Specificity of Pseudomonas aeruginosa serralysin revisited, using biologically active peptides as substrates

The characterization of the specificity of alkaline protease from Pseudomonas aeruginosa has not yet been clearly defined. Some previous results suggested that its specificity was influenced more by amino acids far from the hydrolyzed peptide bond, than by amino acids in P1 or P'1 position. From other data, it was a C-(COOH)-type endoprotease where the preferential amino acid in P1 position was an arginine residue. We have studied the hydrolysis of several biologically active peptides. Many various sites of cleavage have been characterized but no arginine in P1 position was found, despite the presence of arginine in the peptide sequence. In fact P1 and P'1 position could be occupied by various amino acids. It seems unlikely that Pseudomonas alkaline protease may only be considered as a protease specific to arginine in P1 position. On the other hand, we have observed that increase of the peptide chain length led to an important increase of the hydrolysis rate, suggesting an extended number of subsites.     

Specificity of an extracellular proteinase from Brevibacterium linens ATCC 9174 on bovine alpha s1-casein

The specificity of the extracellular proteinase from Brevibacterium linens ATCC 9174 on bovine alpha s1-casein was studied. Hydrolysis was monitored over time by sodium dodecyl sulfate-polyacrylamide gel electrophoresis (PAGE) and urea-PAGE. The major pH 4.6-soluble peptides were isolated by high-performance liquid chromatography and identified by N-terminal amino acid sequencing and mass spectrometry. The time course of peptide formation indicated that His-8-Gln-9, Ser-161-Gly-162, and either Gln-172-Tyr-173 or Phe-23-Phe-24 were the first, second, and third bonds cleaved, respectively. Other cleavage sites included Asn-19-Leu-20, Phe-32-Gly-33, Tyr-104-Lys-105, Leu-142-Ala-143, Phe-150-Arg-151, Gln-152-Phe-153, Leu-169-Gly-170, and Thr-171-Gln-172. The proteinase had a broad specificity for the amino acid residues at the P1 and P'1 positions but showed a preference for hydrophobic residues at the P2, P3, P4, P'2, P'3, and P'4 positions.     

Hydrogen peroxide induced impairment of post-ischemic ventricular function is prevented by the sodium-hydrogen exchange inhibitor HOE 642 (cariporide)

Using highly degenerate, serine-protease-specific PCR primers on a midgut-specific cDNA library it was estimated that a minimum of 24 independent serine proteases were expressed in the midgut of Stomoxys calcitrans. The relative abundance of these 24 independent serine proteases has been estimated by restriction analysis of PCR products, showing that 69% fall into six almost equally abundant groups. Two highly abundant serine protease cDNAs (Ssp1 and Ssp2) were isolated and sequenced. They encode preproenzymes of 272 amino acids (Mr 28521) and 255 amino acids (Mr 27097) with putative signal peptides of 17 amino acids and 16 amino acids, putative activation peptides of 15 amino acids and 10 amino acids and mature enzymes of 239 amino acids (Mr 25322; pI 4.89) and 228 amino acids (Mr 24182; pI 7.59), respectively. Both deduced amino acid sequences contain the Asp/His/Ser catalytic triad and the highly conserved sequences surrounding it. Ssp2 also has the aspartate and two glycine residues in the specificity pocket, marking this as a typical trypsin. The positioning of the residues in the specificity pocket of Sspl is unusual; aspartate and glycine residues are present, which is typical of trypsin, but both are separated from surrounding conserved residues by additional amino acids; the second glycine found in the specificity pocket of trypsin is replaced by a serine, which is typical of chymotrypsin. Although a serine protease, the precise substrate specificity of Sspl remains to be determined. Northern analysis shows that both serine proteases are expressed constitutively with only a 20% change in the levels of expression of Ssp1 and Ssp2 through the digestive cycle, and that expression occurs predominantly in the opaque region of the midgut, the region responsible for secretion of digestive enzymes.     

