Electron transfer-based combination therapy of cisplatin …

Electron-transfer reactions play key roles in diverse processes in chemistry, physics, and biology, ranging from photo-induced reactions (1, 2), electron tunneling in proteins (3), and electron transport in DNA (4) to the ozone hole formation (5) and reductive DNA damage (6, 7). Electron-transfer reactions in molecular systems have therefore been the subject of intense experimental and theoretical studies. Following the pioneering contribution of Zewail (8), the advent of femtosecond time-resolved laser spectroscopy (fs-TRLS; 1fs=10-15s) provided an unprecedented capacity in techniques of observing molecular reactions, including electron transfer. Its application to the study of chemical and biological systems led to the birth of new scientific subfields: femtochemistry and femtobiology (8). Recently, Lu (9) further proposed that integrating ultrafast laser techniques with biomedical methods to advance fundamental understandings and treatments of major human diseases might lead to the opening of a new frontier called femtomedicine. Regarding the latter, we have recently unraveled unique dissociative electron-transfer (DET) mechanisms of reductive DNA damage (6, 7) and several anticancer agents for radiotherapy and chemotherapy (1014). In this paper, we present results of experimental studies on a unique combination therapy based on the DET mechanism of cisplatin to enhance the efficacy of human ovarian-, cervical-, and lung-cancer therapies.

Ovarian and cervical cancers are leading causes of cancer deaths in women, while lung cancer is the deadliest type of cancer for both men and women (15). The standard treatment is surgery followed by platinum-based chemotherapy, which generally yields a positive response in the initial treatment. In many cases, however, cancer cells become refractory with time (1620), and 7090% of patients with serious cancers die of progressive chemoresistant disease (19). Successful treatment strategies are still lacking, especially for ovarian and lung cancers (1520).

Cisplatin [Pt(NH3)2Cl2] (CDDP) is the first and most widely used platinum-based chemotherapy drug (2124). CDDP is now the cornerstone agent in treating a variety of cancers, including ovarian, testicular, cervical, bladder, lung, head and neck, lymphoma, and brain cancers (1618, 2224). However, its application is often limited by toxic side effects and resistance of various cancers (1618, 2224). The cytotoxicity of CDDP is well known to arise from its capacity to damage DNA by binding the cis-[Pt(NH3)2] unit to DNA. The conventional understanding of the initial action of CDDP was the hydrolysis mechanism (17, 2224). Based on this mechanism, many studies in the past 40years attempted to circumvent the drawbacks of CDDP. Over 3,000 CDDP analogues were designed, synthesized, and tested, but only two have been approved by the US Food and Drug Administration: oxaliplatin and carboplatin. This might imply that a precise understanding of the molecular mechanism of the cytotoxicity of CDDP was lacking.

Moreover, recent advances in cancer research have shown that even the most successful targeted therapies lose potency with time; attaining permanent cures for most cancers would require a combination therapy of two or more drugs simultaneously (25). Indeed, the combinations of cisplatin with another therapy/agent have shown promising results in treating lung, cervical, and head and neck cancers (2631). In most cases, however, the combined therapies/agents act via different mechanisms of action or target different pathways, thereby achieving an additive therapeutic effect. Thus, there have been no or limited synergetic effects in the combinations. It is desirable to develop a synergetic combination therapy of low-dose CDDP with another agent.

