Search

1. A new diagonal etectrophoretic technique for determining the order of the lysine-containing tryptic peptides of a protein is described. The protein is converted into its trifluoracetyl derivative, digested enzymatically (or chemically), and the resulting peptides separated by paper electrophoresis. The paper is then treated with ammonia vapour, which re-exposes the ε-amino groups of the lysine residues, and submitted to a second electrophoresis at right angles to the first direction. Peptides containing lysine residues, together with the N-terminal peptide of the protein, are found to lie off a diagonal formed by all other peptides, whence they may be readily purified. A study of these peptides enables the order of the lysine-containing tryptic peptides in the protein to be deduced. 2. The technique has been successfully tested with insulin. 3. When the method was applied to porcine pepsin, the four tryptic peptides isolated were easily ordered and the carboxyl-terminal sequence of the protein shown to be Arg-Gln-Tyr-TyrThr-Val-Phe-Asp-Arg-Ala-Asn-Asn-Lys-Val-Gly-Leu-Ala-Pro-Val-Ala. The thee basic ammo acid residues in the molecule are thus found clustering towards the C-terminus of the polypeptide chain. 4. A common ancestral gene for porcine pepsin and bovine (calf) rennin is suggested by the close homology between the C-terminal sequents of the two proteins. [ABSTRACT FROM AUTHOR]

A new approach of predicting structural classes of protein domain sequences is presented in this paper. Besides the amino acid composition, the composition of several dipeptides, tripeptides, tetrapeptides, pentapeptides and hexapeptides are taken into account based on the stepwise discriminant analysis. The result of jackknife test shows that this new approach can lead to higher predictive sensitivity and specificity for reduced sequence similarity datasets. Considering the dataset PDB40-B constructed by Brenner and colleagues, 75.2% protein domain sequences are correctly assigned in the jackknife test for the four structural classes: all-α, all-β, α/β and α + β, which is improved by 19.4% in jackknife test and 25.5% in resubstitution test, in contrast with the component-coupled algorithm using amino acid composition alone (AAC approach) for the same dataset. In the cross-validation test with dataset PDB40-J constructed by Park and colleagues, more than 80% predictive accuracy is obtained. Furthermore, for the dataset constructed by Chou and Maggiona, the accuracy of 100% and 99.7% can be easily achieved, respectively, in the resubstitution test and in the jackknife test merely taking the composition of dipeptides into account. Therefore, this new method provides an effective tool to extract valuable information from protein sequences, which can be used for the systematic analysis of small or medium size protein sequences. The computer programs used in this paper are available on request. [ABSTRACT FROM AUTHOR]

This paper presents the first purification of the mitochondrial branched‐chain amino acid aminotransferase (BCATm) from sheep placenta. It is a homodimer with an apparent subunit molecular mass of 41 kDa. The enzyme differs from those of the rat and human as it appears to form at least one intermolecular disulfide bond. The sheep BCATm cDNA (1.4 kb) encodes a mature polypeptide of 366 amino acids with a calculated molecular mass of 41 329 Da and a partial mitochondrial targeting sequence of seven amino acids. The sheep BCATm sequence shares higher identity with other mammalian BCATm isoenzymes (82–85%) than with the cytosolic isoenzymes (60%). By Northern blot analysis, a message of 1.7 kb was detected in sheep placenta and skeletal muscle. Measurements of BCAT activity, mRNA and BCATm protein in sheep placenta and skeletal muscle revealed that BCATm is the sole BCAT isoenzyme expressed in placenta, whereas it contributes 57 and 71% of the BCAT activity in tensor fascia latae and masseter muscles from weaned lambs respectively. Skeletal muscle, the main site of branched‐chain amino acid transamination, exhibits significantly lower BCAT activity in sheep than in rat. Our results suggest that the low BCATm mRNA level probably accounts for the low BCAT activity in sheep skeletal muscle, and that the metabolic scheme for branched‐chain amino acid catabolism is specific to each species. [ABSTRACT FROM AUTHOR]

