Search

Cytochrome c oxidase was treated with 1,5-difluoro-2,4-dinitrobenzene at molar ratios (DFDNB:oxidase) varying from 5 to 625. At the lowest ratio, there was virtually no effect of the probe on oxidase activity or on migration of oxidase subunits on sodium dodecyl sulfate--polyacrylamide disc gel electrophoresis. At ratios of 25 and greater, there was loss of oxidase activity and a change of the pattern of subunit migration on sodium dodecyl sulfate electrophoresis. (i) Activity loss was probably a result of severely perturbing the cytochrome c binding site since oxidase activity with a low molecular weight reductant (N,N,N',N'-tetramethylphenylenediamine) was unaltered. Also unaltered were the oxidized, reduced, and carbon monoxide binding spectra of the treated oxidase. (ii) The staining pattern on sodium dodecyl sulfate electrophoresis showed that subunits III and VI disappeared from their normal positions on the gel. A new band of higher molecular weight accompanied their loss from the gel indicating that the two subunits were being cross-linked. Subunits III and VI are thus shown to have two reactive groups within 4.8 A (1 A = 0.1 nm) of one another. This proximity has not been detected with other probes that react with the same groups.

The reagent 2-carboxy-4,6-dinitrochlorobenzene (CDNCB) reacts with the imino, amino and sulfhydryl groups of model compounds. At pH 8.2, sulfhydryl groups react much faster than do amines. N alpha-Acetylhistidine, N alpha-acetyltyrosine and N alpha-acetyltryptophan do not react. Poly(L-Lysine) and poly(DL-lysine) react about 50 times as fast as does N alpha-acetyllysine. A dichloroanalog, 6-carboxy-2,4-dinitro-1,3-dichlorobenzene, shows stepwise reactivity with amines. With bovine pancreatic ribonuclease, which contains no sulfhydryl, CDNCB reacts preferentially with the epsilon-amino of Lys-41 at 450 times the rate with the epsilon-amino of N alpha-acetyllysine. The preferential reactivity at Lys-41 is discussed in relation to the pK of Ly-41, the cationic character of the active site cleft, and the mechanism of RNAase action on substrates.

The near-neighbor relationships of proteins in the myelin membrane were examined using dinitrodifluorobenzene and other cross-linking reagents. When intact cat dorsal column or isolated myelin fragments were treated with cross-linking reagents, up to 20% of the myelin basic protein dimerized. The only other cross-linked product formed in the intact cat dorsal column was a heterodimer consisting of myelin basic protein and either the major or minor proteolipid protein. The remaining myelin proteins, including the major proteolipid protein, were cross-linked into very high molecular weight aggregates. In contrast, when the myelin membrane was dipersed in sodium dodecyl sulfate before the addition of cross-linking reagent, all the proteins remained essentially monomeric, with the exception of myelin basic protein which dimerized to some extent. In the absence of cross-linking reagent, it was shown by radioimmunoassay that small amounts of myelin basic protein dimer and the heterodimer were normally present in sodium dodecyl sulfate-polyacrylamide gels. We found no evidence of intramolecular cross-links between the two peptides formed by N-bromosuccinimide cleavage or between the two peptides formed by cyanogen bromide cleavage of the basic protein monomer. The regions of the myelin basic protein molecule involved in dimerization were also determined by similar cleavage of the cross-linked dimer. A rudimentary model for the structure of basic protein dimer in myelin is presented.

At pH 7, the apoenzyme of carboxymethylated and acylated aspartate aminotransferase reacts selectively with 1,5-difluoro-2,4-dinitrobenzene to form a single intramolecular covalent bond with the epsilon-amino group of the functional lysine residue located within the active centre. On shifting the pH to 9, the second fluorine atom of the bifunctional reagent is substituted with the sterically adjacent side groups of cysteine and tyrosine residues. The modified apoenzyme was subjected to partial proteolysis with pronase, and the digest was used to obtain and isolate the labeled products and to localize amino acid residues involved in the reaction. The established structures of several peptides containing Cys-2,4-dinitrobenzene-Lys and Tyr-2,4-dinitrobenzene-Lys allowed the identification of the amino acid residues involved in the reaction with the bifunctional reagent as Lys 258, Cys 390 and probably Tyr-70. The residues of Cys and Tyr are thus located at a distance of approximately 5 A (the length of the dinitrophenylene bridge) from the lysine residue forming an aldimine bond with pyridoxal 5'-phosphate in the active site.

