Cellular Target of a Rhodium Metalloinsertor is the DNA Base Pair Mismatch.

A Journal of Accepted Article Title:Cellular Target of a Rhodium Metalloinsertor is the DNA Base Pair Mismatch Authors:Kelsey M. Boyle, Adela Nano, Catherine Day, and Jacqueline K. Barton This manuscript has been accepted after peer review and appears as an Accepted Article online prior to editing, proofing, and formal publication of the final Version of Record (VoR). This work is currently citable by using the Digital Object Identifier (DOI) given below. The VoR will be published online in Early View as soon as possible and may be different to this Accepted Article as a result of editing. Readers should obtain the VoR from the journal website shown below when it is published to ensure accuracy of information. The authors are responsible for the content of this Accepted Article. To be cited as: Chem. Eur. J. 10.1002/chem.201900042 Link to VoR: http://dx.doi.org/10.1002/chem.201900042 Supported by 10.1002/chem.201900042 COMMUNICATION Cellular Target of a Rhodium Metalloinsertor is the DNA Base Pair Mismatch Kelsey M. Boyle, Adela Nano, Catherine Day, and Jacqueline K. Barton* Abstract: Defects in DNA mismatch repair (MMR) are commonly found in various cancers, especially in colorectal cancers. Despite the high prevalence of MMR-deficient cancers, mismatch-targeted therapeutics are limited and diagnostic tools are indirect. Here, we examine the cytotoxic properties of a rhodium metalloinsertor, [Rh(phen)(chrysi)(PPO)]2+ (RhPPO) in 27 diverse colorectal cancer cell lines. Despite the low frequency of genomic mismatches and the non-covalent nature of the RhPPO-DNA lesion, RhPPO is on average 5 times more potent than cisplatin. Importantly, the biological target and profile for RhPPO differs from that of cisplatin. A fluorescent metalloinsertor, RhCy3, was used to demonstrate that the cellular target of RhPPO is the DNA mismatch. RhCy3 represents a direct probe for MMR-deficiency and correlates directly with the cytotoxicity of RhPPO across different cell lines. Overall, our studies clearly indicate that RhPPO and RhCy3 are promising anticancer and diagnostic probes for MMR-deficient cancers, respectively. 2+ H 3 C CH 3 H O H 3 N Cl Deficiencies in cellular mismatch repair (MMR) machinery N N Pt are a hallmark of 14% of colorectal cancer cases and up to 20% Rh of all solid tumors.[1,2] Cells with MMR-deficiencies (MMR−) cannot N N H N Cl H 3 repair DNA mismatches or insertions/deletions (indels), leading to N a relative abundance of these lesions in these cells. Mismatches could thus serve as a unique druggable target that has yet to be RhPPO Cisplatin utilized in a clinical setting. Furthermore, these lesions could serve as a target for direct detection and diagnosis of MMR 4+ deficiencies in tumors, something that is commonly measured N+ H indirectly through tests of mutational frequency in microsatellite O N regions instead of number of mismatches.[2b] Such a diagnostic NH N N O O N would further support the promise of a mismatch- and indel- HN Rh N targeted therapeutic agent. Our group and others have addressed N N this need through the development of mismatch-targeted small molecules.[3] Specifically, our group has developed a unique RhCy3 family of metal complexes, called rhodium metalloinsertors, that can selectively target thermodynamically destabilized regions of Figure 1. The binding and structure of a metalloinsertor and cisplatin. As DNA, such as base pair mismatches (Figure 1), small indels, and observed crystallographically in previous studies, a classic metalloinsertor binds abasic sites, making the metal complexes an ideal candidate for selectively to a mismatch in DNA (top left, PDB 3GSK), while cisplatin binds to a d(GpG) site in DNA (top right, PDB 1AIO). The structures of RhPPO (middle targeting and detecting the DNA lesions found in MMR-deficient left), cisplatin (middle right), and RhCy3 (bottom). tumors.