MOLECULAR STRUCTURE ELSEVIER
Journal of Molecular
Ab initio calculation
of the NMR shielding constants for histamine’
A.P. Mazureka3*, J. Cz. Dobrowolskiaqb, J. Sadlej”,” aDrug Institute, 30/34 Cheimska Street, 00-725 Warsaw, Poland hIndustrial Chemistv Research Institute, 8 Rydygiera Street, 01-793 Warsaw, Poland ‘Department of Chemistry, University of Warsaw, 02-093 Warsawl.Poland Received
10 March 1997; accepted
16 April 1997
Abstract The gage-independent atomic orbital (GIAO) approach is used within the coupled Hartree-Fock approximation to compute the ‘H, “C and 15N NMR shielding constants in two tautomeric forms of both the histamine molecule and its protonated form. An analysis of the results shows that the protonation on the end of the chain changes its nitrogen shielding constants of the pyridine and pyrrole type. These changes are much higher for the N(3)-H than for the N( 1)-H tautomer. 0 1997 Elsevier Science B.V. Keywords:
Ab initio; Cationic; Histamine;
Neutral; NMR; Protonated;
1. Introduction Histamine (1 H-imidazole-4(5)-ethanamine, HA) is a neurotransmitter that has been shown to act on a three different receptors, classified as the HI, H2 and H3 receptors [l]. For this reason it has attracted a large amount of research. Its biological activity has stimulated a number of study of its tautomerism and conformation. Histamine contains three nitrogen atoms available for protonation, two in the imidazole ring and one of the side chain. Thus, histamine can exist as an uncharged base with only one ring nitrogen atom protonated, the monocation, where the sidechain amino group is also protonated, and as the dication where all the nitrogen atoms are protonated. * Corresponding author. Tel: 0048 22 412940; fax: 0048 22 410652. ’This paper is dedicated to Professor Henryk Ratajczak on the occasion of his 65th birthday. 0022-2860/97/$17.00
0 1997 Elsevier
At the physiological pH, according to Ganellin and Parsons [I], all three species coexist in the ratio base:monocation:dication of around 1:96:3, respectively . The monocation and the base are able to undergo tautomerism. Each of these two tautomers can exist in three different stable conformations. The first is termed an open or tram, for which the torsion angle 7, is ca. 90” or less and the torsion angle 72 is close to 180”. The N(l)...N(8) distance is usually greater than 4.0 A. The second, gauche conformation is defined by rl greater than 90” and r2 lower than 90”. For this cgnformer the N(l)...N(S) distance is usually ca. 4.0 A or less. Moreover, the third hydrogen bonded one is characterized by 7, and 72 lower than 90” and the N(1). ..N(8) distance is shorter than 3 A. The convention used for the names of the neutral tautomers is shown in Fig. 1. It is known from previous experiments that in the crystal, histamine molecules exist in the truns (open)
B.V. All rights reserved
A.P. Mazurek et al.Nournal of Molecular Structure 436-437
Fig. 1. The notation of the histamine neutral tautomers:(A) (B) N(3)-H.
