Reaction of Tetranitromethane with Lutropin, Oxytocin, and Vasopressin*

Tetranitromethane reaction with intact ovine lutropin and its isolated subunits was studied using spectrophotometric measurements, amino acid analysis, and isolation of tyrosyl peptides. Tyrosyl residues in the /J’ subunit (@37,059) did not react with tetranitromethane in the intact hormone, but were nitrated in the isolated subunit. The sequence and extent of reaction of tetranitromethane with the tyrosyl residues in the 01 subunit was a21 = a92 = 0193 (in intact hormone or isolated subunit) > (~41 (reacted in isolated subunit only) > a30 (reacted in isolated subunit in 8 M urea only). Polymerization was observed as a side reaction in agreement with previous studies. The degree of polymerization appeared to be related to both primary sequence and tertiary structure, and for lutropin had the relation: 01 subunit polymerized) fl than 40%). Polymerization observed with vasopressin was significantly greater than with oxytocin; for these peptides the tyrosine residues in the monomeric product were converted to 3-nitrotyrosine.

t1on ---- let light. The paper was sprayed with a solution of 1 mg of fluoresca-Edman Degradation-The Edman degradation and identification of mine in 100 ml of acetone in a fine mist with an air-driven aerosol dansyl amino acids were done using microadaptations of procedures sprayer, and was dried in warm air for about 10 min and a developing reviewed by Hartley (17). solvent consisting of 0.5% by volume of pyridine in acetone was Nitration Reactions-The nitration reaction was carried out by applied. Both reagents were made up immediately before use. After variations of procedures described by Sokolovsky et Al. (1). The molar spraying with the developing solvent, the paper was dried in warm air ratio of tetranitromethane to tyrosine varied from 1 to 50 (see Table I).
for 30 min. The peptide spots were then circled, cut out, and sewn to Tetranitromethane was dissolved in 0.5 ml of 95% ethanol and added strips of Whatman No. 1 paper of convenient size. These strips were dropwise to the protein solution with rapid stirring. Reaction rate was eluted into appropriate receiving tubes m an enclosed chamber. absorbance in this region are sometimes formed. Table  I shows that the extent of modification in the purified monomer products was more dependent upon the ratio of gtranitromethane to tyrosine than upon protein concentration or reaction temperature, even though these factors affected the relative amounts of monomeric product and cross-linked products obtained. Tyrosyl modification in the purified monomer was largely independent of protein concentration and temperature. The monomeric products were separated from other reaction products by chromatography on Sephadex G-100 (Fig. 2). In the reaction of intact oLH at a tetranitromethane/tyrosine molar ratio of 1, a monomeric form lacking 1 residue of tyrosine was found (Table I). Somewhat higher tetranitromethane/tyrosine ratios gave monomers in which about 2 tyrosine residues were modified, and still higher ratios led to the modification of approximately 3 residues. Separation of the subunits from the isolated monomers in the oLH Preparations 1 to 9, Table I, by countercurrent distribution (18) gave patterns such as that in Fig. 3. Measurement of the 310 to 350 nm absorbance indicated that modification was confined to the (Y subunit. This was also confirmed by amino acid analysis (Table II).
Tetranitromethane treatment of the separated subunits gave quite different results. Reaction of oLHa using tetranitromethane/tyrosine ratios between 5 and 50 gave varying yields of isolated monomeric products in which 4 of the 5 tyrosyl residues were modified. Modification of all 5 tyrosyl residues in oLHa was attained only by using similar reaction conditions in the presence of 8 M urea.
The reaction rate for oLHP was considerably slower under equivalent conditions than that of oLH or oLHcr. A second addition of tetranitromethane was made after 1 hour of reaction, giving an overall tetranitromethane/tyrosine ratio of EFFLUENT VOLUME @Is) "" 50/l. Under these conditions, both tyrosyl residues in oLH@ were modified, giving high yields of the monomeric preparation.
Analyses of the purified monomers for tyrosine and 3nitrotyrosine are shown in Table I. In none of the preparations was 3,5-dinitrotyrosine detected. In many of the preparations the amount of 3-nitrotyrosine formed was not equal to the amount of tyrosine lost during the modification procedure. Cheng and Pierce (5) observed a similar discrepancy, but Sairam et al. (6), who used a normalized calculation, did not report such differences. The discrepancy may be explained at least partially by tyrosyl side reactions which give rise to products other than 3-nitrotyrosine.
