Lysozyme association with nucleic acids.

Lysozyme is well known for the ability to hydrolyze the cell wall of bacteria. Based on the similarity of structure between lysozyme and histones as seen from the results of X-ray crystal structure determinations, we have postulated that binding to nucleic acids may be another biological function of lysozyme. We have therefore begun a systematic study of the interactions of lysozyme and related molecules with nucleic acids, and present here a preliminary report. Binding to DNA and RNA has been demonstrated from gel electrophoresis, enzyme activity, and coprecipitation studies. We suggest that this function of lysozyme will provide an explanation why Lee-Huang et al. (1999) [Proc. Natl. Acad. Sci. USA 96, 2678-2681] were able to call lysozyme a "killer protein" against the AIDS virus, and may provide a new avenue of research on AIDS therapy.

Molecular structure of the amyloid-forming protein kappa I Bre.

The molecular structure of the amyloid-forming Bence-Jones protein kappa I Bre has been determined by X-ray crystallography at 2.0 A resolution. The fragment from the kappa chain of immunoprotein contains 107 amino acid residues, and polymerizes in the crystal form into a giant helical spiral, surrounding a cylinder of water 50 A in diameter with a repeat of 77.56 A, containing 12 kappa molecules, plus another 12 molecules from neighboring parallel spirals. The resulting structure has many features which have been found or suggested from studies on the protein fibrils found in amyloid deposits. From the results of the X-ray crystal structure a hypothesis is presented for the structure and formation of the amyloid fibril.

Structures of monoclinic lysozyme iodide at 1.6 A and of triclinic lysozyme nitrate at 1.1 A.

Hen egg-white lysozyme is one of the most thoroughly studied of enzymes and has been the subject of study by many methods, including X-ray crystallography. The present work extends the X-ray crystallography to higher resolution, includes the positions of the anions, and examines the contacts of the neighbors in greater detail. Data were collected at room temperature on a Rigaku R-axis area detector with rotating-anode X-ray generator to 1.6 A resolution for monoclinic lysozyme iodide at pH 4.0, to 1.8 A for monoclinic lysozyme iodide at pH 8.0, and to 1.1 A resolution for triclinic lysozyme nitrate at pH 4.5. The structures have been refined by SHELX93 with the expected number of anion sites being accounted for. Two regions of the protein have been found to be variable: residues 65-75 and 99-104. Except for 65-75 and 99-104, lysozyme is a very stable molecule with the crystal forms being held together by the electrostatic contacts of the anions and by layers of water molecules. The anion positions can be described as paired half sites, each half being contributed by a different lysozyme molecule. The many different crystal forms of lysozyme may be due to different combinations of the many such half sites on the surface. A hypothesis is presented for lysozyme in the different crystal forms and which may be extended to behavior in solution. Suggestions for future crystallographic research are proposed, involving anions of different shape and charge.

Mechanism of glycogenin self-glucosylation.

Glycogenin, the proposed initiator of mammalian glycogen biosynthesis, transfers glucose residues from UDP-glucose to an oligosaccharide chain attached to Tyr-194 in a self-glucosylation reaction. Mutation of Tyr-194 to either Phe or Thr residues results in the loss of this self-glucosylating activity since the site of oligosaccharide attachment has been lost (Y. Cao, A. M. Mahrenholz, A. A. DePaoli-Roach, and P. J. Roach (1993) J. Biol. Chem. 268, 14687-14693). We describe here that Phe-194 and Thr-194 mutants of glycogenin, as well as wild-type protein, were active in transferring glucose to an exogenous acceptor, maltose, a known inhibitor of the self-glucosylation reaction. The reaction product was exclusively maltotriose with no evidence for further elongation to maltotetraose or maltopentaose. The values of Vmax/Km for maltotriose synthesis for the mutant proteins were 1.5-3.5 times greater than that of the wild type. Analysis of crystals of wild-type glycogenin by X-ray diffraction gives a tetragonal unit cell of a = b = 130 A and c = 174 A in space group I4 with four glycogenin molecules in one asymmetric unit. Considerations of the symmetry and the crystal packing indicate the existence of dimers of glycogenin which may further associate to form a tetramer. The existence of oligomeric forms of glycogenin, together with the idea that glucose transfer to an exogenous acceptor is possible, raises the possibility that the intramolecular self-glucosylation of glycogenin could involve an intersubunit transfer of glucose.

Regeneration of antidigoxin binding activity in an antibody by changes surrounding the original binding defect.