Monoamine oxidases (MAO) A and B show a high degree of amino acid similarity. Apart from the NH2-terminus, which contains an ADP-binding consensus sequence, little is known about their structural features or the sequences involved in the binding of substrates. In the present paper, we have studied the structure/function relationships of MAOs by constructing 18 different chimeric forms of MAO, engineered by moving progressively the junction between the NH2-terminus of one MAO form with the COOH-terminus of its isoenzyme. After transient expression in HEK-293 cells, the properties of these chimeric enzymes were investigated using both selective and nonselective substrates and inhibitors. Whereas exchange of the ADP-binding sequence did not modify the catalytic properties of either MAO isoforms, chimeras with increasing length of the NH2-terminus of MAO-A (up to residue 256) showed a marked decrease in affinity towards the MAO-B substrate phenylethylamine and the inhibitor N-(2-aminoethyl)-5-chloro-2-pyridine carboxamide · HCI (lazabemide) when compared to wild-type MAO-B. No major changes were observed in the kcat, values of these chimeras. From the data obtained, two sequences, i.e. 62–103 and 146–220, appeared of particular importance in constituting the binding site of MAO-B. On the other hand, the catalytic properties and specificity of MAO-A appeared to be relatively insensitive to substitution of both the NH2-- (up to position 112) and COOH-termini (from residue 395) of MAO-A with the corresponding MAO-B sequences. However, further modification of the central 283-residue sequence of MAO-A did not appear compatible with enzymic activity. None of the engineered chimeras showed a shift in specificity from one isoform to the other. [ABSTRACT FROM AUTHOR]

Preparations of bovine lung surfactant D (SP-D) and conglutinin were examined by electron microscopy, gel-filtration and SDS/PAGE. SP-D is composed of non-covalently linked subunits, of 160 kDa, which each contain three, disulphide-linked, 44-kDa polypeptide chains. In the electron microscope a single 160-kDa subunit of SP-D appears as a 45.8 ± 3-nm-long rod connected to a small globular ‘head’. Particles were also seen which correspond to non-covalently linked dimmers, trimers and tetramers of the 160-kDa monomer subunit of SP-D. The tetramer structure contains 12 polypeptide chains and is very similar to the electron microscopy images and model reported by Strang et al. [Strang, C. J., Slayter, US., Lachmann, P. J. and Davis, A. E. (1986) Biochem. J. 236, 3811–389] for bovine conglutinin in which four 160-kDa subunits are disulphide-linked to give a molecule of expected molecular mass of 528 kDa. This study confirmed the findings by Strang et al. in the above paper for intact conglutinin and also emphasized that the rod-like structures, of length 37.6 ± 3.7 nm. Seen in the conglutinin subunits were significantly shorter than those in SP-D despite the close similarity in amino acid sequence (79% identity) and chain length between the two proteins. In addition, a truncated form of conglutinin was found in the conglutinin preparations, due to limited proteolysis of the Arg-Ala bond at position 54 in the 44-kDa chains. These truncated conglutinin chains yield a subunit composed of three shortened, non-disulophide-linked, chains and this subunit appears as a monomer with a rod length of 34.2 ± 2.8 nm in the electron microscope. On gel-filtration, a proportion of the SP-D preparation behaved, as expected, as a molecule with an apparent molecular mass of 600 kDa. The remainder of the SP-D preparation behaved as aggregated material with a molecular mass greater than 900 kDa which yielded no distinct structures in the electron microscope. Intact conglutinin was wluted at a position greater than 900 kDa but yet provided clear electron microscopy images of the tetramer structure described above. Although it is difficult to explain fully the anomalous behaviour of SP-D and conglutinin on gel-filtration, it is proposed that the natural form of both Sp-D and conglutinin is the highest distinct oligomer form seen in the electron microscope, i.e. a tetramer of the 160-kDa subunits in which the four rod-like elements are linked in a tail-to-tail fashion to a central core. [ABSTRACT FROM AUTHOR]

A protein is usually classified into one of the following four structural classes: all α, all β, (a + B) and α/β. In this paper, based on the maximum correlation-coefficient principle, a new formulation is proposed for predicting the structural class of a protein according to its amino acid composition. Calculations have been made for a development set of proteins from which the amino acid compositions for the standard structural classes were derived, and an independent set of proteins which are outside the development set. The former can test the self consistency of a method and the latter can test its extrapolating effectiveness. In both cases, the results showed that the new method gave a considerably higher rate of correct prediction than any of the previous methods, implying that a significant improvement has been achieved by implementing the maximum-correlation-coefficient principle in the new method. [ABSTRACT FROM AUTHOR]