The specificity of prolyl endopeptidase from Flavobacterium meningoseptum: mapping the S' subsites by positional scanning via acyl transfer

The S1'-S3' subsite specificity of prolyl endopeptidase from Flavobacterium meningoseptum was studied by acyl transfer to libraries of amino acid amides and peptides. Whereas the S1' and S3' subsites influence the specificity for the amino component by approximately one order of magnitude, the S2' subsite possesses a markedly higher specificity. Besides the high specificity for hydrophobic residues at P1'-P3', proline was efficiently bound by the S2' and S3' subsites of the enzyme. In contrast, no binding of P1' proline-containing peptides was observed. It could be demonstrated that the specificity of the S' subsite is not restricted to L-amino acids. Effective P'-S' interactions were also found for beta- and gamma-amino acids indicating that the enzyme does not form close contacts to the backbone of P1' and P2' amino acid residues.     

Synthetic substrates for human factor VIIa and factor VIIa-tissue factor

A series of 100 tripeptide fluorogenic substrates has been synthesized. These substrates contain Arg in the P1 position, various amino acids in the P2 and P3 positions, and different 6-amino-1-naphthalenesulfonamides (ANSN) as the detecting group (P'). The 38 compounds possessing the highest initial rates of factor VIIa hydrolysis were evaluated for substrate kinetic parameters in the presence and absence of tissue factor (TF) and by factor Xa. Most of these substrates had a higher kcat/KM (keff) value for the factor VIIa-TF complex than for factor Xa. Substitution of different amino acids in the P2 position showed that substrates with bulkier amino acids such as Leu, Pro, and Val have higher values for KM and kcat than those with smaller amino acids (Gly or Ser). The highest second-order rate constants were found for substrates with Val or Pro in the P2 position. A decrease or increase in volume of the P2 substituent (Gly, Ser, or Leu) resulted in a decrease in this constant. Substrates with the highest keff values have Phe in the P3 position. As the hydrophobicity and volume of the amino acid in the P3 position decreased, the keff was reduced. The efficiency of substrates for hydrolysis by factor VIIa was enhanced by an increase of hydrophobicity in the P' structure. TF enhanced the amidolytic activity of the "family" of 38 substrates with ANSN in the P' position on an average of 58-fold.     

Effects of introduced aspartic and glutamic acid residues on the P'1 substrate specificity, pH dependence and stability of carboxypeptidase Y

Carboxypeptidase Y is a serine carboxypeptidase isolated from Saccharomyces cerevisiae with a preference for C-terminal hydrophobic amino acid residues. In order to alter the inherent substrate specificity of CPD-Y into one for basic amino acid residues in P'1, we have introduced Asp and/or Glu residues at a number of selected positions within the S'1 binding site. The effects of these substitutions on the substrate specificity, pH dependence and protein stability have been evaluated. The results presented here demonstrate that it is possible to obtain significant changes in the substrate preference by introducing charged amino acids into the framework provided by an enzyme with a quite different specificity. The introduced acidic amino acid residues provide a marked pH dependence of the (kcat/Km)FA-A-R-OH/(kcat/Km)FA-A-L-OH ratio. The change in stability upon introduction of Asp/Glu residues can be correlated to the difference in the mean buried surface area between the substituted and the substituting amino acid. Thus, the effects of acidic amino acid residues on the protein stability depend upon whether the introduced amino acid protrudes from the solvent accessible surface as defined by the surrounding residues in the wild type enzyme or is submerged below.     