Through the femtomedicine approach, we have recently obtained the precise molecular mechanisms of cisplatin in combination with radiotherapy (13) and chemotherapy (14) of cancer. First, we found that CDDP is extremely effective for the DET reaction with ultrashort-lived (540fs) prehydrated electrons () generated in radiotherapy to produce a [Pt(NH3)2Cl] or cis-[Pt(NH3)2] radical that then causes DNA strand breaks (13). The latter will lead to apoptosis and final clonogenic cell kill (32). Second, we found that, for chemotherapy, the DET reaction of CDDP occurs preferentially with two neighboring G bases (the most favorable electron donor among the four bases in DNA), while the DET with base A is much weaker and there is no DET with C and T (14). This DET mechanism has directly unraveled the long-existing mystery regarding why Pt drugs result in the preferential binding of the cis-[Pt(NH3)2] to two neighboring G bases in DNA. The high reactivity of cisplatin with electrons has been confirmed subsequently by Kopyra et al. (33) in studying dissociative attachments of low-energy free electrons to gas-phase CDDP. It is well known that in liquid water, free electrons produced by ionizing radiation will rapidly become on the fs timescale and finally form the nonreactive solvated electrons (6, 7, 9). Most recently, quantum chemical calculations by Kuduk-Jaworska et al. (34) have confirmed the DET mechanism of CDDP. The authors presented results on delicate electronic-structure calculations of model systems for cis-and trans-platin with free electrons, hydrated electrons, and water by using density functional theory approach. Their results are consistent with our DET mechanism involving electrons trapped in water, though the hydrolysis of CDDP was not excluded in the study. The DET mechanistic insight into the cytotoxicity mechanism of CDDP obtained through our fs-TRLS studies (13, 14) has potential to improve existing therapies using CDDP and enable new combination treatments for challenging cancers.

Based on the DET mechanism, it is expected that CDDP may be administered in combination with a biological electron donor (mimicking the generated in radiotherapy) to enhance the chemotherapeutic efficacy. Aromatic amines like N,N,N,N-tetramethyl-p-phenylenediamine (TMPD) are well-known biological electron donors (35). Thus, the expected strong DET reaction between cisplatin and TMPD will form reactive radicals to kill cancer cells. In this study, we conducted both absorption and fluorescence spectroscopic measurements to confirm the DET reaction. Furthermore, we measured the DET-induced strand breaks of plasmid DNA by gel electrophoresis. We used a CDDP-sensitive human cervical cancer (HeLa) cell line and highly CDDP-resistant human ovarian cancer (NIH:OVCAR-3) and lung cancer (A549) cell lines to evaluate the combination therapy of CDDP with the exemplary electron-donating molecule TMPD. The NIH:OVCAR-3 cell line has been established by Hamilton and coworkers (3639) as a model system for in vitro and in vivo studies of the resistance of CDDP and other chemotherapy drugs.

We first show time-series changes of the absorption spectrum of TMPD in pure water. A shows that after 100M TMPD was dissolved for a few hours, a pronounced absorption band appeared at 500650nm, which is the well-known characteristic absorption of the cation TMPD+. This band was observed in many studies of photoionization of TMPD in polar solvents and photo-induced DET from TMPD to halogenated molecules as solvents or solutes (40, 41). In contrast to those previous studies, no extra photo-excitation was applied in our present experiments. This indicates the effective autoionization of TMPD into a TMPD+ and a solvated electron in water. As a strong electron-donating agent, TMPD has a low ionization potential of 6.10.1eV in the gas phase (42), which can be lowered by more than 3eV due to the solvent polarization in some polar solvents (40). Thus, the ionization energy of TMPD in some solvents can be lower than the binding energy of the solvated electron, which is approximately 3.23.5eV in water. As a result, the autoionization of TMPD occurs in water. More interestingly, the results in B show that adding CDDP increased the TMPD+ yield, indicating an effective electron transfer from TMPD to CDDP. B also shows that when the samples were irradiated by a laser at 266nm for 1h, the TMPD+ yield was additionally enhanced. This is consistent with the previous observations of photo-induced electron transfer reactions of TMPD (40, 41).

Spectroscopic observations of the DET reaction between cisplatin (CDDP) and TMPD. Absorption spectra (AD) and fluorescence spectra (E, F) of TMPD in water or EtOH with and without the presence of CDDP. In B, spectra are also shown for the samples ...

