This work has been digitalized and published in 2013 by Verlag Zeitschrift für Naturforschung in cooperation with the Max Planck Society for the Advancement of Science under a Creative Commons Attribution4.0 International License.

Dieses Werk wurde im Jahr 2013 vom Verlag Zeitschrift für Naturforschungin Zusammenarbeit mit der Max-Planck-Gesellschaft zur Förderung derWissenschaften e.V. digitalisiert und unter folgender Lizenz veröffentlicht:Creative Commons Namensnennung 4.0 Lizenz.

Mono- and Binuclear Gold(I) Compounds Containing

Deprotonated Purines and Pyrimidines

Flavio Bonati, Alfredo Burini, and Bianca Rosa Pietroni*

Dipartimento di Scienze Chimiche dell’Universitä, Via S. Agostino 1, 1-62032 Camerino

Z. Naturforsch. 40b, 1749-1752 (1985); received June 18, 1985

Deprotonated Purines, Pyrimidines, Mono- and Binuclear Gold(I) Compounds

In the presence of alkali and LAuCl one or two —N H — groups of a purine or pyrimidine base can be transformed into -N (AuL)- groups giving stable and soluble compounds containing one or two two-coordinated gold(I) nuclei; mononuclear L A u-Q ' or binuclear LA u-Q -A uL with a monodentate Q " or an exobidentate O '" ligand, resp. (Q 'H = adenine, guanine, theobromine, theophylline, azaguanine or cytosine; Q H 2 - thymine or uracyl; L = triphenylphosphine).

Introduction

Much interest has recently been dedicated to the

chemistry of gold since this element in the oxidation

state + 1 has been found to provide compounds, such as

(2,3,4,6-tetra-O-acetyl-l-thio-ß-glucopyranosato-S)-

(triethylphosphine)gold(I) or triethylphosphine-

chlorogold(I), successfully used in the treatment of

rheumatoid arthritis [1]. The mode of action of this

or of other drugs actually used or undergoing clinical

tests of gold containing drugs [2] is extremely com­

plex, and many more investigations will be required.

Our investigations in the field of gold chemistry

have been concerned with the interaction between

Au(I) and several nitrogen-containing heterocycles

such as pyrazoles [3], pyrazolones [4], indazoles,

imidazoles [5], triazoles, tetrazole [6], or pyridones

[7]. Isolation and characterization of several N- and

of a few C-(trisubstituted phosphine)gold(I) deriva­

tives of the said molecules was reported. These were

found to be stable and soluble in organic solvents,

thus allowing spectroscopic investigations, whereas

the uncomplexed gold(I) derivatives are often insolu­

ble coordination polymers or, at best, oligomers,

e.g.: l-gold-2-R-imidazole (R = /so-proypl [8] or

phenyl [9]) and tris(pyrazolato-N,N')trigold(I) [10],

respectively.

It was therefore decided to react with gold(I) cer­

tain purine and pyrimidine bases of relevance to

biological systems. Here we report the isolation and

the characterization of stable and generally soluble

mono- or bis-aurated derivatives of purines or py­

rimidines where gold(I)-nitrogen bonds are present.

* Reprint requests to Dr. B. R. Pietroni.

Verlag der Zeitschrift für Naturforschung, D-7400 Tübingen0340 - 5087/85/1200 -1749/$ 01.00/0

Results and Discussion

The reaction between a purine or pyrimidine base

(Q 'H or QH2), LAuCl and alkali was carried out in

homogeneous or heterogeneous phase, as detailed in

the experimental part, according to one of the fol­

lowing patterns:

Q '-H + O H “ + LAuCl =

Q '-A uL + CP + H20

1, 4-8

Q - H 2 + 2 O H ' + 2 LAuCl =

Q(AuL )2 + 2 c r + 2 h 2o

2,3

The colourless, moisture- and air-stable solids

were characterized through elemental analyses, in­

frared and proton NMR spectra (Tables I —III) .

These data confirmed that in several cases molecules

of solvents can be clathrated in the solids, a frequent

observation in the chemistry of the heterocyclic

derivatives of gold e.g.: l-(triphenylphosphinegold)-

6-methyl-2-pyridone [7] or l-tri(cyc/o-hexyl)phos-

phinegold-2-/so-propylimidazole [8]. The ligand L is

triphenylphosphine for several reasons: the starting

compound LAuCl is easily obtained and handled,

the corresponding products are generally soluble in

organic solvents, and the A u—P bond is not as easily

broken as, for example, the A u—As or the A u—Sb

bond of similar compounds having triphenylarsine or

triphenylstibine in the place of L [8, 11].

