Genetic variants of human albumin: structural characterization of allotypes used as references for electrophoretic classification

Eight different types of genetic variants of albumin are observed in the French population. The analysis of electrophoretic patterns of sera containing these variants, performed a three different pHs (8.6, 5.0 and 6.9) after addition of a reference protein (transferrin), allows the identification each variant by a quantitative estimation of its relative mobilities. The accuracy and reproducibility of the technique make it a useful reference method, commonly employed for studying European variants. The samples used as references for five genetic variant types, proalbumins Christchurch and Lille, albumins Vanves, B and Reading, were subjected to sequence analysis to determine the nature and localization of their structural change. Together with the mutations of albumins Gent and Roma previously described, the data presented here make available seven reference specimens for which the structural changes are characterized out of the eight variants known to exist in France.     

Human albumin variants. Nomenclature of allotypes observed in Europe and quantitative estimation of their relative mobilities

In order to clarify the actual nomenclature of the albumin allotypes in French and Italian populations we compared the samples collected in our laboratory to those kindly supplied by Dr. Porta (Italy) with reference to albumin variants classified in starch gel electrophoresis using three buffer systems by Weitkamp. The inherited human albumin variants can be classified on the basis of their relative mobilities on cellulose acetate electrophoresis compatively to human transferrin mobility. The relative mobility of each variant can be expressed by the following ratio: migration distance of the variant versus migration distance of the normal albumin where zero represents the transferrin mobility. Using three buffers system at pH 8.6, 5.0 and 6.9, it is possible to distinguish some albumin variants having a same mobility at alkaline pH and different mobilities at acidic pH. In the european area, eleven albumin variants are distinguishable on the basis of their relative mobilities at different pH: four Fast moving variants: Gent, Vanves (a new variant described here), Reading and CN/BL, and seven Slow moving variants: MI/MI Slow, GE/CT, SO/BS (or D type), Pollibauer, Gainesville, Roma and B type. Thirty-six sera from unrelated subjects with genetic bisalbuminemia were analyzed in our laboratory. Their distribution was as follows: B type (22), Pollibauer (9), SO/BS (2), Gainesville (1), Gent (1) and Vanves type (1). The frequency of bisalbuminemia was 0.35 per 1,000 in a population of 19,949 blood donors.     

Structure and properties of albumin Tokushima and its proteolytic processing by cathepsin B in vitro

Albumin Tokushima is a Japanese genetic variant of human serum albumin. Two homozygous and 6 heterozygous subjects with this variant were found in a family. Albumin Tokushima was purified from sera of the homozygous subjects. Its amino acid composition and amino-terminal sequence were determined and compared with those of a normal serum albumin. Albumin Tokushima with the amino-terminal sequence of Arg-Gly-Val-Phe-His-Arg-Asp-Ala-His-Lys-Ser-Glu-Val-Ala-His-Arg-Phe-Lys- Asp- Leu-Gly-Glu-Glu-Asn-Phe was found to be the same abnormal proalbumin as proalbumin Lille (Abdo, Y. et al. (1981) FEBS Lett. 131, 286-288). The isoelectric points of albumin Tokushima were pH 4.70 and 4.90 as compared with pH 5.05 and 5.25 of a normal serum albumin. Albumin Tokushima was converted to normal serum albumin by purified cathepsin B in vitro. Albumin Tokushima can bind Ni2+ at 4 degrees C but binds little at 37 degrees C.     

Ligand-binding properties of proalbumin Christchurch.

Proalbumin Christchurch, a circulating variant of human serum albumin, is secreted from the liver without cleavage of the hexapeptide situated at the N-terminal end of the peptide chain of proalbumin. We compared ligand-binding properties of proalbumin Christchurch and of normal albumin A from the same individual in order to test the effect of the presence of the hexapeptide. The two albumin forms exhibited similar affinities for palmitate, bilirubin, 8-anilinonaphthalene-1-sulphonate and Bromocresol Green. The patterns of endogenous fatty acids bound to the two forms of albumin were slightly different, although the differences were probably not of physiological significance. From these studies it would appear that the propeptide of proalbumin does not alter the protein conformation in such a way as to alter binding sites for organic anions.     

Calcium-dependent Golgi-vesicle fusion and cathepsin B in the conversion of proalbumin into albumin in rat liver.