Analysis of tissue plasminogen activator specificity using peptidyl fluorogenic substrates

A series of 54 fluorogenic substrates have been synthesized and evaluated for tissue-type plasminogen activator (tPA) hydrolysis in an attempt to create efficient sensitive substrates for tPA and to investigate substrate structure-efficiency correlations. All substrates contain the 6-amino-1-naphthalenesulfonamide (ANSN) leaving group, Arg in the P1 position, various amino acids in the P2 and P3 positions, and various substituents in the sulfonamide moiety of the leaving group (P' position). The majority of substrates have relatively low K(M) values (< 100 microM), reaching as low as 2.6 microM, and reasonably high k(cat) values (up to 3.6 s(-1)). These substrates have higher affinity, higher hydrolysis rates, and higher efficiency for two-chain tPA than for the single-chain form of this enzyme. Analysis of the P3 structure influence on substrate efficiency demonstrates that compounds which contain D-isomers of N-blocked bulky amino acids, such as Phe, Leu, and Val, in this position are more efficient for tPA than substrates with N-unblocked small amino acids (Ser or Pro) in the P3 position. The second-order rate constants and k(cat) values for substrate hydrolysis increase with decreases in the P2 amino acid hydrophobicity in the following manner: Leu < Val and Gly < Ser < Pro. Substrates which contain an ANSN leaving group had a higher affinity for tPA than substrates with p-nitroaniline or 7-amino-4-methylcoumarin leaving groups. Analyses of substrate hydrolysis dependence on the substrate P' structure show that the k(cat) and the second-order rate constants increased with an increase in the size of monoalkyl substituent in the sulfonamide moiety, whereas substrates which contain either glycine methyl ester or a dialkyl group displayed the lowest efficiency for tPA. The substrate Boc-(p-F)Phe-Pro-Arg-ANSNHC2H5 allowed quantitation of tPA at a concentration as low as 1 pM, a concentration significantly lower than the plasma concentration of this protein. Evaluation of the activation of single-chain tPA by factor Xa demonstrates that prothrombinase is approximately 3-fold more efficient in activating sc-tPA than factor Xa alone, increasing the initial rate of activation from 0.0055 nM/s per 1 nM of factor Xa to 0.017 nM/s per 1 nM.     

Substrate specificity of honeydew melon protease D, a plant serine endopeptidase

The substrate specificity of honeydew melon (Cucumis melo var. inodorus Naud) protease D was studied by the use of synthetic substrates and oligopeptides derived from a protein hydrolyzate. The hydrolysis rates of succinyl-(L-Ala)1-3-p-nitroanilide (Suc-(Ala)1-3-pNA) the hydrolysis rate progressively rose in proportion to the increased chain length. Benzyloxycarbonyl-L-tyrosine p-nitrophenyl ester (Z-Tyr-ONp) and benzoyl-L-tyrosine ethyl ester (Bz-Tyr-OEt) were cleaved by honeydew melon protease D, but benzoyl-L-arginine p-nitroanilide (Bz-Arg-pNA), benzyloxycarbonyl-L-lysine p-nitrophenyl ester (Z-Lys-ONp) and tosyl-L-arginine methyl ester (Tos-Arg-OMe) were not hydrolyzed. Contrary to the results obtained by using synthetic substrates, the carboxyl sides of charged amino acid residues were preferentially cleaved by the enzyme in the oligopeptide substrates. The substrates that had charged or polar amino acids at P2 positions were not cleaved. On the other hand, the non-polar amino acid or proline at P2 were favored for hydrolysis. The information concerning the subsite of protease D was obtained and is useful for synthesis of a good substrate. As it is distinct from molecular mass, the substrate specificity of honeydew melon protease D is most analogous to cucumisin [EC 3.4.21.25] among serine proteases from cucurbitaceous plants.     

