Polyacrylic polyhydrazides as novel reagents for detection of antibodies in immunoblotting assays.

By the glycoprotein specific staining method introduced recently (Heimgartner et al., 1989, Anal. Biochem. 181, 182-189) it is also possible to detect an antibody bound to its antigen on a membrane. The antibody is oxidized by periodate prior to incubation. Next, a polyacrylic polyhydrazide is coupled to the aldehyde groups generated in the sugar part of the antibody molecule. A periodate oxidized glycoenzyme, such as horseradish peroxidase, is then coupled to the remaining hydrazide groups of the polymer and incubation with a suitable enzyme substrate visualizes the glycoenzyme-polyhydrazide-antibody-antigen complex. The sensitivity of the detection is most critically dependent on the antibody class and the polyhydrazide reagent. Oxidation conditions are less important and the antibody does not need to be purified prior to periodate oxidation. Under standard conditions, the sensitivity obtained with IgG type antibodies was about ten times lower than with peroxidase-labeled secondary antibodies. However, a similar if not higher sensitivity can be expected with more glycosylated antibodies, such as IgM or IgE, or with chicken antibodies. This approach is advantageous because antibodies of all classes and from all species can be detected with the same reagent, the polyhydrazide, no foreign molecule is introduced in the antibody before its binding to the antigen and no conjugate needs to be prepared in advance.

Picogram cloning and direct in situ sequencing of DNA from gel pieces.

We describe a simple and rapid procedure for cloning and sequencing of DNA fragments separated by gel electrophoresis, using novel hydrophilic gels, Clearose BG, Spreadex, and Poly(NAT), that do not melt at 95 degrees C. For cloning, a band of interest is excised precisely and incubated in an extraction buffer containing 5-10 mM MgCl2 at 70 degrees C for 15-45 min. The eluted DNA is added directly to the plasmid solution. Using a topoisomerase-based ligation system, we were able to transform bacteria with a few picograms of DNA and isolate recombinant clones. For in situ sequencing, the DNA in the gel serves as the template. No treatment before cycle sequencing is necessary for fragments up to 500 bp.

Purification of endo-N-acetyl-beta-D-glucosaminidase H by substrate-affinity chromatography.

Endo-N-acetyl-beta-D-glucosaminidase H (Endo H) was purified to homogeneity (3000-fold) from a culture filtrate to Streptomyces plicatus. The key step was substrate-affinity chromatography, which afforded a 1000-fold purification and yielded a protease- and exoglycosidase-free preparation of Endo H. Proteins from the crude sample were applied to the substrate-affinity column, consisting of yeast-invertase glycopeptides bound to Sepharose-immobilized concanavalin A. After washing off the unbound proteins, Endo H was quantitatively eluted by methyl alpha-D-mannopyranoside. Various conditions were tested to achieve an optimal binding of Endo H to this substrate-affinity gel. After substrate-affinity chromatography, Endo H was separated from the coeluted gylcopeptide substrate and some protein impurities by gel filtration and hydrophobic chromatography.

Preparation of the stabilized glycoenzymes by cross-linking their carbohydrate chains.

Each of the three high-mannose type glycoproteins studied, acid phosphatase, invertase, and glucose oxidase, could be specifically cross-linked through its carbohydrate chains. The procedure involves periodate oxidation of carbohydrate residues followed by reaction of the generated aldehyde groups with adipic acid dihydrazide as a cross-linker. The amount and size as well as solubility of the formed polymers could be efficiently controlled by varying the reaction conditions, i.e., the oxidation degree and the concentrations of glycoproteins, cross-linker, and hydrogen ions during the cross-linking reaction. It was found that the quantity and size of polymers increased with oxidation degree and protein concentration and by lowering the pH. When the protein concentration was above and pH below certain values, depending on the glycoenzyme, insoluble polymers formed. The soluble cross-linked polymers retained a high level of original activity, and the minor decrease in specific activity noticed was shown to occur during the periodate oxidation step. The cross-linked glycoenzymes are much more resistant to denaturation by high temperature and by changes in pH, demonstrating the usefulness of this method in preparation of the stabilized glycoprotein derivatives.

Looking at bands from another side.