In mouse spermatozoa, DNA is compacted by two protamines mP1 and mP2. Protamine mP2 (63 residues) is synthesized in spermatid nuclei as a precursor pmP2 (106 residues) which is subsequently processed at the end of spermiogenesis [Yelick, P. C., Balhorn, R., Johnson, P. A., Corzett, M., Mazrimas, J. A., Kleene, K. C. & Hecht, N. B. (1987) Mol. Cell. Biol. 7, 2173-2179]. Six proteins, three of which were described earlier [Chauvière, M., Martinage, A., Debarle, M., Alimi. E., Sautière, P. & Chevaillier, Ph. (1991) C. R. Acad. Sci. 313, 107-112], have molecular and electrophoretic properties similar to those of pmP2. They were isolated from purified testis nuclei and characterized by amino acid composition, N-terminal sequence and peptide mapping. From the amino acid compositions, it appears that all six proteins are rich in arginine, cysteine and histidine and are closely related to pmP2 and mP2. The N-terminal sequence of each protein overlaps a distinct region of the N-terminal part of pmP2. The C-terminal part of protamine mP2 starting at arginine 15 is common to all proteins as assessed by amino acid compositions and peptide maps, ll these structural data demonstrate that the six isolated proteins are products of pmP2 precursor processing. The six intermediate proteins pmP2/5, pmP2/11, pmP2/16, pmP2/20, pmP2/26 and pmP2/ 32 which contain 102, 96, 91, 87, 81 and 75 residues, respectively, are generated from the pmP2 precursor after N-terminal excision of 4, 10, 15, 19, 25 and 31 residues, respectively. The C-terminal sequence of protamine mP2 is strictly identical to that of its precursor; therefore, no maturation occurs in this part of the molecule. At the present time, the proteolytic pathway involved in the amino-terminal processing leading to the mature form of the protamine mP2 (63 residues) has not been elucidated. However, the different representation of six intermediates in the testis suggests that some stages of processing are faster than others or that some cleavage sites are preferred. The proteins described in this paper could result either from stepwise excision of N-terminal residues or from nonsequential cleavages. [ABSTRACT FROM AUTHOR]

1. The abundance of tyrosine sulfate in membrane proteins was quantified in four different cell lines and compared to that in soluble cellular and secreted proteins. 2. Upon metabolic labelling of HepG2, Ltk-, AtT20 and PC12 cells with [35S]sulfate or [³H]tyrosine, a fraction enriched in integral membrane proteins was found to contain small, but significant, amounts of protein-bound tyrosine sulfate (up to 2.5% of the total cellular plus secreted protein-bound tyrosine sulfate). On the other hand, the frequency of sulfation of tyrosine residues of membrane proteins was within the same order of magnitude as that of secreted proteins, indicating that the low abundance of tyrosine sulfate in membrane proteins was largely a reflection of the low abundance of these proteins themselves. Consistent with this conclusion were the results of an analysis showing that 14 out of 32 selected membrane-spanning proteins contain potential tyrosine sulfation sites. 3. In HepG2 cells, three tyrosine-sulfated integral membrane glycoproteins of molecular mass 100, 125 and 150 kDa were identified. Characterization of the 150-kDa tyrosine-sulfated membrane protein revealed that it was protected from proteolysis in intact cells, suggesting a localization in an intracellular organelle. 4. Together with the results reported in the preceding paper in this journal, our data suggest that tyrosine sulfation occurs in various classes of trans-Golgi-derived proteins, soluble as well as membrane, and extracellularly exposed as well as intracellularly retained, proteins. This suggests that tyrosine sulfation may have a variety of physiological functions, depending on the individual tyrosine-sulfated protein or protein class. [ABSTRACT FROM AUTHOR]