Specific fluorogenic substrates for neprilysin (neutral endopeptidase, EC 3.4.24.11) which are highly resistant to serine- and metalloproteases

Two intramolecularly quenched fluorogenic peptides containing o-aminobenzoyl (Abz) and ethylenediamine 2,4-dinitrophenyl (EDDnp) groups at amino- and carboxyl-terminal amino acid residues, Abz-DArg-Arg-Leu-EDDnp (Abz-DRRL-EDDnp) and Abz-DArg-Arg-Phe-EDDnp (Abz-DRRF-EDDnp), were selectively hydrolyzed by neutral endopeptidase (NEP, enkephalinase, neprilysin, EC 3.4.24.11) at the Arg-Leu and Arg-Phe bonds, respectively. The kinetic parameters for the NEP-catalyzed hydrolysis of Abz-DRRL-EDDnp and Abz-DRRF-EDDnp were K(m) = 2.8 microM, kcat = 5.3 min-1, kcat/K(m) = 2 min-1 microM-1 and K(m) = 5.0 microM, kcat = 7.0 min-1, kcat/K(m) = 1.4 min-1 microM-1, respectively. The high specificity of these substrates was demonstrated by their resistance to hydrolysis by metalloproteases [thermolysin (EC 3.4.24.2), angiotensin-converting enzyme (ACE; EC 3.4.24.15)], serineproteases [trypsin (EC 3.4.21.4), alpha-chymotrypsin (EC 3.4.21.1)] and proteases present in tissue homogenates from kidney, lung, brain and testis. The blocked amino- and carboxyl-terminal amino acids protected these substrates against the action of aminopeptidases, carboxypeptidases and ACE. Furthermore, DR amino acids ensured total protection of Abz-DRRL-EDDnp and Abz-DRRF-EDDnp against the action of thermolysin and trypsin. Leu-EDDnp and Phe-EDDnp were resistant to hydrolysis by alpha-chymotrypsin. The high specificity of these substrates suggests their use for specific NEP assays in crude enzyme preparations.     

A method for identifying the carboxy terminal amino acid of a protein

A highly sensitive new method for identifying the carboxy terminus of a protein was developed. The carboxyl terminal amino acid was racemized by reaction with acetic anhydride. The resulting modified protein was subjected to acid hydrolysis. The hydrolysate was derivatized with (+)-1-(9-fluorenyl)ethyl chloroformate to give fluorescent amino acid diastereomers. The amino acid diastereomers were separated on a reversed-phase column. Only carboxyl terminal amino acids give a D-amino acid. Application of this method was described for the isolation and identification of carboxyl terminal peptides from an enzymatic digest of a protein.     

Synthesis of a mixture of cyclic peptides based on the Bowman-Birk reactive site loop to screen for serine protease inhibitors

A peptide mixture containing 21 peptide sequences has been constructed to test the Bowman-Birk inhibitor reactive-site loop motif as the basis of inhibition for a range of serine proteases. The 21 peptides are all based on an 11 amino acid sequence designed from a Bowman-Birk like inhibitor reactive-site loop. Variation has been introduced at the P1 site of the loop, which has been randomised to include all the natural L-amino acids (except for cysteine), plus the non-natural L-amino acids ornithine and norleucine, The mixture of peptides was screened for specific binding to immobilised porcine pancreatic elastase, subtilisin BPN', alpha-chymotrypsin, trypsin, anhydro-alpha-chymotrypsin and anhydrotrypsin. Five peptides from the mixture bind to alpha-chymotrypsin, two of which also bind to anhydro-alpha-chymotrypsin, and two peptides bind trypsin, neither of which binds to anhydro-trypsin. The competitive inhibition constants (K(i)) and the rates of proteolytic hydrolysis of the individual peptides with their respective enzymes were determined. The rates of hydrolysis were found to vary widely and show little correlation with the K(i) values. In the case of the alpha-chymotrypsin inhibitors, the peptides with the lowest K(i) (0.1-0.05 mM) were the only peptides that bound to anhydro-alpha-chymotrypsin. However, no peptides bound to anhydrotrypsin, suggesting a fundamental difference in the way that alpha-chymotrypsin and trypsin are inhibited by these cyclic peptides.     

