bispecific molecules in the presence of human peripheral blood mononuclear cells significantly slow down tumor growth. This new compact, linker-free bispecific antibody format offers a promising approach for optimizing antibody-based therapies. Mol Cancer Ther; 12(8); 1481–91. ©2013 AACR.

International audience; The natural killer group 2 member D (NKG2D) receptor is a C-type lectin-like activating receptor mainly expressed by cytotoxic immune cells including NK, CD8 + T, γδ T and NKT cells and in some pathological conditions by a subset of CD4 + T cells. It binds a variety of ligands (NKG2DL) whose expressions is finely regulated by stress-related conditions. The NKG2DL/NKG2D axis plays a central and complex role in the regulation of immune responses against diverse cellular threats such as oncogene-mediated transformations or infections. We generated a panel of seven highly specific anti-human NKG2D single-domain antibodies targeting various epitopes. These single-domain antibodies were integrated into bivalent and bispecific antibodies using a versatile plug-and-play Fab-like format. Depending on the context, these Fab-like antibodies exhibited activating or inhibitory effects on the immune response mediated by the NKG2DL/NKG2D axis. In solution, the bivalent anti-NKG2D antibodies that compete with NKG2DL potently blocked the activation of NK cells seeded on immobilized MICA, thus constituting antagonizing candidates. Bispecific anti-NKG2DxHER2 antibodies that concomitantly engage HER2 on tumor cells and NKG2D on NK cells elicited cytotoxicity of unstimulated NK in a tumor-specific manner, regardless of their apparent affinities and epitopes. Importantly, the bispecific antibodies that do not compete with ligands binding retained their full cytotoxic activity in the presence of ligands, a valuable property to circumvent immunosuppressive effects induced by soluble ligands in the microenvironment.

International audience; Antibodies are key tools in biomedical research and medicine. Their binding properties are classically measured in solution and characterized by an affinity. However, in physiological conditions, antibodies can bridge an immune effector cell and an antigen-presenting cell, implying that mechanical forces may apply to the bonds. For example, in antibody-dependent cell cytotoxicity—a major mode of action of therapeutic monoclonal antibodies—the Fab domains bind the antigens on the target cell, whereas the Fc domain binds to the activating receptor CD16 (also known as FcgRIII) of an immune effector cell, in a quasi-bidimensional environment (2D). Therefore, there is a strong need to investigate antigen/antibody binding under force (2D) to better understand and predict antibody activity in vivo. We used two anti-CD16 nanobodies targeting two different epitopes and laminar flow chamber assay to measure the association and dissociation of single bonds formed between microsphere-bound CD16 antigens and surface-bound anti-CD16 nanobodies (or single-domain antibodies), simulating 2D encounters. The two nanobodies exhibit similar 2D association kinetics, characterized by a strong dependence on the molecular encounter duration. However, their 2D dissociation kinetics strongly differ as a function of applied force: one exhibits a slip bond behavior in which off rate increases with force, and the other exhibits a catch-bond behavior in which off rate decreases with force. This is the first time, to our knowledge, that catch-bond behavior was reported for antigen-antibody bond. Quantification of natural killer cells spreading on surfaces coated with the nanobodies provides a comparison between 2D and three-dimensional adhesion in a cellular context, supporting the hypothesis of natural killer cell mechanosensitivity. Our results may also have strong implications for the design of efficient bispecific antibodies for therapeutic applications.

International audience; Mesothelin is a cell-surface glycoprotein restricted to mesothelial cells overexpressed in several types of cancer, including triple-negative breast cancer not responding to trastuzumab or hormone-based therapies. Mesothelin-targeting therapies are currently being developed. However, the identification of patients potentially eligible for such a therapeutic strategy remains challenging. The objective of this study was to perform the radiolabeling and preclinical evaluation of 99mTc-A1 and 99mTc-C6, two antimesothelin single-domain antibody (sdAb)-derived imaging agents. Methods: A1 and C6 were radiolabeled with 99mTc and evaluated in vitro on recombinant protein and cells, as well as in vivo in xenograft mouse models of the triple-negative breast cancer cell lines HCC70 (mesothelin-positive) and MDA-MB-231 (mesothelin-negative). Results: Both 99mTc-A1 and 99mTc-C6 bound mesothelin with high affinity in vitro, with 99mTc-A1 affinity being 2.4-fold higher than that of 99mTc-C6 (dissociation constant, 43.9 ± 4.0 vs. 107 ± 16 nM, P < 0.05). 99mTc-A1 and 99mTc-C6 remained stable in vivo in murine blood (>80% at 2 h) and ex vivo in human blood (>90% at 6 h). In vivo 99mTc-A1 uptake (percentage injected dose) in HCC70 tumors was 5-fold higher than in MDA-MB-231 tumors and 1.5-fold higher than that of 99mTc-C6 (2.34% ± 0.36% vs. 0.48% ± 0.18% and 1.56% ± 0.43%, respectively, P < 0.01) and resulted in elevated tumor-to-background ratios. In vivo competition experiments demonstrated the specificity of 99mTc-A1 uptake in HCC70 tumors. Conclusion: Mesothelin-positive tumors were successfully identified by SPECT using 99mTc-A1 and 99mTc-C6. Considering its superior characteristics, 99mTc-A1 was selected as the most suitable tool for further clinical translation.

