A Novel Organometallic Rhenium Salt: DNA-Binding and Cytotoxicity Studies on Pancreatic, Breast and Lymphoma Cancers

DNA-binding studies of a variety of rhenium(I) tricarbonyl complexes are known. Primarily the rhenium complexes bind to DNA through intercalation or minor groove or both. We synthesized a novel organometallic salt of the type, [Re(µ-H)Re] + [Re(µ-OR) 3 Re] - . The UV-vis and emission spectroscopic titrations and viscosity studies indicate that the salt binds to DNA through partial intercalation. A variety of mono-, di-, and trinuclear rhenium(I) carbonyl complexes are known to exhibit cytotoxicity against numerous cancer cell lines. Examples of the cytotoxicity of tetranuclear rhenium (I) tricarbonyl complexes are rare. We have found that the tetranuclear, novel organometallic salt is highly potent on numerous cancer cells. The IC 50 values (concentrations required to induce 50% cell deaths) are 0.220 µM on BxPC-3 pancreatic cancer cells, 0.298 µM on estrogen receptor positive MCF-7 breast cancer cells, 0.948 µM on triple negative MDA-MB-231 breast cancer cells, and 0.300 µM on U-937 lymphoma cells.


Synthesis of 2
The synthesis of compound 2 has been previously published [12]. In brief, a mixture of Re 2 (CO) 10 and neocuproine in a 1:2 mole ratio was re uxed for 24 h and cooled to room temperature for another 24 h. This mixture was ltered and the reddish brown crystalline minor product 2 was separated by ltration. Complex 2 was characterized spectroscopically and through X-ray crystal structure determination [12] DNA-binding studies Electronic absorption spectroscopy A measured amount of the tetranuclear rhenium complex 2 was dissolved in DMSO. Tris buffer at pH 7.2 (3000 µL) was placed in the reference cuvette. The sample cuvette contained the same volume of buffer.
Both cuvettes were diluted with equal amounts of neat DMSO (reference) and the DMSO solution of the complex (sample). To perform the titrations, a measured amount of a DNA stock solution in Tris buffer, pH 7.2 was added to each cuvette to achieve a desired DNA concentration in each cuvette. Solutions were mixed thoroughly by repeated inversion of the cuvettes. After an incubation time of 15 min, the absorption spectra (225-700 nm) were recorded. Absorption titrations were carried out by gradually increasing the DNA concentration. Any change in the complex concentration due to each titration was negligible. The intrinsic binding constant, K b was determined through an earlier reported method [30].

Emission spectroscopy
Emission intensity measurements were carried out using a Carey Fluorescence Spectrophotometer 5000 at room temperature. Tris-HCl buffer was used as the control to make preliminary adjustments.
Competitive binding experiments were performed in this same buffer. The [DNA]/[EB] ratio was kept at 1.10 and the concentration of 2 was varied. Fluorescence spectra of EB were measured using an excitation wavelength of 360 nm and the emission range was set between 440 and 750 nm. To understand the binding strength magnitude of 2 with DNA, the linear Stern-Volmer equation (eq. 1) is applied [31]: see formula 1 in the supplementary les.
where I 0 and I represent the uorescence intensities in the absence and presence of 2 (quencher), respectively. Q is the concentration of 2. K sv is the quenching constant.
Cyclic Voltammetry (CV) studies All electrochemical studies were performed using a CH Instrument Electrochemical analyzer in a single compartmental cell with a three-electrode con guration. This con guration is comprised of a Pt wire as the auxiliary electrode, a glassy carbon electrode as the working electrode, and Ag/AgCl as the reference electrode. A 1:1 mixture of acetonitrile and Tris buffer (pH 7.2) was used as the solvent, and 0.05 M tetrabutylammonium perchlorate was used as the supporting electrolyte.

Viscosity study
Viscosity measurements were carried out using a 3156 viscometer (Q Glass Company Inc) as reported earlier [30]. In brief, the ow times were measured using a digital stopwatch. The CT-DNA was quantitated by absorbance at 260 nm using a NanoVue UV spectrophotometer. nm and an emission wavelength of 590 ± 10 nm on a multimode microplate reader. Percent uorescence intensity of the DMSO control was determined from the measured relative uorescence units. The IC 50 was determined by nonlinear curve tting to the dose-response curve.
Lymphoma cells (U-937) U937 cells were grown in RPMI-1640 medium, supplemented with 10% (v/v) heat in-activated FBS, 2 mM L-glutamine, 10 units/ml penicillin and 10 mg/ml streptomycin at 37 °C, 5% CO 2 / 95% air in 75-mL asks. These cells were screened on a regular basis for the presence of mycoplasma and were found to be negative. Fresh medium was added twice weekly. The cell density was kept at 2-8 x 10 5 cells per ml. A hemocytometer was used to count cells and cells were subsequently resuspended to the desired concentration. When the growth reached approximately 10 6 cells/ml, the cells were removed from the ask, centrifuged, the pellet was washed with RPMI-1640 and distributed into wells of a 24-well plate at cell densities of 5 x 10 5 /well. Cell treatment with the rhenium complex 2 and viability assays were carried out according to the procedure described above for pancreatic cancer cells.

Synthesis and characterization
We previously reported the synthesis and characterization of the tetranuclear rhenium complex 2 [12]. The molecular structure of 2 is shown in Fig. 1. Both rhenium atoms (Re1 and Re2) in the cation are in +1 states and are connected through a bridging hydride ion. Both rhenium atoms (Re3 and Re4) in the anion are also in +1 oxidation states and are connected through three bridging pentyl alkoxide ions.