In high-percentage gels run in the submerged electrophoresis mode separated DNA bands were bent, regardless of whether the gel was prepared from polyacrylamide, poly-N-acryloyl-tris(hydroxymethyl)amino-methane, or hydroxyethylated agarose. This declination from the vertical axis is the major reason for the appearance of diffuse bands after recording by a camera positioned vertically above the gel. Five major factors which could cause the bending, including an inhomogeneous electric field, electroendosmosis, nonuniform heat dissipation, different effective porosity across the gel thickness, and different conductivity in the gel and electrophoresis buffer, were investigated. The results indicate that the major cause is the difference in the ratio of current density and specific conductivity between the gel and the running buffer. When that ratio was adjusted by empirically optimizing gel ionic composition, the bending could be prevented or greatly diminished. Other results described in this work suggest that electrophoretic systems in which gels are polymerized and run in the same buffer, frequently referred to as "continuous buffer systems," may not provide a continuous, constant electric field. Furthermore, the pH change regularly observed in circulated running buffers could be related to difference in the amount of current transported by buffer anions and cations. A buffer in which current was equally transported by anions and cations showed no detectable pH change after prolonged electrophoresis.

A model of gel electrophoresis.

Electrophoretic migration of macromolecules through gels is described in terms of dislocation of the gel polymers by migrating molecules. The polymer displacement depends on two forces. The first is the electrokinetic force of the migrating macromolecules and the second is the resisting force of the polymers. The molecules migrate in discrete steps, and in each step they pass through one gel layer. "Doors" are formed by a dislocation of the polymer chains in one layer whereas "corridors" are formed by a deformation of the layer, accompanied by dislocation of the polymers in at least one layer above and below. The model is supported by experiments showing a loss of resolution in the gels comprising a branched, polymeric cross-linker, and a faster migration of a 23 kbp than a 9.4 kbp DNA fragment in two completely different matrices, including a novel cross-linked agarose gel.

On the "door-corridor" model of gel electrophoresis. I. Equations describing the relationship between mobility and size of DNA fragments and protein-SDS complexes.

In a gel, electrophoretic mobility of a DNA fragment and a protein-SDS complex is an exponential function of the ratio between two forces, the resisting force of the gel polymers and the electrokinetic force of the migrating macromolecule. It is also proportional to the mobility of unit size of the migrating molecule, mu 1. Each gel gives an optimal resolution of macromolecules of a defined size. That size must be such that the electrokinetic force of the migrating molecule equals the gel resistance force, so that the ratio of mobility and mu 1 becomes e-1. The optima for several poly[N-acryloyl-tris(hydroxymethyl)aminomethane], polyacrylamide and agarose gels are given. Only two constants are sufficient for description of the relationship between mobility and size of DNA fragments and protein-SDS complexes of the sizes close to that optimally resolved. One is mu 1 and the other, ks, is the slope of the straight line which can be obtained from the size versus reciprocal of the mobility plot. The experimentally observed sigmoidal curves in the plots including the logarithm of size follow directly from the developed equations. The use of mu 1 instead of mu 0 may provide a better interpretation of the Ferguson plots.

On the "door-corridor" model of gel electrophoresis. II. Developments related to new gels, capillary gel electrophoresis and gel chromatography.

An excellent separation of small DNA fragments is possible also at a low total polymer concentration. This is illustrated with novel gels containing 1% of the cross-linked agarose polymers. In gel-filled capillaries, preelectrophoresis in the direction opposite to that of separation may create gel discontinuities capable not only of separating, but also of concentrating the migrating molecules. Such capillaries are able to provide an efficiency in the range of 100 x 10(6) theoretical plates per meter. It is proposed that electrophoretic separations in gel-filled capillaries have to be performed at high field strengths in order to achieve the optimal ratio between gel resistance and electrokinetic forces. When considered in terms of the forces exerted on the gel polymers by the migrating macromolecules, gel filtration and gel electrophoresis appear as two profoundly different techniques.

On the "door-corridor" model of gel electrophoresis. III. The gel constant and resistance, and the net charge, friction, diffusion and electrokinetic force of the migrating molecules.

