Characterization of Sulfhydryl Heterogeneity in Human Serum Albumin and Recombinant Human Serum Albumin for Clinical Use

Since human serum albumin has one sulfhydryl group and 17 disulfides, reactive sulfhydryl groups give rise to heterogeneity. The present paper presents a comparison of sulfhydryl heterogeneity in human serum albumin and recombinant human serum albumin for clinical use. Low molecular weight sulfhydryl compounds were identified from both sources. The recombinant albumin had a much higher sulfhydryl content than plasma serum albumin.     

A dynamic model for bilirubin binding to human serum albumin

Site-directed mutagenesis of human serum albumin was used to study the role of various amino acid residues in bilirubin binding. A comparison of thermodynamic, proteolytic, and x-ray crystallographic data from previous studies allowed a small number of amino acid residues in subdomain 2A to be selected as targets for substitution. The following recombinant human serum albumin species were synthesized in the yeast species Pichia pastoris: K195M, K199M, F211V, W214L, R218M, R222M, H242V, R257M, and wild type human serum albumin. The affinity of bilirubin was measured by two independent methods and found to be similar for all human serum albumin species. Examination of the absorption and circular dichroism spectra of bilirubin bound to its high affinity site revealed dramatic differences between the conformations of bilirubin bound to the above human serum albumin species. The absorption and circular dichroism spectra of bilirubin bound to the above human serum albumin species in aqueous solutions saturated with chloroform were also examined. The effect of certain amino acid substitutions on the conformation of bound bilirubin was altered by the addition of chloroform. In total, the present study suggests a dynamic, unusually flexible high affinity binding site for bilirubin on human serum albumin.     

Further evidence for the heterogeneity of serum albumin

Albumin samples from three species (avian, bovine and human) were electrophoresed on gradient polyacrylamide gels in the presence of sodium dodecyl sulphate (SDS-PAGE). The resulting electrophoregram from each sample of serum albumin investigated showed multiple protein bands of a wide range of molecular weights. All seven samples of human serum albumin were found, using gel immunodiffusion, to be contaminated with other proteins. All but one sample was contaminated with proteins such as haptoglobin, alpha 1-glycoprotein, alpha 1-trypsin inhibitor, and prealbumin. This contamination accounts for part of the heterogeneity of these samples. Immunoblots, where the proteins were transferred to nitrocellulose and incubated with antisera, gave a better demonstration of the heterogeneity than Coomassie Blue staining and the immunoblotting procedure appeared to be more sensitive than the gel immunodiffusion technique. The heterogeneity of serum albumin demonstrated by the former technique included that of the monomer which was shown to be contaminated with antithrombin III. The commercial samples of human serum albumin, claimed as pure, were found to vary greatly in their tryptophan content, which also indicated heterogeneity. Heat treatment of human serum albumin with 1% SDS, followed by chromatography on agarose, removed the protein contaminants and with it the tryptophan. The presence of tryptophan in human serum albumin, therefore, indicated the presence of impurities.     

99mTc-neolactosylated human serum albumin for imaging the hepatic asialoglycoprotein receptor

99mTc-labeled diethylenetriaminepentaacetic acid (DTPA)-coupled neogalactosyl human serum albumin (GSA) is used as an imaging agent for asialoglycoprotein receptor of the liver. However, its labeling is inconvenient because it should be incubated for 30 min at 50 degrees C. In addition, the conjugated DTPAs can cause decrease of pI and denaturation of protein. Therefore, we developed an improved agent 99mTc-neolactosyl human serum albumin (LSA) which contains a terminal galactose. LSA was synthesized by conjugating lactose to human serum albumin by the formation of a Schiff's base and successive reduction with sodium cyanoborohydride. The number of conjugated lactose molecules per LSA was 40.7 +/- 12.3. To simplify the labeling procedure, we used a direct labeling method that adopts a high affinity 99mTc binding site concept in antibody labeling. The produced LSA was reduced by beta-mercaptoethanol to generate sulfhydryl groups and purified by PD-10 size-exclusion column. The number of generated sulfhydryl groups per LSA was 21.9 +/- 3.0. Medronate and stannous chloride were added to the reduced LSA and freeze-dried. Finally, 99mTc-pertechnetate (37 MBq, 1 mL) was added to the vial and incubated for 10 min at room temperature. The labeling efficiency of 99mTc-LSA was higher than 98%, and the stability in human serum at 37 degrees C for 24 h was over 90%. Biodistribution study using balb/c mice and imaging study using SD rats showed high initial liver uptake and slow increase in the intestine due to hepatobiliary excretion after metabolism in the hepatocytes. Negligible spleen uptake was found while 99mTc-tin colloid showed significant amount of spleen uptake due to reticuloendothelial uptake. In conclusion, an improved agent, 99mTc-LSA, for imaging asialoglycoprotein receptor of the liver was successfully developed which showed a simple labeling procedure, high labeling efficiency, high stability, and high initial liver uptake.     

