Search

The structure of recombinant human serum albumin derived from Pichia pastoris (rHSA) was analyzed in detail. Complete amino acid analysis was performed by the phenyl isothiocyanate precolumn labeling method. The amino terminal sequence was determined by the Edman degradation. The carboxyl terminal amino acid was determined by digestion with carboxypeptidase, and the carboxyl terminal peptide fragment was analyzed by electrospray mass spectrometry. The peptide fragments of rHSA digested with Lysylendopeptidase, Endoproteinase Glu-C, or Endoproteinase Asp-N were analyzed by electrospray mass spectrometry. The complete amino acid composition, the terminal sequences and the complete amino acid sequence of rHSA agreed with the primary structure deduced from its cDNA. The elution pattern of reduced and carboxymethylated rHSA digested with Lysylendopeptidase and the elution pattern of intact rHSA digested with pepsin were respectively similar to those of plasma-derived human serum albumin (pHSA). The pattern of CD spectrum of rHSA was identical in both shape and magnitude to that of pHSA. 1H-NMR spectra of rHSA and pHSA in deuterium oxide showed the same signal patterns in the observed region (delta 10.5-0.5 ppm). Cross peaks assigned to the alpha proton-beta proton of Asp-1 (delta 4.2/2.8 ppm) and the delta proton-epsilon proton of lysine residues (delta 2.8-3.2/1.4-2.0) showed the same cross peak patterns and chemical shifts in two-dimensional phase sensitive double-quantum filtered 1H-1H correlation spectra of rHSA and pHSA.

We developed enzyme immunoassays for human anti‐Pichia pastoris components (PPC) IgG and anti‐PPC IgE antibody titers. Anti‐PPC IgG antibody assay were performed using antigen‐coated plate and antihuman IgG peroxidase conjugate. The intra‐ and interassay coefficients of variation (CV) of anti‐PPC IgG antibody were 1.83–2.51% and 1.97–2.76%, respectively. The anti‐PPC IgE antibody assay was performed using an anti‐IgE monoclonal antibody‐coated plate, biotin‐labeled PPC and avidin‐labeled peroxidase, which was not subject to interference by the high titer of anti‐PPC IgG antibody. The intra‐ and interassay CV were 3.83–5.34% and 3.56–5.84%, respectively. We determined and compared anti‐PPC IgG antibody titers in the 40 normal individuals. We confirmed that a high titer of anti‐PPC IgG antibody is contained in all normal human sera and that these antibodies are directed primarily to mannan by immunoblotting analysis. The ratio of the maximum to minimum anti‐PPC IgG antibody titers in normal individuals was > 8,000. Anti‐PPC IgG antibody titers did not correlate with the age. However, we did not detect anti‐PPC IgE antibody in normal individuals. J. Clin. Lab. Anal. 11:196–201, 1997. © 1997 Wiley‐Liss, Inc.

Antithrombin III (ATIII) is a member of the serpin superfamily and a major regulator of the blood coagulation cascade. To express recombinant human ATIII (rATIII) in the methylotrophic yeast Pichia pastoris, we constructed an rATIII expression plasmid which contained the ATIII cDNA encoding mature protein region connected with the truncated mAOX2 promoter and the SUC2 secretion signal, introduced it into the P. pastoris genome, and screened for a single copy transformant. The secretion of rATIII from the transformant reached a level of 320 IU/L in the culture broth at 169 h. From the culture-supernatant, rATIII was purified to over 99% by heparin-affinity chromatography and other column chromatography methods. We characterized rATIII and compared it with human plasma-derived ATIII (pATIII). The purified rATIII possessed correct N-terminal amino acid sequence, and its molecular weight by SDS-PAGE of 56,000 Da was slightly different from the 58,000 Da of pATIII. Sequence and mass spectrometry analysis of BrCN fragments revealed that posttranslational modifications had occurred in rATIII. O-linked mannosylation was found at Ser 3 and Thr 9, and in some rATIII molecules, modification with O-linked mannosyl-mannose had probably occurred at Thr 386, close to the reactive center. Although the heparin-binding affinity of rATIII was 10-fold higher than that of pATIII, its inhibitory activity against thrombin was only half. As the conformation of rATIII and pATIII by circular dichroism spectroscopy was similar, O-glycosylation in the reactive center loop was assumed to be mainly responsible for the decreased inhibitory activity. pATIII can inactivate thrombin through formation of a stable thrombin-ATIII complex, but rATIII modified with O-glycosylation in the reactive center loop may act as a substrate rather than an inhibitor of thrombin.