1. An enzyme from rat liver that converts proalbumin into albumin is described. Partial purification, inhibitor studies and the conditions for maximum activity suggest that the enzyme is cathepsin B. 2. A membrane-bound enzyme, located mainly in lysosomes, also converts proalbumin into albumin. This appears to be a membrane-bound form of cathepsin B. 3. Isolated Golgi vesicles, incubated under conditions suitable for cathepsin B, convert endogenous proalbumin into albumin. 4. This conversion in Golgi vesicles has an absolute requirement for Ca2+ at micromolar concentrations. Mg2+ does not affect or substitute for Ca2+. Both the proalbumin and the albumin formed from it are intravesicular. 5. Converting activity is enhanced by pretreatment with the known chemical fusogen, poly(ethyleneglycol). 6. Vesicles preincubated at pH above 7 in the presence of dithiothreitol show a marked fall in converting activity. This can be partially restored by incubation with native vesicles. These results suggest that vesicle fusion is a requirement for conversion of proalbumin into albumin.     

Structural characterization, stability and fatty acid-binding properties of two French genetic variants of human serum albumin

Four bisalbuminemic, unrelated persons were found in Bretagne, France, and their variant and normal albumins were isolated by DEAE ion exchange chromatography, reduced, carboxymethylated and treated with CNBr. Comparative two-dimensional electrophoresis of the CNBr digests showed that three of the variants were modified in fragment CB4, whereas the fourth had an abnormal fragment CB1. These fragments were isolated, digested with trypsin and mapped by reverse-phase HPLC. Sequencing of altered tryptic peptides showed that the three variants modified in CB4 were caused by the same, previously unreported, amino acid substitution: Asp314-->Val (albumin Brest). The fourth, however, was a proalbumin variant with the change Arg-2-->Cys (albumin Ildut). Both amino acid substitutions can be explained by point mutations in the structural gene: GAT-->GTT (albumin Brest) and CGT-->TGT (albumin Ildut). The proalbumin Ildut is very unstable and already in vivo it is to a large extent cleaved posttranslationally to Arg-Albumin and normal albumin. Furthermore, we observed that during a lengthy isolation procedure the remaining proalbumin was changed to Arg-Albumin or proalbumin lacking Arg-6. In addition, part of normal albumin had lost Asp1. Gas chromatographic investigations using isolated proteins indicated that albumin Brest has improved in vivo fatty acid-binding properties, whereas the structural modification(s) of albumin Ildut does not affect fatty acid binding.     

Identification of a calcium-dependent microsomal proteinase responsible for monobasic cleavage of chicken proalbumin

The location and nature of the endoproteolytic activity involved in processing of proproteins has been studied in chicken liver microsomes. A membrane-bound, calcium-dependent proteinase was found to cleave chicken proalbumin with a monobasic cleavage site approx. 10-times faster than human proalbumin, which has a dibasic cleavage site. The mutant (human) proalbumin Christchurch (Arg(-1)----Gln), with a potential monobasic site, was not processed. The enzyme, which had a pH optimum of between 5.0 and 7.0, was not inhibited by serine or aspartyl proteinase inhibitors but was affected by some inhibitors of cysteine proteinases. The convertase was specifically inhibited by the reactive centre variant alpha 1-antitrypsin Pittsburgh, but not by normal alpha 1-antitrypsin.     

Proalbumin to albumin conversion by a proinsulin processing endopeptidase of insulin secretory granules