By contrast, the observed results for 100M TMPD in pure EtOH are shown in C. No changes in the absorption spectrum were observed up to 48h or longer, giving direct evidence of no autoionization of TMPD in EtOH. This is consistent with the observation that the ionization energy of TMPD in alcohols is higher than in water, which is approximately 4.75eV in methanol (40). Thus, the autoionization of TMPD cannot occur in EtOH. Interestingly, however, adding 100M CDDP caused no changes in the absorption spectrum of 100M TMPD in EtOH initially, but the strong absorption band of the TMPD+ at 500650nm appeared after a few hours (D). The results for mixing 100M TMPD with various concentrations of CDDP are plotted in Fig.S1, showing that the delay time to observe the pronounced TMPD+ absorption band decreased with rising CDDP concentrations. An additional absorption band at 300350nm was also observed in the spectra shown in A,B, and D and Fig. S1. This extra band appeared simultaneously with that of TMPD+ at 500650nm, both having an identical growth rate either in pure water or in mixtures with CDDP in H2O or EtOH. Thus, it may be attributed to TMPD+ as well. These data clearly demonstrate the effective DET reaction between TMPD and CDDP in EtOH even with no autoionization of TMPD.

The steady-state fluorescence spectra of TMPD excited at 266nm in pure H2O and EtOH with and without the presence of CDDP are shown in E and F. It is seen that for TMPD in pure H2O (E), there were two emission peaks at 370 and approximately 425430nm. With the addition of CDDP, the peak at 370nm became dominant, while the peak at 425nm was significantly depleted. These peaks can be reasonably attributed to the fluorescence peaks of TMPD+ and neutral TMPD, respectively (43). This attribution is further confirmed by the results shown in F: The fluorescence spectrum of TMPD in pure EtOH exhibited only a peak at approximately 420nm, while a new peak around 370nm appeared as CDDP was introduced. Again, the fluorescence spectra in E and F confirm the autoionization of TMPD in water but not in pure EtOH and the DET reaction of TMPD with CDDP in both H2O and EtOH solvents.

Moreover, under the condition that the CDDP concentration (e.g., 2mM) is far larger than the TMPD concentration (100M) (Fig.S1), the DET reaction between TMPD and CDDP, leading to the formation of the ion pair and the radical:

can be described by pseudo-first order reaction kinetics. Here, the reaction rate constant k can be determined by the absorption intensity I variation of TMPD at 261nm with reaction time t. As detailed in SI Text and Fig.S2, the obtained reaction rate constant is k=1.70.210-2M-1s-1. This indicates that without extra excitation, the DET reaction between ground-state TMPD and CDDP (Eq.1) can effectively occur on the time scale of hours.

To examine the above hypothesis, we measured plasmid DNA damage induced by cisplatin only and its combination with TMPD by agarose gel electrophoresis (7). The gel image and double-strand break (DSB) yields for plasmid DNA incubated with CDDP alone at various concentrations and in combination with 100M TMPD for 24h are shown in . The DSB yields were determined from the DSB peak areas in the gel densitograms, which are shown in Fig.S3. The intrastrand cross-link is the well-known form of the CDDP-DNA adduct, but the platinated DNA is invisible in the gel image because the fluorescence emission of the DNA-binding dye (EtBr) is quenched. This resulted in a much weaker band of the supercoiled DNA for CDDP-treated DNA than the untreated DNA (control). Of particular interest, however, is our result shown in and Fig. S3: Cisplatin directly induces DSBs of the DNA. In fact, CDDP-induced DSBs in replicating yeast, Escherichia coli, and mammalian cells had been reported previously (44, 45), but they were thought to arise from the repair process of DNA cross-links in which DSBs act as an intermediate step. Because no repair could be involved in the extracted and purified plasmid DNA in the present experiments, our observation of DNA DSBs gives direct evidence that DSB is an intrinsically damaging form of cisplatin interacting with DNA. More remarkably, the combination of CDDP with TMPD increased the DNA DSB yield by a factor of approximately 3.5 (). This result provides strong evidence of the expected enhancement in DNA DSBs induced by reactive radicals due to the DET reaction.

Gel electrophoresis measurements of strand breaks in plasmid DNA treated by cisplatin alone and in combination with TMPD: (A) gel image, where the bands for supercoiled (SC) DNA, open circular DNA with single-strand breaks (SSBs), and linear DNA with ...