The door-corridor model of gel electrophoresis enabled an estimation of the net charge of DNA molecules run in various gels. When the runs were carried out in Trisacetate-EDTA buffer having a concentration from 10 to 120 mM, the net charge in 1% agarose gel varied from 1.1 to 0.58 e per base pair. The friction between migrating molecules and gel fibers was dependent on the gel type and concentration, as well as the electric field strength and temperature during electrophoresis. In the 123 to 1,474 bp size range, the friction in 1% agarose changed from 1.91 to 378.03 x 10(-10) N.m-1.s. It was found that the friction per 123 bp DNA segment is not constant, but raises with size. The gel resistance force increases at higher electric field strengths, indicating that elastic forces govern the migration of macromolecules through gels. In the gels studied, the friction, and therefore thermal diffusion, of DNA and protein-SDS complexes scale with from 2.20 to 2.32 power of size. The ratio of thermally induced diffusion and velocity in various gels shows that there is a profound reduction of diffusion compared to velocity with increasing DNA size. This is directly linked to the high exponent relating friction and size. The high resolving power of gel electrophoresis can be correlated to the difference between the frictional coefficients of a diffusing and migrating macromolecule.

An apparatus for submerged gel electrophoresis.

A novel apparatus for submerged gel electrophoresis is described in detail. It includes an upper buffer compartment, a lower buffer compartment, and a horizontal plate between the two compartments. The horizontal plate is a heat exchanger connected to an external heater/cooler. Buffer circulates between the two compartments through openings in the horizontal plate. In the upper compartment two separated openings are positioned on each side of the horizontal plate between the side walls and long vertical barriers. The barriers initially direct the flow of buffer and define the electric field on the sides of the upper compartment. The electric field is confined essentially into a rectangular box, defined on the ends by the end walls, on the sides by the barriers, on the bottom by the cooling plate, and on the top by the air. Since the volume of buffer is smaller in the electrode compartment than in the reservoir under the cooling plate, this design enables formation of a substantially uniform electric field without creating too high a current. To enhance uniformity of the electric field, anode and cathode consist each of two platinum wires positioned one above the other at a distance of 6 mm. The electrodes can be placed parallel to the sides and perpendicular to the buffer flow or parallel to the ends and the flow of buffer. The stream of buffer in the upper compartment is regulated by two dams, perpendicular to the long barriers, on each end of the horizontal plate.(ABSTRACT TRUNCATED AT 250 WORDS)

Reversible and irreversible cross-linking of immunoglobulin heavy chains through their carbohydrate residues.

After periodate oxidation and incubation with a dihydrazide, cross-linking of the two heavy chains of immunoglobulins G from several species proceeds specifically through their oligosaccharides. We have used malonic acid dihydrazide, adipic acid dihydrazide and dithiodipropionic acid dihydrazide. The last compound is introduced in this work as a cleavable-carbohydrate-specific cross-linker. It was found that in rabbit and human immunoglobulins the degree of cross-linking was strongly dependent on the oxidation conditions but only very weakly dependent on the concentration and size of the dihydrazides. Papain cleavage of the cross-linked rabbit IgG indicated that the cross-linking occurred predominantly, if not exclusively, in the Fc region, probably through the two glycans linked to Asn-297 in the CH2 domain of each of the two heavy chains. The immunoglobulins from sheep, pig, goat and guinea pig show a comparable cross-linking pattern, indicating that the sugar chains from these immunoglobulins have a spatial structure closely related to that of rabbit and human IgG. When dithiodipropionic acid dihydrazide was used as the cross-linker, the cross-link could be cleaved by mercaptoethanol.

Electrophoresis of DNA restriction fragments in poly-N-acryloyl-tris gels.

Poly-N-acryloyl-tris(hydroxymethyl)aminomethane (NAT) gels were evaluated as a matrix for DNA electrophoresis. The resolution of DNA restriction fragments in three poly(NAT)-N,N'-methylenebisacrylamide (Bis) gels (4, 5, and 6%) was compared with the resolution in polyacrylamide (AA)-Bis gels of the same percentage. Poly(NAT) gels were found to give a substantially improved separation of DNA fragments larger than 200 bp. In contrast to poly(AA) gels, DNA fragments of up to 4 kbp were well resolved in the new matrix. By pulse-field electrophoresis the useful separation range of poly(NAT) gels was expanded to at least 23 kbp. For DNA fragments below 10 kbp, the resolution was better than that in a 0.7% agarose gel. Thus poly(NAT) gels are most suitable for the electrophoretic separation of DNA molecules whose size is out of the optimal fractionation range of poly(AA) or agarose gels.