The reactivity of the molecules investigated is not

related to the number of nitrogen-bonded hydrogen

atoms available for the substitution by the LAu

group; in all the cases but two only mono-auration is

observed. Since no C-auration is observerd, even if

only one NH is present — as with theophylline or

theobromine (Q 'H) — a mono-aurated compound

1750 F. Bonati et al. ■ Mono- and Binuclear Gold(I) Compounds

Table I. Analytical and other data for compounds 1—8a. nh2 M ? m m

N ^ l p u . u n %N r CHru n u n u . 1Com- Meth- Yield m.p. Elemental analyses [%]b J 2 c f=H6 H2° J ch2c i2 h 2o i J c 6h6 2pound od [%] [°-C] C H N

1 A 73 227-229 48.32 3.94 6.31

2 A 77 258-260(47.93)44.49

(3.86)3.22

(6.71)2.59

3 A 79 268-270(44.04)48.78

(3.34)3.72

(2.44)2.40

4a A 72 271-273(48.73)45.70

(3.64)3.18

(2.47)11.80

4b B 51 271-273(45.86)45.83

(3.35)3.52

(11.63)11.47

5a A 42 >350(45.86)44.39

(3.35)3.11

(11.63)10.99

5b B 61 >350(44.67)45.02

(3.26)3.33

(11.32)11.47

6 A 73 320-321(44.67)42.74

(3.26)3.14

(11.32)13.70

7 B 82 267-268(42.66)46.78

(3.09)3.48

(13.57)8.56

8 B 78 273-274(47.03) 47.46(47.03)

(3.47) 3.53

(3.47)

(8.77) 8.98

(8.77)

a The methods A and B are described in the experimen­tal part; b calculated values in brackets.

M M M

1 2

NHj 0

--------N u r\ ^ l l -------- N 1L 1 j l ' 2 H2 ° L J L I] ' Y H2 °

H ,N N N I 2 I

M rl

4a,4b 5a, 5b

oM

xn' 'n— nL jl II '7 n2L

M 'NN H

6

9 CH3 H3c j?N '^ r , ----N ---- N'J I J J II J

0 0 N^NI I

C H j CH3

7 8

M = AuPPh,

Table II. Proton nuclear magnetic resonance data for compound 1—8ab.

Compound Aromatic protons Other protons

1 7.90-7.10 m [19]; 5.55 d [1] j = 7 6.25-6.35 s, br [2]d; 3.65 s [2]d2 7.82-7.42 m [30]; 7.30 s [1] 5.75 s [2]d; 3.30 s [2]d; 1.72 s [3]3 7.90-7.30 m [37]; 5.45 d [1] j - 7 3.30 s [3]d4a, 4b 8.09 s [1]; 7.90 s [1]; 7.90-7.45 m [15] 6.70 s [2]d; 3.35 s [l]d5a, 5bc 7.90-7.30 m [16] 6.38 s [2]d; 3.35 s, br [2]d6 8-7.30 m [15] 10.5 s [l]d; 6.38 s [2]d; 3.32 s [l]d7 8-7.40 m [16] 3.92 s [3]; 3.44 s [3]8 7.90-7.30 m [16] 3.63 s [3]; 3.47 s [3]

a The data were recorded on a Varian instrument operating at 90 MHz using TMS as reference; they are in 6 units. Coupling constants in Hz; s = singlet, d = doublet, m = multiplet; [ ] denotes relative intensities; b the spectra were recorded in (CD3)2SO solution in all the cases but 8 (CDC13); 0 very little soluble in (CD 3)2SO; d disappears upon deuteration.

Compound v(OH; NH) 1500—1700 cm 1 region

1 3390 w, br 1620 s; 1595 m; 1585 m; 1570 s; 1565 s; 1510 m2 not evident 1642w; 1620s; 1610s; 1583w; 1572m; 1562m; 1595s3 not evident 1630 s; 1600 w; 1590 s; 1580 w; 1540 s4a, 4b 3660-3050 s, br 1630 s; 1590 s; 1550 m5a, 5b 3680-2500 s, br 1665 s; 1610 s; 1550 m6 3320 s, br; 3160 s, br 1685 s; 1670 s; 1615 m; 1560 w; 1529 w7 - 1660 s; 1650 s; 1625 s; 1585 m; 1545 m; 1530 m8 - 1690 s; 1640 s; 1530 s

Table III. dataa.

Selected infrared

Recorded as Nujol mull in the NaCl region with a Perkin- Elmer 297 instrument.