1. The abundance of tyrosine sulfate in secretory proteins and in various classes of cellular proteins has been quantified and compared to protein-bound carbohydrate sulfate. 2. HepG2 cells and fibroblasts, two cell types showing only the constitutive pathway of secretion, and PC12 cells, which show both the constitutive and the regulated pathway of secretion, were subjected to pulse-chase and/ or long-term labelling with [35S]sulfate and [³H]tyrosine, followed by analysis of proteins in the cells and medium. Under both conditions of labelling, 65-92% of the protein-bound tyrosine sulfate and 44-84% of the protein bound carbohydrate sulfate were found to be secretory. In HepG2 cells, the frequency of sulfation of tyrosine residues, which can be determined independently from protein abundance and the rate of protein synthesis, was 8-22 times higher in proteins secreted into the medium than in cellular proteins. 3. All cell lines studied contained significant amounts, not only of carbohydrate sulfate, but also of tyrosine sulfate in specific cellular proteins. As shown for fibroblasts, these tyrosine-sulfated proteins were retained within the cells for at least 100 min of chase following a pulse with [35S]sulfate and were almost completely recovered in a light membrane fraction after subcellular fractionation. 4. Lysosomes were found to contain small, but significant, amounts of protein-bound tyrosine sulfate in addition to protein-bound carbohydrate sulfate. Protein-bound tyrosine sulfate in lysosomes reached a peak at 20 min of chase and rapidly disappeared thereafter, whereas protein-bound carbohydrate sulfate accumulated after 20 min of chase. Examination of the known sequences of eleven lysosomal enzymes revealed the presence of potential tyrosine sulfation sites in five of them. 5. Our results show that secretory proteins are the most abundant, but not exclusive, in vivo substrates for tyrosine sulfation and suggest the presence of soluble tyrosine-sulfated proteins in lysosomes and other, as yet unidentified, organelles of the secretory pathway. In the following paper in this journal we describe the abundance of tyrosine sulfate in integral membrane proteins. [ABSTRACT FROM AUTHOR]

This paper experimentally verifies the idea presented earlier that the contact of nonpolar clusters located on the surface of protein molecules with water destabilizes proteins. It is demonstrated that protein stabilization can be achieved by artificial hydrophilization of the surface area of protein globules by chemical modification. Two experimental systems are studied for the verification of the hydrophilization approach. 1. The surface tyrosine residues of trypsin are transformed to aminotyrosines using a two-step modification procedure: nitration by tetranitromethane followed by reduction with sodium dithionite. The modified enzyme is much more stable against irreversible thermoinactivation: the stabilizing effect increases with the number of aminotyrosine residues in trypsin and the modified enzyme can become even 100 times more stable than the native one. 2. α-Chymotrypsin is covalently modified by treatment with anhydrides or chloroanhydrides of aromatic carboxytic acids. As a result, different numbers of additional carboxylic groups (up to five depending on the structure of the modifying reagent) are introduced into each Lys residue modified. Acylation of all available amino groups of α-chymotrypsin by cyclic anhydrides of pyromellitic and mellitic acids results in a substantial hydrophilization of the protein as estimated by partitioning in an aqueous Ficoll-400/Dextran-70 biphasic system. These modified enzyme preparations are extremely stable against irreversible thermal inactivation at elevated temperatures (65-98 °C); their thermostability is practically equal to the stability of proteolytic enzymes from extremely thermophilic bacteria, the most stable proteinases known to date. [ABSTRACT FROM AUTHOR]

The identities of the protein phosphatases involved in the regulation of hepatic glycolysis, gluconeogenesis and aromatic amino acid breakdown were investigated using 6-phosphofructo-1-kinase, fructose-1,6-biphosphatase, L-pyruvate kinase, phenylalanine hydroxylase and the bifunctional enzyme 6-phosphofructo-2-kinase/fructose-2,6-bisphosphatase as substrates. Purified preparations of protein phosphatases-1, 2A, 2B and 2C exhibited activity towards all five substrates in vitro, although phosphatases-1 and 2B were only weakly active. Studies in liver extracts using inhibitor-2 and trifluoperazine, which inhibit protein phosphatase-1 and 2B, respectively, confirmed that these phosphatases are unlikely to be important in dephosphorylating these substrates in vivo. Sequential fractionation of rat liver extracts by anion-exchange chromatography and gel-filtration failed to resolve any protein phosphatases acting on each substrate, apart from protein phosphatases-2A and 2C. The present results, together with those described in the following paper (in this journal) indicate that under the assay conditions used, protein phosphatase-2A is the most powerful phosphatase acting on each substrate, although protein phosphatase-2C contributes a significant percentage of the activity towards 6-phophofructor-1-kinase. No clear evidence was obtained for a role of metabolites in the regulation of dephosphorylation of the five substrates. This study reinforces our contention that only a few serine-specific and threonine-specific protein phosphatase catalytic subunits participate in cellular regulation. [ABSTRACT FROM AUTHOR]