Phage display is a well-established procedure to isolate binders against a wide variety of antigens that can be performed on purified antigens, but also on intact cells. As selection steps are performed in vitro, it is possible to focus the outcome of the selection on relevant epitopes by performing some additional steps, such as depletion or competitive elutions. However in practice, the efficiency of these steps is often limited and can lead to inconsistent results. We have designed a new selection method named masked selection, based on the blockade of unwanted epitopes to favor the targeting of relevant ones. We demonstrate the efficiency and flexibility of this method by selecting single-domain antibodies against a specific portion of a fusion protein, by selecting binders against several members of the seven transmembrane receptor family using transfected HEK cells, or by selecting binders against unknown breast cancer markers not expressed on normal samples. The relevance of this approach for antibody-based therapies was further validated by the identification of four of these markers, Epithelial cell adhesion molecule, Transferrin receptor 1, Metastasis cell adhesion molecule, and Sushi containing domain 2, using immunoprecipitation and mass spectrometry. This new phage display strategy can be applied to any type of antibody fragments or alternative scaffolds, and is especially suited for the rapid discovery and identification of cell surface markers.

9 pags, 4 figs, 3 tabs, Cystic fibrosis is associated with pancreatic insufficiency and acidic intraluminal conditions that limit the action of pancreatic enzyme replacement therapy, especially that of lipase. Directed evolution combined with rational design was used in the aim of improving the performances of the human pancreatic lipase at acidic pH. We set up a method for screening thousands of lipase variants for activity at low pH. A single round of random mutagenesis yielded one lipase variant with an activity at acidic pH enhanced by ∼50 on medium- and long-chain triglycerides. Sequence analysis revealed two substitutions (E179G/N406S) located in specific regions, the hydrophobic groove accommodating the sn-1 chain of the triglyceride (E179G) and the surface loop that is likely to mediate lipase/colipase interaction in the presence of lipids (N406S). Interestingly, these two substitutions shifted the chain-length specificity of lipase toward medium- and long-chain triglycerides. Combination of those two mutations with a promising one at the entrance of the catalytic cavity (K80E) negatively affected the lipase activity at neutral pH but not that at acidic pH. Our results provide a basis for the design of improved lipase at acidic pH and identify for the first time key residues associated with chain-length specificity. © The Author 2010., This work was supported by the French Cystic Fibrosis Foundation and the French Research Medical Foundation (D.Y.C).

Colipase is a key element in the lipase-catalyzed hydrolysis of dietary lipids. Although devoid of enzymatic activity, colipase promotes the pancreatic lipase activity in physiological intestinal conditions by anchoring the enzyme at the surface of lipid droplets. Analysis of structures of NMR colipase models and simulations of their interactions with various lipid aggregates, lipid droplet, and bile salt micelle, were carried out to determine and to map the lipid binding sites on colipase. We show that the micelle and the oil droplet bind to the same side of colipase 3D structure, mainly the hydrophobic fingers. Moreover, it appears that, although colipase has a single direction of interaction with a lipid interface, it does not bind in a specific way but rather oscillates between different positions. Indeed, different NMR models of colipase insert different fragments of sequence in the interface, either simultaneously or independently. This supports the idea that colipase finger plasticity may be crucial to adapt the lipase activity to different lipid aggregates. Proteins 2008. © 2008 Wiley-Liss, Inc.

Resume L’obesite, en augmentation croissante dans les pays industrialises comme dans les pays emergents, est devenue un probleme de sante publique majeur. Si l’on considere que les lipides alimentaires du fait de leur valeur energetique elevee jouent un role important dans le developpement de cette pathologie, reduire la digestion des graisses en inhibant la lipase pancreatique est maintenant considere comme une nouvelle approche dans le traitement de l’obesite. Le domaine C-terminal isole de la lipase pancreatique en agissant comme un leurre proteique vis-a-vis de la colipase offre une nouvelle facon d’inhiber de facon tres specifique la lipolyse intestinale. In vivo l’administration de ce domaine a des rats soumis a un regime riche en lipides entraine une diminution significative de la prise de poids associee a une augmentation du taux d’excretion des triglycerides dans les feces et une amelioration du statut lipidique du foie. Le domaine C-terminal de la lipase peut etre envisage comme un agent therapeutique potentiel dans le cadre du traitement de l’obesite.

Previous in vitro experiments, as well as acute assays in rat showed that the C-terminal domain (CT-domain) of porcine pancreatic lipase behaves as a potent specific noncovalent inhibitor of pancreatic lipase. Nevertheless, the potential use of the CT-domain as a therapeutic tool against obesity in humans requires further investigation and would be best achieved using the human CT-domain. In the present study, we investigated the inhibitory effects of the recombinant human CT-domain, in vivo, upon chronic administration to rats fed a high-fat diet.The long-term in vivo study requiring relatively high amounts of the human CT-domain, the domain was overexpressed in Escherichia coli as inclusion bodies and an efficient refolding protocol was designed. The inhibitory effect of the recombinant human CT-domain on the activity of pancreatic lipase from different species was first investigated in vitro. Then chronic assays were performed for 4 weeks in rats fed a high-fat diet with or without a daily dose of 1.2 mg of CT-domain per kilogram rat. The time course of food intake, body weight, plasma parameters, liver lipids, faecal output of fat and total cholesterol were measured.A high yield of correctly folded recombinant human CT-domain was obtained using our refolding process, as evidenced by the capability of the recombinant domain to inhibit human horse and porcine pancreatic lipases in vitro. The recombinant human CT-domain had no influence on the food intake, but significantly reduced the body weight gain. As compared to control rats, higher amounts of total fat (mainly triglycerides and monoglycerides) and total cholesterol were found in the faeces of the rats treated with the CT-domain. Finally, a decrease in liver triglycerides and nonesterified cholesterol was observed while no significant effect could be detected on the plasma parameters.These results demonstrated that the CT-domain efficiently reduces in vivo, lipolysis and subsequently body weight gain in rat fed a high-fat diet. The CT-domain could, therefore, be effective in preventing obesity.