A lysate of purified insulin secretory granules, which contains two types of proinsulin processing activity (type 1, Arg-Arg-directed and type II, Lys-Arg-directed (Davidson, H.W., Rhodes, C.J., and Hutton, J. C. (1988) Nature 333, 93-96), was found to process proalbumin by specific proteolytic cleavage of the COOH-terminal side of the Arg-2-Arg-1 sequence. The subcellular distribution of proalbumin processing activity in insulinoma tissue paralleled that for proinsulin conversion and occurred principally in a secretory granule fraction. Cleavage appeared to result from the Arg-Arg-directed type 1 proinsulin processing endo-peptidase. It was Ca2+-dependent (K0.5 activation = 1.0-1.5 mM Ca2+), unaffected by group-specific inhibitors of serine, cysteinyl, or aspartyl proteinases, and had an acidic pH optimum (5.5). Active-site inhibitor studies showed this activity had a preference for dibasic over monobasic amino acid sequences and indicated that the sequence of the dibasic site was an important determinant of the susceptibility of the substrate to cleavage. The activity did not process the proalbumin Christchurch mutant (Arg-2-Arg-1 to Arg-2-Gln-1). It was inhibited by the variant alpha 1-antitrypsin Pittsburgh (Met358 to Arg358; K0.5 = 100 nM) but not by other related proteins normally co-secreted with albumin from hepatocytes, namely alpha 1-antitrypsin M, alpha 2-macroglobulin, or antithrombin III. The insulin secretory granule proalbumin processing activity was indistinguishable from a proalbumin endopeptidase reported in rat liver membranes and similar to the yeast KEX-2 protease. These findings suggest that a highly conserved set of proprotein endopeptidases exists, which are specific for a dibasic sequence but broadly specific for proprotein substrates. Such enzymic activities appear to be active within both the constitutive and regulated pathways of secretion. Intraorganellar Ca2+ and pH appear to play a key role in regulating their activities.     

Bovine microsomal albumin: Amino terminal sequence of bovine proalbumin

Bovine liver microsomes contain an albumin having an apparent isoelectric point approximately 0.3 pH unit in excess of bovine serum albumin. Sequence analysis of the purified protein shows that the first ten residues at the amino terminus are: $\text{Arg-Gly-Val-Phe-Arg-Arg}$-Asp-Thr-His-Lys. The data suggest that the hexapeptide (underlined), identical to that found in proalbumin from rat liver, is attached to the amino terminus of bovine serum albumin (the last four residues). By analogy with the rat liver system, this protein therefore is bovine proalbumin, a precursor of bovine serum albumin.     

Putative convertase involved in the proteolytic conversion of rat proalbumin to serum albumin

We have found a proteolytic activity in Golgi membranes which efficiently converts [35S]methionine-labeled proalbumin, isolated from pulse-labeled rat hepatocytes in culture, to serum albumin in an in vitro assay system. The proalbumin-converting activity was dependent on Ca2+ and the maximum activity was observed at pH 5.5-6.0. Since the enzyme activity was found to be resistant not only to both leupeptin and E-64 but also to thiol-blocking reagents, it is unlikely that cathepsin B is involved in the proteolytic conversion of proalbumin occurring in the Golgi complex.     

Evidence for the coupling of biosynthesis and secretion of serum albumin in the rat. The effect of colchicine on albumin production

1. By using isotopic-dilution techniques it was found that colchicine causes a slight increase in the proalbumin content of liver, from 0.63+/-0.06 to 0.83+/-0.10mg/g of liver, but has no effect on albumin content (0.50+/-0.05mg/g of liver). All the proalbumin and 67% of the albumin is found in vesicles from which they are liberated by detergents. 2. Colchicine inhibits secretion of albumin, decreases the rate of conversion of proalbumin into albumin and decreases the rate of incorporation of l-[1-(14)C]leucine into proalbumin. 3. Balance studies in vivo show that all the (14)C appearing in serum albumin can be accounted for by the flow of (14)C through the proalbumin, in the presence or absence of colchicine. 4. When cycloheximide is given to the rats, 2min after [(14)C]leucine, further synthesis of protein stops. The label in proalbumin disappears and the proalbumin content of the liver falls, so as to account for the albumin appearing in the plasma. This occurs both in the presence and in the absence of colchicine. By contrast, there is little change in liver albumin. Studies with isolated perfused livers are in agreement with the above. Lumicolchicine has no effect on any of these systems at doses at which colchicine exerts its action. 5. These results suggest that biosynthesis and conversion of proalbumin into albumin, and secretion of serum albumin are controlled at each step.     

