Spectroscopy of Anticancer Ruthenium Complexes and Transferrin

August 2006 Spectroscopic Analysis of Three Anticancer Ruthenium (II) Complexes and Transferrin: No evidence for Specific or Non-Specific Binding Introduction Research on ruthenium-complexes has grown over the years as scientists become aware of potential applications in a number of innovative fields. Chief among the attractive qualities of several categories of Ru-complexes are their promising photochemical properties and their usefulness as fluorescent probes (see Appendix). No less intriguing are the anticancer properties shown by many ruthenium(III) and a number of ruthenium(II) complexes synthesized over the past two decades1,2,3. Representative Ru(III) complexes studied in vivo exhibit inhibition of DNA replication, reduction of RNA synthesis, mutagenic activity, and induction of the SOS repair mechanism3. In the case of many Ru(III) complexes, the “activation by reduction” mechanism is invoked in which the inert complexes are reduced in situ to ruthenium(II), the species thought to be responsible for direct attack on DNA nucleobases3. Renzo Cini at the University of Siena has synthesized three novel ruthenium(II) complexes (Figure 1) and sent them to our lab for spectroscopic analysis. In addition to the ruthenium oxidation state less encountered in literature on anticancer Ru-complexes, Fig. 1: Ru-847.9 Ru-866.8 Cis-[Ru(PPh3)2(TPYM)2] PPh3= triphenylphosphine, TPYM= 2-thio-1,3,pyrimidine. Molecular Weight= 847.9 1 (“Ru-847.9”) HO CH 2 OH O Trans, trans, trans-[RuCl2(PPh3)2(THZ)2] THZ= thiazole-1,3. MW= 866.8 4 (“Ru-866.8”) OH N N Ph 3 P S NH N Ru 2+ S • PPh 3 N HN O CH 2 N 2 Cl• 2.75 H2O OH N HO OH Cis-Ru(PPh3)2(HTPR)2]Cl2x2.75H2O: **same ligands as the above diagram but with the cis octaherdral arrangement rather than the depicted trans configuration HTPR=6-thiopurine-riboside MW=1314.7 5 (“Ru-1314.7”) 2 these complexes are especially interesting because the ligands themselves have medicinal properties. Phosphines have shown selective cytotoxic and anticancer properties1. Along with antiviral properties, thiopyrimidines possess photochemical properties that may prove useful in the design of new photodynamic cancer therapies1. Thiopurines are currently used as antileukemic and antiviral agents1. Likewise, thiazoles are present in a number of active drugs4. Complexation of anticancer drugs to a metal atom is beneficial for a number of reasons. The inertness of certain metal-ligand linkages protects the ligand drugs against enzymatic degradation1. Secondly, the metal complex may have a better hydrophobicity or hydrophilicity than the free ligands1. Thirdly, the ligands’ activity can be synergistically reinforced by the metal’s own anticancer properties1,4. Lastly, the metal complex has the potential to release the active drug ligands in a target organ through several biological processes1. I. UV/Vis Absorption of the Ruthenium(II) Complexes All three Ru(II) compounds display signature visible and/or ultraviolet absorption peaks. Ru-847.9 dissolved in chloroform or a buffer solution with 15% DMSO shows a well-defined peak in the visible range at 333 nm or 327 nm respectively. In a buffer solution with less DMSO (4%), a corresponding peak becomes somewhat obscured, yet is nonetheless discernable (Figure 2). This peak at approximately 330 nm lies in the typical range of ligand to metal charge transfers (LMCT) found in other octahedral and nonoctahedral ruthenium complexes2,6,7,8,9, and therefore can be attributed to such. Figure 2 Ru-847.9 in various solvents displays an absorption maximum at ~330 nm. --- 56.4 μM in chloroform. Blank: chloroform. --- 20.0 μM in buffer solution and DMSO (15%). Blank: buffer solution and DMSO (15%). ---20.0 μM in buffer solution and DMSO (4%). Blank: buffer solution and DMSO (4%). 