Effect of PGE2 on thymocyte proliferation induced by Con A or IL-4 + PMA

A vector projection method is proposed to predict the cleavability of oligopeptides by extended-specificity site proteases. For an enzyme with eight specificity subsites the substrate octapeptide can be uniquely expressed as a vector in an 8-dimensional space, whose eight bases correspond to the amino acids at the eight subsites, P4, P3, P2, P1, P1', P2', P3', and P4', respectively. The component of such a characteristic vector on each of the eight bases is defined as the frequency of an amino acid occurring at a given site. These frequencies were derived from a set of octapeptides known to be cleaved by HIV protease. The cleavability of an octapeptide can then be estimated from the projection of its characteristic vector on an idealized, optimally cleavable vector. The high ratio of correct prediction vs. total prediction for the data in both the training and the testing sets indicates that the new method is self-consistent and efficient. It provides a rapid and accurate algorithm for analyzing the specificity of any multi-subsite enzyme for which there is no coupling between subsites. In particular, it is useful for predicting the cleavability of an oligopeptide by either HIV-1 or HIV-2 protease, and hence offers a supplementary means for finding effective inhibitors of HIV protease as potential drugs against AIDS.     

An endo-acting proline-specific oligopeptidase from Treponema denticola ATCC 35405: evidence of hydrolysis of human bioactive peptides

An endo-acting proline-specific oligopeptidase (prolyl oligopeptidase [POPase], EC 3.4.21.26) was purified to homogeneity from the Triton X-100 extracts of cells of Treponema denticola ATCC 35405 (a human oral spirochete) by a procedure that comprised five successive fast protein liquid chromatography steps. The POPase is a cell-associated 75- to 77-kDa protein with an isoelectric point of ca. 6.5. The enzyme hydrolyzed (optimum pH 6.5) the Pro-pNA bond in carbobenzoxy-Gly-Pro-p-nitroanilide (Z-Gly-Pro-pNA) and bonds at the carboxyl side of proline in several human bioactive peptides, such as bradykinin, substance P, neurotensin, angiotensins, oxytocin, vasopressin, and human endothelin fragment 22-38. The minimum hydrolyzable peptide size was tetrapeptide P3P2P1P'1, while the maximum substrate size was ca. 3 kDa. An imino acid residue in position P1 was absolutely necessary. The hydrolysis of Z-Gly-Pro-pNA was potently inhibited by the following, with the Ki(app) (in micromolar) in parentheses: insulin B-chain (0.7), human endothelin-1 (0.5), neuropeptide Y (1.7), substance P (32.0), T-kinin (4.0), neurotensin (5.0), and bradykinin (16.0). Chemical modification and inhibition studies suggest that the POPase is a serine endopeptidase whose activity depends on the catalytic triad of COOH ... Ser ... His but not on a metal. The amino acid sequence around the putative active-site serine is Gly-Gly-Ser-Asn-Pro-Gly. The enzyme is suggested to contain a reactive cysteinyl residue near the active site. Amino acid residues 4 to 24 of the first 24 N-terminal residues showed a homology of 71% with the POPase precursor from Flavobacterium meningosepticum and considerable homology with the Aeromonas hydrophila POPase. The ready hydrolysis of human bioactive peptides at bonds involving an imino acid residue suggests that enzymes like POPase may contribute to the chronicity of periodontal infections by participating in the peptidolytic processing of those peptides.     

Elastin degradation by matrix metalloproteinases. Cleavage site specificity and mechanisms of elastolysis