F. Bonati et al. • Mono- and Binuclear G o ld (I) Compounds 1751

Q' —AuL, 8 or 7, was obtained. The assignment gi­

ven in the Figure is supported by the disappearance

of the NH absorptions and the shifting of the C = 0

vibrations in the infrared spectra as well as by the

presence of two methyl and of one =C H — signal in

the NMR spectra. In the case of the theophyllinato

complex the gold atom is assumed to be in the less

hindered 7-position rather than in the 9-position,

where it would be flanked by the 3-methyl group. In

the first preparation of this theophyllinato complex

the NMR spectrum of the crude product showed the

presence of both isomers (the 9-isomer being ca. 1/4),

but the following preparations failed to yield the

same result.

In the case of uracyl and of its C-methylated

homologue, thymine, two LAu replace the hydrogen

in both the NH groups, yielding 3 and 2, bulky

molecules where water and benzene or water and

dichloromethane are clathrated. Here, too, disap­

pearance of NH bands, shifting of the C = 0 infrared

vibrations, and the presence of the required number

of methyl and CH signals rule out C-auration and are

in agreement with the formulae given in the figure.

Many structures are possible, in principle, for the

monoaurated derivatives of adenine, guanine, aza-

guanine, or cytosine (Q'H). For these purines sever­

al tautomeric forms [12] may be written in each of

which a nitrogen bonded hydrogen may be replaced

by an LAu group. Actually, in the said bases only

mono-aurated derivatives have been isolated for

which the NMR spectra show an NH2 group. This

spectral evidence (together with the presence of a

C = 0 stretching group where required) rules out the

possibility of a substitution on the 1-position of cy­

tosine and adenine and on the 3-position of guanine

and azaguanine. On this basis cytosine must be a 3-

aurated derivative, while in the case of adenine two

positions are still available, N(7) and N(9), the latter

being preferred because of less steric hindrance [12].

Three more types of substitutions are possible in the

case of guanine and of azaguanine (N (l), N(7) or

N(9)): they are all equally possible on the data avail­

able. Therefore, in the absence of an X-ray crystal

structure determination, no definitive formula can be

assigned to 5 or 6, although in the Figure they are

depicted as 1-substituted derivatives because this is

the most acidic position available [13].

In conclusion, one or two LA u— moieties can be

attached to a purine or pyrimidine giving stable and

soluble compounds with A u—N bonds and two-coor­

dinated Au(I) where the corresponding deproto-

nated nitrogen base may behave as a monodentate or

as an exobidentate (i.e. bidentate and bridging)

ligand. Previously only a few adducts of gold(I)

chloride with purine or nucleoside bases were re­

ported: insoluble Au(nucl)2Cl (nucl is guanosine or

inosine) [14], to which a two-coordinated

N—>Au —Cl arrangement was assigned [15], or the

adduct (cytosine)AuCl, investigated only in solution

by proton NMR [16]. Our results show that the in­

teraction between gold(I) and purine or pyrimidine

bases includes auration and is not limited to adduct

formation.

Experimental

Elemental analyses were performed by our Mi- croanalytical Laboratory (Perkin Elmer 240 Instru­ment) or by Mr. A. Canu (University of Sassari). Evaporation was always carried out under reduced

pressure.

Method A

To a methanol solution (20 ml) of cytosine (0.150 g; 1.35 mM), 1% sodium hydroxide (5.40 ml; 1.35 mM) in the same solvent and Ph3PAuCl (0.735 g; 1.48 mM) were added. After 8 h stirring at 40 °C the unreacted substrate was filtered, the solu­tion evaporated to dryness, and the residue washed repeatedly with benzene, hexane and water to give the analytical sample 1 .

In the same way were prepared compounds 5 a and6 using as starting materials guanine and azaguanine, respectively. The compound 6 precipitated from methanol solution and was purified by several wash­ings with water, methanol, benzene, and dichloro­methane. The compound 2 was obtained by reaction

of thymine at r.t. and cristallization from CH2C12/ hexane of the residue left after evaporation. The

compound 3 was obtained by reaction of uracyl at r.t.; the residue after evaporation was extracted with CH2C12 and the evaporated residue was dissolved in benzene from which clathrated compound 3 precipi­tated nearly at once. The compound 4a was obtained as a precipitate from adenine at r.t. and was purified by means of several washings with water, methanol

and then hexane.

Method B

A dichloromethane solution (20 ml) of theo­bromine (0.150 g; 0.83 mM) and Ph3PAuCl (0.453 g;

1752 F. Bonati et al. • Mono- and Binuclear Gold(I) Compounds

0.92 mM) was added to an ice-cold suspension of tetra-n-butylammonium hydrogen sulphate (0.283 g;

0.83 mM) in 0.83 ml of 2 N aqueous sodium hydro­xide. After 4 h stirring at r.t. the organic layer was

separated, washed with water till neutral washing, dried over sodium sulphate and evaporated to dry­

ness. The residue was washed several times with ben­

zene giving the analytical sample 7.The compound 8 was prepared as above starting

from theophylline and Ph3PAuCl. The residue was purified by cristallization from benzene/hexane.