Summary of recombinant human serum albumin development.

The methylotrophic yeast Pichia pastoris is well known as a host strain for the production of a variety of heterologous proteins. We have used P. pastoris for the production of recombinant human serum albumin (rHSA), for which we have developed efficient and specialized downstream processes. Results from structural analysis suggest that purified rHSA possesses an identical conformation to plasma derived human albumin (pdHA) and no difference from pdHA has been observed in neo-antigenicity. Host-cell-derived impurities (i.e. Pichia yeast component, DNA and mannan) have been evaluated in the purification process as well as in the drug substance and relevant specifications established. The efficacy and safety of rHSA have been tested in clinical studies and no difference from pdHA has been found in comparative study. Such studies have confirmed rHSA to have high efficacy with little or no adverse reaction.     

RNA-hydrolyzing activity of human serum albumin and its recombinant analogue

The comparative analysis of RNA-hydrolyzing activity of albumin from human serum and albumin expressed in methylotrophic yeast Pichia pastoris has been carried out. The rate of polyribonucleotide phosphodiester bond cleavage in the presence of recombinant albumin has been found to be similar to that of the reaction mediated by the native protein. According to ³¹P NMR data, RNA hydrolysis follows the mechanism of intermolecular trans-esterification to yield 2′,3′-cyclophosphodiester reaction products that are further slowly hydrolyzed to form nucleoside-3′- and nucleoside-2′-phosphates. Analysis of pH dependence suggests an acid–base mechanism of catalysis. The catalytic activity and substrate specificity of albumin in RNA hydrolysis distinguish it from human ribonucleases.     

Size of a human serum albumin molecule in solution]

The size of a human serum albumin molecule in aqueous solution containing 150 mM NaCl was studied using small-angle neutron scattering. The molecular radius of gyration was estimated to be 27.4 +/- 0.35 A. The compact sphere should have a smaller radius of gyration, whereas the popular human serum albumin model, a "cigar" 136 A long, should correspond to a greater radius of gyration. Possible shapes of the human serum albumin molecule which are in accordance with the results obtained, are the following: an extended ellipsoid less than 110 A of length or a nonsymmetrical oblate ellipsoid with a diameter of 85 A. The oblate ellipsoid might be close to the heart"-shaped structure of the crystalline human serum albumin molecule. The size of the albumin molecule does not change significantly as pH increases to 8.9. The possibility of the dynamic coexistence of various human serum albumin conformers in solution is discussed.     

The mitogenic activity of type III bacterial Ig binding proteins (protein G) for human peripheral blood lymphocytes is not related to their ability to react with human serum albumin or IgG

The mitogenic potential of bacterial IgG Fc binding proteins for human PBL is controversial. Wild type and recombinant type III IgG Fc binding proteins induce a wide spectrum of proliferative responses ranging from non-mitogenic to potent responses. To understand the reason for these differences, three recombinant forms of a type III IgG Fc binding protein derived from a single human group C streptococcal strain, 26RP66, were generated. Form I bound human IgG and human serum albumin, form II bound IgG alone and form III bound human serum albumin alone. These functionally distinct forms were compared with the corresponding wild type preparation from the same strain for mitogenic potential. A mitogenic response was induced only with the form I recombinant or the native wild type protein. These proteins shared the functional characteristics of binding human serum albumin and IgG. Mixtures of the IgG binding (form II) and human serum albumin binding fragments (form III) failed to reconstitute the mitogenic potential of the full length proteins. These results demonstrate that the type III IgG Fc binding protein has mitogenic potential for human PBL that is not related to its ability to react with human serum albumin or IgG.     