Insoluble elastin was used as a substrate to characterize the peptide bond specificities of human (HME) and mouse macrophage elastase (MME) and to compare these enzymes with other mammalian metalloproteinases and serine elastases. New amino termini detected by protein sequence analysis in insoluble elastin following proteolytic digestion reveal the P'1 residues in the carboxyl-terminal direction from the scissile bond. The relative proportion of each amino acid in this position reflects the proteolytic preference of the elastolytic enzyme. The predominant amino acids detected by protein sequence analysis following cleavage of insoluble elastin with HME, MME, and 92-kDa gelatinase were Leu, Ile, Ala, Gly, and Val. HME and MME were similar in their substrate specificity and showed a stronger preference for Leu/Ile than did the 92-kDa enzyme. Fibroblast collagenase showed no activity toward elastin. The amino acid residues detected in insoluble elastin following hydrolysis with porcine pancreatic elastase and human neutrophil elastase were predominantly Gly and Ala, with lesser amounts of Val, Phe, Ile, and Leu. There were interesting specificity differences between the two enzymes, however. For both the serine and matrix metalloproteinases, catalysis of peptide bond cleavage in insoluble elastin was characterized by temperature effects and water requirements typical of common enzyme-catalyzed reactions, even those involving soluble substrates. In contrast to what has been observed for collagen, the energy requirements for elastolysis were not extraordinary, consistent with cleavage sites in elastin being readily accessible to enzymatic attack.     

Differentiation of lysine/glutamine in peptide sequence analysis by electrospray ionization sequential mass spectrometry coupled with a quadrupole ion trap

Electrospray ionization coupled to a quadrupole ion trap mass spectrometer is used to differentiate between the isobaric amino acids lysine and glutamine in sequence analysis of peptides. Collision-induced dissociation is used for fragmentation. Several isobaric peptides with one or more lysines or glutamines at different positions were investigated. The ambiguous amino acid either in the peptide chain or at the C- or N-terminus can be clearly identified based on specific side chain fragment ions resulting from MS3 or MS4 of B- and Y"-fragment ions.     

Mapping the functional domains of bacteriophage lambda integrase protein

Bacteriophage lambda encodes a site-specific recombination system that promotes the movement of the phage genome into and out of the host bacterial chromosome. The phage-encoded integrase (Int) is composed of 356 amino acid residues and carries out the required strand exchanges by means of a type I topoisomerase activity. Int also contains two distinct DNA-binding domains that interact with two different, specific sequences (arm-type and core-type sites) on DNA. In order to help understand the mechanism of site-specific recombination, we have used a genetic approach to isolate mutants defective in different steps in the recombination reaction. We developed a genetic screen for Int mutants that are defective in catalyzing excisive recombination in vivo. These mutants were screened for proficiency in binding to the P'123 arm-type sites using the bacteriophage P22 challenge-phage assays. In all, 78 such mutants were isolated and the mutational changes mapped and sequenced. These mutants have been further characterized (1) for their ability to bind the P'1 and P'123 arm-type sites and for their ability to form the attL complex in vivo, (2) for negative dominance in vitro, (3) for the presence of type I topoisomerase activity, and (4) for the ability to resolve artificially constructed recombination intermediates. We found that (1) residues in a stretch of 88 amino acids in the middle of the protein may be involved in Int-Int interactions, (2) a region around Arg212 is involved in the catalytic site, (3) residues near the carboxyl terminus play a role in enhancing Int binding to its arm-type sites, possibly by interacting with the small amino-terminal region that has been shown to be responsible for specific recognition of the arm-type sites, and (4) residues at the very carboxyl end of the protein may be involved in modulating the cleavage or religation activities of the Int protein.     

Prediction of substrate-specific pockets in cyclosporin synthetase

Amino acid sequence comparisons between domains of cyclosporin synthetase have been used to identify regions of the sequence which are responsible for the recognition and binding of the individual amino acids. Using a limited set of selection rules it was possible to identify three amino acid positions in the subdomain sequences which are responsible for amino acid specificity. Homology with the firefly luciferase protein shows that these three key residues are close to each other and line the surface of a putative specific substrate binding pocket located on the amino acyl-adenylation subdomain. These results allow us to predict a large number of cyclosporin synthetase mutants which could be used to synthesise alternative cyclosporin-like peptides.     

Liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry of omega- and (omega-1)-hydroxylated metabolites of elaidic and oleic acids in human and rat liver microsomes

A peptide that inhibits the human cholesteryl ester transfer protein (CETP) was isolated from hog plasma by ultracentrifugation, two sequential column chromatographies and electroelution from gels. Molecular weight of the peptide was determined to be approximately 3 kDa on the SDS-PAGE. The peptide contained 28 amino acids with an identical sequence to the amino terminus of hog apolipoprotein-CIII except two amino acid residues: -Pro-Glu- at the fifth and sixth amino acids from the amino terminus in the isolated peptide, in contrast to -Leu-Leu- in hog apo-CIII. A peptide synthesized chemically according to the amino acid sequence of the peptide (designated P28) showed approximately the same degree of CETP inhibitory activity as the isolated peptide. Synthetic peptides with different number of amino acids were also tested for CETP inhibition. Among the peptides, the one with 20 amino acid residues (P20) from the amino terminus showed the highest inhibitory activity against the CETP. The peptide appeared to be associated with the hog high-density lipoproteins (HDL), as determined by immunoblot analysis using antibody against P28. The CETP-inhibitory activity of the peptide was examined in vivo using diet-induced hypercholesterolemic rabbits. When the peptide was injected into the rabbits (7-9 mg/kg body weight), approximately 75% CETP activity disappeared from the plasma in 1 h after the injection and the effect lasted up to 30 h. The inhibition of CETP in vivo led to a concomitant decrease in total plasma cholesterol level up to 30% and an increase in the level of HDL-cholesterol up to 32%. The cholesterol concentrations in the rabbit plasma gradually recovered to the initial level after 48 h.     

The association of herpes simplex virus type 2 (HSV-2), Haemophilus ducreyi, and syphilis with HIV infection in young men in northern Thailand

BACKGROUND: Protein antigens are presented to cytotoxic T lymphocytes as small peptides (approximately 9-10 amino acids long) bound to class I molecules of the major histocompatibility complex. The identification of tumor-associated antigens and specific peptide epitopes (i.e., antigenic determinants) may be useful in the development of anticancer vaccines. The generation of a cytotoxic T-cell response to one peptide epitope (amino acids 146-154) of human prostate-specific antigen (PSA) has been reported. PURPOSE: Our aim was to identify novel PSA peptides capable of eliciting specific cytotoxic T-cell responses. METHODS: Candidate peptides were identified on the basis of the following criteria: 1) they contained consensus amino acid motifs for binding to HLA-A2, which is the most common type of class I molecule; 2) they lacked strong homology with PSA-related kallikrein proteins; and 3) they were capable of stabilizing HLA-A2 class I molecules on the surface of human T2 (transport deletion mutant) cells, which are defective in antigen presentation. T-cell lines capable of killing (i.e., lysing) T2 target cells that had been pulsed with specific PSA peptides were generated from three different males (two disease-free individuals and one patient with prostate cancer) by incubating peripheral blood mononuclear cells with the peptides and interleukin 2. Specific cell lysis was monitored by the release of radioactivity from target cells that had been labeled with [111In]oxyquinoline. RESULTS: Two novel PSA peptides capable of eliciting cytotoxic T-cell responses were identified; these peptides were designated PSA-1 (amino acids 41-150) and PSA-3 (amino acids 154-163). Four different cytotoxic T-cell lines were generated in response to these peptides-three against PSA-3 and one against PSA-1. Specific lysis of peptide-pulsed T2 cells by the T-cell lines was blocked by the addition of a monoclonal antibody directed against class I molecules. The T-cell lines were also capable of lysing PSA-positive, HLA-A2-positive LNCaP cells (human prostate carcinoma cells). The specificity of LNCaP cell lysis was shown by the following: 1) the inability of added human K562 (chronic myelogenous leukemia) cells to inhibit it, 2) the ability of added anti-HLA-A2 antibodies to block it, and 3) the inability of the T-cell lines to induce substantial lysis of PSA-negative, HLA-A2-positive human cancer cells. IMPLICATIONS: Our studies form a rational basis for the use of PSA peptides or recombinant vectors encoding PSA in the development of anticancer vaccine immunotherapy protocols for patients with prostate cancer.