conformation of N(l)-H tautomer . Some authors agree that in solution it appears to be equal proportions of truns and gauche conformers of the N(3)-H tautomer [4,5]; whereas some authors state that under physiological pH histamine almost entirely exists as a monocation of N(3)-H tautomer . Parallel to the experimental studies, many calculations by ab initio schemes of varying sophistication have been carried out [6-81. Recently, it is also possible to take into account the effect of the environment, such as a solvent, by the molecular mechanics calculations [9,10], or by the ab initio calculation simulating the water medium [ll] based on the selfconsistent-reaction field method . In Ref. [l l] extensive ab initio calculations were undertaken for histamine tautomers at the MBPT 2/6-31G level in vacuum and water medium and at the MBPT 2/6-316”” level in vacuum. The most stable tautomer in water medium appeared to be the N(3)-H monocation with a strong intramolecular hydrogen bonding between the ethylamine side-chain and the N(1) atom of imidazole ring. In this structure the proton involved in the hydrogen bond is localized closer to the N(8) than to the N(1) atom. Modern NMR spectroscopy has proven to be an exceptionally powerful technique, able to provide the solution of many problems in chemistry and biology. However, the problem of correct assignment of the signals and the understanding of the relationship between the chemical shifts and molecular structure is quite difficult for larger molecules. The ab initio calculations of the NMR chemical shifts are presently accurate enough to analyze and predict the spectra of the medium size molecules. Few methods based on the coupled-Hartree-Fock perturbation-theory equations have been developed and successfully applied to such problems [ 13- 151. Here we use the GIAO-CHF approach (gage-independent
atomic orbital) formulated by Wolinski et al. . The GIAO approach internally extends the basis set with higher angular momentum orbitals, which are necessary for the correct description of the perturbed system. Therefore, calculations with the double-c basis set provide quite good results for organic molecules using the GIAO method [ 16,171. The NMR method is extremely sensitive to small changes in the molecular electronic structures, as the nuclear shielding constant u is determined by the electronic distribution around the nucleus of interest. With the help of two- and three-dimensional NMR methods, the NMR studies of biochemical molecules can often provide important information about the structure and the function of residues. It should become increasingly valuable in developing a general framework for interpretation of the chemical shifts that covers the complicated tautomeric equilibrium of histamine and the influence of environmental effects such as protonation and hydrogen bonding. Development of such a framework requires a more precise knowledge of the chemical shifts of the imidazole ring in each of the individual tautomeric forms. This information could be obtained from theoretical calculations. The aim of this study has been to calculate the ‘H, 13C and “N shielding constants for the two tautomers of histamine and two tautomers of its protonated form found as the most stable species in Ref. [l 11. Here, we provide direct calculations of the chemical shifts of the tautomeric forms of histamine and discuss the implications of the results for the interpretation of the ‘H, 13C and “N chemical shift data for histamine.
The GIAO-CHF approach for the calculations of the NMR chemical shifts was used incorporating the recent improvements of the calculation scheme introduced by Wolit’iski et al. . The basis set we have used for the shielding constant calculations was the DZP CGTO basis of Hansen and Bouman (DZP-HB) . It is composed of (31/l) A0 contracted to [2slp] for the hydrogen atom and (721/221/l) A0 contracted to [3s3pld] for the C, N atoms. This basis set has been previously found to
Fig. 2. The N( I)-H (A, B) and N(3)-H (C, D) forms of neutral and protonated histamine.
be efficient in the chemical shift calculations for the carbon, nitrogen and the oxygen atoms in several molecules [16,17,19,20]. According to Ref. [ 111, we have chosen the most
N( I)-H neutral tautomer MBPT 2/6-3 1G**
R(Nl-C2) R(N 1-C5) R(N 1-H9) R(C2-HIO) R(C2-N3) R(N3-H9) R(C5-C6) R(C5C4) R(C6-C7) R(C7-N8) R(Nl3-Hlh) R(Nl3-H17) cu(CZNlC5) ol(C2NI H9) ol(C5NIH9) cu(C2N3H9) o(C2N3C4) cx(C6C5C4) o(C5C6C7) or(C6C7N8) o(Hl6Nl3Hl7)
stable four tautomers of histamine in vacuum: the neutral and cationic form of the N(l)-H tautomer, and the neutral and cationic form of the N(3)-H tautomer (see Fig. 2). It is known that the chemical shift is a sensitive function of the molecular geometry [21-231. A small variation in intermolecular distances or in the molecular conformations was used to change the chemical shifts. For this reason the chemical shift calculations are usually performed on the experimental geometry. Because for histamine conformers there is no experimental data, we are using theoretically predicted geometries of all species. We have performed two series of calculations. The first one, with MBPT 2/6-3 1G** optimized geometrical parameters is taken from Ref. [ll] and the NMR shielding constants calculated in DZP-HB basis set . The second one with both geometry optimization and shielding constants performed in the DZP-HB basis set. The differences between the results obtained in the first and the second series reflect the influence of the geometry changes on the shielding constants.
