Numerous amino acid sequences of lipases are known, primarily deduced from DNA sequencing. A total of 58 entries are listed in the Swiss-Prot: Bos tauru (bovine) pregastric, pancreatic lipase; Rattus norvegi-cus (rat) hepatic, pancreatic, lingual, pancreatic lipase related protein; Mus musculus (mouse) hepatic, pancr-reatic lipase; canis familiaris (dog) gastric, pancreatic lipase-related protein; Homo sapiens (human) hepatic, pancreatic, gastric, pancreatic lipase-related protein; Cavia porcellus (guinea pig) pancreatic; Oryctolagus cuniculus (rabbit) pancreatic; Equus caballus (horse)
pancreatic; Sus scrofa (pig) pancreatic; Candida rugosa (yeast) (formerly Candida cylindracea); C. albicans (yeast); Moraxella sp; Geotrichum candidum (Oospora lactis); Myocastor coypus (Coypu) (Nutria) pancreatic, pancreatic lipase-related protein; Chromobacterium vis-cosum; Rhizomucor miehei; Rhizopus oryzae (R. dele-mar); R. niveus; Staphylococcus epidermidis; S. hyicus; S. aureus; Saccharomyces cerevisiae (baker's yeast); Photohabdus luminescens (Xenorhabdus luminescens); Yarrowia lipolytica (Candida lipolytica); Aeromonas hydrophila extracellular lipase; Burkholderia cepacia (Pseudomonas cepacia); Ps. aeruginosa lactonizing lipase; Pseudomonas sp. (strain KWI-56); Ps. fluores-cens; Ps. glumae; Ps. fragi; Pseudomonas sp. (strain 109) lactonizing lipase; Psychrobacter immobilis; Canadida antarctica (yeast) (Trichosporon oryzae); Bacillus subtilis; Vibrio cholerae lactonizing lipase.
The crystal structures of a number of lipases have been solved, including those of Bos taurus (bovine) pancreatic lipase; Sus scrofa (pig) pancreatic lipase; Rattus norvegicus (rat) pancreatic lipase-related protein; Homo sapiens (human) gastric, pancreatic lipase, lipase complexed with colipase and phospholipid, lipase complexed with colipase and inhibited by unde-cane phosphonate methyl ester; Cavia porcellus (guinea pig) pancreatic lipase related protein; Equus caballus (horse) pancreatic lipase; Canis familiaris (dog) pancreatic lipase related protein; Chromobacterium viso-sum lipase; Candida antarctica lipase with phosphonate inhibitor; Rhizopus niveus lipase; R. dele-mar lipase; Candida rugosa (formerly C. cylindracea) lipase, lipase complexed with (IR)-methyl hexyl phos-phonate, lipase complexed with doecanesulfonate, lipase complexed with hexadecanesulfonate, lipase complexed with hexadecanesulfonate, lipase com-plexed with (IS)-menthyl hexyl phosphonate; Pseudomonas cepacia lipase; Ps. glumae lipase; Candida antarctica lipase; Rhizomucor miehei lipase, complex with diethylphosophate, complexed with n-hexylphosphonate ethyl ester; Geotrichum candidum strain atcc 34614 lipase; and Humicola lanuginosa lipase.
The porcine pancreatic enzyme is a glycoprotein composed of 450 amino acids with a calculated MW of 50,084 daltons. The enzyme consists of six disulfide bridges, Cys4-Cys10, Cys91-Cys102, Cys238-262, Cys286-297, Cys300-Cys305, and Cys434-450, with two free cysteines, Cys 104 and Cys 182 (5-8) (see also Swiss-Prot). The porcine pancreatic enzyme is
N-glycosylated at Asn167. The glycosylation site Asn167-Gly168-Thr169 in the porcine enzyme is conserved in the human and canine lipases, and the glycan structure of porcine pancreatic lipase has been elucidated (9). In contrast to the human, porcine, and canine pancreatic lipases, the horse, ovine, and bovine enzymes are not glycosylated. Heterogeneity of the gly-can moiety at Asn167 gives rise to 4 isoforms, LA1, LA2, Lb, and LC, with the less anionic LB the major form (70%) in the porcine enzyme (10). Unlike most of the pancreatic enzymes which are secreted as proenzymes and further activated by proteolytic cleavage in the small intestine, pancreatic lipases are directly secreted as a 50-kDa active enzyme.