According to the signal hypothesis, a signal sequence, once having initiated export of a growing protein chain across the rough endoplasmic reticulum, is cleaved from the mature protein at a specific site. It has long been known that some part of the cleavage specificity resides in the last residue of the signal sequence, which invariably is one with a small, uncharged side-chain, but no further specific patterns of amino acids near the point of cleavage have been discovered so far. In this paper, some such patterns, based on a sample of 78 eukaryotic signal sequences, are presented and discussed, and a first attempt at formulating rules for the prediction of cleavage sites is made. [ABSTRACT FROM AUTHOR]

The brain specific S-100 protein is a mixture of two components S-100a and S-100 b with a subunit composition αβ or β2 respectively. The amino acid sequence of the β subunit has been previously determined. This paper presents the sequence of the α subunit in the S-100a protein. The α subunit consists of 93 amino-acid residues and has a relative molecular mass of 10400. The sequence shows extensive homology (58 %) with that of the β-subunit and shares an apparent calcium binding site in the C-terminal half of the molecule, suggesting a close evolutionary relationship between these subunits. [ABSTRACT FROM AUTHOR]

We have developed a semi-empirical method to determine the subunit molecular weight of a protein. The method is a minor modification of the Edman degradation and is based on a simple chemical procedure to modify specifically the N-terminal NH2 group. Initially, all NH2 groups, including the ε-NH2 groups of lysine residues and the N-terminal α-NH2 group, are reacted with phenylisothiocyanate and the protein derivative is subjected to one step of the Edman degradation. The newly exposed N terminus is then reacted with radioactively labelled phenylisothiocyanate. A value for the subunit molecular weight can be obtained from an analysis of the incorporated radioactivity and the amount of the protein in the sample. The molecular weight of five different proteins have been determined by this method. Our method is particularly useful for proteins containing lipid or sugar components and also for relatively small peptides. The procedure described in this paper for the specific modification of the N terminus has been found to be a powerful tool for protein sequencing. [ABSTRACT FROM AUTHOR]

Sheep-serum transferrin shows marked polymorphism and more than 20 alleles have been identified although only 4 or 5 of these have a frequency higher than 1%. Each of the alleles has two bands in starch-gel electrophoresis, corresponding to a major and a minor fraction. This paper describes the isolation and partial characterisation of the two fractions from the transferrin of sheep homozygous for Tf B. The purification consisted of: (a) precipitation by ammonium sulphate, (b) chromatography on CM-cellulose and, finally (c) chromatography on DEAE-Sephadex. The purification procedure had no effect on the electrophoretic mobility of the two fractions and they appeared homogeneous by starch-gel and polyamide-gel electrophoresis and by immuno-electrophoresis. The amino-acid compositions of both fractions were very similar and the sequences of the first eight amino-acid residues: Ser-Pro-Glu-Lys-Thr-Val-Arg-Trp-were identical for both bands. These results, and the fact that the two fractions are found in all genetic variants and always have the same relative mobility strongly suggest that the differences do not lie in the polypeptide chain. From the results of hydrolysis by neuraminidase and assay of sialic acid, the major and minor fractions most probably contain two and three sialic acid residues (exclusively N-acetylneuraminic acid) respectively, thus explaining the different electrophoretic mobilities. The sialic-acid content has been calculated on the basis of a molecular weight of 77500 as determined both by low-speed equilibrium ultracentrifugation and by Archibald's method. The following physico-chemical constants have been established: absorption coefficient at 280nm, A1 mg/cm³1 cm = 1.25 ± 0.01 cm² · mg-1; partial specific volume, &vmacr; = 0.734 cm³ · g-1; sedimentation coefficient s20.wo = 5.15 S; refractive-index increment, dn/dc = 0.173 cm³ · g-1. [ABSTRACT FROM AUTHOR]

Trypanosoma brucei contains an ATP-dependent phosphofructokinase (PFK), located in its glycosomes, which are peroxisome-like organelles sequestering the majority of its glycolytic enzymes. In this paper, we report the cloning and sequencing of the single-copy gene encoding this enzyme. Its amino-acid sequence is more similar to pyrophosphate (PPi)-dependent PFKs than to other ATP-dependent PFKs. a phylogenetic analysis suggests that the enzyme must have been derived from a PPi-dependent ancestral PFK, which changed its phospho-donor specificity during evolution. The enzyme is no longer capable of using PPi as phospho substrate, nor can it catalyze the reverse reaction as PPi-PFKs generally can. Moreover, the presence of a high pyrophosphatase activity in the cell renders it unlikely that PPi can function as free-energy source in present-day trypanosomes. It remains to be determined which mutations were responsible for the chage in phospho-substrate specificity of the trypanosomatid PFK. As a result of its particular evolutionary hitory, the T. brucei PFK shows many structural differences, even at the active site, when compared with other ATP-dependent PFKs. These differences offer great potential for the structure-based design of trypanocidal drugs. [ABSTRACT FROM AUTHOR]