A panel of antibodies which differ in their L chain structures and which bind to structurally defined haptens, would be useful in investigating L chain structure and function. In a previous study, chain recombinant antibody CR24 (26-10 H, 45-20 lambda) was produced by hybridoma-hybridoma fusion. Although both parental antibodies bound digoxin with high affinity, CR24 lacked detectable digoxin-binding activity. Hybridoma CR24 was subsequently fused with H chain-loss hybridomas in order to produce a panel of antibodies composed of 26-10 H chains and 26-10 "like" L chains. Two antibodies produced were CR260 which demonstrated digoxin-binding activity and CR256 which did not. CR260 and CR256 expressed only one amino acid difference (Pro to Leu at L-96). This difference resulted in the CR256 binding defect. In this report, two new antidigoxin antibodies are described. One, SR2E7, contained the Pro to Leu (L-96) defect, but still bound digoxin. Binding affinities and binding specificity patterns, as well as complete VL DNA sequence and corresponding protein sequence of the new digoxin binding antibody L chains (SR2E7 and SR1C7) are presented. Both kappa L chains are highly homologous to the 26-10 kappa L chain as well as the BALB/c germline gene K5.1. These results suggest that antibodies which are initially defective in binding activity can be cured by changing specific amino acids involved in determining the binding-site structure. Molecular modelling studies of the binding-site region were completed to address L chain structural changes induced by specific amino acid substitutions.

The x-ray crystal structure refinements of normal human transthyretin and the amyloidogenic Val-30-->Met variant to 1.7-A resolution.

The x-ray crystal structures of normal human transthyretin (prealbumin) and the amyloidogenic Val-30-Met variant have been refined at 1.7-A resolution to R-values of 0.168 and 0.179, respectively, for 19,882 and 20,362 reflections (Fobs > 2.0 sigma). Standard deviations for stereochemical parameters are 0.018 and 0.022 A for bond distances, 0.030 and 0.038 A for angle distances, and 0.035 and 0.070 A for planar 1-4 distances. The newly refined normal structure shows improvement over the original structure of Blake and Swan (Blake, C. C. F., and Swan, I. D. A. (1971) J. Mol. Biol. 61, 217-224) in stereochemistry and in the conformation of the loop regions. Residues Arg-103, Thr-123, Asn-124, and Pro-125 have now been resolved, and residues 1-9 and 126-127 have been modeled with the aid of simulated annealing refinement. The functional form of transthyretin is a tetramer, having a cylindrical cavity which will bind thyroxine and an exterior binding site for the complex of retinol with retinol-binding protein. The monomer is a beta barrel flattened to become more like a sandwich with residue 30 in the interior. The methionyl for valyl substitution forces the beta sheets of the monomer as much as 1 A apart, resulting in a distortion of the thyroxine-binding cavity, in agreement with the independent observations that the Met-30 variant has low affinity for thyroxine.

X-ray crystal structure of the Ala-109-->Thr variant of human transthyretin which produces euthyroid hyperthyroxinemia.

The structure of the Ala-109-->Thr mutation of human transthyretin, a nonamyloidogenic variant with enhanced thyroxine binding, has been determined by x-ray diffraction to a resolution of 1.7 A. The model, including 175 solvent water molecules, has been refined by constrained least squares to an R-value of 0.157. The standard deviations for protein geometry are 0.016 A for bond distances, 0.5 degree for bond angles, 0.031 A for 1-4 distances, and 0.005 A for deviations of planar groups from their least squares plane. The estimated error in protein atomic coordinates is 0.12 A. Residue 109 extends inward between the two beta sheets which form the major component of the monomer, as does the side chain of residue 30 in the amyloidogenic Met-30 variant. Comparison of the Thr-109 structure with that of the normal shows that the extra atoms of the threonine fit into empty space between sheets and make no extensive changes to the molecular conformation. The substitution at 109 causes small local changes in the secondary structure of the A, G, and H strands resulting in a shift of residues 15-17, 108-110, and 117 in each monomer. The thyroxine-binding sites of the Thr-109 and Met-30 variants and of the normal protein are compared, and the results suggest that the variation in affinity for thyroxine between the three proteins may arise from differences in the size of the binding pocket.

Alteration in molecular structure which results in disease: the Met-30 variant of human plasma transthyretin.

The structure of a variant transthyretin has been determined by X-ray crystallography at 2.3 A resolution in order to investigate those changes which lead to amyloid formation. This variant transthyretin, in which the internal valyl residue at position 30 is replaced by methionyl, is associated with the most common form of familial amyloidotic polyneuropathy (FAP). Comparison to the known structure of the normal transthyretin tetramer shows that the bulkier methionine residue 30 which lies between the nearly orthogonal beta sheets of the dimer, results in the sheets being displaced an average of 0.4 A. The internal structure of the sheets and of the monomer-monomer interface is maintained. Such global changes may affect the metabolic properties and the tendency towards polymerization of the mutant protein. These findings may form a basis for understanding other amyloid-deposition diseases.

Preparation and crystallization of human transthyretin (prealbumin) variants.