We further investigated whether TMPD in combination with CDDP modulates the sensitivity of human cancer cells to CDDP. Based on spectroscopic results (), the concentrated stock solution (40mM) of TMPD must be prepared in pure EtOH to prevent the loss of electrons from TMPD. We compared the effects of CDDP and TMPD alone and in combination on the viability of CDDP-sensitive cervical cancer (HeLa) cells and highly CDDP-resistant ovarian cancer (NIH:OVCAR-3) and lung cancer (A549) cells. As shown in A and B, 100M of TMPD alone caused only a small cell-killing effect (approximately 10%) of the treated cells. For HeLa cells (A), a 24-h treatment with CDDP alone decreased the cell survival rate in a dose-dependent manner: A CDDP concentration as high as 60M was required to kill the cells completely. Interestingly, the addition of 100M TMPD to CDDP greatly enhanced the killing of cancer cells in a synergistic manner: At CDDP concentrations of 30M, nearly all of the HeLa cells were killed. Even more interesting were the results for highly CDDP-resistant ovarian cancer cells (B): About 40% of the treated NIH:OVCAR-3 cells survived from the treatment of CDDP alone, even at very high concentrations of 200300M. This result confirms the strong resistance of NIH:OVCAR-3 cells to CDDP (3639). Strikingly, we found that the combination of CDDP with 100M TMPD dramatically enhanced the killing of NIH:OVCAR-3 cells: The killing rate increased from approximately 50% to 95% at 100M CDDP and from 60% to 100% at 200M CDDP. Similar enhancements in killing A549 cells by combination of CDDP with TMPD are shown in Fig.S4. Also shown in and Fig. S4 are the results of fractional effect analysis (28), which is one of the most straightforward methods to evaluate the synergetic effect; the effects of CDDP and TMPD alone were simply multiplied and compared with the observed effect of the combination at the same concentration of the single agents. The results clearly show that the observed effect was significantly greater than the product of the effects of individual agents. As also shown in TableS1, moreover, the values of CDDP IC50 (the concentration required to kill 50% of untreated cells) for these treated cancer cells are reduced by factors of 24 by combination with TMPD, which are about 50% the values expected for the additive effect of the two agents. From these results, we conclude that the combination treatment showed strong synergistic effects in killing cervical, ovarian, and lung cancer cells and removed the resistance of CDDP-resistant ovarian and lung cancer cells to CDDP.

MTT cell viability assays of cancer cells with treatments of cisplatin alone and in combination with TMPD:(A) HeLa cells; (B) NIH:OVCAR-3 cells. The cells were treated by vehicle alone (0.2% EtOH),various concentrations of CDDP alone,and their combinations ...

To show that the DET reaction between TMPD and CDDP indeed led to observed synergetic killing of cancer cells, we additionally tested the combination by using a stock solution of TMPD prepared in pure H2O for 24h as a negative control. According to spectroscopic results in , the autoionization (i.e., electron knockout from TMPD to form the TMPD+) had mainly occurred prior to adding to the cell culture medium and therefore no or few DET reactions were expected to occur between CDDP and the so-prepared (ionized) TMPD. The results for cell viability measurements on cells so treated are shown in Fig.S5, which indeed shows that the enhancements in cell killing of CDDP by the presence of ionized TMPD (mainly TMPD+) were negligibly small compared with those with fresh TMPD prepared in EtOH. These electron-knockout results therefore confirm that the DET killing mechanism was responsible for strong synergetic effects in killing cancer cells in the combination of CDDP with unoxidized TMPD.

To study whether the TMPD-enhanced cytotoxicity of CDDP was due to the induction of apoptosis, we detected caspase activation and nuclear morphology of cancer cells treated with CDDP alone and in combination with TMPD. As shown in , the combination treatment resulted in a significant activation of caspase-3 and -7 in the NIH:OVCAR-3 cells, which was evident from the significant enhancement in green fluorescence. Also, the results from the Hoechest 33342 staining showed that the combination significantly enhanced the nuclear fragmentation/condensation of treated cells. From these images, we estimated that the combination of CDDP with TMPD led to an enhancement in apoptosis by a factor of 35 compared to treatment with CDDP alone.

Fluorescence microscopy observation of NIH:OVCAR-3 cells undergoing apoptosis induced by treatments indicated for 10h. Hoechest 33342 staining (blue fluorescence) detects the nuclear fragmentation/condensation, while green fluorescence of FLICA ...