Table I Selected bond lengths and bond angles for the N( 1)-H and the N(3)-H tautomers 3 I G** and SCF/DZP-HB Parameters
of histamine calculated
at two levels of theory: MBPT 2/6-
N(3)-H neutral tautomer SCF/DZP-HB
1.3499 1.3714 0.9954 1.071 I I .2906 _
I .3500 I .3733 1.0002 I .0778 I .2907 _
I .4984 I .3562 I .5302 I .4622 1.0018 I .0006
1.4987 I .3560 1.5282 I .4626 I .0082 I .0068
107.2 128.3 124.2
107.1 128.4 124.1 _
105.2 131.9 114.3 110.9 107.1
105.2 131.8 114.4 111.0 106.3
MBPT 216-3 I G**
1.2917 I .3776
1.2918 I .3791
I .0779 I .3453 0.9926 1.5004 I .3547 I .5387 1.4507 I.0015 I.001 I
1.0779 1.3463 0.9976 I .5OOl I .3548 1.5371 I.4515 I .0075 I .0070
106.0 _ _
126.4 106.7 128.2 114.3 116.1 107.1
125.5 106.7 128.2 114.4 116.4 105.9
Table 2 The calculated Atom
A.P. Mazurek et alJJourna1 of Molecular Structure 436-437
c (in ppm) for the N( 1)-H tautomer:
103.8 56.2 -22.5 67.5 63.4 170.4 155.6 242.9 22.5 24.8 25.2 29.7 29.8 29.4 29.7 31.8 31.5
neutral and cation form
Neutral form MBPT2/6-31G** Nl c2 N3 c4 c5 C6 c7 N8 H9 HlO HI1 H12 H13 HI4 H15 H16 H17 H18
SCF/DZP-HB 103.3 59.9 -22.4 67.1 63.2 169.8 154.8 240.1 22.4 24.1 25.1 29.5 29.7 29.3 29.6 31.6 31.3 _
We used the Gaussian 94  code for the geometry optimization procedure based on the Berny optimization method. All geometrical parameters were varied until convergence was obtained as indicated by the changes of the total molecular energy that were smaller than a presumed threshold of 1.0 x 10” Hartree. The calculations were performed on an IBM RISC 6000 and on six-processor SGI Power Challenge computers.
3. Results and discussion In Table 1, we present selected bond lengths and bond angles for the N(l)-H and the N(3)-H neutral tautomers to illustrate the effect of the basis set and method on the geometry parameters. (The optimized (MBPT/6-3 lG** and SCF/DZP-HB) molecular geometries are available from the authors upon request.) As one can see from this data, the optimized geometrical parameters obtained using 6-3 1G** and DZP-HB basis set are very close. The protonation of the NH2 group in the cation species obviously changes the N-H bond lengths, while the other geometrical parameters vary to a smaller extent.
Cation form MBPT 2/6-31G** 108.3 49.3 -28.4 67.1 76.0 173.7 153.6 234.6 24.2 24.2 24.9 28.6 29.5 28.5 28.3 26.3 28.4 27.8
SCF/DZP-HB 107.8 48.9 -27.5 67.3 75.7 173.2 153.5 232.9 24.0 24.1 24.8 28.5 29.3 28.4 28.2 26.1 28.1 27.6
Table 2 and Table 3 present the shielding constants u for all the nuclei in two histamine tautomers in their neutral and cation forms. Let us first to point out here that the influence of geometry on the calculated shielding constants is relatively small. All the computed shielding constants for two sets of geometries of histamine differ by less than 0.5%. Our next step is to analyze the changes in the nitrogen shielding constants due to protonation of the NH2 group. Protonation and deprotonation processes are important in chemistry and biochemistry. Since nitrogenous sites in molecules often play an important role in such processes, “N NMR provides a powerful tool for the investigation of protonationdeprotonation equilibria. First, the nitrogen shielding of amino group NH2 shows some decrease upon protonation for both tautomers: N(l)-H and N(3)-H. Second, the larger effects of protonation of NH2 group are observed for nitrogen atoms in the ring. We noticed that the calculations show clearly much higher shielding of the nitrogen nucleus in the pyrrole-type nitrogen atom of imidazole (N-H) compared with that in the pyridine type nitrogen atom (-N=). This is in agreement with the experiential data where it was found that these two types of nitrogen of imidazole ring differ by ca. 80 ppm [25-271.