Naturally occurring variants of human serum transthyretin (prealbumin) have been prepared by recombinant DNA methods and crystallized from ammonium sulfate solutions to give crystals suitable for x-ray crystallographic analysis. Included are variants which are known to be associated with familial amyloidotic polyneuropathy. Dyes which have been used as histochemical stains to identify amyloid tissue deposits: Congo Red, Methylene Blue and Bromophenol Blue, have been co-crystallized with the transthyretin variants. Congo Red was found to be very selective while Methylene Blue actually assisted in the formation of crystals. All crystal forms which were examined were isomorphous to the structure of normal transthyretin.

A V kappa-J kappa junctional change in an antidigoxin recombinant antibody destroys digoxin-binding activity.

A set of high affinity antidigoxin antibodies were previously identified with high homologous V kappa 1A L chain sequences but were associated with two entirely different VH regions and two dramatically different specificities for digoxin analogs. Antibodies 40-20, 40-60, 40-90, and 40-100 displayed similar binding specificities but differed from that of antibody 26-10. In a previous study using somatic cell fusion for Ig chain recombination we demonstrated that a recombinant antibody consisting of the H chain of antibody 26-10 and the L chain of antibody 40-20 retained digoxin binding and the 26-10 Id, but displayed a binding specificity pattern dominated by the 26-10 H chain donor. In the present study we produced three additional chain recombinant antibodies that contain the 26-10 H chain recombined with each of the L chains of antibodies 40-60, 40-90, and 40-100. All four recombinants expressed the 26-10 Id indistinguishably from the 26-10 antibody. Two of the recombinants (using the 40-60 and 40-90 L chains) bind digoxin; however, the recombinant using the 40-100 L chain failed to bind digoxin. Complete sequence analyses of the 40-20, 40-60, 40-90, and 40-100 VH and VL regions were performed. Antibodies 40-90 and 40-100 have identical VH region sequences but differed only in their L chains at position 96 (proline/leucine). This single difference at the VK-JK junction abolished digoxin binding in the context of one H chain (26-10), but does not cause a significant change in binding in association with the "normal" parental chains 40-90 and 40-100. Thus, structurally closely related VL regions can recombine with different VH regions to form digoxin binding sites of different specificity; in one binding site the identity of a L chain junctional residue is critical whereas in the second binding site that residue is unimportant. Molecular modeling studies revealed major differences between calculated binding site structures for 26-10 when leucine is substituted for proline at position 96 in the 26-10 VL region.

New evaluation of potential methylmercury scavengers.

A biological assay was developed to evaluate rapidly the relative efficacy of marketed and experimental mercurial scavengers. Rat liver mitochondrial protein (1.0 mg) was titrated against methyl-mercuric chloride to the inhibitory level of mitochondrial respiration. Respiration induced by adenosine 5'-diphosphate with succinate (plus rotenone) as the substrate was inhibited consistently by 20.7 +/- 3.9 nmoles of methylmercury/mg of protein. Adenosine 5'-diphosphate-stimulated respiration (State 3) was restored with dimercaprol, penicillamine, and cysteine but not with serine. The antagonists glutathione, 3-mercapto-propionic acid, 2-mercaptoethanol, dithiothreitol, thioglucose, mercaptosuccinic acid, and thiosalicylic acid and mercaptosuccinic acid. Sodium sulfide, thioacetamide, and ethylenediaminetetraacetic acid were completely inactive. Substitution of glutamate (plus malate) for succinate (plus rotenone) as the substrate did not alter the responses significantly. The rat liver mitochondrial assay provides preliminary information about the efficacy and toxicity of water-soluble thiols. Investigations utilizing encapsulated water- and lipid-soluble mercaptans are in progress.

Mercapto steroids in protection against mercury and lead poisoning.

When thiocholesterol is administered as liposomes, it provides significant protection against methylmercuric chloride in mice when given in three intraperitoneal injections, 0.5 hr before and 2 and 8 hr after the methylmercuric chloride. Thiositosterol, 5 alpha-cholestane-2 beta, 3 alpha-dithiol, and 5 beta-cholane-3 beta, 24-dithiol also are active, but 3 alpha-mercapto-5 alpha-pregnan-20-one, 6 beta-mercapto-5 alpha-cholestane-3 beta, 5 alpha-diol, 3 beta-mercapto-5 beta-cholanic acid, and adamantanethiol are ineffective under these conditions. Adamantanethiol is somewhat effective when administered in soybean oil. Cholestanyl amine was treated with acetylthiosuccinic anhydride to give the half amide; cleavage with hydroxylamine liberated the thiol group. This product is active against both methylmercuric chloride and lead nitrate.

Synthesis and evaluation of sulfur-containing steroids against methylmercuric chloride toxicity.

Sulfur-containing steroids, analogs, and derivatives were synthesized for evaluation in mice suffering acute toxicity from methylmercuric chloride. Steroids were administered by intraperitoneal injection, by stomach tube feeding, or by absorption through the tail skin. Thiocholesterol and the thiocholanoic acids were effective if given prior to poisoning. The thiosteroids were significantly more effective than penicillamine or dimercaprol under these conditions.