International audience; Mesothelin, a cancer biomarker overexpressed in tumors of epithelial origin, is a target for nanotechnology-based diagnostic, therapeutic, and prognostic applications. The currently available anti-mesothelin antibodies present limitations, including low penetration due to large size and/or lack of in vivo stability. Single domain antibodies (sdAbs) or nanobodies (Nbs) provide powerful solutions to these specific problems. We generated a phage-display library of Nbs that were amplified from B cells of a llama that was immunized with human recombinant mesothelin. Two nanobodies (Nb A1 and Nb C6) were selected on the basis of affinity (K(D) = 15 and 30 nM, respectively). Nb A1 was further modified by adding either a cysteine to permit maleimide-based bioconjugations or a sequence for the site-specific metabolic addition of a biotin in vivo. Both systems of conjugation (thiol-maleimide and streptavidin/biotin) were used to characterize and validate Nb A1 and to functionalize nanoparticles. We showed that anti-mesothelin Nb A1 could detect native and denatured mesothelin in various diagnostic applications, including flow cytometry, western blotting, immunofluorescence, and optical imaging. In conclusion, anti-mesothelin Nbs are novel, cost-effective, small, and single domain reagents with high affinity and specificity for the tumor-associated antigen mesothelin, which can be simply bioengineered for attachment to nanoparticles or modified surfaces using multiple bioconjugation strategies. These anti-mesothelin Nbs can be useful in both conventional and nanotechnology-based diagnostic, therapeutic and prognostic biomedical applications.

Although structurally similar to pancreatic lipase (PL), the key enzyme of intestinal fat digestion, pancreatic lipase-related protein type 2 (PLRP2) differs from PL in certain functional properties. Notably, PLRP2 has a broader substrate specificity than PL, and unlike that of PL, its activity is not restored by colipase in the presence of bile salts. In the studies presented here, the activation mechanism of horse PLRP2 was studied through active site-directed inhibition experiments, and the results demonstrate fundamental differences with that of PL. The opening of the horse PLRP2 flap occurs as soon as bile salt monomers are present, is accelerated in the presence of micelles, and does not require the presence of colipase. Moreover, in contrast to PL, horse PLRP2 is able to directly interact with a bile salt micelle to form an active binary complex, without the micelle being presented by colipase, as evidenced by molecular sieving experiments. These findings, together with the sensitivity of the horse PLRP2 flap to partial proteolysis, are indicative of a higher flexibility of the flap of horse PLRP2 relative to PL. From these results, it can be concluded that PLRP2 can adopt an active conformation in the intestine, which could be important for the further understanding of the physiological role of PLRP2. Finally, this work emphasizes the essential role of colipase in lipase catalysis at the lipid-water interface in the presence of bile.

International audience; Trastuzumab is established as treatment of HER2high metastatic breast cancers but many limitations impair its efficacy. Here, we report the design of a Fab-like bispecific antibody (HER2bsFab) that displays a moderate affinity for HER2 and a unique, specific and high affinity for FcγRIII. In vitro characterization showed that ADCC was the major mechanism of action of HER2bsFab as no significant HER2-driven effect was observed. HER2bsFab mediated ADCC at picomolar concentration against HER2high, HER2low as well as trastuzumab-refractive cell lines. In vivo HER2bsFab potently inhibited HER2high tumor growth by recruitment of mouse FcγRIII and IV-positive resident effector cells and more importantly, exhibited a net superiority over trastuzumab at inhibiting HER2low tumor growth. Moreover, FcγRIIIA-engagement by HER2bsFab was independent of V/F158 polymorphism and induced a stronger NK cells activation in response to target cell recognition. Thus, taking advantage of its epitope specificity and affinity for HER2 and FcγRIIIA, HER2bsFab exhibits potent anti-tumor activity against HER2low tumors while evading most of trastuzumab Fc-linked limitations thereby potentially enlarging the number of patients eligible for breast cancer immunotherapy.

In vertebrates, dietary fat digestion mainly results from the combined effect of pancreatic lipase, colipase, and bile. It has been proposed that in vivo lipase adsorption on oil-water emulsion is mediated by a preformed lipase-colipase-mixed micelle complex. The main lipase-colipase binding site is located on the C-terminal domain of the enzyme. We report here that in vitro the isolated C-terminal domain behaves as a potent noncovalent inhibitor of lipase and that the inhibitory effect is triggered by the presence of micelles. Lipase inhibition results from the formation of a nonproductive C-terminal domain-colipase-micelle ternary complex, which competes for colipase with the active lipase-colipase-micelle ternary complex, thus diverting colipase from its lipase-anchoring function. The formation of such a complex has been evidenced by molecular sieving experiments. This nonproductive complex lowers the amount of active lipase thus reducing lipolysis. Preliminary experiments performed in rats show that the C-terminal domain also behaves as an inhibitor in vivo and thus could be considered a potential new tool for specifically reducing intestinal lipolysis.

Colipase is a small protein (10 kDa), which acts as a protein cofactor for the pancreatic lipase. Various models of the activated ternary complex (lipase–colipase–bile salt micelles) have been proposed using detergent micelles, but no structural information has been established with bile salt micelles. We have investigated the organization of sodium taurodeoxycholate (NaTDC) micelles and their interactions with pig and horse colipases by homonuclear nuclear magnetic resonance (NMR) spectroscopy. The NMR data supply evidence that the folding of horse colipase is similar to that already described for pig colipase. Intermolecular nuclear Overhauser effects have shown that two conserved aromatic residues interact with NaTDC micelles.