To further confirm the above results, we used flow cytometry to quantify apoptotic cells and analyze cell cycle profiles. A landmark of cellular self-destruction by apoptosis is the activation of nucleases that eventually degrade the nuclear DNA into fragments. Detection of these fragments is relatively straightforward to quantify apoptotic cells and can be conducted using the APO-BrdU TUNEL Assay Kit. As shown in for the NIH:OVCAR-3 cells, the 24-h treatment of 100M TMPD alone resulted in a small percentage of DNA fragmentation (approximately 0.3%), while the treatments with CDDP alone increased DNA fragmentation to 2.6% at 30M and 3.2% at 50M. In contrast, the DNA fragmentation contents were significantly increased to 5.0% and 16.9% for 30 and 50M CDDP in combination with 100M TMPD, respectively, 25 times those expected for the additive effect of the two agents. Similar enhancements for HeLa cells treated with various concentrations of CDDP plus 100M TMPD are shown in Fig.S6. These data clearly show that the combination of CDDP with TMPD resulted in large synergetic enhancements in apoptosis of both cisplatin-sensitive and -resistant cancer cells.

APO-BrdU DNA fragmentation assay of NIH:OVCAR-3 cells with treatments indicated for 24h. Cells in the region above the line in each histogram are BrdU-positive cells (cells exhibiting DNA fragmentation); the line was drawn based on the samples ...

We also performed cell cycle analysis of cancer cells treated by the combination of cisplatin with TMPD. The results for NIH:OVCAR-3 cells are shown in Fig.S7 and SI Text. Overall, the observed results in and , and Figs. S6 and S7 have clearly demonstrated a synergetic increase in apoptosis of the cells treated by the combination of cisplatin with TMPD.

One of the practical ways to improve the therapeutic efficacy of cisplatin is to combine it with a molecular promoter in a synergetic manner. According to the DET mechanism (13, 14), CDDP is an extremely reactive compound with electrons. As it enters a cell, CDDP may react with e--donating components such as amino acids in proteins other than G bases in DNA, leading to the loss of its target to DNA and, hence, the cytotoxic activity. This loss in capacity to damage DNA is likely to result in insensitivity of cancer cells to CDDP (17). Indeed, intracellular CDDP inactivation by glutathione has been proposed as a mechanism of CDDP resistance (16, 46, 47). Glutathione is a cellular antioxidant, preventing damage to important cellular components caused by reactive radicals (48). It reduces disulfide bonds formed within cytoplasmic proteins to cysteines by serving as an electron donor. Thus, glutathione may direct cisplatin to target protein sites relatively distant from DNA in the cell. Reversely, the resistance may be circumvented by activating CDDP with an e--donating compound like TMPD to produce reactive radicals that then lead to DNA damage and cell death.

In the present model study, our results from spectroscopic measurements confirm that the exemplary e--donating compound, TMPD, indeed has a strong DET reaction with CDDP, consistent with the expectation from the DET mechanism of CDDP (13, 14, 34). Furthermore, our gel electrophoresis results on plasmid DNA show that the DET reaction indeed leads to DNA double-strand breaks. Correspondingly, the results from cell viability and apoptosis measurements show that the combination of CDDP with TMPD significantly reduced the doses of CDDP required to kill both CDDP-sensitive cervical cancer cells and highly CDDP-resistant ovarian cancer cells in a synergetic manner. And most strikingly, we found that adding TMPD led to a complete killing of highly CDDP-resistant ovarian cancer cells. It might be considered that the effect of TMPD may involve not the reversal of CDDP resistance but a simple enhancement of toxicity of CDDP. As confirmed in B, however, even for NIH:OVCAR-3 cells treated with CDDP alone in extremely high concentrations of 200300M (at which CDDP is supposed to have an extremely high toxicity), a large fraction (40%) of the cancer cells still survived. Only with TMPD did we observe that NIH:OVCAR-3 cells became sensitive to CDDP. It is more likely that the DET reaction of CDDP with TMPD competed with or suppressed the reaction causing CDDP resistance (e.g., the reaction of CDDP with glutathione or cysteines in proteins in the cell). As a result, the highly CDDP-resistant cancer cells became sensitive to CDDP and the drug resistance was circumvented. It should also be noted that TMPD also enhanced the killing of CDDP-sensitive HeLa cells. However, this is not surprising because the DET reaction between CDDP and TMPD in CDDP-sensitive cells will also increase the yield of reactive radicals that lead to DNA damage and apoptosis.