A. P. Muzurek
Table 3 The calculated Atom
Nl c2 N3 c4 CS C6 c7 NU H9 HI0 HI1 HI2 HI3 HI4 HIS HI6 HI7 HI8
o (in ppm) for the N(3)-H tautomer:
(I 997) 435-441
neutral and cation form
Neutral form MBPT 2/6-3 1G**
-18.7 54.8 I IS.7 83.2 47.2 163.7 153.5 242.6 24.6 24.8 25.5 30.0 29.8 29.6 23.6 29.8 31.9 _
-20.0 54.5 114.7 82.6 47.4 162.9 152.2 240.7 24.4 24.7 25.4 29.9 29.7 29.4 29.7 29.9 31.8 _
Although the protonation site is on the end of the chain, the shielding constants of both N(1) and N(3) imidazole nitrogen atoms undergo the changes of 5 and 9%, respectively, for the N( 1)-H tautomer, and 20 and 25% for the N(3)-H tautomer. When the protonation of the NH2 group of the N(3)-H tautomer takes place and the pyridine-type nitrogen atom acts as a hydrogen-bond acceptor, one can observe a significant increase of the N(1) atom in agreement with previous findings shielding, [21,22]. Thus the shielding constant for the pyridine type ring nitrogen N(l), acting as a H-bond acceptor in the N(3)-H tautomer, is higher after protonation (by ca. 20 ppm), whereas the shielding constant for the pyrrole type nitrogen, acting as the H-bond donor, is lower by ca. 23 ppm after protonation. When we now compare the pyridine type nitrogen shielding constant in the N(3)-H cation tautomer with the N( 1) shielding constant in the N(l)-H cation tautomer, we find that the protonation effect on the pyridine nitrogen shielding is much stronger than those due to the hydrogen bonding of the pyridine type nitrogen, but both follow the same pattern which is typical for the involvement of the lone-pair electrons of the pyridin type nitrogen atom in the bonding.
Cation form MBPT 2/6-31G** +0.6 45.6 92.9 73.5 51.4 174.3 151.1 225.5 22.8 24.4 23.8 28.8 29.0 28.5 28.6 27.8 27.8 14.8
SCF/DZP-HB -0.4 44.5 90.9 72.2 so. 1 173.1 150.2 223.1 22.6 23.6 24.2 28.6 28.8 28.2 28.3 27.7 27.5 14.2
The shielding constants for all the carbon atoms in the ring are ca. 60 ppm, while u for the chain carbon atoms are ca. 160 ppm. From Tables 2 and 3 it is seen that only three carbon shielding constants for C(2), C(4) and C(5) atoms are strongly dependent on the protonation of NH2 group. After protonation the shielding constants increases for C(5) but decreases for C(2) and C(4) atoms. In the N( 1)-H tautomer the biggest changes (the increase) are on C(5), while in the N(3)-H tautomer on C(2) (the decrease). Our theoretical results presented in Tables 2 and 3 can be directly compared with chemical sift measurements in the gas phase where environment effects are negligible. However, there is no such experimental data available. To our knowledge, direct observation of the tautomeric forms of histamine in the NMR spectra is difficult, and this is why we were obliged to compare our results with the low temperatures NMR spectra for histidine . Histidine contains the imidazole ring and differs from histamine in the side chain by -COOH. The two tautomeric forms of the histidine, which are equivalent to the N( 1)-H and N(3)-H forms of histamine, are the most abundant species in solutions. The chemical shifts of 144.6 ppm for pyridine type ‘5N and of