The catalytic activity of most lipases depends on the aggregation state of their substrates. It is supposed that lipase activation requires the unmasking and structuring of the enzyme's active site through conformational changes involving the presence of oil-in-water droplets. This phenomenon has been called interfacial activation. Here, we report the crystal structure of the pancreatic activated lipase/colipase/micelle complex as determined using the D2O/H2O contrast variation low resolution neutron diffraction method. We find that a disk-shaped micelle interacts extensively with the concave face of colipase (CL) and the distal tip of the C-terminal domain of lipase away from the active site of the enzyme. Such interaction appears to help stabilizing the lipase–CL interaction. Consequently, we conclude that lipase activation is not interfacial but occurs in the aqueous phase and it is mediated by CL and a micelle.

Besides the active pancreatic lipase (PL) which plays a major role in dietary fat digestion, the presence of a pancreatic lipase related protein 1 (PLRP1) displaying a very low lipolytic activity has been reported in vertebrates. It has been suggested that the reduced lipolytic activity of PLRP1 results from specific features of the N-terminal domain of the protein. Therefore, based on sequence comparison between PL and PLRP1 and modelling experiments, several residues located in the vicinity of the active site pocket of both enzymes have been mutated. In this paper, we report that, as regards to PL, two substitutions in positions 179 and 181 in PLRP1 account for the very low lipolytic activity of the protein. Indeed, substituting these residues (V179 and A181) in PLRP1 for those found in PL (A179 and P181), restores a significant lipolytic activity for PLRP1.

Antibody-dependent cell-mediated cytotoxicity, one of the most prominent modes of action of antitumor antibodies, suffers from important limitations due to the need for optimal interactions with Fcγ receptors. In this work, we report the design of a new bispecific antibody format, compact and linker-free, based on the use of llama single-domain antibodies that are capable of circumventing most of these limitations. This bispecific antibody format was created by fusing single-domain antibodies directed against the carcinoembryonic antigen and the activating FcγRIIIa receptor to human Cκ and CH1 immunoglobulin G1 domains, acting as a natural dimerization motif. In vitro and in vivo characterization of these Fab-like bispecific molecules revealed favorable features for further development as a therapeutic molecule. They are easy to produce in Escherichia coli, very stable, and elicit potent lysis of tumor cells by human natural killer cells at picomolar concentrations. Unlike conventional antibodies, they do not engage inhibitory FcγRIIb receptor, do not compete with serum immunoglobulins G for receptor binding, and their cytotoxic activity is independent of Fc glycosylation and FcγRIIIa polymorphism. As opposed to anti-CD3 bispecific antitumor antibodies, they do not engage regulatory T cells as these latter cells do not express FcγRIII. Studies in nonobese diabetic/severe combined immunodeficient gamma mice xenografted with carcinoembryonic antigen–positive tumor cells showed that Fab-like bispecific molecules in the presence of human peripheral blood mononuclear cells significantly slow down tumor growth. This new compact, linker-free bispecific antibody format offers a promising approach for optimizing antibody-based therapies. Mol Cancer Ther; 12(8); 1481–91. ©2013 AACR.

The crystal structure of the ternary porcine lipase-colipase-tetra ethylene glycol monooctyl ether (TGME) complex has been determined at 2.8 A resolution. The crystals belong to the cubic space group F23 with a = 289.1 A and display a strong pseudo-symmetry corresponding to a P23 lattice. Unexpectedly, the crystalline two-domain lipase is found in its open configuration. This indicates that in the presence of colipase, pure micelles of the nonionic detergent TGME are able to activate the enzyme; a process that includes the movement of an N-terminal domain loop (the flap). The effects of TGME and colipase have been confirmed by chemical modification of the active site serine residue using diisopropyl p-nitrophenylphosphate (E600). In addition, the presence of a TGME molecule tightly bound to the active site pocket shows that TGME acts as a substrate analog, thus possibly explaining the inhibitory effect of this nonionic detergent on emulsified substrate hydrolysis at submicellar concentrations. A comparison of the lipase-colipase interactions between our porcine complex and the human-porcine complex (van Tilbeurgh, H., Egloff, M.-P., Martinez, C., Rugani, N., Verger, R., and Cambillau, C. (1993) Nature 362, 814-820) indicates that except for one salt bridge interaction, they are conserved. Analysis of the superimposed complexes shows a 5.4° rotation on the relative position of the N-terminal domains excepting the flap that moves in a concerted fashion with the C-terminal domain. This flexibility may be important for the binding of the complex to the water-lipid interface.

Pancreatic colipase plays an essential role in the intestinal fat digestion by anchoring lipase on lipid/water interfaces in the presence of bile salts. In contrast to other species, two molecular forms of colipase, A and B, have been found in horse. The two corresponding cDNAs were isolated from a horse pancreatic library and their nucleotide sequences were determined. Moreover, for the first time, active colipase has been obtained after transfection of COS cells by either colipase A or B cDNA.

Pancreatic lipase (EC 3.1.1.3) plays a key role in dietary fat digestion by converting triacylglycerols into 2-monoacylglycerols and free fatty acids in the intestine. Although the crystallographic structures of the human pancreatic lipase and of a human lipase-porcine colipase complex have been solved, no refined structure of pancreatic lipase has yet been published. The crystal structure of the horse enzyme was solved by the molecular replacement method from the model of the human pancreatic lipase and subsequently refined to 2.3 A resolution. The final model contains two molecules of 449 amino acid residues each in the asymmetric unit, 705 well-defined water molecules and two calcium ions. The two molecules in the asymmetric unit of the orthorhombic crystals are related by a 2-fold non-crystallographic symmetry axis as in the case of the human lipase. However, the association between the two molecules in their respective crystal forms is different. The overall molecular structure of the horse lipase is very similar to that of the human enzyme. The horse lipase is made up of two well-defined domains. The N-terminal domain which bears the active centre has a typical alpha/beta hydrolase fold topology. The C-terminal domain which is devoted to colipase binding has a beta-sheet sandwich topology. Comparison of equivalent C alpha atom positions between the final model of the horse lipase and the human lipase structure shows only slight differences which are mainly located in the C-terminal domain. The horse enzyme possesses the common features of the known mammalian and microbial lipases, in particular the "flap" covering the catalytic triad. In addition to more precise information concerning these features, the elucidation of the horse lipase crystal structure allowed us to better understand the structural basis of the kinetic behaviour of pancreatic lipases towards a soluble substrate, p-nitrophenyl acetate, and the different sensitivity of these enzymes towards limited proteolysis.