Moreover, the tumor microenvironment is well known to be associated with hypoxia in cancer biology. This has multiple consequences for tumor progression and treatment outcome (17, 31). However, hypoxia could be an advantage if a therapy activated only in hypoxic cells were designed; this would provide a method of killing cancer cells while having no or little harm to healthy cells in normal tissue. Thus, the unique presence of hypoxia in human tumors could provide an important target for selective cancer therapy (49). In our unique design, the combination of TMPD and cisplatin should predominantly target hypoxic cancer cells, while the remaining oxic cells will only be killed with the low dose of cisplatin. This is due to the consideration that DET will be mainly effective in a hypoxic environment, while in an oxic environment the electron-donating agent will easily become oxidized and lose its reaction capability with cisplatin. In fact, this has partially been demonstrated by our results of negative control experiments on cancer cells treated by the combination of ionized (oxidized) TMPD with cisplatin. Thus, a preferential targeting of tumor cells is expected to be achieved; further studies will be needed.

Finally, due to the wide application of CDDP in treating various human cancers, the strategy developed from this study may have potential for improving the treatment of multiple types of cancer beyond ovarian, cervical, and lung cancers. Furthermore, a similar structural feature exists in all clinically active platinum-based chemotherapy agents, and the DET reaction mechanism is expected to operate for all such Pt-based agents, such as oxaliplatin and carboplatin (13, 14). Thus, this combination strategy is expected to be applicable to all Pt-based chemotherapy. Through an increasing number of successful case studies, our results have demonstrated the potential of femtomedicine as an exciting new frontier to yield breakthroughs in understanding fundamental biological processes and improving the therapeutic efficacy of human diseases such as cancer.

1. Kavarnos GJ, Turro NJ. Photosensitization by reversible electron-transfer-theories, experimental-evidence, and examples. Chem Rev. 1986;86:401449.

2. Marcus RA. Electron transfer reactions in chemistry: Theory and experiment (Nobel Lecture) Rev Mod Phys. 1993;65:599610.

5. Lu QB. Cosmic-ray-driven electron-induced reactions of halogenated molecules adsorbed on ice surfaces: Implications for atmospheric ozone depletion and global climate change. Phys Rep. 2010;487:141167.

31. Lehnert S. Biomolecular Action of Ionizing Radiation. New York: Taylor & Francis Ltd.; 2007.

35. Jurtshu KP, Jr., McQuitty DN, Riley WH. Use of 3,3-diaminobenzidine as a biochemical electron-donor for studies on terminal cytochrome-oxidase activity in azotobacter-vinelandii. Curr Microbiol. 1979;2:349354.

40. Hirata Y, Ichikawa M, Mataga N. Monophotonic ionization from the vibrationally unrelaxed excited singlet state of N,N,N,N-tetramethyl-p-phenylenediamine in acetonitrile solution. J Phys Chem. 1990;94:38723874.

41. Shimamori H, Musasa H. Spectroscopic observation of contact ion pairs formed from photolyzed TMPD and TMB in the presence of halogenated compounds in nonpolar solvents. J Phys Chem. 1995;99:1435914364.

42. Egdell R, Green JC, Rao CNR. Photoelectron spectra of substituted benzenes. Chem Phys Lett. 1975;33:600607.

43. Fujiwara H, Fukumoto H, Fukumura H, Masuhara H. Dynamics of excited and ionic states of N,N,N,N-tetramethyl-p-phenylenediaminein poly(methyl methacrylate) underablation condition. Res Chem Intermediat. 1998;24:879892.

See the article here:
Electron transfer-based combination therapy of cisplatin ...