The extremely high level expression of human serum albumin in the milk of transgenic mice

The recombinant production of human serum albumin has been challenging due to the low unit price and huge amount needed, for the commercial production of rhSA at an economically feasible level, It will be well worth the effort to exploit new method for the extremely high level expression of rhSA. To this end, here a hybrid gene locus strategy was employed, a 37?Kb mWAP-hSA hybrid gene locus was constructed and used as mammary gland specific expression vector, in which the 3?Kb genomic coding sequence in the 24?Kb mouse whey acidic protein (mWAP) gene locus was substituted by the 16?Kb genomic coding sequence of human serum albumin (hSA), exactly from the start codon to the end codon. Corresponding transgenic mice were generated and rhSA was secreted into the milk at an extremely high level of 11.9?g/L. Our transgenic mice carrying the mWAP-hSA hybrid gene locus represent a model system for the cost-effective production of human serum albumin.     

Interaction of prostaglandin E2 with human serum albumin

Using equilibrium dialysis, protein fluorescence and fluorescent probing as well as chemical modification, the interaction of prostaglandin E2 with human serum albumin was studied. The serum albumin molecule has a highly specific prostaglandin E2-binding site. The binding of prostaglandin causes conformational rearrangements in the protein molecule. The amino group of serum albumin is involved in the interaction with prostaglandin E2. Prolonged exposure of prostaglandin E2 to serum albumin causes partial irreversible binding of prostaglandin molecules to the protein.     

Inhibition of protein tyrosine phosphatases by low-molecular-weight S-nitrosothiols and S-nitrosylated human serum albumin

The homogeneous recombinant mammalian protein tyrosine phosphatase 1B (PTP1B) and Yersinia protein tyrosine phosphatase (PTPase) are inactivated by a series of low-molecular-weight S-nitrosothiols. These compounds exhibited different inhibitory activities in a time- and concentration-dependent manner with second-order rate constants (k(inact)/K(I)) ranging from 37 to 113 M(-1) min(-1) against mammalian PTP1B and from 66 to 613 M(-1) min(-1) against Yersinia PTPase. Furthermore, the inactivation of Yersinia PTPase by S-nitrosylated protein:S-nitroso human serum albumin was investigated. Both single-S-nitrosylated and poly-S-nitrosylated human serum albumin show good inhibitory ability to Yersinia PTPase. The second-order rate constants are 472 and 1188 M(-1) min(-1), respectively. This result indicates a possibility that S-nitrosylated albumin in vivo may function as an inhibitor for a variety of cysteine-dependent enzymes.     

Interaction of anesthetic supplement thiopental with human serum albumin

Thiopental (TPL) is a commonly used barbiturate anesthetic. Its binding with human serum albumin (HSA) was studied to explore the anesthetic-induced protein dysfunction. The basic binding interaction was studied by UV-absorption and fluorescence spectroscopy. An increase in the binding affinity (K) and in the number of binding sites (n) with the increasing albumin concentration was observed. The interaction was conformation-dependent and the highest for the F isomer of HSA, which implicates its slow elimination. The mode of binding was characterized using various thermodynamic parameters. Domain II of HSA was found to possess a high affinity binding site for TPL. The effect of micro-metal ions on the binding affinity was also investigated. The molecular distance, r, between donor (HSA) and acceptor (TPL) was estimated by fluorescence resonance energy transfer (FRET). Correlation between the stability of the TPL-N and TPL-F complexes and drug distribution is discussed. The structural changes in the protein investigated by circular dichroism (CD) and Fourier transform infrared (FT-IR) spectroscopy reflect perturbation of the albumin molecule and provide an explanation for the heterogeneity of action of this anesthetic.     

Expression of human serum albumin in metylotrophic yeast Pichia pastoris and its structural and functional analysis

The stable strain of methylotrophic yeast Pichia pastoris secreting human serum albumin into cultural medium was obtained. Optimal conditions for expression of the protein were determined. We characterized the recombinant protein by mass spectrometry and circular dichroism and analyzed its catalytic activity.     

A high-capacity hydrophobic adsorbent for human serum albumin

A simple method, based on salting out hydrophobic interaction chromatography, for the efficient removal of trace amounts of serum albumin from partially purified protein preparations is described. The method is also successfully applied for the purification of albumin from Cohn fraction IV, a by-product obtained from the commercial fractionation of human serum proteins by the ethanol precipitation procedure. About 70% of the adsorbed albumin can be eluted by buffer of low ionic strength and can thus be lyophilized directly, if required. The adsorbent can be used for several cycles of adsorption and desorption without affecting its selectivity or capacity. Its adsorption properties and capacity for serum albumin are compared with those of the commercially available adsorbent Blue Sepharose CL-6B.     