[4–7] Rhodium metalloinsertors have been rigorously studied in To understand more fully the potential clinical applicability of several pairs of matched cancer cell lines which differ primarily in rhodium metalloinsertors, we examined our most potent and the presence or absence of functioning MMR machinery.[8,9] In selective metalloinsertor, [Rh(phen)(chrysi)(PPO)]Cl2 (RhPPO, every matched pairing, metalloinsertors are significantly more Figure 1), across 27 CRC cell lines (Table S1).[12,13] These cell cytotoxic towards the MMR− cell line compared to their MMR- lines represent a diverse set of tumors, spanning the four proficient (MMR+) counterpart (a feature referred to as selectivity). subtypes of CRC and both MMR− and MMR+ phenotypes.[14,15] These results demonstrate that rhodium metalloinsertors can The toxicities of RhPPO, which selectively targets MMR target MMR deficiencies in cells, however they do not prove the deficiencies, and the non-selective FDA-approved specific cellular target of metalloinsertors to be DNA mismatches. chemotherapeutic cisplatin, which covalently binds the abundant Furthermore, these cell pairings are not reflective of the diversity d(GpG) motifs present in all DNA (Figure 1), were assessed in of clinical colorectal cancer (CRC) cases; in reality, the this cell line panel using a luciferase-based luminescence assay differences between tumors in two different patients or healthy which measures ATP from living cells. Dose-response curves and and cancerous tissues in a single patient will be far greater than corresponding IC50 values (50% inhibitory concentration) were just the presence or absence of a single MMR protein.[10,11] determined for each therapeutic and are shown in Figure 2 and Table S2. The IC50 values of RhPPO in different cell lines span nearly three orders of magnitude, ranging from 63 ± 3 nM to 18 ± [a] K. M. Boyle, Dr. A. Nano, C. Day, Prof. J. K. Barton 3 μM. Despite the low number of genomic mismatches, RhPPO Division of Chemistry and Chemical Engineering is more potent than cisplatin in nearly every cell line, with the IC50 California Institute of Technology values of RhPPO being on average 5 times lower than those of 1200 East California Boulevard, MC 127-72, Pasadena, California cisplatin (2.9 μM vs. 13.2 μM, respectively). This result is 91125 (USA) Email: jkbarton@caltech.edu remarkable considering that DNA mismatches are significantly less abundant than d(GpG) sites and metalloinsertors interact Supporting information for this article is given via a link at the end of the only through non-covalent stacking with these mismatches. The document. potency of a therapeutic in cell culture has long been considered tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved. COMMUNICATION a key predictor of its clinical success; potent therapeutics have sensitivities of some cell lines are contrary to what we expected the potential for low dosing conditions, based solely on MMR status; some MMR− cell lines (DLD-1, HCT15, and CW2) show minimal sensitivity to RhPPO, whereas some MMR+ cell lines (HT29, WiDr, Ls123, and Colo205) show high sensitivity to RhPPO. Overall RhPPO shows moderate selectivity towards the MMR− cell lines (average IC50 of 2.5 μM) compared to MMR+ cell lines (average IC50 of 3.0 μM), and the selectivity increases further when looking only at cell lines with deficiencies in MLH1 or MSH2, the two most essential MMR proteins (average IC50 of 2.1 μM, Figure S1). These results are promising and follow the expected trend, however they are not as significant as anticipated (Figure S1). The range observed for both MMR− and MMR+ cell lines can be rationalized; unlike in matched cell lines, cell lines in this panel differ in mutations and regulation of many proteins.