International audience; Pancreatic cancer is a devastating disease with the worst mortality rate and an overall 5-year survival rate lower than 5%. In the U.S., this disease is the fourth leading cause of death and represents 6% of all cancer-related deaths. Gemcitabine, the current standard first-line treatment, offers marginal benefits to patients in terms of symptom control and prolongation of life. Since 1996, about 20 randomized phase III trials have been performed to improve the efficacy of gemcitabine, with little success regarding a significant improvement in survival outcomes. The need for novel therapeutic strategies, such as target therapy, is obvious. Monoclonal antibodies have finally come of age as therapeutics and several molecules are now approved for cancer therapies. This review aims to give a general view on the clinical results obtained so far by antibodies for the treatment of pancreatic cancer and describes the most promising avenues toward a significant improvement in the treatment of this frustrating disease.

The complete sequence of the horse pancreatic lipase was elucidated by combining polypeptide chain and cDNA sequencing. Among the structural features of horse lipase, it is worth mentioning that Lys373 is not conserved. This residue, which is present in human, porcine and canine lipases, has been assumed to be involved in p-nitrophenyl acetate hydrolysis by pancreatic lipases. Kinetic investigation of the p-nitrophenyl acetate hydrolysis by the various pancreatic lipases and by the C-terminal domain (336-449) of human lipase reveals that this hydrolysis is the result of the superimposition of independent events; a specific linear hydrolysis occurring at the active site of lipase, a fast acylation depending on the presence of Lys373 and a non-specific hydrolysis most likely occurring in the C-terminal domain of the enzyme. This finding definitely proves that pancreatic lipase bears only one active site and raises the question of a covalent catalysis by pancreatic lipases. Moreover, based on sequence comparison with the above-mentioned pancreatic lipases, three residues located in the C-terminal domain, Lys349, Lys398 and Lys419, are proposed as possible candidates for lipase/colipase binding.

After a selective cleavage of a lipase/colipase cross-linked complex, the colipase has been shown to be bound to a 5 kDa lipase fragment identified as the C-terminal extremity of the chain extending from residue 403 to the C-terminus (Cys 449). The colipase binding site on lipase is therefore localized in a restricted contact area. Moreover, from sequence comparison of lipase from various species, an acidic residue, Glu 440, is likely to be involved in ion pairing with colipase.

In the intestine, the hydrolysis of triglycerides by pancreatic lipase is performed only in the presence of colipase, whose function is to anchor lipase to the bile-salt-coated lipid interface. Biochemical and crystallographic data on porcine and human lipases have shown that the molecule is made of two well-delimited domains. In order to get more information on the role of the domains in catalysis and colipase binding, we performed limited proteolysis on lipase from various species and obtained different patterns of cleavage. In the case of porcine and human lipases, only the C-terminal domain (12 kDa) could be obtained after chymotryptic attack, whereas in the horse enzyme the cleavage of the Leu410-Thr411 bond gave rise to a large N-terminal (45 kDa) and a small C-terminal (4 kDa) fragment. The isolated porcine and human C-terminal domains were completely inactive towards emulsified tributyrin, though were able to bind colipase. Conversely, the horse 45 kDa fragment retained the lipase activity but failed to correctly bind colipase. This work definitely proves that catalysis and colipase binding are separate events involving topographically distinct regions of the molecule and focuses attention on the role of the C-terminal domain in colipase binding.

International audience; Within the scope of improving the efficiency of pancreatic enzyme replacement therapy in cystic fibrosis, the feasibility of shifting the pH-activity profile of pancreatic lipase toward acidic values was investigated by site specific mutagenesis in different regions of the catalytic cavity. We have shown that introducing a negative charge close to the catalytic histidine induced a shift of the pH optimum toward acidic values but strongly reduced the lipase activity. On the other hand, a negative charge in the entrance of the catalytic cleft gives rise to a lipase with improved properties and twice more active than the native enzyme at acidic pH

Using the Tb3+ luminescence technique, we showed that bovine subunit III, a defective pancreatic serine endopeptidase-like protease, possessed a single metal ion binding site able to bind Tb3+ with a high affinity comparable to that of porcine elastase. The topology of the metal ion binding site in subunit III is predicted from sequence homologies and modeling experiments based on the known crystallographic three-dimensional structures of the equivalent sites in porcine elastase 1 and bovine beta-trypsin. Moreover, the Tb3+ luminescence technique in parallel to a 19F NMR investigation, allowed us to measure the binding of a very potent specific inhibitor of porcine elastase (trifluoroacetyl-L-lysyl-alanyl p-trifluoromethylphenylanilide) to bovine subunit III. These results confirm that, although devoid of any specific activity, subunit III might possess a conformation close to that of an active enzyme and further support the analogy between subunit III and an elastase-like family.

