Definitive identification of cysteine and glutathione complexes of bismuth by mass spectrometry: assessing the biochemical fate of bismuth pharmaceutical agents

Abstract

Solutions containing BiCl₃, bismuth subsalicylate or Bi(NO₃)₃ with L-cysteine, DL-homocysteine, D-methionine or glutathione have been examined by electrospray mass spectrometry. Prominent peaks are assigned to bismuth complexes of these biomolecules and provide insight towards understanding the bioactivity of bismuth compounds.

---

Despite a long history of medicinal applications for many bismuth compounds, the mechanisms of bioactivity are not understood.^1–4 The thiophilicity of bismuth⁵ has prompted speculation that sulfur-based biomolecules represent the primary target for pharmaceuticals such as ‘colloidal bismuth subcitrate’ (CBS) and ‘bismuth subsalicylate’ (BSS). Bismuth complexes of such ligands have been characterised by ¹³C NMR spectroscopy^6–11 and X-ray absorption spectroscopy,¹² but structural assignments and formulations are speculative, as the compounds are difficult to isolate. Electrospray mass spectral (ESI-MS) data provide definitive assignments for fragments and molecular species arising from solutions of such systems,¹³ as demonstrated below (summarised in Table 1) for solutions containing BiCl₃, BSS or Bi(NO₃)₃ with L-cysteine (CYS), DL-homocysteine (HCYS) and glutathione (GSH).†

Table 1 Prominent cation peaks (m/z), relative peak intensities (%) and assignments observed for in situ ESI-MS of reaction mixtures containing bismuth salts with CYS, HCYS or GSH (pH 1–4)

m/z	Assignment	Relative Int. (%)	MS/MS m/z	Assignment
CYS
449	1	45	328	2
328	2	100	241	BiS
			209	Bi
241	BiS	40
209	Bi	10
HCYS
818	3	5	342	5
477	4	100	342	5
			209	Bi
342	5	55	241	BiS
			209	Bi
209	Bi	35
GSH
821	6	30	514	4
514	7	100	385	5
385	8	35	209	Bi
308	HGSH	90

---

Irrespective of reaction stoichiometry (1∶1, 1∶3, 1∶5), reaction mixtures involving CYS show prominent peaks at m/z 449 and 328 (Fig. 1), that are assigned to the monocationic bismuth complexes 1 and 2,‡ respectively. Formation of the bisthiolate complex (1) is consistent with previously reported solid state structures for aminothiolate derivatives.¹⁴ Cation 2 can be described as a bismuth complex containing the dianionic conjugate base of CYS and is modelled on the crystallographically characterised dimethyl derivative, penicillaminatobismuth chloride.¹⁵ In addition, tandem mass spectra of m/z 445 (1) gave 328 (2), and m/z 328 (2) gave m/z 241 (BiS⁺) and 209 (Bi⁺).

Fig. 1 A representative ESI-MS of a solution containing BSS and CYS.

ESI-MS of reaction mixtures containing BiCl₃ or BSS and HCYS in aqueous solution, shown in Fig. 2, contain peaks at m/z 818, 477 and 342, which correspond to the tethered dibismuth cation (3) as well as the bis- (4) and mono-thiolate (5) complexes, respectively. ESI-MS/MS of ion 3 (m/z 818) gives a fragment at m/z 342 (5), m/z 445 (4) fragments to give m/z 342 (5) and m/z 209 (Bi⁺), and m/z 342 (5) gives 241 (BiS⁺) and m/z 209 (Bi⁺).

Fig. 2 A representative ESI-MS of a solution containing BSS and HCYS.

ESI-MS of solutions containing BSS or Bi(NO₃)₃ with GSH (Fig. 3) reveal analogous complexes to those observed for the bismuth–cysteine mixtures, in contrast to observations for the antimony–glutathione system.¹⁶ These observations were also independent of stoichiometry (1∶1 or 1∶2). Peaks at m/z 821, 514 and 385 are assigned to 6, 7, and 8, respectively. ESI-MS/MS of cation 6 gives fragments at m/z 514 (7), which releases glutamic acid to give m/z 385 (8). The same cations are observed using a MALDI source.§

Fig. 3 A representative ESI-MS of a solution containing Bi(NO₃)₃ and GSH.

Bismuth–methionine complexes could not be observed in solutions of BiCl₃, Bi(NO₃)₃ or BSS with D-methionine, despite previous reports of such complexes,¹⁷ suggesting the importance of the thiolate anchor in the formation of bismuth complexes.

ESI-MS of reaction mixtures provide definitive identification data for bismuth complexes involving CYS, HCYS and GSH. The proposed formulations are important in the quest to understand the bioactivity of bismuth compounds,^15,18,19 and are consistent with the established series of tris- 9, bis- 10 and mono- 11 thiolate complexes that have been previously isolated and comprehensively characterised.^14,20,21

The interactions of bismuth with weakly donating functional groups of the hetero-bifunctional ligands in these complexes are made possible by the thermal and hydrolytic stability of the sulfur–bismuth bond which serves as an anchor for the ligand.^5,21 In this context, the observations described above provide support for the thiolation of bismuth as the primary biochemical fate of bismuth pharmaceuticals.