A.P. Mazurek et al./Joumal of Molecular Structure 436-437
197.9 ppm for pyrrole type 15N were observed in room temperature in the water . As the temperature is lowered to - 55°C the signals appear at 129.1 and 210.8 ppm, and they are expected for an individual tautomeric form of histidine. Nonaqueous solvents have been show to affect the 15N chemical shifts of both the >N-H type and =N: type nitrogens [26,28]. These results agree with our calculations for histamine in two important respects. Firstly, the calculated difference of 15N shielding constant for the N-H type and :N: type atoms of 125 ppm for N(l)-H tautomer and the same difference of 133 ppm for N(3)-H tautomer for isolated species could be compared with that difference of ca. 82 ppm found for histidine in solutions . Secondly, according to our calculations for isolated species the pyridine-type nitrogen atom, acting as the H-bond acceptor, in the N(3)-H tautomer is shifted upfield by 18 ppm, whereas the pyrrole-type nitrogen, acting as the H-bond donor, is shifted downfield by 23 ppm. Experimental chemical shift induced by the H-bond with the imidazole ring nitrogen atom is ca. 10 ppm [26,28]. The magnitude of the experimental data are smaller compared to the calculated values of the shielding constants. It should be noted that our calculated values for the nitrogen atoms at the individual forms of histamine refer only to the isolated histamine tautomers, while the experiment was usually carried out in the aqueous or alcoholic histidine solutions. Both nitrogen types, however, show substantial solvent shifts . Understanding the effects of solvents on the 15N shifts of imidazole ring nitrogens is an area that further investigation. With carefully warrants designed experiments and our calculated data we believe that the chemical shifts are the best parameters for determining tautomeric equilibrium constants for histamine.
higher. The changes caused by protonation in the N(3)-H are much higher than in the N( 1)-H tautomer. Thus, the intramolecular effects in the shielding tensors are known from the present calculations. This is challenging for experiments in the liquid and gas phases for histamine.
Acknowledgements We would like to thank Dr K. Woliriski for the Texas program package for GIAO calculations. This work was partly supported by the grant of Committee for Scientific Research BW-1343/20/96 (to J. Sadlej) and a general grant for the Drug Institute.
References 111C.R. Ganellin, PI [31 [41 1% [61 [71 PI
4. Conclusions With the ab initio GIAO-CHF calculations the reliable values of molecular shielding constants can be obtained for histamine molecules. According to our calculations, protonation of the N(l)-H and the N(3)-H tautomer leads to deshielding of the N(3) nitrogen while the N(1) shielding constants are