The reaction of the title compound with human serum albumin has been examined at various concentrations of the sulfonate. Kinetic data suggest that there are two highly reactive lysine amino groups on the protein, five lysine residues which are less reactive and an undetermined number of additional nucleophilic groups that react very slowly with the reagent at pH 7.5. One of the rapidly reacting lysines is tentatively identified as lysine-199 in the protein sequence. Fluorine NMR experiments indicate the presence of tight binding sites on the protein for the sulfonate which are not near reactive functional groups.

When one or two drops of a dilute, non-irritating solution of 2,4-dinitrochlorobenzene (DNCB) is applied to a small area of skin of the intact guinea pig, about 20 per cent of the applied material, or some derivative of it, is soon excreted in urine. In normal, as well as in specifically sensitized guinea pigs, DNCB at the site of local application becomes rapidly bound to skin protein through primary chemical bonds. Twenty-four hours after application roughly half of the material present at the local skin site is still extractable with organic solvents. Of the non-extractable dinitrophenyl groups, about 99 per cent are in epidermis, and about 85 per cent are substituted in epsilon-NH(2) groups of lysine residues. Only traces of bound dinitrophenyl groups were observed in the corium. It is uncertain whether these are formed in situ, or are experimental contaminants, or are migratory epidermally formed conjugates. Even when DNCB is injected intradermally it combines predominantly with overlying epidermis and with epidermal components of hair follicles, but only slightly with corium. The 2,4-dinitrophenyl conjugates which are localized in the deeper, viable half of the epidermis, close to the epidermal-dermal junction, are inferred to be the agents responsible for specifically evoking the allergic response in sensitized animals. Conjugates which are situated in the outer, cornified half of the epidermis are shown to be incapable of eliciting the allergic response. The results furnish a basis for interpreting a common pattern of lesions in allergic contact dermatitis as it occurs spontaneously in man.

2,4-dinitrophenylsulfenyl chloride (DSCl) and 2,4-dinitrophenylthiocyanate (DSCN) elicited allergic reactions of the delayed type when applied to the skin of guinea pigs and of human beings who had been sensitized by prior exposure to 2,4-dinitrofluorobenzene (DF). DSCl and DSCN, together with 2,4-dinitrobenzene sulfonate (DSO(3)), constitute a clearly defined group of allergenic dinitrophenyl compounds in that they all combined with skin protein in vivo through reaction with cysteine or cystine. In vitro, these compounds combine with free SH groups, and with -S-S- groups of hair and epidermis, but not with -S-S- groups of oxidized glutathione or of bovine gamma globulins. DSO(3), DSCl, and DSCN did not react with amino groups in vivo, but did react with protein amino groups in vitro at pH values of about 10. Another group of dinitrophenyl compounds (DF, DCl, and DBr) previously had been shown to combine with lysine epsilon-NH(2) groups of epidermal proteins. In the present work it was found that these compounds do not react with the disulfide groups of these proteins, either in vivo or in vitro. Moreover, they did not seem to react with SH groups of viable skin, although they are highly reactive with sulfhydryl in vitro. This apparent discrepancy between reactivity with SH groups in vitro and in vivo may be due to the fact that the chromatographic technique employed was relatively insensitive for the sulfhydryl derivative. When a compound of either group was applied to the skin surface, dinitrophenyl-amino acids were recovered from the epidermis but not from the dermis. The results are discussed from the viewpoint of the epidermal localization of dinitrophenyl-protein conjugates.