The porcine lipase comprises two domains (7). The larger, N-terminal domain (residues 1-336) has an a/ft structure, containing the catalytic triad: Ser153, Asp177, and His264. The C-terminal domain (residues 337-450) assumes a ft-sandwich, and contains the binding site for colipase. In the human pancreatic lipase, the central parallel ft-sheet of the a/ft-domain has nine strands cross-connected by a-helices or loops, whereas the C-terminal ft-sandwich is formed by two layers of ft-sheet, each of four antiparallel strands (11). The crystal structure of the horse enzyme shows close resemblance (8). A significant discovery in the x-ray crystallographic investigation is the existence of a surface loop or flap covering the active site. This "lid" must be repositioned via a conformational change during interfacial activation to allow the substrate to enter the active site. In a lipase-colipase system, the lid, together with the colipase, forms a continuous hydro-phobic surface which serves as the lipid binding site of the complex at the interface.
Microbial lipases have been extensively studied owing to their potential industrial applications. The phyco-mycete fungus Rhizomucor miehei (formally Mucor miehei) produces an extracellular lipase that hydrolyzes a broad spectrum of lipid substrates found in animal plants. The lipase is synthesized as a precursor containing a signal peptide and a propeptide, in addition to 269 amino acid residues of the mature enzyme which has a calculated MW of 29,472 daltons (12). The secreted lipases were purified in two isoforms, A (pi = 3.9) and B (pi = 4.3), because of partial degly-cosylation in posttranslational processing (13). The Geotrichum candidum lipase also exists in two isoforms, I (pi = 4.56) and II (pi = 4.46) (14). Both have the same chain length of 544 amino acid residues (exclud ing the signal peptide of 13 residues), with a calculated MW of 59,085, containing two and three N-glycosyla-tion sites, respectively (15). However, lipases I and II, with an overall 84% identity in the two sequences, are encoded by separate genes, lip1 and lip2 (16, 17). The asporogenic yeast Candida rugosa (formerly C. cylin-dracea) produces extracellular lipase in multiple forms, five of which have been cloned and sequenced: lipase I (pI 4.5), II (pI 4.9), III (pI 5.1), IV (pI 5.7), and V (pI 5.5) (18, 19). All five genes code for 57-kDa proteins of 534 amino acid residues.
The Rhizomucor miehei lipase, similar to pancreatic lipases, is an a/ft-type protein. The enzyme molecule consists of a central parallel eight-stranded ft-sheet folded onto a highly amphipathic N-terminal helix (Fig. 2) (20-22). All the connecting loops are right-handed, and hence located on one side of the ft-sheet. Three disulfide bonds, Cys29-Cys268, Cys40-Cys43, and Cys235-Cys244, provide a global stabilization to the protein. The active center containing the catalytic triad (Ser144- • -His257- • -Asp203) is buried under a helical "lid" that is amphipathic in nature. When the enzyme is adsorbed at the lipid-water interface, the lid undergoes conformational changes to roll back from the active site. The hydrophobic side of the lid is exposed to the interface, thereby expanding the nonpolar surface area surrounding the active site. This stabilization of the open form of the lipase at the interface in effect creates a catalytically active enzyme to attack the triacylglycerol molecule with the lipid phase.
The Geotrichum candidum lipase has an a/ft structure with a central mixed ft-sheet formed by 11 strands, seven of which are parallel (23, 24). The helices and loops connecting the strands are found on both sides of the ft-sheet. Two disulfide bonds are present: Cys61-Cys105, and Cys276-Cys288. The active site, consisting of the catalytic triad Ser217-His463-Glu354 is concealed by two nearly parallel a helices.
The general molecular architecture has been observed in lipases from other microbial sources, including Candida rugosa (25), C. antarctica (26), Rhizopus delmar (27), Pseudomonas glumae (28), and Ps. aeruginosa (29). The major difference between the molecular structures of the pancreatic and the micro-bial enzymes is that the former contains a C-terminal domain serving as the binding site for colipase which is required for facilitating the interfacial catalysis in the presence of bile salts, whereas the latter is devoid of this domain. It should also be noted that lipases do not all subscribe to the phenomenon of interfacial activation. For example, the Psedomonas aeruginosa lipase lacks a lid-like helical loop structure covering the
Figure 2 A stereo view of Candida rugosa lipase (PDBid: 1trh).
active site, and does not show interfacial activation (30). A lipase isolated from guinea pig was found to carry a deletion in the lid sequence, and shows no interfacial activation (31). In addition to providing accessibility of the active site for the substrate, the opening of the lid causes a concomitant conforma-tional change for proper positioning of the residues forming the oxygen hole during catalysis (22, 25).
Was this article helpful?