Two distinct chloride ion requirements in the constitutive protein secretory pathway

The role of chloride ions in regulated secretion is well described but remains poorly characterised in the constitutive system. In the liver, newly synthesised proalbumin is transported to the trans Golgi network where it is converted to albumin by a furin protease and then immediately secreted. We used this acid-dependent hydrolysis and the measurement of specific protein secretion rates to examine the H+ and Cl- ion dependence of albumin synthesis and secretion, a major constitutive protein secretory event in all mammals. Using permeabilised primary rat hepatocytes we show that ordinarily chloride ions are essential for the processing of proalbumin to albumin. However Cl- is not required for transport which continues but releases solely proalbumin. Prior treatment of the cells with Tris (used as a membrane-permeable weak base to neutralise Golgi luminal pH) both eliminated the formation of albumin and very greatly reduced secretion. After washing out Tris, both authentic secretion and processing could be restarted if Cl-, ATP, GTP, cAMP, Ca2+ and cytosolic proteins were added. Hence a requirement for chloride in transport, in addition to processing, can be uncovered by first neutralising pH gradients. Furthermore, the chloride channel blocker DIDS (4,4-diisothiocyanostilbene 2,2-disulphonic acid) reversibly inhibited the constitutive secretory pathway. However, the total mass of proalbumin detectable in DIDS-treated cells fell to 36% of control while the fraction processed to albumin remained almost constant. This clearly dissociates a large part of the Cl- requirement of the constitutive protein secretory pathway from the function of known liver Golgi Cl- channels.     

Selective processing of proalbumin determined by site-specific mutagenesis

Rat proalbumin is cleaved at the dibasic pair Arg-Arg and converted into a mature form with Glu at the NH2 terminus. In the present study site-directed mutagenesis of the albumin cDNA was designed to generate proalbumin variants in which Glu1 was substituted with various amino acid residues. The expression plasmids constructed were transfected into COS-1 cells, and the intracellular processing of proalbumins expressed was examined by labeling experiments. Substitution of Glu1----Ser allowed the expressed proalbumin to be processed as observed for the wild-type precursor. However, replacement of Glu1 with a hydrophobic residue (Val, Leu or Ile) resulted in no processing of proalbumin, despite retaining the same cleavage signal Arg-Arg as above. The results indicate that the residue at position 1 adjacent to the dibasic pair is also important for recognition by the proalbumin-processing enzyme.     

A new proalbumin variant: albumin Jaffna (-1 Arg----Leu)

Albumin Jaffna is an electrophoretically slowly moving genetic variant of human serum albumin found in two members of a Tamil family from Jaffna (Northern Sri Lanka), both heterozygous for the abnormal protein. Sequential analysis of albumin Jaffna, purified from serum by ion exchange chromatography on DEAE Sephadex and Mono Q columns, revealed that this variant is a new abnormal proalbumin, arising from a -1 Arg----Leu substitution, which prevents the proteolytic removal of the N-terminal hexapeptide and allows the mutated proalbumin to enter the circulation. The presence of two additional positive charges is in keeping with the decreased electrophoretic mobility of albumin Jaffna, as well as with its isoelectric point of 5.01, determined by chromatofocusing on a Mono P column. The variant is selectively cleaved by trypsin in vitro, leaving leucin -1 as N-terminal residue.     

Novel human proalbumin variant with intact dibasic sequence facilitates identification of its converting enzyme

We describe here the identification of a new genetic variant of human proalbumin with an N-terminal sequence of Arg-Gly-Val-Phe-Arg-Arg-Val-Ala-His-Lys-. Proalbumin Blenheim (10%) and mature albumin Blenheim (38%) with an initial sequence of Val-Ala-His-Lys-make up nearly half the serum albumin in affected individuals. Despite retaining an intact dibasic processing site, proalbumin Blenheim (1 Asp----Val) enters the circulation unprocessed. The observed ratio of proalbumin to albumin can be accounted for by proteolysis in the periphery. Employed as a potential substrate, proalbumin Blenheim provides a unique means of identifying the physiologically relevant proalbumin convertase. In vitro studies showed that the variant is readily cleaved by trypsin. However, it is not cleaved by the proposed proalbumin convertase, a membrane-bound Ca2+-dependent proteinase prepared from rat liver Golgi vesicles, which gives authentic cleavage of normal human proalbumin.     