Such products have been partially characterized in the present study by spectral studies and analysis of modified tyrosyl peptides.  Isolation and Analysis of Tetranitromethane-modified Tyrosine Peptides-The countercurrent distribution separations indicated that no modification of the /3 subunit occurred in the intact hormone. Since both tyrosine residues in this subunit were nitrated completely in the isolated subunit no further study of the tryptic peptides from the nitrated p subunit was undertaken.
Isolation of nitrated peptides was confined to those of the (Y subunit in various preparations. Fig. 4 shows the sequence of oLHn as determined by Liu et al. (8) with the theoretical tryptic peptides enumerated from the NH, terminus as Tl through T14, in which T4-T5, T6, and T13 are the peptides containing tyrosyl residues. Because the Lys-Pro sequence is resistant to cleavage, peptides T4 and T5 were never observed and this region was isolated as the larger fragment T4-T5. Fig. 5 shows a typical set of maps of the tryptic digest of oLHa, giving the location of various identified peptides, and showing the locations of the tyrosyl-containing peptides from a fully nitrated a subunit and of the parent unmodified tyrosyl-containing peptides. In some of the preparations a peptide, T-13', was found which proved to be due to COOH-terminal heterogeneity in the original oLH preparations. T-13' was derived from a small fraction of molecules which had lost the last 3 amino acid residues (-His-Lys-Ser-OH). This heterogeneity appears to be the result of a chymotryptic-like degradation obtained in this batch of pituitary glands. Although COOH and NH,-terminal heterogeneity has been reported in previous lutropin preparations (8, 9, 21-23) this appears to be the first instance involving these residues (cr94 to 96).
Peptide T4-T5 was found in the unmodified form (36%) only in the "monoatyrosyl" Preparations 1 and 2; peptides T13 and T13' were modified in all preparations studied. Analyses of modified peptides T4-T5, T6, and T13, were always consistent with their theoretical composition except with respect to tyrosyl and 3-nitrotyrosyl residues. The isolation of the theoretical tryptic peptides in good yield from these preparations, the absence of cross-linked peptides detectable on the peptide maps, and the expected separation of the subunits from the monomeric oLH with countercurrent distribution indicated that very little interchain or intrasubunit cross-linking occurred in these monomeric products. Analyses for tyrosine and 3-nitrotyrosine in these peptides from several of the preparations are given in Table III. These data show that 1 of the tyrosyl residues in T4-T5 and tyrosyl 41 in T6 are protected from tetranitromethane modification in the intact molecule, but tyrosy141 is readily accessible to modification in reactions using the separated oLHcv subunit.
Modification of tyrosyl residues in peptides T4-T5 and T13 occurred in varying degrees in all of these preparations, and the extent of modification of these residues in the intact molecule depended on the tetranitromethane/tyrosine ratio. Table III shows that the appearance of 3-nitrotyrosine did not equal the loss of tyrosine, and thus the loss of tyrosine on amino acid analysis was the best measure of tyrosyl residue modification. The present experiments do not allow a determination of the difference in accessibility between the 2 adjacent tyrosyl residues in T13, a92 and 93, if such a difference exists. However, the even increase in the amount of tyrosyl modification in peptides T13 and T13' with increasing tetranitromethane concentration suggests that these residues may be equally accessible to tetranitromethane modification. Analysis of the modification of tyrosyl residues a21 and cr30 in peptide T4-T5 was readily evaluated in all of the preparations analyzed because Tyr a21 was the NH, terminus of the peptide. Analyses of this peptide from various preparations before and after a single Edman degradation afforded a determination of the extent of modification of the separate residues (see also Ref. 5). Preparation 1 showed that Tyr a21 was accessible to tetranitromethane modification in the intact molecule, while Tyr a30 was inaccessible under