We analyzed and compared the physicochemical and immunochemical properties of recombinant human serum albumin (rHSA) from Pichia pastoris with those of plasma-derived human serum albumin (pHSA). The second virial coefficient of rHSA, obtained from colloid osmotic pressure measurements at pH 6.7 +/- 0.1 was not significantly different from that of pHSA (P > 0.05). A 25% rHSA solution exhibited Newtonian flow, and the viscosity of 25% rHSA at 20 +/- 0.02 degrees C was not significantly different from that of 25% pHSA (P > 0.05). We analyzed the long- and medium-chain fatty acid composition of rHSA by reverse-phase HPLC using 9-anthryldiazomethane as the fluorescent labeling reagent. The total amount of fatty acid was higher for pHSA than for rHSA. The fatty acid composition of the rHSA preparation was the same as that of the pHSA preparation. However, the amounts of palmitic acid (C16:0) and stearic acid (C18:0) in rHSA were much lower than those in pHSA. Interestingly, we found that P. pastoris produced linolenic acid (C18:3) because it was detected in rHSA. The immunochemical properties of rHSA were analyzed by a parallel line assay method using anti-pHSA polyclonal antibody, and were identical to those of pHSA (P > 0.05).

We have developed a recombinant human serum albumin (rHSA) from Pichia pastoris which expresses high levels of heterologous proteins. rHSA is used clinically in high concentration (approximately 250 mg/ml in a 50 mL vial). We had to consider not only proteins from host cells as impurities but also mannan, which exhibits harmful effects on humans. However, the analysis of mannan in biopharmaceuticals produced from yeast has not been reported. Contaminating mannans in the final product were one important index to assess the clinical safety of rHSA. We have developed a highly sensitive enzyme immunoassay (EIA), utilizing an avidin-biotin system, for the detection of either the protein or mannan polysaccharide components from P. pastoris components (PPC) in rHSA. In addition, we used anion exchange chromatography with pulsed amperometric detection (AE-PAD) for monosaccharide analysis of glycoconjugates for the detection of mannan from PPC in rHSA. The detection limits of the EIA for PPC (PPC EIA) and the AE-PAD were 1 ng of protein/250 mg of rHSA and 180 ng of mannose/mg of rHSA, respectively. The mannan content in partially purified rHSA as determined by the AE-PAD was about same as the PPC content as determined by the PPC EIA. We showed that the PPC EIA and the AE-PAD are useful methods for the purity analysis of biopharmaceuticals produced from yeast.

The ligand binding properties and esterase-like activity of recombinant human serum albumin (rHSA) expressed by Pichia pastoris were compared with those of plasma derived human albumin (pHSA). The binding of long fatty acid ions was determined by the equilibrium partition method using radiolabeled palmitate. The association constants and the number of binding sites of diazepam, salicylate and warfarin were determined by specific and nonspecific binding models. The high affinity binding of bilirubin was kinetically determined from the oxidation rate of free bililubin in the binding mixture. The binding parameters of these five ligands obtained with rHSA were within the same range observed with pHSA preparations. The kinetic parameters for hydrolytic activity of rHSA toward p-nitrophenyl acetate was also similar to pHSA. These results indicate that rHSA and pHSA have the same functional property.

We purified a 58 kDa serine protease from culture-supernatant of Pichia pastoris and found that the NH2-terminal amino acid sequence of this protease is closely homologous to that of mature protein of Saccharomyces cerevisiae carboxypeptidase Y (CPY), which is encoded by the PRC1 gene. Using the S. cerevisiae PRC1 gene as a hybridization probe, a cross-hybridizing fragment of P. pastoris genomic DNA was identified and the gene, PRC1, encoding CPY, was cloned. The open reading frame of the P. pastoris PRC1 gene consists of 1569 bp encoding a protein of 523 amino acids. The molecular mass of the protein is calculated to be 59.44 kDa without sugar chains. The protein comprises 20 amino acids of pre (signal)-peptide, 87 amino acids of pro-peptide and 416 amino acids of mature peptide, and has four N-glycosylation sites. The NH2-terminal amino acid sequence of mature peptide is completely identical with that of the protease purified from the culture-supernatant. There is 61% identity between the amino acid sequences of P. pastoris Prc1p and S. cerevisiae Prc1p. Chromosomal disruption of the PRC1 gene resulted in the loss of CPY activity. Over-expression of the PRC1 gene under regulation of the P. pastoris AOX1 promoter resulted in accumulation of a large amount of active CPY in the intracellular fraction, and secretion of a slightly larger molecule that is thought to be pro-CPY. The nucleotide sequence data reported in this paper will appear in the EMBL Nucleotide Sequence Databases under the Accession Number X87987.