We thank the Natural Sciences and Engineering Research Council of Canada, MDS Sciex, the Killam Foundation, the Canada Research Chairs Program, the Canada Foundation for Innovation and the Nova Scotia Research and Innovation Trust Fund for funding, and the Dalhousie Mass Spectrometry Laboratory and MDS Sciex for use of instrumentation.

Notes and references

1. P. J. Sadler, H. Li and H. Sun, Coord. Chem. Rev., 1999, 185–186, 689–709.
2. P. J. Sadler and Z. Guo, Pure Appl. Chem., 1998, 70, 863–871.
3. G. G. Briand and N. Burford, Chem. Rev., 1999, 99, 2601–2657.
4. J. Reglinski, Chemistry of Arsenic, Antimony and Bismuth, ed. N. C. Norman, Blackie Academic & Professional, London, 1998, pp. 403–440.
5. G. G. Briand and N. Burford, Adv. Inorg. Chem., 2000, 50, 285–357.
6. G. Alonzo, N. Bertazzi and M. Consiglio, Inorg. Chim. Acta, 1984, 85, L35–L37.
7. H. Sun, H. Li, A. B. Mason, R. C. Woodworth and P. J. Sadler, J. Biol. Chem., 2001, 276, 8829–8835.
8. L. Zhang, K. Y. Szeto, W. B. Wong, T. T. Loh, P. J. Sadler and H. Sun, Biochemistry, 2001, 40, 13281–13287.
9. A. Napoli, Ann. Chim., 1982, 72, 575–583.
10. H. Sun, H. Li and P. J. Sadler, J. Inorg. Biochem., 1995, 59, 190.
11. P. J. Sadler, H. Sun and H. Li, Chem. Eur. J., 1996, 2, 701–708.
12. H. Sun, H. Li, I. Harvey and P. J. Sadler, J. Biol. Chem., 1999, 274, 29094–29101.
13. N. Burford, M. D. Eelman and T. S. Cameron, Chem. Commun., 2002, 1402–1403.
14. G. G. Briand, N. Burford, T. S. Cameron and W. Kwiatkowski, J. Am. Chem. Soc., 1998, 120, 11374–11379.
15. W. A. Herrmann, E. Herdtweck and L. Pajdla, Chem. Ber., 1993, 126, 895–898.
16. H. Sun, S. C. Yan and W. S. Cheng, Eur. J. Biochem., 2000, 267, 5450–5457.
17. C. A. McAuliffe, J. V. Quagliano and L. M. Vallarino, Inorg. Chem., 1966, 5, 1996–2003.
18. P. V. Radheshwar, R. Dev and G. H. Cady, J. Inorg. Nucl. Chem., 1972, 34, 3913–3915.
19. S. P. Summers, K. A. Abboud, S. R. Farrah and G. J. Palenik, Inorg. Chem., 1994, 33, 88–92.
20. G. G Briand, N. Burford and T. S. Cameron, Chem. Commun., 2000, 13–14.
21. L. Agocs, G. G. Briand, N. Burford, T. S. Cameron, W. Kwiatkowski and K. N. Robertson, Inorg. Chem., 1997, 36, 2855–2860.

---

Footnotes

† Samples were injected directly at a flow rate of 5 μl min⁻¹ into the electrospray source of a VG Micromass Quattro triple quadrupole mass spectrometer or an API 3000 PE Sciex triple quadrupole mass spectrometer, with a source temperature of 385 K and skimmer cone voltage of 50 V. MS/MS spectra were obtained using argon as a collision gas with a collision energy equal to 50 eV. The argon pressure was sufficient to reduce the intensity of the main beam by 1%. BSS (2.93 mmol) and CYS (13.5 mmol) [or BiCl₃ (3.52 mmol) and CYS (10.6 mmol) or CYS (3.49 mmol)] in distilled water (100 mL) stirred overnight at RT and suction filtered. BSS (3.88 mmol) and HCYS (3.91 mmol) or HCYS (6.05 mmol) [or BiCl₃ (3.52 mmol) and HCYS (10.6 mmol)] in distilled water (150 mL) stirred for 6 h and suction filtered. Bi(NO₃)₃ (3.63 mmol) and GSH (3.71 mmol) or GSH (7.52 mmol) [or BSS (3.20 mmol) and GSH (3.11 mmol)] in distilled water (150 mL) stirred for 2 h and suction filtered. BSS (3.01 mmol) and D-methionine (2.99 mmol) [or BiCl₃ (3.21 mmol) or Bi(NO₃)₃ (2.86 mmol) in distilled water, stirred for 3 h and suction filtered.

‡ Molecular drawings represent monocationic species and illustrate connectivity only; drawings of these complexes aimed at describing bonding features (e.g. Lewis) are not meaningful or are misleading.

§ MALDI MS was performed on a Micromass MALDI LR mass spectrometer using 0.01 g of α-cyano-4-hydroxycinnamic acid in a 50/50 mixture of acetonitrile and ethanol as the matrix.

---