0.2 cm cuvettes buffer solution: (25 mM Tris-base and 10 mM NaHCO3 in deionized-distilled H2O, pH 7.4) Similarly, the peak at approximately 378 nm for Ru-1314.7 (Figure 3) is likely a ligand to metal charge transfer band. Ru-866.8’s absorption peak is at a significantly 2 3 shorter wavelength (at approximately 280 nm, Figure 3), and is probably instead due to ligand absorption, or some combination of metal and ligand absorption that cannot be determined without additional spectroscopic study. 276.02 nm 281.36 nm Figure 3 378.02 nm Ru-866.8 and Ru1314.7 --- 20 µM Ru-866.8 in buffer solution and DMSO (4%) --- 20 µM Ru-1314.7 in buffer solution and DMS0 (4%) Blank: buffer solution and DMSO (4%) 0.2 cm cuvettes buffer solution: same as noted in Fig. 2 II. Transferrin Increased selectivity of a drug reduces systemic toxicity by keeping a high, localized Figure 4 concentration of the drug at target tissue but a low 10 concentration in other regions . The linkage of the drug to biological macromolecules that possess high selectivity for well-defined biological targets is a means to appropriate drug-targeting. The iron(III)-transport protein transferrin has been looked to as a possible transport vehicle for ruthenium-based drugs because of its abundance in the plasma and because of ruthenium’s location just below iron on the periodic table. Human serum transferrin, a glycoprotein of 80 kilodaltons, binds ferric ions and transports them to the hemoglobinsynthesizing immature red blood cells11 or to the liver for storage as ferritin and hemosiderin12. A bicarbonate or carbonate ion is necessary to promote the hinge-like closing of the protein domains around the metal ions13. Preliminary XMonoferric transferrin with ferric ray crystallographic studies12 and amino acid 14 and carbonate ions16 sequencing indicate that the protein is a single polypeptide chain composed of two homologous domains each containing a metal3 4 binding site. Each metal site contains a distorted octahedron with two tyrosines, one histidine, one aspartate, and the synergistic, bidentate bicarbonate anion as ligands13. Despite similarity in overall structure and function, the two sites have been shown to differ in kinetics and pH dependence of binding, thermal stability, iron-donating properties and other characteristics12,11. Most trivalent metals, Ru3+ included15, and a few divalent metals bind specifically to the iron-binding sites of the apoprotein12. Transferrin enters endosomic compartments within the cell via the transferrin cycle, and could thus theoretically usher the ruthenium-based drugs across the cell membrane into an acidic compartment where the low pH would favor disassociation from the protein10. The elevated nutrient requirements, higher membrane permeability, and increased angiogenesis of tumor cells17, coupled with the disproportionately large number of transferrin receptors on the surface of quickly proliferating tumor cells2,10,18, could result in selective, elevated uptake of the drug-protein adducts by these target tissues. Apo-transferrin’s distinguishing feature in the ultraviolet and visible light range is an absorption peak at 280nm with a shoulder at 290 nm. Holo-transferrin (diferric transferrin) presents a similar curve over this ultraviolet range, yet is unique in its visible light absorption peak centered on ~465 nm (Figure 5). This latter peak of holo-Tf is assigned to a phenolate to iron(III) charge transfer19. 