Biosynthesis of soluble carnitine acetyltransferases from the yeast Candida tropicalis.

Soluble carnitine acetyltransferase from Candida tropicalis is synthesized as a 76 kDa precursor, which is monomeric and possesses no or very little carnitine acetyltransferase activity. Maturation of the enzyme begins with proteolytic processing of the 76 kDa precursor to 64 and 57 kDa subunits. The processed subunits subsequently associate into two kinds of active oligomers; the 57 kDa subunits are assembled into a tetramer and the 64 kDa subunits into an octamer. Formation of these oligomers depends apparently on growth conditions, since both oligomers were present in cells grown in continuous culture, but cells grown batchwise contained only the tetrameric form of carnitine acetyltransferase.

Polyacrylic polyhydrazides as reagents for detection of glycoproteins.

Glycoproteins immobilized on membranes can be detected with high selectivity and sensitivity by the four-step procedure described in this work. The glycoproteins are first oxidized by sodium periodate and then polyacrylic polyhydrazides are coupled to the aldehyde groups generated in the sugar part of the glycoproteins. In the third step, a glycoenzyme, such as horseradish peroxidase, is coupled to the remaining hydrazide groups on the polymer through the aldehydes formed in its glycan chains. In the last step, the visualization of glycoproteins is achieved through the reaction product of the bound glycoenzyme. The sensitivity of the glycoprotein detection is most critically dependent on the hydrazide reagent. Thus, dihydrazides were not satisfactory, a trihydrazide was better, and polyhydrazides were the best. Two different polyhydrazides were used. One was based on acrylamide and the other on N-acryloyl-tris(hydroxymethyl)aminomethane. The second one proved to be superior because it gave higher sensitivity with no detectable background staining. We have also investigated the influence of various reaction conditions on staining of glycoproteins having oligomannose and N-acetyllactosamine type glycan chains. Some of them, invertase and fetuin, could be detected with sensitivity similar to that of silver staining in gels and colloidal gold staining on the membranes. The detection of small quantities of Endo H-deglycosylated glycoproteins was possible under standard conditions only if several N-acetylglucosamine residues remained bound to the protein.

Poly-N-acryloyl-Tris gels as anticonvection media for electrophoresis and isoelectric focusing.

We describe in detail the synthesis of an acrylic monomer, N-acryloyl-tris(hydroxy-methyl)aminomethane (NAT), which was successfully used for the preparation of gels for electrophoresis and isoelectric focusing. The polymerization kinetics and transparency of the poly(NAT) gels crosslinked by N,N'-methylenebisacrylamide (Bis) are also shown. Poly(NAT)-Bis gradient (4-24%) gel resolves proteins according to their size. The exclusion limit of this gel is slightly over 3 X 10(6), which is more than threefold higher than the exclusion limit of the polyacrylamide gradient gel of the same concentration. The gel made of 6% NAT and 3% Bis represents a suitable matrix for isoelectric focusing. These results demonstrate that poly(NAT)-Bis gels could be advantageously used in those applications where the extensive sieving by the polyacrylamide matrix is not desir desirable.

Study of the carbohydrate part of yeast acid phosphatase.

It has been found that the carbohydrate part of acid phosphatase from yeast Saccharomyces cerevisiae consists of 16 N-glycosidically linked carbohydrate chains containing from 14 to about 150 mannose units. The presence of very small amounts of O-glycosidically linked chains was indicated. Acetolysis studies pointed to a high similarity in the structure of acid phosphatase and mannan carbohydrate chains. A new method is described for cross-linking of acid phosphatase specifically via carbohydrate chains. The possibility to cross-link the enzyme subunits intramolecularly is in accordance with the suggestion that carbohydrate chains play a role in subunit associations.

Physicochemical and kinetic properties of acid phosphatase from Saccharomyces cerevisiae.