The compounds 4 b and 5 b were obtained in the

same way starting from adenine on guanine and Ph3PAuCl, resp. The first product was cristallized

from dichloromethane/hexane and the former, insol­

uble in dichloromethane, were filtered and purified by several washings with dichloromethane and

hexane.

We thank the “Ministero della Pubblica Is- truzione” and the “Consiglio Nazionale delle Ricerche” for financial support.

Note added in proof: After this work was accepted,

a paper became available to us in which several pal­

lad ium ^), one rhodium(I), and some gold(I) deriva­tives were reported (Y. Rosopulos, U. Nagel, and W.

Beck, Chem. Ber. 118,931 (1985)). Amongst all these complexes only two are in common with those reported by us, though prepared in a different way. They are 1,3- bis(triphenylphosphinegold)uracyl and (triphenyl- phosphine)(adeninato-N(9))gold(I) for which an X-

ray crystal structure is also reported. The results of the German group agree fully with ours, and complete a

fieW of investigation where only scattered results had

been available.

[1] a) R. J. Puddephatt, “The Chemistry of Gold”; Elsevier, Amsterdam, The Netherlands 1978, pages 248-253;b) D. H. Brown and W. E. Smith, Chem. Soc. Rev. 9, 217 (1980); B. M. Sutton and R. G. Franz, Proceed­ings of a Symposium “Bioinorganic Chemistry of Gold Coordination Compounds”, Nov. 16—17, 1981, Philadelphia, Penn., Smith, Kline and French Laboratories, 1983; K. C. Dash and H. Schmidbaur, in H. Sigel (ed.): Metal Ions in Biological Systems, Vol. 14, 179 (1982).

[2] S. J. Lippard (ed.): “Platinum, Gold, and other Metal Chemotherapeutic Agents” , ACS Symposium Series 209, American Chemical Society, Washington D .C., U.S.A. 1983.

[3] F. Bonati, Chim. Ind. (Milan) 62, 323 (1980); G. Min- ghetti, G. Banditelli, and F. Bonati, Inorg. Chem. 18, 658 (1979); F. Bonati, G. Minghetti, and G. Banditel­li, J. Chem. Soc., Chem. Commun. 1974, 88 ; G. Ban­ditelli, A. L. Bandini, G. Minghetti, and F. Bonati. Can. J. Chem. 59, 1241 (1981); J. R. Lechat, R. H. de Almeida Santos, G. Banditelli, and F. Bonati, Cryst. Struct. Commun. 11, 471 (1982).

[4] F. Bonati, A. Burini, B. R. Pietroni, and M. Felici. J. Organomet. Chem. 274, 275 (1984).

[5] F. Bonati, M. Felici, B. R. Pietroni, and A. Burini, Gazz. Chim. Ital. 112, 5 (1982).

[6] F. Bonati, A. Burini, M. Felici, and B. R. Pietroni, Gazz. Chim. Ital. 113, 105 (1983).

[7] F. Bonati, A. Burini, B. R. Pietroni, and B. Bovio, J. Organomet. Chem., accepted.

[8] B. Bovio, F. Bonati, A. Burini. and B. R. Pietroni, Z. Naturforsch. 39b, 1747 (1984).

[9] D. Leonesi, A. Lorenzotti, A. Cingolani, and F. Bonati, Gazz. Chim. Ital. I l l , 483 (1981).

[10] B. Bovio, F. Bonati, and G. Banditelli, Inorg. Chim. Acta 87, 25 (1984).

[11] Ref. [la], page 54.[12] D. J. Hodgson. Progr. Inorg. Chem. 23, 211 (1977).[13] Ref. [12], p. 221; T. J. Kistenmaker, Acta Crystallogr.

B30, 1610 (1974); T. J. Kistenmaker, L. G. Marzilli, and C. H. Chang, J. Am. Chem. Soc. 95, 5817 (1973).

[14] N. Hadjiliadis, G. Pneumaticakis, and R. Basosi, J. Inorg. Biochem. 14, 115 (1981).

[15] G. H. M. Calis and N. Hadjiliadis. Inorg. Chim. Acta 79, 241 (1983); ibid. 91, 203 (1984).

[16] M. Bressan, R. Ettorre, and P. Rigo, J. Magn. Reson. 26, 43 (1977).