279.41 nm Figure 5 467.07 nm Holo-Transferrin vs. Apo-Transferrin Blanks: buffer solution --- Holo-transferrin (diferric transferrin) purchased from Sigma: 10 µM in buffer --- Apo-Transferrin purchased from Sigma: 10 µM in buffer 0.2 cm cuvettes buffer solution: same as noted in Fig. 2 Two spectroscopic signals indicate the occupancy of the two metal-binding sites of transferrin by ferric ions. The first is the increase in absorption at 280 nm in going from apotransferrin [A280 (1%) = 10.9]11 to holo-transferrin [A280 (1%) = 14.0]11 (Figure 4 5 5). When iron binds to transferrin, a conformational change occurs in which amino acid ligands from the three different regions of the protein become coordinated to the metal20. The second piece of evidence indicating specific metal binding is the LMCT absorption peak at ~465 nm seen in the holo-transferrin spectrum. A corresponding absorption peak can be induced by adding an appropriate source of ferric ions to an apotransferrin solution. For example, a 1.5 mM solution of Fe(NTA)2 of pH 4.0 was made in line with previously published results11. The addition of two molar equivalents of this Fe(NTA)2 solution to apo-transferrin in the presence of bicarbonate ion produces an absorption peak analogous to that of holo-transferrin (Figure 6). Figure 6 Diferric Transferrin vs. Apo-Transferrin Blanks: Buffer solution as in Figure 2 --- Holo-Tf purchased from Sigma: 20 µM protein in buffer solution ---Diferric Transferrin: 40 µM Fe(NTA)2 plus 20 µM apo-Tf in buffer solution ---Apo-Tf: 40 µM protein in buffer solution 1 cm masked cuvettes 1 1 Fe(NTA)2 was prepared by dissolving dissolving 15.0 µmol iron filings in 5 mL of 6.0 M HCl over night. A small volume of this iron solution (0.77 mL) was added to 3.0 mL nitrilotriacetate solution (30 µmol NTA in 5 mL deionized-distilled H2O). The pH of this Fe(NTA)2 solution was adjusted to 4.0 with KOH. 5 6 III. Do the three Ruthenium Complexes Bind to Transferrin? When two or even three molar equivalents of Ru-847.9 are introduced to apo-Tf under the identical conditions used for introduction of Fe(NTA)2 to apo-Tf, spectroscopic signs of specific binding are not observed (Figure 7). The peak at ~329 nm for the sample of apo-Tf, buffer solution, and two equivalents of Ru-847.9 versus a blank of buffer solution (pink curve) can be attributed to Ru-847.9 unbound to transferrin (see Fig. 2). Furthermore, when Ru-847.9 is placed in the blank cuvette, the resulting transferrin spectrum (Figure 7, red curve) fails to show the characteristic features that distinguish a diferric transferrin spectrum from that of apo-Tf. First and foremost, a change in the protein’s absorption in the visible spectrum corresponding to that seen with the addition of ferric ions (i.e., holo-Tf’s absorption peak at ~465 nm) is not observed. Secondly, the transferrin absorbance values at 280 nm for apo-Tf in the presence of Ru847.9 are lower than that for apo-Tf alone. Diferric transferrin has a higher absorbance at 280 nm than apo-Tf, and one would expect the same pattern if ruthenium were binding to transferrin. Together, these two observations seem to indicate that the ruthenium complex does not occupy the iron-binding sites of transferrin. Figure 7 328.98 nm Ru-847.9 and Transferrin --- Apo-Tf: 10 µM in buffer solution. Blank: buffer solution --- Tf plus 2 molar equivalents of Ru: 10 µM Apo-Tf, 20 µM Ru-847.9, buffer solution. Blank: buffer solution. --- Tf plus 2 molar equivalents of Ru, corrected for free Ru: same sample as pink curve. Blank: 20 µM Ru-847.9 in buffer solution. --- Tf plus 3 molar equivalents of Ru, corrected for free Ru: 10 µM apo-Tf, 30 µM Ru-847.9, buffer solution. Blank: 30µM Ru-847.9 in buffer solution. 0.2 cm cuvettes Buffer solution as in Figure 2 Similarly, the spectra of Ru-866.8 and Ru-1314.7 in solution with apo-Tf fail to conclusively indicate specific binding (Figure 8 and 9). 6 7 Figure 8 Ru-866.8 and Transferrin --- Apo-Tf: 10 µM in buffer solution. Blank: buffer solution --- Tf plus 2 molar equivalents of Ru: 10 µM Apo-Tf, 20 µM Ru-866.8, buffer solution. Blank: buffer solution. --- Tf plus 2 molar equivalents of Ru, corrected for free Ru: same sample as green curve. Blank: 20 µM Ru-866.8 in buffer solution. --- Tf plus 3 molar equivalents of Ru, corrected for free Ru: 10 µM apo-Tf, 30 µM Ru-866.8, buffer solution. Blank: 30µM Ru-866.8 in buffer solution. 