the same conditions. The data for Preparation 11 (Table III) showed that this residue is also inaccessible in the separated oLHa subunit. Tyr a21 and Tyr Fro. 5. Maps of oLHa tryptic peptides: pH 6.5 high voltage modified form. pH 6.5 high voltage electrophoresis neutral band electrophoresis (HVE) mobilities normalized with respect to aspartate removed before chromatography. b, isolation of T4-T5, intact and (M,,,); pH 2.1 high voltage electrophoresis mobilities normalized with tetranitromethane-modified: high voltage electrophoresis, pH 6.5, 3 respect to NO-DNS-Arg(&m+nra , ). specific conditions as indicated. kv, 120 min U~PSUS BAWP chromatography. Glycopeptides TlO, T13, Descending chromatography as indicated in 1-butanol/acetic acid/ and T13' are immobile on chromatography. Dashed area is cut out and watedpyridine 15/3/12/10 (BAWP). a, isolation of basic peptides, sewed to another sheet for pH 2.1 high voltage electrophoresis in c. c, including T6: high voltage electrophoresis, pH 6.5, 3 kv, 50 min versus isolation of T13 and T13': high voltage electrophoresis, pH 6.5, 3 kv, BAWP chromatography, 18 hours, 25". Acidic glycopeptides not 120 min, strip removed from b after chromatography versus high separated; T4-T5 incompletely separated from its tetranitromethanevoltage electrophoresis, pH 2.1, 2 kv, 180 min.  T6 T13 T13'  T4-T5  T6 T13  T4-T5  T6 T13  T4-T5  T6 T13  (100%)  T4  (~30 were both modified extensively when the reaction of the found by amino acid analysis. High absorbance in the 310 to oLHa subunit was carried out in 8 M urea, as shown in 350 nm region correlated with the disappearance of tyrosine Preparation 15. but not with the appearance of 3nitrotyrosine. The possibility Spectral Studies-Comparison of the absorbance spectra of that the products absorbing in the 310 to 350 nm region are purified monomer oLH derivatives, of the modified separated generated by performic acid oxidation of the nitrated protein subunits, and of the isolated modified peptides with that given was eliminated by examination of the visible spectrum of by 3-nitrotyrosine confirmed the conclusion that tetranitro-Preparation 15 before and after treatment, as shown in Fig. 7~. methane treatment results in the formation of other products in In addition, analyses for tyrosine and 3-nitrotyrosine before addition to 3-nitrotyrosine. Fig. 6 shows the spectrum of T6 and after performic acid treatment were virtually the same. It from Preparation 11 compared with spectra of 3-nitrotyrosine thus appears that these products are not the result of subseand the unmodified peptide. The spectrum of the modified quent treatment, and must be formed by different reactions of peptide shows significant absorbance in the 310 to 350 nm the tyrosyl residues. region which correlated well with the amount of 3-nitrotyrosine While the amounts of the differently absorbing products were variable even for repeated preparations at the same tetranitromethane/tyrosine ratio, relatively more of the 310 to 350 nm absorbing products were formed when low tetranitromethane/tyrosine ratios were used, while the yield of X-nitrotyrosine increased in the (Y subunit when the structure was unfolded in 8.0 M urea. High yields of 3-nitrotyrosyl derivatives were also obtained in the reaction of tetranitromethane with the /3 subunit without urea denaturation, Fig. lb. The spectrum of peptide T4-T5 isolated from "monoatyrosyl" Preparation 1 is shown in Fig. 8a, together with a spectrum made after a single Edman degradation. Analyses of these peptides (Table III) revealed 1 intact tyrosyl residue both before and after the Edman degradation. Dansylation showed no identifiable NH,-terminal amino acid in the intact peptide. Dansyl phenylalanine appears in good yield after one Edman degradation.
The yellow color shown by this peptide on the map upon exposure to ammonia vapor was less intense than that shown by similar amounts of other peptides containing true 3-nitrotyrosyl residues, and the peptide also gave an atypical reaction with fluorescamine or ninhydrin. It was revealed on fluorescamine-treated maps by a strong absorbance of ultraviolet light, instead of the yellow-green fluorescent spot characteristic of the fluorescamine reaction products.