[10,11] Accordingly, there are several factors that could obscure the strong MMR− selectivity we expected based on our hypothesis. We investigated two such factors that seemed likely to influence metalloinsertor toxicity: cellular uptake and the number of lesions in genomic DNA that can be targeted by metalloinsertors. Cell lines can exhibit different uptake and efflux properties towards small molecule therapeutic, therefore differences in uptake between cell lines may explain the wide cytotoxicity range of RhPPO.[18,19] We measured the whole cell uptake of RhPPO after 24 hours in various cell lines by ICP-MS to determine if the whole cell uptake of RhPPO correlated with cytotoxicity (Figure 3 and Figure S2). A significant correlation (Pearson’s r = -0.63, p < 0.01) was observed between increasing RhPPO uptake and decreasing IC50. Furthermore, several of the results contrary to our hypothesis (i.e. high IC50 in MMR− cells, low IC50 in MMR+ cells) are clarified by this assay; the three MMR− cell lines least sensitive to RhPPO (DLD-1, HT29, CW2) show the lowest cellular 40 uptake and two of the most sensitive MMR+ cell lines (Ls123 and Colo205) exhibit the highest cellular uptakes. For these cell lines, 30 high or low cellular uptake of RhPPO likely obscures the 20 selectivity that would normally be observed on the basis of MMR 10 status alone. While this correlation between uptake and 0 cytotoxicity is intuitive, it is of note that there are few reported studies correlating cellular uptake and cytotoxicity of a small molecule therapeutic across different cell lines.[20] More commonly, reports examine the correlation of cellular uptake and cytotoxicity of different therapeutics in a single cell line or look only at a relatively small number of cell lines.[21] Therefore, our results comparing cytotoxicity and cellular uptake suggest that uptake may often play a non-negligible role in the cytotoxicity differences of a small molecule therapeutic between cell lines. While a correlation between cytotoxicity and uptake is expected for any small molecule therapeutic, a correlation between cytotoxicity and DNA binding would only be expected if DNA were the relevant biological target of the therapeutic being studied. As discussed previously, inactivation of MMR proteins confers the cells with an increased level of uncorrected Figure 2. Cytotoxicity of RhPPO and cisplatin in 27 colorectal cancer cell lines. Dose response curves of RhPPO and cisplatin in CRC cell lines (top). Dose mismatches and indels that propagate into mutations upon response curves of RhPPO in MMR− and MMR+ CRC cell lines (middle top). replication.[3] The number of these lesions in the genome can Direct IC50 comparison of RhPPO and cisplatin in 27 CRC cell lines (middle fluctuate between cell lines, for instance mutations (an indirect bottom) and boxplot representation, with an average IC50 for RhPPO of 3.02 measure of mismatches and indels) occur at different rates in cell μM and a median of 1.34 μM, and an average IC50 for cisplatin of 13.89 μM and a median of 11.62 μM (bottom). lines deficient in different MMR proteins.[22] The number of these lesions present in the genomic DNA (gDNA) of a cell could low off-target effects, and minimal solubility issues.[16,17] Therefore, influence differences in potency of RhPPO, which targets these destabilized DNA features, in different cell lines.[5,7] Currently, the high potency of RhPPO highlights its great therapeutic there are few techniques developed for the detection and potential. Furthermore, in the cell lines least sensitive to cisplatin sensitive measurement of destabilized lesions in gDNA, such as (Colo205, HT29, and WiDr), RhPPO is over 100 times more polymerase chain reaction (PCR) amplification, but these potent than cisplatin (Table S2), suggesting it could be a techniques are destructive and time-consuming.