International audience; Colipase is a key element in lipase-catalyzed dietary lipids hydrolysis. Although devoid of enzymatic activity, colipase promotes pancreatic lipase activity in the physiological intestinal conditions by anchoring the enzyme on the surface of lipid droplets. Polarization modulation infrared reflection absorption spectroscopy combined with Brewster angle microscopy studies was performed on colipase alone and in various lipid environments to obtain a global view of both conformation and orientation and to assess lipid perturbations. We clearly show that colipase fully inserts into a dilaurin monolayer and promotes the formation of lipid/protein domains, whereas in a phospholipid environment its insertion is only partial, limited to the polar head group. In a mixed 70% phosphatidylcholine/30% dilaurin environment, colipase adsorbs to but does not penetrate deeply into the film. It triggers the formation of diglyceride domains under which it would form a rather uniform layer. We also clearly demonstrate that colipase adopts a preferred orientation when dilaurin is present at the interface. In contrast, at a neutral phospholipid interface, the infrared spectra suggest an isotropic orientation of colipase which could explain its incapacity to reverse the inhibitory effects of these lipids on the lipase activity

Pancreatic lipase is a soluble globular protein that must undergo structural modifications before it can hydrolyze oil droplets coated with bile salts. The binding of colipase and movement of the lipase lid open access to the active site. Mechanisms triggering lid mobility are unclear. The *KNILSQIVDIDGI* fragment of the lid of the human pancreatic lipase is predicted by molecular modeling to be a tilted peptide. Tilted peptides are hydrophobicity motifs involved in membrane fusion and more globally in perturbations of hydrophobic/hydrophilic interfaces. Analysis of this lid fragment predicts no clear consensus of secondary structure that suggests that its structure is not strongly sequence determined and could vary with environment. Point mutations were designed to modify the hydrophobicity profile of the [240–252] fragment and their consequences on the lipase-mediated catalysis were tested. Two mutants, in which the tilted peptide motif was lost, also have poor activity on bile salt-coated oil droplets and cannot be reactivated by colipase. Conversely, one mutant in which a different tilted peptide is created retains colipase dependence. These results suggest that the tilted hydrophobicity pattern of the [240–252] fragment is neither important for colipase binding to lipase, nor for interfacial binding but is important to trigger the maximal catalytic efficiency of lipase in the presence of bile salt.

Human pancreatic lipase (HPL, triacylglycerol acylhydrolase, EC 3.1.1.3) is a carboxyl esterase which hydrolyzes insoluble emulsified triglycerides and is essential for the efficient digestion of dietary fats. Though the three-dimensional structure of this enzyme has been determined, monitoring the conformational changes that may accompany the binding of various substrates and inhibitors is still of interest. Because of its sensitivity and ease of use, fluorescence spectroscopy of the intrinsic Trp residues is ideally suited for this purpose. However, the presence of seven Trp residues spread all over the HPL structure renders the interpretation of the fluorescence changes difficult with respect to the identification and location of the conformational or environmental changes taking place at the various Trp residues. In this context, the aim of this work was to investigate the contribution of the individual Trp residues to the fluorescence properties of HPL. To this end, we analyzed the steady-state and time-resolved fluorescence parameters of five single-point mutants in which one Trp residue was substituted with a weakly fluorescent Phe residue. In addition to the Trp residues at positions 30, 86, and 252, strategically located with respect to the active site, we also mutated Trp residues at positions 17 and 402, as representative residues of the HPL N- and C-terminal domains, respectively. Taken together, our data suggested that the solvent-exposed Trp30 residue contributed to at least 44% of the overall fluorescence of wild-type HPL. Moreover, we found that the long-lived fluorescence lifetime (6.77 ns) of wild-type HPL could be specifically attributed to Trp30, a feature that enables selective monitoring of its environmental changes. Additionally, Trp residues at positions 17 and 402 strongly contributed to the 1.61 ns lifetime of HPL, while Trp residues at positions 86 and 252 contributed to the 0.29 ns lifetime.

The physiological role of the two lipase-related proteins, PLRP1 and PLRP2, still remains obscure although some propositions have been made concerning PLRP2. In this paper, we report the presence of high amounts of PLRP2 in adult horse pancreas whereas no PLRP1 could be detected. As well, a non-parallel expression of PLRP2 and PLRP1 is observed in adult cat and dog, since no PLRP2 could be detected in these two species. In adult ox, neither PLRP2 nor PLRP1 could be found. These findings are in favor of a different regulation of the expression of the genes encoding pancreatic lipase and the related proteins according to the species. The cDNA encoding horse PLRP2 has been cloned and the protein expressed in insect cells. Both native and recombinant PLRP2 display the same catalytic properties. They possess a moderate lipase activity, inhibited by bile salts and not restored by colipase. Interestingly, they differ from PLRP2 from other species by their very low phospholipase activity indicating that PLRP2 could not be considered as a general phospholipase as previously postulated. This work highlights the variability of the properties of PLRP2 and rises the question of the physiological function of this protein in adult according to the species.

In the duodenum, pancreatic lipase (PL) develops its activity on triglycerides by binding to the bile-emulsified oil droplets in the presence of its protein cofactor pancreatic colipase (PC). The neutron crystal structure of a PC-PL-micelle complex (Hermoso, J., Pignol, D., Penel, S., Roth, M., Chapus, C., and Fontecilla-Camps, J. C. (1997) EMBO J. 16, 5531-5536) has suggested that the stabilization of the enzyme in its active conformation and its adsorption to the emulsified oil droplets are mediated by a preformed lipase-colipase-micelle complex. Here, we correlate the ability of different amphypathic compounds to activate PL, with their association with PC-PL in solution. The method of small angle neutron scattering with D(2)O/H(2)O contrast variation was used to characterize a solution containing PC-PL complex and taurodeoxycholate micelles. The resulting radius of gyration (56 A) and the match point of the solution indicate the formation of a ternary complex that is similar to the one observed in the neutron crystal structure. In addition, we show that either bile salts, lysophospholipids, or nonionic detergents that form micelles with radii of gyration ranging from 13 to 26 A are able to bind to the PC-PL complex, whereas smaller micelles or nonmicellar compounds are not. This further supports the notion of a micelle size-dependent affinity process for lipase activation in vivo.