Synthesis of alpha-fetoprotein and albumin by fetal mouse liver cultured in chemically defined medium

Fetal mouse liver synthesizes two major secretory proteins: α-fetoprotein and albumin. The relative proportions of these proteins change during development. Fetal mouse liver secretes predominantly α-fetoprotein, and the neonatal mouse liver secretes predominantly albumin. α-Fetoprotein and albumin synthesis were studied in vitro using an organ culture system and a chemically defined medium (BGJ_b). This medium does not support hepatocellular replication but maintains protein synthesis at high levels for prolonged periods of culture. Patterns of protein synthesis were analyzed as functions of gestational age and time in culture. The ratio of α-fetoprotein/albumin decreases from 1.4 on gestational Day 14 to 0.4 on gestational Day 18. However, when liver cultures were begun on any given gestational day, the ratio of α-fetoprotein/albumin remains constant for as long as 8 days in culture. Thus, developmental changes in α-fetoprotein and albumin synthesis are arrested under the conditions of this culture system. Fetal mouse liver secretes two electrophoretically distinguishable forms of albumin. One form is similar in mobility to albumin from adult mouse serum; the other more electropositive species is similar in mobility to proalbumin isolated from adult mouse liver microsomes and can be converted to albumin by mild trypsin treatment.     

Genetic polymorphism of plasma vitamin D-binding protein (GC) in Australian goats

Polymorphism at the GC locus in goats was detected using isoelectric focusing (pH 4.5-5.4) and immunoblotting with antiserum to human GC. Three variants, designated A, B and C in order of decreasing mobility to the anode, were detected and were shown to be controlled by three codominant alleles, GCA, GCB and GCC. GCA and GCB occurred in all four breeds (Australian and Texan Angora, Cashmere and Dairy) with GCA being the most common and having gene frequencies ranging from 0.851 to 0.993. GCC was found only in Australian Angora and Cashmere animals. The products of the three GC alleles had isoelectric points in the range pH 4.63-4.95 and M(r) of approximately 54,375. The major isoforms of the three alleles were shown to contain sialic acid. Linkage between the GC and albumin loci was unable to be demonstrated due to the low frequency of ALBA (0.02) in the Cashmere breed.     

In vivo effect of colchicine on hepatic protein synthesis and on the conversion of proalbumin to serum albumin

Treatment of rats with 0.5-25 mumol/100 g body weight of colchicine for 1 h or more caused an inhibition of hepatic protein synthesis. This effect was not seen if animals were exposed to colchicine for less than 1 h. The delayed inhibition of protein synthesis affected both secretory and nonsecretory proteins. Treatment with colchicine (15 mumol/100 g) for 1 h or more caused the RNA content of membrane-bound polysomes to fall but did not change the polysomal profile of this fraction. By contrast, the total RNA content in the free polysome cell fraction was increased, and this was due to the presence of more ribosomal monomers and dimers. Electron microscope examination of the livers from rats treated for 3 h with colchicine showed an accumulation of secretory vesicles within the hepatocytes and a general distention of the endoplasmic reticulum. Administration of radioactive L-leucine to the rats led to an incorporation of radioactivity into two forms of intracellular albumin which were precipitable with antiserum to rat serum albumin but which were separable by diethylaminoethyl-cellulose chromatography. One form has arginine at the amino-terminal position and is proalbumin, and the other form, which more closely resembles serum albumin chromatographically, has glutamic acid at its amino terminus. Only proalbumin was found in rough and smooth endoplasmic reticulum fractions and in a Golgi cell fraction wich corresponds morphologically to mostly empty and partially filled secretory vesicles. However, in other Golgi cell fractions which were filled with secretory products, both radioactive proalbumin and serum albumin were found. This indicates that proalbumin is converted to serum albumin in these secretory vesicles just before exocytosis. Colchicine delayed the discharge of radioactive albumin from these filled secretory vesicles and caused an accumulation of both proalbumin and serum albumin within these cell fractions.     

Weakly basic amines inhibit the proteolytic conversion of proalbumin to serum albumin in cultured rat hepatocytes

Effects of weak amines on the proteolytic conversion of proalbumin to serum albumin were studied in primary culture of rat hepatocytes. In control culture proalbumin was converted to serum albumin before discharge into the medium. However, in the presence of chloroquine the conversion to serum albumin was inhibited and proalbumin per se was released into medium. A similar inhibition of the processing was also observed in the presence of other amines such as methylamine and NH4Cl. Thus weak amines mimic the carboxylic ionophore monensin with regard to the effect on proalbumin conversion [Oda & Ikehara (1982) Biochem. Biophys. Res. Commun. 105, 766-772]. Since proteolytic conversion of proalbumin is believed to occur at the Golgi complex, these results suggest that weakly basic amines perturb the Golgi complex in addition to lysosomes and endosomes.