[23] Fluorescence- particularly useful therapeutic for treatment of clinically based probes have been widely used to visualize and quantify challenging cisplatin-resistant tumors. dynamic processes in live cells via interaction with various When considering MMR status, a wide range of sensitivities biological targets and are great candidates for screening of is observed for both MMR− and MMR+ cell lines (Figure 2). The damaged DNA.[24] )Mμ( 05CI 2oCaC 502oloC D023oloC 876oloC 2WC 1-DLD 8992CCH 51-TCH 611TCH 92TH 21MK oVoL 4301SL 321SL T471SL 805H-ICN 617H-ICN OKR 6111WS 3641WS 304WS 84WS 084WS 026WS 738WS 849WS rDiW 1.2 1 0.8 0.6 0.4 0.2 0 0.01 0.1 1 10 100 ytilibaiV MMR+ MMR− [RhPPO] (μM) nitalpsiC OPPhR 1.2 1 0.8 0.6 0.4 0.2 0 0.01 0.1 1 10 100 p < 0.01 0 10 20 30 IC50 (µM) ytilibaiV 10.1002/chem.201900042 Cisplatin RhPPO [Metal] (μM) tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved. COMMUNICATION consider MMR+ cells to have regular MMR expression and MMR− to underexpress MMR proteins. If we randomly chose only two 1.5 cell lines from our panel, one MMR+ and one 1 0.5 0 -0.5 -1 -1.5 0 3 6 9 12 Figure 3. A correlation between whole cell uptake and IC50 for RhPPO. A correlation of −0.63 is observed. MMR− cell lines are shown in red and MMR+ are shown in blue, with select cell lines labeled. As such, our group recently reported a bifunctional fluorescent probe, RhCy3, which exhibits a fluorescent light-up effect upon interaction with thermodynamically destabilized mismatches in gDNA (Figure 1).[25] The fluorescence of RhCy3 is an exceptional readout on the relative number of destabilized lesions in gDNA and an excellent predictor of the relative number of targetable DNA lesion for RhPPO, which is structurally similar. Here we use this probe to better understand the cytotoxic effect of RhPPO on a panel of cancer cell lines, but these studies also demonstrate the powerful detection and diagnostics properties of RhCy3 in MMR− cancers. We performed fluorescence titrations with RhCy3 and increasing amounts of gDNA extracted from a test set of eight cell lines that span deficiencies in different MMR genes (Table S2).[26] As can be seen Figure 4. RhCy3 fluorescence assay with gDNA. Full fluorescence titrations of in Figure 4, a correlation (r = -0.52) was observed between gDNA extracted from eight different cancer cell lines, with [gDNA] as per base increasing RhCy3 fluorescence and decreasing IC50 of RhPPO. pairs, I as the emission integral from 548 – 675 nm as scalar function measured By removing the potential outlier, DU145 (the only cell line tested after each addition of DNA, and I0 is the emission integral of RhCy3 solution mutated in two MMR proteins), the correlation improves without gDNA (top). Correlation of r = −0.52 is observed between IC50 value dramatically, (r = -0.81, p <0.05), suggesting other factors may and max I/I0 for all cell lines (bottom). Removal of a potential outlier (circled) influence the cytotoxicity of RhPPO or fluorescence of RhCy3 in leads to correlation of r = −0.81 between IC50 value and max I/I0 for all cell lines. DU145. This strong correlation between the IC50 of RhPPO and the fluorescence of the reporter RhCy3 confirms that the effective MMR−, we could observe every possible trend. Comparing RKO biological target of rhodium metalloinsertors is, in fact, DNA (MMR−, IC50: 120 nM) and Colo320DM (MMR+, IC50: 18.0 μM) lesions such as mismatches and indels, and that differences in would suggest RhPPO is dramatically more toxic in MMR− cells, the number of these lesions between different cell lines controls however comparing CW2 (MMR−, IC50: 9.2 μM) and Colo205 cytotoxicity of metalloinsertor therapeutics. Notably, there is a (MMR+, IC50: 63 nM) would suggest the opposite trend, with clear relationship between the identity of MMR protein