0.2 cm cuvettes Buffer as in Fig. 2 Figure 9 377.15 nm Ru-1314.7 and Transferrin --- Apo-Tf: 10 µM in buffer solution. Blank: buffer solution --- Tf plus 2 molar equivalents of Ru: 10 µM Apo-Tf, 20 µM Ru-1314.7, buffer solution. Blank: buffer solution. --- Tf plus 2 molar equivalents of Ru, corrected for free Ru: same sample as green curve. Blank: 20 µM Ru-1314.7 in buffer solution. --- Tf plus 3 molar equivalents of Ru, corrected for free Ru: 10 µM apo-Tf, 30 µM Ru-1314.7, buffer solution. Blank: 30µM Ru-1314.7 in buffer solution. 0.2 cm cuvettes Buffer as in Fig. 2 7 8 IV. Do the Ru-complexes display fluorescent properties that could be employed to investigate possible binding to transferrin? Although the above absorption spectra appear to indicate that the three ruthenium(II)-complexes do not bind specifically to transferrin, several previous studies by other labs have confirmed the ability of various Ru(III)-complexes to bind tightly to several plasma proteins including transferrin2,3,9,10,17,18,20. All but one20 of these referenced ruthenium(III) complexes are coordinated to chloride ions; and, many of the authors of these studies assert that the loss of at least one3,9 or two21 coordinated chlorides is a prerequisite for further reactions with proteins. Perhaps our three Ru(II) molecules, two of which are not coordinated to halogens, are unable to access the appropriate binding sites of transferrin as a result of their retention of ligands. In addition, the fact that the protein-bound ruthenium complexes remain in the Ru(III) oxidation state upon binding to transferrin3 may indicate that our Ru(II)-complexes need to be oxidized prior to binding. Ru(III) ions bound to NTA ligands have been shown to bind specifically at the iron-binding sites of transferrin15. As with iron, ruthenium(III) administered to apotransferrin as Ru(III)-NTA binds in a two to one ratio if bicarbonate ion is available. The resulting UV difference spectrum of Ru(III)2Tf versus apo-Tf reflects that usually seen for a number of metallotranferrins15. However, Ru(III)-complexes have also been shown to bind to transferrin at exposed imidazole rings of histidines3,9,17,20. Unlike iron which has a maximum binding ratio of two iron atoms to one transferrin molecule, ruthenium(III)-complexes can bind with up to five3 or six20 equivalents per transferrin. That specific binding at the iron-binding sites proper is not occurring between the ruthenium-complexes and transferrin is further demonstrated by the fact that ruthenium modified transferrin retains its ability to bind up to two iron(III) equivalents when Fe(NTA)2 is administered20. Documentation of the binding of Ru(III)-complexes to the histidines of transferrin offered hope that our ruthenium(II)-complexes might possess the ability to likewise bind to transferrin. Perhaps fluorescence spectra could elucidate these histidine interactions and show evidence of binding that our absorption spectra failed to indicate. We were optimistic that our three ruthenium(II)-complexes would display useful fluorescent properties because other ruthenium complexes cited in literature have intriguing and valuable fluorescent properties. For example, the emission spectra of ruthenium(III) dendrimers, ruthenium(III) polypyridyl complexes, and cyano-bridged trinuclear polypyridyl complexes of Ru(II) employed as artificial, luminescent chromophores are useful tools for understanding the excited state dynamics of these molecules6,7,22,23. Likewise, the emission spectra of several oligonuclear polypyridyl Ru(II) complexes helps elucidate the structure and charge transfer behaviors of these molecules24. Furthermore, unsymmetrical Ru-ligand complexes such as Ru(dcby)(bpy)2 are used as anisotropy probes because of their emission properties and long lifetimes25. Lastly, Ru(bpy)32+-pendant dendrimers can serve as sensors for the explosive trinitrotoluene (TNT) based on the quenching of their fluorescence by TNT26. Ru-847.9 dissolved in chloroform has an absorption peak at ~333 nm and a larger peak at ~250 nm (Figure 10). However, an emission spectrum of 0.38 mM Ru-847.9 8 9 dissolved in chloroform and excited at 250 nm failed to result in emission intensity greater than the background intensity of light scattered by the cuvette holder (Figure 11). 