Similar products have been reported in studies using tyrosyl-30 containing di-and tripeptides (4). 20 In contrast to the above peptide isolated from Preparation 1, the T4-T5 peptide isolated from Preparation 15 (carried out in IO 8 M urea) gave the visible spectrum shown in Fig. 8b, which 0": also shows the spectrum after a single Edman degradation. 04 Yields of the 3-nitrotyrosyl derivatives in the same peptides were much higher in Preparation 15 than in Preparation 1 02 (Table III), and the relative amount of 310 to 350 nm o I absorbance was correspondingly less. This suggests that relaxation of the tertiary structure in the presence of urea may favor the production of the 3-nitrotyrosyl derivative. Studies on Oxytocin and Vusopressin-Effects of tertiary structure on the course of the reaction of tetranitromethane 9 with the tyrosyl residues in oLH suggested the examination of 2 O"' the reaction with small peptides where tertiary structure s 30 may not exert as much influence. Oxytocin and vasopressin are g 2 o cyclic, disulfide-containing octapeptides, each of which contains a tyrosine residue in position 2. Sequences around the d: tyrosine are: H .Cys-Tyr-Ile-and H .Cys-Tyr-Phe-for oxytocin 06 and vasopressin, respectively. Their reaction with tetrani-04 tromethane permitted three observations: 1. The range and yield of polymeric and monomeric prod-02 ucts were very different for these two structurally similar 01 peptides.
2. The size distribution of products produced from either peptide, as measured by gel chromatography, was relatively independent of the tetranitromethane/tyrosine ratio employed. 3. The production of 3-nitrotyrosine derivatives was limited to the monomeric forms.
0 01 Fig. 9 shows elution profiles on Sephadex G-25 gel filtration  (Table IV). native oxytocin or vasopressin monomer peak elution area. Oxytocin, The major product formed in the isolated monomer fraction a, tetranitromethane (mol/mol) = 1.5/l; 6, tetranitromethane (mol/ mol) = 25/l; Vasopressin, c, tetranitromethane (mol/mol) = 1.5/l; d, was a peptide containing the 3-nitrotyrosyl derivative (Table   tetranitromethane (mol/mol) = 25/l. V). In marked contrast, the polymeric fractions showed no detectable 3-nitrotyrosine, although solution suspension in identical with that of nitroformate, a side product always alkaline media indicated substantial yellow-colored material present in these reactions (24). (presumptive nitration). The soluble material in the high molecular weight fractions showed higher 340 to 428 nm DISCUSSION absorption ratios, with spectral patterns similar to the peptide The accessibilities of oLH tyrosyl residues found in this shown in Fig. 8~. Neither 3-nitrotyrosine nor unmodified study are in general agreement with those determined for tyrosine were present in these fractions. This suggests that bovine lutropin (5). They are also consistent with the ratio of nitration via the unidentified derivative with an absorption iodination of specific tyrosyl residues in porcine lutropin (25). maximum near 340 nm may favor polymerization.
The study of Sairam et al. (6) on nitration of ovine lutropin, The absorbance of each fraction from the soluble products in which they report the accessibility of all tyrosyl residues was measured at three wavelengths (Fig. 9). Elution profiles except a21 and /359 are in disagreement with both the present are similar with tetranitromethane/tyrosine ratios of 1.5/l or studies and those cited above. Since both studies (Ref. 6 and 25/l for either oxytocin or vasopressin, but the difference the present report) deal with lutropin from the same species, between the two peptides is dramatic. In each preparation all the difference must lie in the methods employed, or interpretafractions except C and F gave the expected analysis for tion of the results, or both. Sairam et al. (6) used analytical oxytocin or vasopressin, except that tyrosine was absent data from peptides partially isolated from digests of the intact (Table IV). Analyses of the monomeric products (Fractions C) hormone, whereas the present studies have used data only from are presented in Table V. The spectra of these fractions showed pure peptides isolated from the subunits. an absorbance peak at 428 nm consistent with the amount of The present studies on oLH, considered together with those 3-nitrotyrosine found by amino acid analysis. Fractions F are of Cheng and Pierce (5) and Combarnous and Maghuin-Rogisat the included volume of the column (salt peak) and their ter (25) on bLH and pLH, respectively, show that the topology absorbance spectra showed a sharp peak at 340 nm which was around the specific tyrosyl residues, as revealed by their