and RhCy3 RhPPO being dramatically less toxic in MMR− cells. Overall, we fluorescence output, showing RhCy3 can serve as a direct believe the large range of IC50 values observed here serves as a detection method of destabilized lesions in mismatch-repair point of caution for researchers performing in vitro studies in a deficient tumors and a potential diagnostic for these cancers limited number of cell lines. Cell line selection can unintentionally (Supplemental Discussion and Figure S3). but dramatically influence the trends a researcher observes in The results presented here highlight some interesting their studies, therefore we encourage researchers to perform considerations for in vitro studies performed in any laboratory. We these studies with larger panels of cell lines and to supplement observed a large range of IC50 values spanning nearly three them using matched cell lines, which reduce the inter-cell line orders of magnitude for a single small molecule therapeutic variation and allow one to observe the effect of a therapeutic on a across 27 cell lines. This result alone has significant implications specific target. for in vitro experiments. Many studies examine a therapeutic of In summary, the experiments described here underscore interest in a single cell line or one cell line from several types of the therapeutic and diagnostic potentials of mismatch-targeting cancer (colorectal, ovarian, etc.), but a single cell line cannot small molecules. The potency of RhPPO across diverse cell lines represent cancer or any subtype of cancer as a whole. It is also spans nearly three orders of magnitude and shows selectivity common to compare cytotoxicity in unmatched cell lines that differ towards MMR-deficient cancer cells. RhPPO is on average 5 in the expression level of a protein of interest (regular expression, times more potent than cisplatin, despite having a less abundant overexpression, and underexpression). Our results here suggest target to which it binds non-covalently. Overall, these results show that using only a small number of unmatched cell lines may RhPPO is a potent and promising therapeutic agent for colorectal produce misleading results. For example, in this study we could cancers, and in vivo experiments involving mouse models are in progress. Significantly, the fluorescent probe RhCy3 provides )05CI( goL HCT15 CW2 7 DLD-1 6 5 Ls123 4 3 Colo205 2 1 Uptake (ng Rh/mg protein) 0 0 100 200 300 400 500 600 I/I 0 DU-145 AN3-CA HCT116O HCT116 DLD-1 HEC-1A SW480 HCT116N [gDNA] µM 0.5 0 -0.5 -1 -1.5 2.5 3.5 4.5 5.5 6.5 )05CI(goL 10.1002/chem.201900042 I/I at 600 µM gDNA 0 tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved. 10.1002/chem.201900042 COMMUNICATION clear evidence that destabilized DNA regions, such as membrane integrity, each cell suspension was sonicated for 20 s at 40% mismatches, represent the cellular target for the metalloinsertors, amplitude on a Qsonica Ultrasonic sonicator, then frozen and lyophilized and that targeting these lesions leads to cell death. As such, for 72 h. The resulting cell particulate was suspending in 1 mL of 6% nitric acid and heated at 110 °C for 8 h to facilitate total digestion prior to ICP- RhCy3 opens up great perspectives for development of new MS analysis. Each sample was then diluted to 2% nitric acid and methods in direct detection and fast quantification of destabilized centrifuged to separate any undigested cell components. The solutions lesions in genomic DNA and as a fluorescent diagnostic tool for were analyzed for Rh content on an Agilent 8800 Triple Quadrupole ICP- MMR deficient cancers. MS. The concentration of Rh in each sample was determined by comparison to a standard curve ranging from 0.01 to 100 ppb. Rh concentrations were normalized to the protein content of each sample determined by BCA assay. The measurements were