Acid phosphatase from yeast Saccharomyces cerevisiae was purified, and its physicochemical and kinetic properties were investigated. The sedimentation coefficient has been determined to be s0(20,w) = 13.6 S. The diffusion constant has been found to be 3.9 X 10(-7) cm2s-1, and the calculated partial specific volume was v = 0.663 cm3/g. From these data, a molecular weight of 252,000 was calculated. Electrophoresis on gel slabs, with a linear concentration gradient of polyacrylamide (4-30%), showed size heterogeneity of the native enzyme preparation and indicated an apparent molecular weight in the range of 170,000 to 360,000. In the presence of sodium dodecyl sulfate, the molecular weight was in the range of 82,000 to 165,000, indicating dimeric structure of the native enzyme, which was confirmed by cross-linking experiments. Isoelectric focusing demonstrated charge heterogeneity of enzyme preparation. From CD spectrum it was calculated that the enzyme contains about 29% of alpha-helical structure. Excitation at 278 nm gave an emission fluorescence spectrum with a maximum at 340 nm. Amino acid analysis revealed a high content of aspartic acid, serine, and threonine. Glycine is found as the NH2-terminal amino acid. Initial velocity dependence on substrate concentration, as well as on pH, and thermostability studies indicated the presence of at least two enzyme forms in the preparation.

Characterization of a soluble carnitine acetyltransferase from Candida tropicalis.

Carnitine acetyltransferase was purified from the cytoplasmic fraction of Candida tropicalis grown on alkanes in continuous culture. By ion-exchange chromatography the enzyme was resolved in two fractions with the same specific activity of 80 U/mg. The molecular mass of both enzyme forms, determined by non-denaturing gradient gel electrophoresis, was 540 kDa. After SDS electrophoresis only one band of 64 kDa was detected indicating that both enzymes are oligomers each containing eight subunits. Isoelectric focusing in agarose under non-denaturing conditions demonstrated the presence of at least four different charged species in the pH range between 5.6 and 6.7. After isoelectric focusing in 9 M urea/1% Nonidet P-40 gels, both enzyme forms were resolved into four bands. Peptide mapping, performed by cyanogen bromide cleavage of polypeptides separated by denaturing isoelectric focusing followed by second-dimension SDS electrophoresis, revealed a very high degree of homology between these polypeptides. The presence of the octameric form of carnitine acetyltransferase already in the starting material was demonstrated by non-denaturing gradient gel electrophoresis and immunoblotting. Antibodies against carnitine acetyltransferase from C. tropicalis ATCC 32113 formed precipitation lines with extracts from several Candida species but not with extracts of Candida utilis, Candida ethanothermophilum and an another strain of C. tropicalis.

Homology among multiple extracellular peroxidases from Phanerochaete chrysosporium.

The extracellular peroxidases of Phanerochaete chrysosporium were separated into 21 proteins by analytical isoelectric focusing. Fifteen of these enzymes oxidized veratryl alcohol (lignin peroxidases) in the presence of H2O2. Six enzymes were Mn(II)-dependent peroxidases. The Mn(II)-dependent enzymes appeared and reached their maximal activity earlier than the lignin peroxidases in the cultures. Peptide mapping, amino acid analysis, and reaction against specific antibodies showed that all the Mn(II)-dependent peroxidases were probably products of one gene. A great degree of homology was also present among the various lignin peroxidases.

Structure and function of yeast acid phosphatase.

A pronounced heterogeneity of purified acid phosphatase has been observed. The size heterogeneity which was demonstrated by electrophoretic methods is in agreement with the finding that the enzyme preparation contains families of molecules differing in the carbohydrate content (38% - 70%). The charge heterogeneity was shown by isoelectric focusing indicating the existence of several slightly different protein chains in the enzyme preparation. It was found that the enzyme is a dimer with a mean molecular weight of 252000 Daltons. There are 16 N-glycosidically linked carbohydrate chains, differing in size from 15-150 mannose units, and a very small amount of O-glycosidically linked mannose. Properties of acid phosphatase deglycosilated by treatment with beta-endo-N-acetylglucosaminidase, were compared with the native enzyme and it was found that the carbohydrate chains do not play a direct role in the catalytic activity of the enzyme, but stabilize the tridimensional structure of the molecule. A drastic increase of sensitivity of the deglycosilated enzyme against proteolysis was found.