Natural abundance nitrogen-15 nuclear magnetic resonance spectroscopy. Medium effects on the nitrogen-15 chemical shifts of small peptides

Abstract

Medium effects upon the 15N chemical shifts of some small peptides are reported, to allow the relative importance of solvent, pH, and sequence effects to be delineated. A change in the pH of a solution of carnosine (P-alanylhistidine) from 0.4 to 11 .O causes the His 15N resonance to shift downfield by 8.5 p.p.m. Changing the solvent from dimethyl sulphoxide (DMSO) to trifluoroacetic acid (TFA) causes downfield shifts of ca. 4 p.p.m. for the peptide nitrogen resonances of some N-acetyldipeptides, and explains why the 15N shift observed for polyglycine in TFA solution is to low field of that predicted earlier for a glycine residue of a peptide in DMSO solution. The magnitude of these pH and solvent effects is such that they may mask any sequence effects upon peptide 15N shifts, which are reported to be 1-4 p.p.m. for dipeptides in aqueous solution.College, Mile End Road, London E l 4NS MAGNETIC resonance spectra of the 15N nucleus are being used increasingly in the study of peptides.1-9 Two important features of these studies are the conclusions of Roberts and his co-workers5 and conclusions from this l a b ~r a t o r y . ~ The former group found that the chemical shifts of the peptide 15N nuclei in simple dipeptides (aqueous solutions; pH range 5.0-6.1) were slightly sensitive to the nature of both C-terminal and N-terminal units. Thus it is formally possible that sequence information may be obtained by 15N n.m.r., a non-destructive method. Our ~t u d i e s , ~ however, indicated that there was no detectable sequence information from 15N chemical shifts of N-acetyldipeptides, at least in dimethyl sulphoxide (DMSO) solution. Accordingly we have undertaken a study of medium effects (including both solvent and pH effects) upon peptide 15N chemical shifts for some simple model systems, to clarify the situation.Assignment of 15N Resonances.-The first stage in the investigation of peptides by 15N n.m.r. is the assignment of the resonances to specific amino-acid units. We shall assume initially that sequence effects upon the l5N shifts are small and that in the absence of any differential effects upon the amide nitrogens in the peptide (caused by solvation, hydrogen bonding, or conformation changes), the relative I5N shifts of the amide nitrogens are controlled mainly by the nature of the amino-acid unit itself.Previously published 15N chemical shifts of aminoacids,l N-formyl a r n i n o -a ~i d s , ~, ~~ N-acetyl amino-acid~,~ and amino-acid methyl ester hydrochlorides allow estimation of the effect of substitution at the amino-acid a-carbon atom upon the 15N shifts. In the manner previously adopted 1*11 we assume an additivity relationship for these substituent effects similar to that of Grant and Pau112 for 13C chemical shifts. The approach is

Solvent Effects on Nitrogen Chemical Shifts

“…The amide 15 N NMR chemical shift is sensitive to hydrogen bonding, with a most pronounced effect upon the carbonyl oxygen hydrogen bonding with proton donors, yielding a decrease in electron density of the adjacent nitrogen and accordingly a shift increase [41,[187][188][189]. Deshielding up to 10.5 ppm have been reported for model peptides upon protonation with trifluoroacetic acid [177], typical changes …”

“…38) change by 3.3-9.8 ppm [177]. For N-acetyldiglycine (303), the AcGly 1 and Gly 2 amide nitrogens were deshielded by 9.8 and 3.8 ppm, respectively.…”

“…38; Table 31) indicated [177] that upon ionization of the carboxyl group of di-(302) and triglycine (304), pK 1 % 3.2, their C terminal amide nitrogen atoms got deshielded by 6.2 ppm and 5.9 ppm, respectively. The amide nitrogen of the central glycine residue, Gly 2 , in triglycine (304) was only slightly deshielded by 1.1 ppm, indicating that the influence of the deprotonation of the carboxyl group is distance dependent.…”

“…The C terminal amide nitrogen of triglycine (304) got deshielded by 6.6 ppm upon deprotonation of its cationic (pH 1.5) form by adjustment to pH 11.9 [178], whereas its central amide nitrogen got shielded by 0.4 ppm, and its α-amino group (pK 2 ¼ 8.1) by 12.9 ppm. For the C terminal Gly 3 residue in glutathione (307) and glutathione disulfide (308), pK 2 ¼ 3.6, the amide 15 N resonances were deshielded by 5.9 ppm upon increasing the pH from 0.4 to 12.5 and 12.0 [177]. As a consequence of deprotonation of the side-chain sulfhydryl group (pK 4 ¼ 9.6) of glutathione (307), its Cys 2 nitrogen got slightly more deshielded than that of glutathione disulfide (308), i.e., by 4.3 versus 1.4 ppm, when comparing their chemical shifts at pH from 0.4 and 12.5.…”

NMR investigations on alanyl‐[15% 13C, 95% 15N]‐proline: 15N chemical shifts and 13C15N coupling constants

“…the ionization of the carboxylate group and the deprotonation of the amino group) is not surprising since a downfield shift is expected if a deprotonation process occurs at a position y to the nitrogen of interest. 16 The quite different magnitude of both shifts is obviously due to the different influence of an appearing negative charge or of a disappearing positive charge on the electronic structure of the peptide bond which, to a large degree, determines the value of the nitrogen chemical shift. Coupling constants.…”