repeated two times Experimental Section using two biological replicates for each cancer cell line. Genomic DNA extraction and purification. The genomic DNA was Materials. All commercially available reagents were used as received. The extracted and purified using PureLink® Genomic DNA Kits following the metalloinsertors [Rh(phen)(chrysi)(PPO)]Cl2 (RhPPO) and RhCy3 were manufacturer’s protocol. Prior to DNA extraction, the cells subjected to synthesized and purified following published protocols.12,25 Cell culture genomic DNA extraction were seeded and grown in their respective media, supplements, and PureLink™ Genomic DNA Mini Kits were cellular medium at < 5 x 106 cells/mL. The lysates were prepared by purchased from Life Technologies (Carlsbad, CA). CellTiter-Glo® removing the growth medium from the culture plate and cells were Luminescent Cell Viability Assay kits were purchased from Promega harvested by trypsinization then re-suspended in 200 μL PBS. (Madison, WI). BCA Protein Assay Kits were purchased from Pierce ProteinaseK (20 μL) and RNase (20 μL) were added to the sample, mixed (Waltham, MA). Cell lines used in the experiment were purchased from by vortexing and incubated at room temperature for 2 min. 200 μL of ATCC (Manassas,VA) or provided by collaborators at AMGEN (Thousand PureLink® Genomic Lysis/Binding Buffer were added, mixed and vortexed Oaks, CA). to obtain a homogenous solution. The samples were incubated at 55 °C for 10 min to promote digestion then 200 μL of 96-100% ethanol was Cell Culture. The specific growth conditions of each cell line, including the added to the lysate which was further mixed by vortexing for 5 s. The DNA type of medium and added supplements, can be found in Table S1. In was washed by adding 500 μL of Wash Buffer 1 then Wash Buffer 2 general, cell lines were grown in RPMI 1640, DMEM, McCoy’s 5A, or provided by the kit, followed by DNA eluting process using the spin Ham’s F-12K media supplemented with 10% FBS (20% FBS for the cell columns. The spin columns were eluted with sterile MilliQ water (200μl) line CaCo2), 100 units/mL penicillin, 100 units/mL streptomycin. Cells two times to recover a maximum of genomic DNA. The samples were were grown in tissue culture treated flasks at 37 °C under a humidified 5% lyophilized and the dry DNA was solubilized in Tris buffer solution (5 mM CO2 atmosphere. Tris, 50 mM NaCl, pH = 8.0) in order to obtain a highly concentrated CellTiter-Glo Viability Assay. CellTiter-Glo Luminescent Cell Viability solution. The concentration of gDNA solutions were determined using a Assays were performed following the protocols provided in the kit. Briefly, NanoDrop 2000 Spectrophotometer (Thermo Scientific) by pippeting 2 μL cell lines were plated at a density of 10,000 cells in 100 μL media per well of the sample solution. The samples purities were determined by obtaining in an opaque, tissue culture treated 96-well plate and allowed to adhere the absorbance ratios at A260/A280 nm and A230/A260 nm using a for 24 h. One of two compounds, RhPPO or cisplatin, was added to each NanoDrop 2000 Spectrophotometer and are reported in Table S3). The well at a final concentration of 0-150 μM, and the cells were allowed to concentrations of the stock solutions of gDNA used during the incubate with the therapeutic for 72 h. After incubation with a therapeutic fluorescence titrations were adjusted at 3140 ng/μl (4.7 mM base pairs agent, the cell solutions were treated with an equal volume of the CellTiter DNA) in Tris buffer (200 mM NaCl, 5 mM Tris, pH 8.1). Glo reagent, which contains beetle luciferin and a recombinant luciferase. Fluorescence titrations with genomic DNA. Luminescence spectra The luciferase can catalyze a reaction between the luciferin and ATP were recorded using a QE Pro High Performance Spectrometer with a provided by viable cells to