249.02 nm Figure 10 Absorption Spectrum 0.38 mM Ru-847.9 dissolved in chloroform 333.8 nm blank: chloroform 0.2 cm cuvettes Figure 11 Emission Spectra: λexcitation= 250 nm ---0.38 mM Ru-847.9 dissolved in chloroform ---empty cuvette holder 0.2 cm cuvettes, 250 nm filter on emission *For all fluorescence spectra, both the excitation and emission slits are 10.0 nm. Scan speed is 200 nm/sec. An emission spectrum of the same sample excited at 333 nm results in an intensity maximum at 421.99 nm (Figure 12). However, the intensity of this signal is not sufficiently higher than that caused by the solvent alone and other background noise. In an effort to determine how to maximize the emission signal of the ruthenium complex, an excitation spectrum was analyzed at an emission wavelength of 420 nm. This excitation spectrum indicated that the intensity at 420 nm is stronger with excitation 9 10 at 290 nm (Figure 13). Therefore, an emission spectrum of Ru-849.7 in chloroform with an excitation wavelength of 290 nm was collected (Figure 14). However, while the intensity of the signal does increase as hoped, it remains insufficiently greater than the emission intensity of the solvent alone. The fluorescence signal of the ruthenium(II) molecule, even at this relatively high concentration, is inadequate in intensity and would not serve as a conclusive probe in our attempts to elucidate the interactions between transferrin and the ruthenium complex. Figure 12 421.99 nm Emission Spectra: λexcitation= 333 nm (0.2 cm cuvettes) ---0.38 mM Ru-847.9 dissolved in chloroform ---chloroform blank ---empty cuvette holder (no sample) ---nothing at all (no sample, no cuvette, no cuvette holder) Figure 13 290 nm Fig. 11 Excitation Spectra: λemission= 420 nm (0.2 cm cuvettes) ---0.38 mM Ru-847.9 dissolved in chloroform ---chloroform blank ---empty cuvette holder (no sample) ---nothing at all (no sample, no cuvette, no cuvette holder) 10 11 422.43 nm Figure 14 Fig. 12 Emission Spectra: λexcitation= 290 nm (0.2 cm cuvettes) ---0.38 mM Ru-847.9 dissolved in chloroform ---chloroform blank ---empty cuvette holder (no sample) ---nothing at all (no sample, no cuvette, no cuvette holder) Conclusion Absorption spectra of the three ruthenium(II) complexes in solution with apotransferrin and bicarbonate ion do not show the spectroscopic evidence of specific Tfmetal binding that Fe2-transferrin displays. It is our belief that these ruthenium complexes are not binding specifically at the metal-binding sites of transferrin. The fluorescence emission of these three ruthenium(II) complexes at 420 nm is quite weak, presumably due to ligand quenching. It is impossible therefore, to use fluorescence as a means of binding to these secondary sites. 11 12 Appendix A Brief Survey of the Literature: Additional Applications of Ruthenium Complexes Artificial Photosynthesis • Ruthenium containing dendrimers (perfectly branched synthetic macromolecules with several side chains radiating from a single core) serve as light-harvesting chromophores that channel absorbed energy to a single reaction center in a fashion analogous to that of natural photosynthetic systems. Rutheniumpolypyridine complexes have ideal luminescence and redox properties for such systems and therefore have the potential to be used in light-emitting diodes, signal amplifiers, fluorescent sensors, frequency converters, and other photonic devices22. • Ruthenium polypyridyl complexes take the place of chlorophylls in artificial photosynthesis. The absorption of light by [Ru-(bpy)3]2+ results from metal to