[Clinical importance of the determination of the hydrophobic forms of serum carcinoembryonic antigen (CEA) in metastatic cancer of the colon]. / Intérêt clinique de la détermination des formes hydrophobes de l'antigène carcinoembryonnaire (ACE) sérique dans les cancers métastatiques du côlon.

Thirty-four patients with metastatic colon cancer were treated with 5 fluorouracil and folinic acid. The follow-up of disease was evaluated by tomodensitometric CT-scan analysis and by serum CEA determination. In addition, a study of the different CEA molecular forms separated by Triton X114 partitioning, immunoprecipitation and immunoblotting was completed. Concerning the aqueous phase, no relationship appeared between the pattern of CEA species and the outcome of chemotherapy. Opposingly, the analysis of the hydrophobic phase gave results closely correlated to chemotherapeutic response. In 19/34 patients, the hydrophobic CEA forms were absent or weakly expressed; out of these patients, 16/19 underwent a successful response to chemotherapy regimen. Opposingly, all of the remaining 15 patients expressing high levels of hydrophobic CEA species were non-responders. The present study thus gives new means for predicting the outcome of 5 fluorouracil-folinic acid chemotherapy by screening the molecular CEA forms expressed in the serum of patients with metastatic colon cancer.

Carcinoembryonic antigen has a different molecular weight in normal colon and in cancer cells due to N-glycosylation differences.

Carcinoembryonic antigen, an apical membrane glycoprotein expressed in normal human colonic epithelial cells, colonic polyps, tumor, and tissue culture cell lines originating from colonic adenocarcinomas, is generally considered to have a molecular weight of 180,000. Using sodium dodecyl sulfate-polyacrylamide gel electrophoresis associated with immunoprecipitation or immunoblotting with both monoclonal (Mab 517 and Mab 601) and polyclonal antibodies, we observed that carcinoembryonic antigen was actually expressed as two discrete apparent molecular weight forms in normal tissues: a broad band averaging at Mr 200,000 and a sharp band at Mr 130,000. This constituted the phenotype of the normal colon. In cancer cells we detected a single band at Mr 170,000 or lower. This variation was mainly the consequence of a modification of the glycosylation pattern of the molecule since deglycosylation by N-glycanase or biosynthesis in the presence of tunicamycin always produced a single molecular weight form, whether or not the source of tissue was normal or cancerous. By close inspection of benign, moderately transformed, and carcinomatous human colonic polyps we noticed that this shift in the molecular weight of carcinoembryonic antigen preceded the detection of other cancer markers such as nonspecific cross-reacting antigen at Mr 95,000 or the histological modifications leading to malignant diagnosis. Carcinoembryonic antigen constitutes, therefore, an important model with which to study the modifications of the glycosylation pattern induced during cancer biogenesis.

Porcine pancreatic alpha-amylase: a model for structure--function studies of homodepolymerases.

The amino acid sequence of the porcine pancreatic alpha-amylase chain (496 residues) contains four regions (96-101, 193-201, 233-236 and 295-300) which are highly homologous in amylases of different origins. These regions all belong to the N-terminal domain of the enzyme. Limited proteolysis by subtilisin allows a cut to be made at bond 369-370. Purified fragments indicate that both N- and C-terminal domains are required for amylolytic activity. Kinetic studies and reaction product analysis using oligomaltosides, their nitrophenylated derivatives and amylose as the substrate allowed us to establish: 1) the energy profile of the 5 subsites and, especially, that subsite number 3 is catalytic; 2) that 2 molecules of either maltotriose or its o-nitrophenylated analog or maltose bind to the active site at high substrate concentration. Such a subsite occupancy was confirmed by fluorescence quenching studies. Finally the hydrolysis of p-nitrophenylmaltoside was studied as a function of pH. In contrast to starch hydrolysis, the initial velocity plots for nitrophenol and p-nitrophenylglucoside liberation both gave a narrow pH-activity peak with a maximum value around pH 5.5. All data provide strong evidence for the participation of 2 carboxylic residues in the catalysis.

The determination of subsite binding energies of porcine pancreatic alpha-amylase by comparing hydrolytic activity towards substrates.