Among the polar interactions occurring in pancreatic lipase/colipase binding, only one ion pair involving lysine 400 on lipase and glutamic acid 45 on colipase has been described. These residues are strictly conserved among species, suggesting that the ion pair is likely to play an important role. Therefore, in order to prevent this interaction, mutations intended to neutralize or inverse the charge of these residues have been introduced in the cDNAs encoding horse lipase and colipase. The recombinant proteins have been expressed in insect cells, and their catalytic properties have been investigated. In all cases, preventing the formation of the correct ion pair Lys400/Glu45 leads to lipase-colipase complexes of reduced affinity unable to perform an efficient catalysis, notably in the presence of bile salt micelles. Diethyl p-nitrophenyl phosphate inhibition experiments with either mutant lipase or mutant colipase indicate a poor stabilization of the lipase flap. These results suggest that the ion pair plays a critical role in the active conformation of the lipase-colipase-micelle ternary complex by contributing to a correct orientation of colipase relative to lipase resulting in a proper opening of the flap.

The organization of the pancreatic lipase in two well defined domains has been correlated to a specific function for each domain, catalytic activity for the N-terminal domain and colipase binding for the C-terminal domain. In order to see if such an organization implies that the two domains can behave as separate entities, we expressed the N- and C-terminal domains in insect cells. The recombinant proteins secreted in the cell supernatants present the expected molecular properties. However, whereas the C-terminal domain retains its function of colipase binding, the N-terminal domain appears to be unable to ensure catalysis. The lack of activity of the recombinant N-terminal domain could result either from a (partially) incorrect folding or from an incapacity to function by itself. These results suggest that, although both are structurally well defined, the two domains of the pancreatic lipase behave differently when they are expressed as separate entities.

Subunit III, a defective serine endopeptidase lacking the typical N-terminal hydrophobic dipeptide is secreted by the pancreas of ruminant species as part of the bovine ternary complex procarboxypeptidase A-S6. Two monoclinic crystal forms were obtained and subsequently used to solve its X-ray structure. The highest resolution model of subunit III was refined at 1.7 A resolution to a crystallographic R-factor of 18.4%, with r.m.s. bond deviations from ideality of 0.012 A. About 80% of the model presents the characteristic architecture of trypsin-like proteases. The remaining zones, however, have well-defined, unique conformations. The regions from residues 70 to 80 and from 140 to 155 present maximum distances of 16 and 18 A relative to serine proteases and zymogens. Comparisons with the structures of porcine elastase 1 and chymotrypsinogen A indicate that the specific binding pocket of subunit III adopts a zymogen-like conformation and thus provide a basis for its inactivity. In general, the structural analysis of subunit III strongly suggests that it corresponds to a truncated version of a new class of highly structured elastase-like zymogen molecules. Based on the structures of subunit III and elastase 1, it is concluded that large concerted movements are necessary for the activation of zymogen E.

A complete microcalorimetric investigation of the interactions between the native subunits of the bovine pancreatic procarboxypeptidase A-S6 ternary complex has been performed. All the association constants and thermodynamic parameters associated with the reactions forming the various complexes have been determined. The influence of pH and ionic strength on the binding reactions has been investigated. Interestingly, the affinity between the subunits is not significantly modified by varying the ionic strength. In this respect, an enthalpy/ entropy compensatory effect is observed for the binding of subunit III to subunit I when the ionic strength is increased, suggesting a physiological function for the association. The various pathways for formation of the ternary complex have been studied. Binding of subunit II (or III) to subunit I, the central element of the ternary complex, does not significantly modify the affinity of the other subunit for subunit I. From a thermodynamic point of view, the same final state is obtained whatever the pathway of ternary complex formation. This study is the first step of a kinetic investigation of the associated subunits.

A spectrofluorimetric investigation of the interactions between the subunits of the pancreatic bovine procarboxypeptidase A ternary complex was carried out after covalent insertion of a fluorescent probe at the active center of one of the constituent subunits. The specific insertion of an anthraniloyl group at the active center of subunit II free or bound to subunit I, after its conversion into chymotrypsin II, allowed us to determine the value of the dissociation constant between subunit I and anthraniloyl-chymotrypsin II (Kd = 0.7 +/- 0.1 x 10(-7) M) and between subunit III and the binary complex subunit I-anthraniloyl-chymotrypsin II (Kd = 1.6 +/- 0.3 x 10(-7) M). Moreover, the influence of the association on the flexibility of the active center of chymotrypsin II was deduced from fluorescence polarization measurements and rotational correlation time determination of anthraniloyl-chymotrypsin II free or bound to subunit I. The anthraniloyl group has no motion independently of the whole chymotrypsin II molecule and the binding of subunit I to anthraniloyl-chymotrypsin II results in an increase of the rigidity of the active site in the latter protein.

In contrast to procarboxypeptidase B which has always been reported to be secreted by the pancreas as a monomer, procarboxypeptidase A occurs as a monomer and/or associated to one or two functionally different proteins, depending on the species. Recent studies showed that, in the human pancreatic secretion, procarboxypeptidase A is mainly secreted as a 44 kDa protein involved in at least three different binary complexes. As previously reported, two of these complexes associated procarboxypeptidase A to either a glycosylated truncated protease E or zymogen E. In this paper, we identified proelastase 2 as the partner of procarboxypeptidase A in the third complex, thus reporting for the first time the occurrence of a proelastase 2/procarboxypeptidase A binary complex in vertebrates. Moreover, from N-terminal sequence analyses, the 44 kDa procarboxypeptidase A involved in these complexes was identified as being of the A1 type. Only one type of procarboxypeptidase B, the B1 type, has been detected in the analyzed pancreatic juices, thus emphasizing the previously observed genetic differences between individuals.