DNA-binding studies
The electronic absorption spectrum of 2 exhibits intense absorptions centered at 235 and 280 nm which are assigned to intraligand π-π* transitions because similar absorptions are observed for the uncoordinated neocuproine ligand. The lower-energy absorption shoulders in the range 350-425 nm are assigned to spin-allowed metal-to-ligand charge-transfer (MLCT) dπ → π* transitions. When a complex binds with DNA through intercalation, it usually results in hypochromism (reduction in absorbance) and bathochromism (red shift) due to strong stacking interaction between aromatic chromophore and the base pairs of DNA. Addition of increasing amounts of CT-DNA to a xed concentration of 2 resulted in both hypochromism and bathochromism for the band at 280 nm (see Fig. 2) suggesting that the complex binds to DNA through intercalation. The intrinsic binding constant K b of the complex 2 to DNA was determined by monitoring the changes of absorbance at 280 nm with increasing concentration of DNA (see Fig. 2) and plotting [DNA]/( 0f ) vs.
To further ascertain the mode of interaction of 2, the ethidium bromide (EB) displacement assay was carried out [34]. The intrinsic uorescence intensity of DNA is very low, and that of EB in Tris buffer is also not high due to quenching by the solvent molecules. However, EB emits intense uorescence in the presence of DNA, due to its strong intercalation between the adjacent DNA pairs. Thus, EB can be used to probe the interaction of 2 with DNA. The emission spectra of EB bound to DNA in the presence of 2 is shown in Fig. 3. The addition of 2 to DNA pretreated with EB causes obvious reduction in emission intensity at 608 nm, indicating that 2 competes with EB in binding to DNA. This observation is often the characteristic of intercalation [35]. The quenching constant, K sv (1.12 x 10 4 ) is determined by plotting I 0 /I versus the concentrations of 2 (see Fig. 3 inset). The apparent binding constant K app is calculated according to eq. 2: see formula 2 in the supplementary les.  [34]. In Stern-Volmer equation (eq. 1), I becomes I 0 /2 at a 50% reduction of the uorescence intensity. Thus, [complex 2] becomes 1/K sv . As a result, K app equals to K EB [EB] x K sv or 1.0 x 10 5 . This is signi cantly lower than the K EB value for EB.
The cyclic voltammograms of 2 in the oxidation region (A) and the reduction region (B) in the absence (black) and presence (red) of DNA are shown in Fig. 4. The oxidation peak around 1.6 V was shifted slightly toward more oxidizing direction (A) while the two reduction peaks between 1.3 and 1.5 V were unchanged (B) in the presence of DNA. Since oxidation peaks are metal-centered and reduction peaks are ligand-based, the observations here suggest some kind of interactions between the complex and DNA; further, the interaction may be at or close to the metal site, possibly through electrostatic attraction, while the ligands are farther away from DNA with no in uence from the DNA interaction.
Viscosity measurement is regarded as the least ambiguous and most precise method of studying the binding mode of metal complexes with DNA in the absence of X-ray crystallography and NMR spectroscopy. It is suggested that an intercalative binding causes a signi cant increase in DNA viscosity due to the lengthening of the DNA helix.
As illustrated in Fig. 5, upon increasing the concentrations of 2, the relative viscosities of 2 increase less steadily than those of EB. These results suggest that the complex 2 less strongly intercalates between the base pairs of DNA.

Cytotoxicity Studies
Pancreatic cancer cells (BxPC-3). The ability to inhibit the cell growth by 2 was evaluated in vitro against BxPC-3 human pancreatic cancer cell lines. An IC 50 value of 0.22 µM was obtained through MTT assay (see Fig. 6). In combination with other anticancer drugs both cisplatin and oxaliplatin are used as combination drugs. The leading pancreatic cancer drug FOLFIRINOX is a mixture of folinic acid, 5uorouracil, irinotecan and oxaliplatin. Previously we demonstrated that tricarbonyl rhenium (I) polypyridyl complexes are very active on BxPC-3 pancreatic cancer cell lines (IC 50 < 5 µM) [38,39]. Recently several other tricarbonyl rhenium (I) complexes have been reported to be highly cytotoxic (IC 50 in the range 4 -9 µM) against a panel of pancreatic cancer cell lines [23,24]. Cytotoxicity of a few metal complexes against numerous pancreatic cancer cell lines are also known [40][41][42].
Lymphoma cells (U-937). The inhibition of cell growth by 2 was evaluated on U-937 lymphoma cells. The IC 50 value was derived from MTT assay which is shown in Fig. 9 The organometallic salt 2 is highly active on pancreatic and breast cancer and lymphoma cells. To the best of our knowledge, no such metal complexes like 2 are known to be highly potent on pancreatic cancer cells. It is known that rhenium complexes are less toxic than platinum metal complexes. One of the ingredients of FOLFIRINOX is oxaliplatin which may be substituted with 2 to yield a better pancreatic cancer drug. Pancreatic cancer has been the second leading cause of cancer-related deaths in the USA.
Due to its chemoresistance, it has been designated as a recalcitrant cancer. Also, 2 may nd applications in the treatment of breast and lymphoma cancers.     Pancreatic cancer cell lines, BxPC-3 exposed to various concentrations of 2 in DMSO. Control represents cells exposed to DMSO.

Figure 7
Breast cancer cell lines, MCF-7 exposed to various concentrations of 2 in DMSO. Control represents cells exposed to DMSO.

Figure 8
Breast cancer cell lines, MDA-MB-231 exposed to various concentrations of 2 in DMSO. Control represents cells exposed to DMSO.