We have previously isolated a form of protein phosphatase-1 (PP1M) from avian smooth muscle myofibrils that is composed of the catalytic subunit of PP1 (PP1C) bound to an M-complex consisting of 110-kDa (M110) and 21-kDa (M21) subunits. The interaction of PP1C with an N-terminal region of the M110 subunit enhances the dephosphorylation of myosin and suppresses the dephosphorylation of other substrates [Alessi, D. R., MacDougall, L. K., Sola, M. M., Ikebe, M. & Cohen, P. (1992) Eur. J. Biochem. 210, 1023–1035; Chen, Y. H., Chen, M. X., Alessi, D. R., Campbell, D. G., Shanahan, C., Cohen, P. & Cohen, P. T. W. (1994) FEBS Lett. 356, 51–56; Johnson, D. F., Moorhead, G., Caudwell, F. B., Cohen, P., Chen, Y. H., Chen, M. X. & Cohen, P. T. W. (1996) Eur. J. Biochem. 239, 317–325]. In this paper, we establish that PP1M accounts for nearly all the myosin phosphatase activity in myofibrils, that the M110 and M21 subunits are present at similar concentrations in the myofibritlar fraction, and that these subunits are entirely bound to PP1. We demonstrate that the M21 subunit does not interact with PP1C, but with the C-terminal 72 residues of the M110 m subunit, a region which is 43 % identical to residues 87–161 of the M21 subunit. A fragment of the M21 subunit, M21-(M1–L146), which lacks the C-terminal leucine zipper, also bound to the M110 subunit, but two other fragments M21-(M1–E110) and M21-(E110–K186) did not. The M110 and M21 subunits were both found to be myosin-binding proteins. The C-terminal 291 residues of the M110 subunit, but not the C-terminal 72 residues, bound to myosin, but the N-terminal fragments M110-(M1—E309) and M110-(M1—S477) did not. Thus, the region of the M110 subunit that stimulates the dephosphorylation of myosin by PP1C is distinct from the region that targets PP1M to myosin. Remarkably, each myosin dimer was capable of binding about 20 tool M21 subunit and many of the M21-binding sites were located in the myosin rod domain. The potential significance of this observation is discussed. [ABSTRACT FROM AUTHOR]

For the elucidation of the amino-acid sequence of the carboxyl-terminal region of soybean trypsin inhibitor (Kunitz), fragments C and D were digested with trypsin, and the resulting peptides were separated by ion-exchange chromatography on Dowex 50X2 or by gel filtration on Bio-Gel P-4. Further fractionation and purification of the peptides were performed by ion-exchange chromatography on Dowex 1X2, by gel filtration on Bio-Gel P-2 or by high-voltage paper electrophoresis at pH 1.9 and 3.6. Three peptides were obtained in pure form from fragment C and ten peptides from fragment D, and their amino-acid sequences were determined by the direct Edman method and by carboxy- peptidase digestion technique. Overlapping peptides necessary for the alignment of the tryptic peptides from fragment D were obtained from a chymotryptic hydrolysate of fragment D. Nine main peptides and rune minor peptides were obtained. The amino acid composition and the partial amino-acid sequence of the 18 chymotryptic peptides of fragment D made it possible to establish the amino-acid sequence of the carboxyl-terminal region of the inhibitor. The complete amino-acid sequence of soybean trypsin inhibitor (Kunitz) deduced here was compared with that of Bowman-Birk inhibitor, another well-known soybean proteinase inhibitor. [ABSTRACT FROM AUTHOR]

A protein is generally classified into one of the following four structural classes: all α, all β, α+β and α/β. In this paper, based on the weighting to the 20 constituent amino acids, a new method is proposed for predicting the structural class of a protein according to its amino acid composition. The 20 weighting parameters, which reflect the different properties of the 20 constituent amino acids, have been obtained from a training set of proteins through the linear-programming approach. The rate of correct prediction for a training set of proteins by means of the new method was 100%, whereas the highest rate of previous methods was 82.8 %. Furthermore, the results showed that the more numerous training proteins, the more effective the new method. [ABSTRACT FROM AUTHOR]