M.E. Parsons (Eds.), Pharmacology of Histamine Receptors, Wright PSG, Bristol, 1982, p. 12. C.R. Ganellin, Journal of Pharmarcy and Pharmacology 25 (1973) 787. J.J. Bonnet, J.A. Ibers, Journal of the American Chemical Society 95 (1973) 3429. S.R. Bym, C.W. Grater, S.L. Midloud, Journal of the Organic Chemistry 41 (1976) 2283. B. Vogelsanger, P.G. Godfrey, R.D. Brown, Journal of the American Chemical Society 113 (1991) 7864. S. Topiol, H. Weinstein, R. Osman, Journal of Medicinal Chemistry 27 (1984) 1531. S. Topiol, Journal of Computational Chemistry 8 (1987) 142. (a) Y.G. Romero-Souchez, F.J. Hemandez-Laguna, Journal of Molecular Structure (Theochem) 123 (1985) 431; (b) A. Hemandez-Laguna, Y.G. Smeyers, L.A. Martero, Y.G. Smeyers (Eds.), Trends in Applied Theoretical Chemistry, Kluwer, Dortrecht, 1992; (c) A. Hemandez-Laguna, Z. CruzRodriguez, Y.G. Smeyers, G.A. Arteca, J.L.M. Abboud, 0. Tapie, Journal of Molecular Structure (Theochem) 335 (1995) 77. B. Vogelsanger, P.D. Godfrey, R.D. Brown, Journal of the American Chemical Society 113 (1991) 7864. P.I. Nagy, G.J. Duront, W.P. Hoss, D.A. Smith, Journal of the American Chemical Society 116 (1994) 4898. G. Karpiriska, J. Cz. Dobrowolski, A.P. Mazurek, Journal of Molecular Structure (Theochem) 369 (1996)137. (a) C. Chipot, L.G. Gorb, J.L. Rivail, Journal of Physical Chemistry 98 (1994) 1601; (b) V. Dillet, D. Rinaldi, J.L. Rivail, Journal of Physics and Chemistry 98 (1994)5034. W. Kutzelnigg, U. Fleischer, M. Schindler, NMR. Basic Principles and Progress 23 (1990) 165. D.B. Chesnut, Annual Reports on NMR Spectroscopy 21 (1989) 51.
A.P. Mazurek et al./Joumal of Molecular Structure 436-437 [IS] K. Woliiiski, J.F. Hinton, P. Pulay, Journal of the American Chemical Society 112 (1990) 825 1. [ 161 A. Barszczewicz, M. Jaszutiski, K. Jackowski, Chemical and Physics Letters 203 (1993) 404. [I71 A. Barszczewicz, M. Jaszunski, L. Stefaniak, Chemical and Physics Letters 186 (1991) 313. [ 181 A.E. Hansen, T.D. Bouman, Journal of Chemical Physics 82 (1985) SO35. [ 191 K. Jackowski. A. Barszczewicz, K. Wozniak, Solid State NMR 2 (1993) 265. 1201 M. Pecul, K. Jackowski, K. Wozniak, J. Sadlej, Solid State NMR (1997) (in press).  C.J. Jameson. H.J. Osten, G.A. Webb, Annual Reports on NMR Spectroscopy 17 (1986) I.  D.B. Chesnut, D.W. Wright, Journal of Computational Chemistry 12 (1991) 54.  K. Jackowski, A. Les, Journal of Molecular Structure (Theochem) 33 1 (I 995) 295.
 M.J. Frisch, G.W. Trucks, H.B. Schlegel. P.M.W. Gill, B.G. Johnson, M.A. Robb, J.R. Cheeseman, T.A. Keith, G.A. Petersson, J.A. Montgomery, K. Raghavachari, M.A. AlLaham, V.G. Zakrzewski, J.V. Ortiz, J.B. Foresman, J. Cioslowski, B.B. Stefanov, A. Nanayakkara, M. Challacombe. C.Y. Peng, P.Y. Ayala. W. Chen, M.W. Wong, J.L. Andres. E.S. Replogle, R. Gomperts, R.L. Martin, D.J. Fox, J.S. Binkley, D.J. Defrees, J. Baker, J.P. Stewart, M. HeadGordon, C. Gonzales, J.A. Pople, Gaussian 94 (Revision D.4). Gaussian, Pittsburgh, PA, 1995. [2.5] 1.1. Schuster, J.D. Roberts, Journal of Organic Chemistry 44 (1979) 3864.  H. Sierzputowska-Gracz, M. Wiewiorowski. L. Kozerski, W. van Philipsborn, Nucleic Acids Research 12 (1984) 6247.  M. Witanowski, L. Stefaniak, G.A. Webb, Annual Reports on NMR Spectroscopy 18 (1986) 1.  S. Fair-Jones, W.Y.L. Wong, W.G. Gutheil, W.W. Bachovchin, Journal of the American Chemical Society 1 IS (1993) 6813.