The active centre of porcine pancreatic alpha-amylase contains five subsites. Their occupancy has been studied using as a substrate maltooligosaccharide of various chain lengths (maltose up to maltoheptaose), some of their p- and o-nitrophenylated derivatives, and 412-residue amylose. Quantitative analysis of the digestion products allowed the determination of the subsite occupancy for the various productive complexes, the bond cleavage frequency and respective kcati (where i is the binding mode). The catalytic efficiency (kcat/Km) increases with chain length from maltose (2 M-1 X S-1) up to amylose (1.06 X 10(7) M-1 X S-1). The kinetic parameters of p-nitrophenylmaltoside hydrolysis are quite close to those of maltose, and the ortho compound behaves as maltotriose. Determination of binding energy of glucose residue at the various subsites calculated according to the method of Hiromi et al. (Hiromi, K., Nitta, Y., Numata, C. and Ono, S. (1973) Biochim. Biophys. Acta 302, 362-375) did not give consistent results. A method is proposed based on certain properties of porcine pancreatic alpha-amylase, especially the non-interaction of the p-nitrophenyl moiety of the maltose derivative with subsites 1 and 2, and the o-nitrophenyl group which interacts in a similar way to a glucose residue at the reducing end, and on the grounds that the amylase-amylose complexes are of the productive type. In addition, binding energy differences were calculated from substrates with the same chain length. The subsite energy profile is characterized by a low value at subsite 3 which confirms this subsite as the catalytic one. Another consequence is that the hydrolysis rate constant of productive complexes (kintn) (where n is the number of glucose or glucose equivalent residues for a given substrate) varies with chain length which is in conflict with the hypothesis of Hiromi et al.

On porcine pancreatic alpha-amylase action: kinetic evidence for the binding of two maltooligosaccharide molecules (maltose, maltotriose and o-nitrophenylmaltoside) by inhibition studies. Correlation with the five-subsite energy profile.

Hydrolysis of small substrates (maltose, maltotriose and o-nitrophenylmaltoside) catalysed by porcine pancreatic alpha-amylase was studied from a kinetic viewpoint over a wide range of substrate concentrations. Non-linear double-reciprocal plots are obtained at high maltose, maltotriose and o-nitrophenylmaltoside concentrations indicating typical substrate inhibition. These results are consistent with the successive binding of two molecules of substrate per enzyme molecule with dissociation constants Ks1 and Ks2. The Hill plot, log [v/(V-v)] versus log [S], is clearly biphasic and allows the dissociation constants of the ES1 and ES2 complexes to be calculated. Maltose and maltotriose are inhibitors of the amylase-catalysed amylose and o-nitrophenylmaltoside hydrolysis. The inhibition is of the competitive type. The (apparent) inhibition constant Kiapp varies with the inhibitor concentration. These results are also consistent with the successive binding of at least two molecules of maltose or maltotriose per amylase molecule with the dissociation constants Ki1 and Ki2. These inhibition studies show that small substrates and large polymeric ones are hydrolysed at the same catalytic site(s). The values of the dissociation constants Ks1 and Ki1 of the maltose-amylase complexes are identical. According to the five-subsite energy profile previously determined, at low concentration, maltose (as substrate and as inhibitor) binds to the same two sites (4,5) or (3,4), maltotriose (as substrate and as inhibitor) and o-nitrophenyl-maltoside (as substrate) bind to the same three subsites (3,4,5). The dissociation constants Ks2 and Ki2 determined at high substrate and inhibitor concentration are consistent with the binding of the second ligand molecule at a single subsite. The binding mode of the second molecule of maltose (substrate) and o-nitrophenylmaltoside remains uncertain, very likely because of the inaccuracy due to simplifications in the calculations of the subsite binding energies. No binding site(s) outside the catalytic one has been taken into account in this model.

Subsite profile of the active center of porcine pancreatic alpha-amylase. Kinetic studies using maltooligosaccharides as substrates.

The hydrolysis of several maltooligosaccharides catalysed by porcine pancreatic alpha-amylase was performed in order to determine their kinetic parameters. Maltose behaves as a substrate. Molecular activity (Ko) increases with chain length up to maltopentaose, remaining practically unchanged from maltopentaose to maltoheptaose. Maltose shows the highest Km value while the one for maltotriose is the lowest. Only maltose and maltotriose were directly cleaved to glucose. From Km and Ko values, the binding energy of the total complexes and of the productive ones respectively were calculated. The binding energy at each subsite was determined assuming that each substrate forms a single productive complex and that maltose and maltotriose differ in their binding mode from higher oligosaccharides. The model was checked by calculating theorical Km and Ko. Theorical values agree reasonably well with experimental ones.

[Purification and characterization of bovine pancreatic alpha-amylase]. / Purification et caractérisation de l' alpha-amylase pancréatique de boeuf.

A method for the purification of alpha-amylase from Bovine pancreas homogenate is described. This enzyme has the same molecular weight as the Porcine isoenzymes but its isoelectric point is more basic. Its amino acid composition is similar to that of Porcine, Ovine, Rat and Mouse amylases.