ligand charge transfer (MLCT)9,10. The excited states that result are reached with high efficiency, are quite stable, and have sufficiently long lifetimes that facilitate the desired chemical reactions. The appeal of these complexes is due to multiple MLCT transitions, vibronic progressions, and solvent broadening which allow for the absorption of light over a wide range of wavelengths. Alternatively, chemical changes can systematically tune the range at which light is absorbed7. • Cyano-bridged trinuclear polypyridyl complexes of Ru(II) are of interest as lightharvesting devices and sensitizers in photovoltaic cells23. Understanding Other Biological Processes • A ruthenium-labeled cytochrome c derivative allows for the study of electron transfer from cytochrome bc1 to cytochrome c (Cc), a step in the electron transport chains of mitochondria and many respiratory and photosynthetic prokaryotes. Labeling the ruthenium-Cc derivative, [Ruz-39-Cc], with Ru(bpz)2(dmb), a complex able to photooxidize the ferrous heme in Cc, permits the study of electron transfer in the forward, physiological direction24. • Oligonuclear polypyridyl Ru(II) complexes model long-range electron and energy transfer processes known to be important steps in respiration, photosynthesis, and DNA oxidative cleavage25. • Ruthenium(II) myoglobins are employed in the study of metal ion substitution in heme proteins and the binding of carbon monoxide and dioxygen to myoglobins26. Probes • Because of their long-lived lifetimes, unsymmetrical Ru-ligand complexes such as [Ru(dcby)(bpy)2] can be used as anisotropy probes for protein hydrodynamics and immunoassays of high-molecular-weight antigens27. • [Ru(bpy)3]2+-pendant dendrimers can serve as TNT sensors based on the quenching of their fluorescence by TNT. The high quantum yield, photostability, and redox power of [Ru(bpy)3]2+ make it a very promising fluorescence probe28. 12 13 • The photoreactive reagent AzRu, containing ruthenium and a photoactivable azido group, interacts specifically with Ca2+ binding proteins regardless of the proteins’ catalytic mechanisms. AzRu thus provides a method of identification of these proteins and characterization of their binding sites29. Works Cited: 1. Cini, R., G. Tamasi, et al. (2003). "Study of Ruthenium(II) complexes with Anticancer Drugs as Ligands. Design of Metal-Based Phototherapeutic Agents." Inorganic Chemistry 42(24): 8038-8052. 2. Kratz, F., M. Hartmann, et al. (1994). "The Binding Properties of Two Antitumor Ruthenium(III) Complexes to Apotransferrin." The Journal of Biological Chemistry 269(4): 2581-2588. 3. Messori, L., F. G. Vilchez, et al. (2000). "Binding of antitumor ruthenium III complexes to Plasma Proteins." Metal-Based Drugs 7(6): 335-341. 4. Pifferi, C. and R. Cini (1998). "Synthesis and solution Studies of ruthenium(II) complexes with thiazole and antileukaemic drug thiopurines. Crystal structure of trans-dichlorotris(1,3-thiazole)(triphenylphosphine)ruthenium(II)." Journal of the Chemical Society, Dalton Transactions: 2679-2688. 5. Cini, R., R. Bozzi, et al. (1993). "Synthesis and NMR Studies of Ruthenium(II) Mercaptopurine Riboside complexes. X-ray Structure of Bis(9-β-Dribofuranosyl-6-mercaptopurine)bis-(triphenylphosphine)ruthenium(II) Chloride." Journal of the Chemical Society, Chemical Communications: 899-901. 6. Bhasikuttan, A. C., M. Suzuki, et al. (2002). "Ultrafast Fluorescence Detection in Tris(2,2’-bipyridine)ruthenium(II) Complex in Solution: Relaxation Dynamics Involving Higher Excited States." Journal of the American Chemical Society 124(28): 8398-8405. 7. Meyer, T. J. (1989). "Chemical Approaches to Artificial Photosynthesis." Acc. Chem. Res. 22(5): 163-170. 8. Brown, G. M., J. E. Sutton, et al. (1978). "Binding of a Tetraammineruthenium Complex to Imidazole and Purine Derivatives under Equilibrium Conditions." Journal of the American Chemical Society 100(9): 2767-2774. 