Unlike the pancreatic endopeptidase zymogens, procarboxypeptidase A is activated very slowly in vitro. The activation proceeds through the removal of about 100 amino acids away from the N-terminus of the chain. The cleavage of the susceptible bond(s) in monomeric and aggregated forms of bovine procarboxypeptidase A by catalytic amounts of trypsin was found to be very fast. However, as in the case of the porcine zymogen, the expression of the carboxypeptidase activity was considerably delayed by the inhibitory effect of the activation peptide which remains bound to the enzyme molecule after the trypsin treatment of the zymogen. α-Carboxypeptidase A was mainly formed under the relatively mild conditions used, indicating that the Arg−1–Ala+1 bond is probably the first to be cleaved during in vitro activation. The bovine carboxypeptidase activity was immediately and reversibly expressed upon dimethylmaleylation of the activation mixtures. This expression does not require full dissociation of the enzyme-peptide complex but merely a suitable change in its quaternary structure resulting from a modification of some electrostatic interactions upon dimethylmaleylation. Separation of bovine carboxypeptidase A from its activation peptide was only achieved upon filtration of the dimethylmaleylated mixtures in the presence of 6 M urea. The bovine activation peptide contains at least 93 amino acids compared to the 94 amino acids found by other authors for the rat and porcine peptides and sequencing of the first 53 amino acids showed a 75–85% homology with the latter two peptides.

Reversible condensation of the ternary complex form of bovine pancreatic procarboxypeptidase A with 2,3-dimethyl maleic anhydride was investigated at pH 9.0 and low concentration of reagent over the acylable amino groups. After subsequent modification of only a few lysyl residues, subunit III was found to have been released from the quaternary structure leading to the separation of an apparently native protein devoid of any contaminating subunit II, while dissociation of the remaining binary complex occurred upon further addition of the anhydride. This observation suggests that the electrostatic interactions existing between subunits I and III are more rapidly weakened than those between subunits I and II, probably because fewer lysyl residues are involved and/or there is greater accessibility to the chemical reagant. Although completely inactive on the specific substrates of trypsin, chymotrypsin and elastase, subunit III hydrolyzed p-nitrophenyl acetate at a rate similar to that of chymotrypsin but without any burst of p-nitrophenol, which indicates that the weakly functional active site of the subunit is not quite comparable to that of serine protease zymogens. Subunit III already has some of the functional characteristics of the corresponding active enzymes.

Up to now, a non-covalent ternary complex in which the pro-carboxypeptidase A (subunit I) is associated to two functionally different proteins (subunits II and III) has only been found in the pancreas of ruminant species. In the other species studied so far, the pro-carboxypeptidase A is secreted either as a monomer or as a binary association with a functionally different protein. Subunit I is the immediate precursor of carboxypeptidase A. Subunit II is a chymotrypsinogen of the C-type, involved, like subunit I, in the degradation of proteins and peptides. Although closely related to the pancreatic serine endopeptidases, subunit III appears to be devoid of any specific enzymatic activity. Information about the spatial organization of the subunits in the ternary complex has been deduced from the sequential dissociation of the complex. In contrast to the mechanism of activation of subunits I and II, which is independent of their aggregation state, the catalytic properties of the resulting enzymes are sensitive to their aggregation state. Moreover, the structural basis of inactivity of subunit III as well as the physiological role of the ternary complex are also discussed in this review.

The existence of procarboxypeptidase A, in the form of a non-covalent ternary complex containing the apparently inactive serine protease (subunit III), has so far been observed only in the ox pancreas. Evidence, obtained in the present study, shows that a ternary complex of procarboxypeptidase A, with a subunit III highly homologous with that of the bovine complex, is also present in two other ruminant species, sheep and goat. The biological significance of these complex, is also present in two other ruminant species, sheep and goat. The biological significance of these complex forms of procarboxypeptidase A and the consistently high biosynthesis level of the apparently inactive subunit III in all three ruminant species is still unknown. Yet the synthesis of subunit III is not related to the animal diet since in the horse, which is a non-ruminant herbivorous animal, the procarboxypeptidase A is monmeric. Reassociation assays between either bovine subunits II or III and monomeric as well as binary forms of procarboxypeptidase A from various species show that, unlike subunit II, the recognition site for subunit III is highly conserved in all the procarboxypeptidases A and that bovine subunit II is different from porcine chymotrypsinogen C with regard to association.

Subunit III (BSIII) of the bovine ternary complex of procarboxypeptidase A-S6 (PCPA-S6), a defective serine endopeptidase-like protein, actively synthesized by the pancreas of some ruminant species, is highly homologous to human protease E (HPE). Both proteins possess the same atypical disulfide bridge in position 98–99b. They are structurally related to porcine elastase 1 and human elastase 2 (about 56% identity). However, in contrast to those two enzymes which have an overall positive net charge, BSIII and HPE are negatively charged. Three-dimensional models of BSIII and HPE have been constructed from the crystallographic structure of porcine pancreatic elastase 1. The inhibitor-binding site for TFAI in these three proteins seems to be very similar; the atypical disulfide bridge does not seem to be involved in this binding site. The specific structural features of BSIII and HPE strongly support the assumption that BSIII is a truncated protease E and that both proteins belong to a separate serine endopeptidase family.