Recombinant mouse/human chimeric monoclonal antibody A10 (ch-A10) and its Fab fragment (ch-Fab) react with carcinoembryonic antigen on various gastrointestinal carcinomas. We performed biodistribution studies with 125I-labeled ch-A10 and ch-Fab in an antigen-positive human pancreatic carcinoma (BxPC-3) xenograft model. We also evaluated the anti-tumor effect of 131I-labeled ch-A10, and studied the detection of BxPC-3 xenografts with 123I-labeled ch-Fab in whole body scintigraphy. In comparative biodistribution studies, the tumor uptake of 125I-labeled ch-A10 was significantly greater than that of 125I-labeled ch-Fab 24 h post-injection. However, the tumor-to-blood ratio was 46.8 for ch-Fab at 24 h post-injection, while it was only 1.4 for ch-A10. Microautoradiography studies showed that ch-Fab penetrated more uniformly into the tumor nodules than did ch-A10. In mice given a therapeutic dose of 131I-labeled ch-A10, a significant inhibition of tumor growth was seen, while control 131-I-labeled human IgG did not affect tumor growth. Leukocyte toxicity was observed within 3 weeks after injection of 131I-labeled ch-A10, but leukocyte counts recovered to normal levels at 8 weeks post-injection. In whole-body scintigraphy, clear and rapid tumor imaging was obtained with 200 microCi of 123I-labeled ch-Fab 24 h post-injection. These results suggest that radioiodine-labeled chimeric A10 antibodies could potentially be useful candidates for radioimmunotherapy and radioimmunodetection of pancreatic carcinomas.

A large-scale purification method for hepatitis B surface antigen produced in a recombinant yeast (Saccharomyces cerevisiae) was established. The resulting HBsAg was greater than 99% pure and suitable for vaccine use. The yeast-derived HBsAg was structurally and biochemically similar to plasma-derived HBsAg. The anti-HBs antibody producing potency of the yeast-derived vaccine in mice was significantly higher than that of the plasma-derived vaccine. The yeast-derived vaccine induced protective antibody against hepatitis B virus of either adr or ayw subtype in a chimpanzee efficacy study. These observations demonstrate the usefulness of the yeast-derived vaccine as a second-generation hepatitis B vaccine.

An antibody against colistin (CL), an antibiotic effective for gramnegative bacteria, was produced in rabbits immunized with a colistinprotein conjugate. The conjugate was prepared by a novel and convenient procedure devised to couple an amino group of CL to thiol groups of bovine serum albumin (BSA) introduced by thiol exchange reduction of its disulfide bonds with dithiothreitol, using N-(m-maIeimidobenzoyloxy) succinimide (MBS) as a cross-linker. Enzyme labeling of CL with β-D-galactosidase was performed by utilizing another crosslinker, N(γ-maleimidobutyryIoxy)succinimide, by means of a convenient labeling method. A double antibody enzyme immunoassay of CL, which could determine as little as 30 ng/mL of CL, was developed using labeled CL and anti-CL antiserum. With this assay, drug levels were easily determined in fish tissue after CL administration. The enzyme immunoassay should provide a useful tool for detection and quantitation of residual drugs in foods and related products.

Antiserum against a strain of the rice blast fungus Pyricularia oryzae was elicited in rabbits immunized with its cell fragments emulsified with incomplete Freund's adjuvant. The fragments were also used as solid-phase antigens. A highly sensitive, competitive type enzyme-linked immunosorbent assay for P. oryzae was developed by using these two preparations as the immune reagents together with the use of beta-D-galactosidase-labeled anti-rabbit IgG as the tracer. Cross-reactivity of nine different strains of P. oryzae were measured by the assay. Sensitivity and accuracy of the assay was improved by choosing the cell fragments of the least cross-reactive strain as the solid-phase antigen. The improved method was successfully applied for sensitive and accurate assay of all ten strains of P. oryzae with the common measuring range between 1 and 100 ng per tube. Other species of microorganisms had little reactivity in this immunoassay indicating that the assay is specific to P. oryzae group microorganisms.