9. Vilchez, F. G., R. Vilaplana, et al. (1998). "Solution studies of the antitumor 13 14 complex dichloro 1,2-propylendiaminetetraacetate ruthenium (III) and of its interactions with proteins." Journal of Inorganic Biochemistry 71: 45-51. 10. Messori, L., F. Kratz, et al. (1996). "The Interaction of the Antitumor Complexes Na[trans-RuCl4(DMSO)(Im)] and Na[trans-RuCl4(DMSO)(Ind)] with Apotransferrin: A Spectroscopic Study." Metal-Based Drugs 3(1): 1-9. 11. Harris, D. C. and P. Aisen (1975). "Iron-Donating Properties of Transferrin." Biochemistry 14(2): 262-268. 12. O'Hara, P. B. and S. H. Koenig (1986). "Electron Spin Resonance and Magnetic Relaxation Studies of Gadolinium(III) Complexes with Human Transferrin." Biochemistry 25(6): 1445-1450. 13. Messori, L., G. Dal Poggetto, et al. (1997). "The pH dependent properties of metallotransferrins: a comparative study." BioMetals 10: 303-313. 14. O'Hara, P., S. M. Yeh, et al. (1981). "Distance between Metal-Binding Sites in Transferrin: Energy Transfer from Bound Terbium(III) to Iron(III) or Manganese(III)." Biochemistry 20: 4704-4708. 15. Kratz, F. and L. Messori (1993). "Spectral Characterization of Ruthenium(III) Transferrin." Journal of Inorganic Biochemistry 49(2): 79-82. 16. Yang, H.-W., R. T. A. MacGillivray, et al. (2000). "Human Serum Transferrin", RCSB Protein Data Bank (online). 17. Clarke, M. J., F. Zhu, et al. (1999). "Non-Platinum Chemotherapeutic Metallopharmaceuticals." Chemical Reviews 99(9): 2511-2533. 18. Kratz, F., B. K. Keppler, et al. (1996). "Comparison of the Antiproliferative Activity of two Antitumor Ruthenium(III) Complexes with their Apotransferrin and Transferrin-Bound Forms in a Human Colon Cancer Cell Line." Metal-Based Drugs 3(1): 15-23. 19. Zatta, P., L. Messori, et al. (2005). "The C2 Variant of Human Serum Transferrin Retains the Iron Binding Properties of the Native Protein." Biochimica et Biophysics 1741: 264-270. 20. Martin, D. M., N. D. Chasteen, et al. (1991). "Fluorescence and Kinetic Properties of Ru(III) (NH3)5 modified Transferrin." Biochimica et Biophysics 1076(252-258). 21. Messori, L., P. Orioli, et al. (2000). "A spectroscopic study of the reaction of NAMI, a novel ruthenium(III) anti-neoplastic complex, with bovine serum albumin." European Journal of Biochemistry 267: 1206-1213. 14 15 22. Adronov, A. and J. M. J. Fréchet. (2000). "Light-harvesting dendrimers." Chemical Communications: 1701-1710. 23. Matsui, K., M. K. Nazeeruddin, et al. (1992). "Transient Absorptions Due to Mixed Valence Species in the Excited-State Absorption Spectra of Cyano-Bridged Trinuclear Polypyridyl Complexes of Ru(II)." Journal of Physical Chemistry 96: 10587-10590. 24. Engstrom, G., R. Rajagukguk, et al. (2003). "Design of a Ruthenium-Labeled Cytochrome c Derivative to Study Electron Transfer with the Cytochrome bc1 Complex." Biochemistry 42(2816-2824). 25. Ishow, E., A. Gourdon, et al. (1999). "“Synthesis, Mass Spectrometry, and Spectroscopic Properties of a Dinuclear Ruthenium Complex Comprising a 20 Å Long Fully Aromatic Bridging Ligand." Inorganic Chemistry 38(7): 1504-1510. 26. Paulson, D. R., A. W. Addison, et al. (1979). "Preparation of Ruthenium(II) and Ruthenium(III) Myoglobin and the Reaction of Dioxygen, and Carbon Monoxide, with Ruthenium(II) Myoglobin." Journal of Biological Chemistry 254(15): 70027006. 27. Szmacinski, H., E. Terpetschnig, et al. (1996). "Synthesis and evaluation of Rucomplexes as anisotropy probes for protein hydrodynamics and immunoassays of high-molecular-weight antigens." Biophysical Chemistry 62: 109-120. 28. Glazier, S., J. A. Barron, et al. (2003). "Quenching Dynamics of the Photoluminescence of [Ru(bpy)3]2+-Pendant PAMAM Dendrimers by Nitro Aromatics and Other Materials." Macromolecules 36(4): 1272-1278. 29. Israelson, A., L. Arzoine, et al. (2005). "A Photoactivable Probe for Calcium Binding Proteins." Chemistry and Biology 12: 1169-1178. 15