Magnetic dye-affinity beads for human serum albumin purification

Cibacron Blue F3GA was covalently attached onto magnetic poly(vinyl alcohol) (mPVAL) beads (100-150 μm in diameter) for human serum albumin (HSA) adsorption from human plasma. Despite low nonspecific adsorption of HSA on mPVAL beads, Cibacron Blue F3GA attachment significantly increased the HSA adsorption. The maximum HSA adsorption was observed at pH 5.0. Higher HSA adsorption was observed from human plasma. Desorption of HSA from mPVAL beads was achieved by medium containing 1.0 M KSCN at pH 8.0. To test the efficiency of albumin adsorption from human serum, before and after albumin adsorption was demonstrated with sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) analyses. HSA molecules could be reversibly adsorbed and desorbed 10 times with the magnetic beads without noticeable loss in their HSA adsorption capacity.     

Tryptophan content of serum albumin

Tryptophan content of serum albumin was determined spectrophotometrically. The method used for the determination of tryptophan gave consistent results. Results show that the tryptophan contents of bovine and human serum albumin are significantly different from chicken serum albumin. Bovine and human serum albumins, however, are not significantly different from each other. A large difference in tryptophan content was found between two samples of chicken serum albumin. This suggests that the tryptophan content of serum albumin may not be constant for any given species. For these reasons, tryptophan content should not be used to estimate the molecular weight of serum albumin.     

Expression of human serum albumin in methylotrophic yeast Pichia pastoris and its structural and functional analysis

The stable strain of methylotrophic yeast Pichia pastoris secreting human serum albumin into cultural medium was obtained. Optimal conditions for expression of the protein were determined. We characterized the recombinant protein by mass spectrometry and circular dichroism and analyzed its catalytic activity.     

Bilirubin binding properties of pigeon serum albumin and its comparison with human serum albumin

Binding of bilirubin (BR) to pigeon serum albumin (PgSA) was studied by absorption, fluorescence and CD spectroscopy and results were compared with those obtained with human serum albumin (HSA). PgSA was found to be structurally similar to HSA as judged by near- and far-UV CD spectra. However, PgSA lacks tryptophan. Binding of BR to PgSA showed relatively weaker interaction compared to HSA in terms of binding affinity, induced red shift in the absorption spectrum of BR and CD spectral characteristics of BR-albumin complexes. Photoirradiation results of BR-albumin complexes also showed PgSA-bound BR more labile compared to HSA-bound BR.     

Large fragments of human serum albumin.

Large fragments of human serum albumin were produced by treatment of the native protein with pepsin at pH3.5. Published sequences of human albumin [Behrens, Spiekerman & Brown (1975) Fed. Proc. Fed. Am. Soc. Exp. Biol. 34, 591; Meloun, Moravek & Kostka (1975) FEBSLett.58, 134-137]were used to locate the fragments in the primary structure. The fragments support both the sequence and proposed disulphide-linkage pattern (Behrens et al., 1975). As the pH of a solution of albumin is lowered from pH4 to pH3.5, the protein undergoes a reversible conformational change known as the N-F transition. The distribution of large fragments of human albumin digested with pepsin in the above pH region was critically dependent on pH. It appeared that this distribution was dependent on the conformation of the protein at low pH, rather than the activity of pepsin. The yields of the large fragments produced by peptic digestion at different values of pH suggested that the C-terminal region of albumin unfolds or separates from the rest of the molecule during the N-F transition. The similarity of peptic fragments of human and bovine albumin produced under identical conditions supports the proposed similar tertiary structure of these molecules.     

Human serum albumin: microheterogeneity and hemin-binding ability]

Microheterogeneity of human serum albumin was studied by isoelectric focussing in ampholines and the borax-borate buffer-mannite system within the pH range 4.0--6.0 and by fractional precipitation in 3 M KCl. Albumin was found to have the same degree of heterogeneity in all separating systems. However, during focussing in ampholines microheterogeneity is partially due to the albumin binding with the ampholines. In other systems similar binding was not observed. Studies of the albumin-hemin complex showed that different fractions bind hemin in a different degree, maximal binding being observed within the pH range of 5.0--5.2. Defatting does not affect the distribution of hemin within the fractions of serum albumin.