produce a luminescence that is proportional to back-thinned, TE-cooled CCD detector controlled by the OceanView data the number of viable cells. Luminescence was recorded on a FlexStation acquisition and analysis software package (Ocean Optics, Inc.). Sample 3 Multi-Mode Plate Reader with integration time of 0.500 seconds. Percent excitation was provided by a 455 nm LED (Thorlabs model M455L2). The viability was determined by the ratio of the luminescence of therapeutic- fluorescence titrations in this study were performed with genomic DNA treated cells compared to untreated cells. IC50 values were determined by extracted from eight cancer cell lines characterized by different fitting the cell viability curve to a sigmoidal curve in OriginPro v 8.5 and phenotypes (HCT116N, HCT116O, HCT116, DLD-1, HEC-1A, SW480, using the resultant parameters to calculate the concentration at which 50% AN3-CA, DU-145). The emission spectra were recorded in Tris buffer of cells were viable. Each therapeutic dose was performed in triplicate and solution (5 mM Tris, 200 mM NaCl, pH = 7.4) at 25 °C using a water each experiment was repeated 2-3 times to confirm reproducible viability circulation system. Excitation wavelength was λEx = 455 nm and emission curves. For statistical analyses, cell lines from a common patient (DLD- integral was reported after each addition of genomic DNA, as a scalar 1/HCT15, HT29/WiDr, SW480/SW620) were averaged and counted as a function from 548 to 675 nm. The measurements were repeated three single cell line to avoid double-counting cancer from a single patient. times using three biological replicates for each cancer cell line. ICP-MS Assay for Whole Cell Uptake of RhPPO. Whole cell uptake experiments were performed following previously published protocols with slight modifications.13 Briefly, cells were plated at a density of 1,000,000 cells in 3 mL media per well in a 6-well plate and allowed to adhere for 24 Acknowledgements h. Cells were then treated with RhPPO to a final concentration of 0.5 μM. This concentration was selected to be great enough to ensure Rh detection by ICP-MS, but low enough to avoid significant cell death in sensitive cell We gratefully acknowledge Dr. Julie Bailis, for providing the cell lines (which could lead to challenges in data analysis). Cells were allowed lines used in this study. We thank NIH and the Moore Foundation to incubate for 24 h with the metalloinsertor, as we had previously for financial support and the Department of Defense for observed that metalloinsertor uptake plateaus in both HCT116N and supporting K.M.B. through the National Defense Science & HCT116O cells by 24 h.13 For adherent cell lines, the rhodium-containing Engineering Graduate Fellowship program. medium was aspirated from each well after 24 h and each well was washed 2x with 1 mL of phosphate-buffered saline (PBS) then harvested by trypsinization and transferred to centrifuge tubes. For mixed or Keywords: antitumor agents • cellular uptake • DNA recognition suspended cell lines, the rhodium-containing medium was transferred to a • cytotoxicity centrifuge tube before the PBS rinses and trypsinization. Harvested cells were centrifuged at 1500 rpm for 5 minutes. The supernatant was decanted and the cell pellet was suspended in 1 mL PBS. Centrifugation [1] (a) I. I. Arzimanoglou, F. Gilbert, H. R. K. Barber Cancer 1998, 82, and PBS washing was repeated three times total (for mixed/suspension 1808-1820. (b) T. A. Kunkel, D. A. Erie, Annu. Rev. Genet. 2015, 49, cell lines, the suspended and trypsinized aliquots were combined during 291–313. the second wash). An aliquot from the final suspension was reserved and [2] (a) J. Guinney, R. Dienstmann, X. Wang, A. De Reyniès, A. Schlicker, analyzed for protein content using a Pierce BCA Protein Assay Kit C. Soneson, L. Marisa, P. Roepman, G. Nyamundanda, P. Angelino, following the manufacturers instructions. To lyse the cells and destroy et al., Nat. Med. 2015, 21, 1350–1356. (b) S. B. Hatch, H. M. Lightfoot Jr., C. P. Garwackie, D. T. Moore, B. F. Calvo, J. T. Woosley, J. tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved. 10.1002/chem.201900042 COMMUNICATION Sciarrotta, W. K. Funkhouser, R. A. Farber, Clin. Can. Res., 2005, 11, S. A. Danielsen, M. Bjørnslett, L. A. Meza-Zepeda, M. Eknæs, G. E. 2180-2187 Lind, et al., Mol. Cancer 2017, 16, 1–16. [3] (a) S. K. Fung, T. Zou, B. Cao, T. Chen, W.-P. To, C. Yang, C.-N. Lok, [15] J. F. Linnekamp, S. R. Van Hooff, P. R. Prasetyanti, R. Kandimalla, C.-M. Che, Nat. Comm. 2016, 7, 10655. (b) A. Granzhan, N. Kotera, J. Y. Buikhuisen, E. Fessler, P. Ramesh, K. A. S. T. Lee, G. G. W. M.-P. Teulade-Fichou, Chem. Soc. Rev., 2014, 42, 3630-3665. Bochove, J. H. De Jong, et al., Cell Death Differ. 2018, 25, 616–633. [4] B. A. Jackson, J. K. Barton, Biochemistry 2000, 39, 6176–6182. [16] M. P. Gleeson, A. Hersey, D. Montanari, J. Overington, Nat. Rev. [5] B. A. Jackson, V. Y. Alekseyev, J. K. Barton, Biochemistry 1999, 38, Drug Discov. 2011, 10, 197–208. 4655–4662. [17] C. Lipinski, Am. Pharm. Rev. 2002, 5, 82–85. [6] B. M. Zeglis, V. C. Pierre, J. T. Kaiser, J. K. Barton, Biochemistry [18] J. Mueller, I. Kretzschmar, R. Volkmer, P. Boisguerin, Bioconjug. 2009, 48, 4247–4253. Chem. 2008, 19, 2363–2374. [7] B. M. Zeglis, J. A. Boland, J. K. Barton, Biochemisrty 2009, 48, 839– [19] R. Krishna, L. D. Mayer, Pharm. Pharmacol. Commun. 1999, 5, 511– 849. 517. [8] J. R. Hart, O. Glebov, R. J. Ernst, I. R. Kirsch, J. K. Barton, Proc. Natl. [20] A. Dayton, K. Selvendiran, M. L. Kuppusamy, B. K. Rivera, S. Meduru, Acad. Sci. 2006, 103, 15359–15363. T. Kálai, K. Hideg, P. Kuppusamy, Cancer Biol. Ther. 2010, 10, 1027– [9] J. M. Bailis, M. L. Gordon, J. L. Gurgel, A. C. Komor, J. K. Barton, I. 1032. R. Kirsch, PLoS One 2013, 8, e78726. [21] K. J. Davis, J. A. Carrall, B. Lai, J. R. Aldrich-Wright, S. F. Ralph, C. [10] C. Tomasetti, L. Marchionni, M. A. Nowak, G. Parmigiani, B. T. Dillon, Dalt. Trans. 2012, 41, 9417–9426. Vogelstein, Proc. Natl. Acad. Sci. 2015, 112, 118–123. [22] W. E. Glaab, K. R. Tindall, Carcinogenesis 1997, 18, 1–8. [11] I. Martincorena, K. M. Raine, M. Gerstung, K. J. Dawson, K. Haase, [23] A. F. El-Yazbi, G. R. Loppnow, Anal. Chem. 2013, 85, 4321−4327. P. Van Loo, H. Davies, M. R. Stratton, P. J. Campbell, Cell 2017, 171, [24] H.-W. Liu, L. Chen, C. Xu, Z. Li, H. Zhang, X.-B. Zhang, W. Tan, 1029–1041. Chem. Soc. Rev. 2018, 47, 7140-7180 [12] A. C. Komor, J. K. Barton, J. Am. Chem. Soc. 2014, 136, 14160– [25] A. Nano, A. N. Boynton, J. K. Barton, J. Am. Chem. Soc. 2017, 139, 14172. 17301–17304. [13] K. M. Boyle, J. K. Barton, J. Am. Chem. Soc. 2018, 140, 5612–5624. [26] J. M. Bailis, A. G. Weidmann, N. F. Mariano, J. K. Barton, Proc. Natl. [14] K. C. G. Berg, P. W. Eide, I. A. Eilertsen, B. Johannessen, J. Bruun, Acad. Sci. 2017, 114, 6948–6953. tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved. 10.1002/chem.201900042 COMMUNICATION Entry for the Table of Contents (Please choose one layout) Layout 2: COMMUNICATION Kelsey M. Boyle, Adela Nano, Catherine Day, and Jacqueline K. Barton* Page No. – Page No. Cellular Target of a Rhodium Metalloinsertor is the DNA Base Pair Mismatch Ta rgeting DNA Mismatches: Rhodium metalloinsertors are a versatile family of complexes th at selectively bind to DNA mismatches in vi tro. A range of cytotoxicities is found for the metalloinsertor with higher potency than cispl atin across a panel of colorectal cell lines, and we use a fluorescent analogue to confirm that the DNA mismatch is the primary biological target for the metalloinsertor in cells. tpircsunaM detpeccA Chemistry - A European Journal This article is protected by copyright. All rights reserved.