Accurate Measurement of Formaldehyde–Induced DNA–Protein

Article pubs.acs.org/crt

Cite This: Chem. Res. Toxicol. 2018, 31, 350−357

Accurate Measurement of Formaldehyde-Induced DNA−Protein Cross-Links by High-Resolution Orbitrap Mass Spectrometry Chih-Wei Liu, Xu Tian, Hadley J. Hartwell, Jiapeng Leng, Liang Chi, Kun Lu,* and James A. Swenberg* Department of Environmental Sciences and Engineering, University of North Carolina at Chapel Hill, Chapel Hill, North Carolina 27599, United States ABSTRACT: Genomic instability caused by DNA−protein crosslink (DPCs)-induced DNA damage is implicated in disease pathogenesis, aging, and cancer development. The covalent linkages between DNA and protein are induced by chemical reactions catalyzed by the endogenous metabolic intermediates and exogenous agents, such as aldehydes, chemotherapeutic agents, and ionizing radiation. Formaldehyde has been classified as a genotoxic carcinogen. In addition, endogenous formaldehyde-induced DPCs may increase the risks of bone marrow toxicity and leukemia. There is a need to develop an effective detection method for DPC analysis, including the structural differentiation of endogenous and exogenous formaldehyde-induced DPCs. To this end, our group previously reported a useful liquid chromatography-selected reaction monitoring (LC-SRM) approach coupled with stable isotope labeling and low mass resolution-triple quadrupole mass spectrometry. In the present work, we further demonstrate an accurate quantification method using a high-resolution, accurate-mass Orbitrap mass spectrometer for the measurement of the covalent linkage between 2′-deoxyguanosine (dG) and cysteine (Cys), specifically termed dG-Me-Cys, one kind of linkages derived from the formaldehyde-induced DPCs. This quantification method with a wide dynamic range of at least 3 orders generates an interference-free spectrum for unbiased and unambiguous quantification, resulting in good intra- and interday precisions and accuracies with less than 10% variations. The endogenous and exogenous amounts of dG-Me-Cys in a human cell line treated with formaldehyde are analyzed by our new methodology. The quantification strategy demonstrated in this study can be widely applied to characterize and quantify other DPC linkages induced by formaldehyde or other chemical agents.

■

INTRODUCTION DNA damage caused by toxic chemicals generated from either endogenous metabolic pathways or exogenous environmental stimuli is implicated in disease pathogenesis and cancer development.1−5 The types of DNA damage include DNA single- and double-strand breaks, DNA adducts, abasic sites (Ap site), Ap site-mediated interstrand DNA−DNA cross-links, and DNA−protein cross-links (DPCs), among others.5−8 Within the bulky lesions caused by DPCs, the covalent bond formed between DNA and protein is induced by chemical reactions catalyzed by either endogenous metabolic intermediates3,9,10 or exogenous agents.11−13 DPCs form obstacles that affect normal DNA−protein interactions during DNA replication and transcription, which may result in the accelerated aging and cancer development.14 While it is very important to understand the molecular mechanisms involved in DPCs repair, currently very little is known. Recently, it was reported that DNAdependent proteases, Wss1 and SPRTN, might play an important role for DPC removal, protecting cells from DPC genotoxicity.1,15 Historically, the bottleneck for studying DPCs has been due to the nonselective methods used for isolation and detection. © 2018 American Chemical Society

Traditional DNA/protein precipitation methods are incapable of selectively recognizing and quantifying a specific structural DPC linkage from a complex DPC pool. Recently, the advancement in mass spectrometry-based detection and quantification methods has allowed for the characterization of specific structural DPCs that were directly induced by exogenous cross-linking agents, such as nitrogen mustards13 or diepoxybutane.16 Most importantly, formaldehyde, a wellknown cross-linking agent, has been classified as a carcinogen in animals and humans.9,17,18 Humans are exposed to formaldehyde through a wide array of industrial and environmental sources, as well as from endogenous formation during various cellular processes.19 Until recently, it has been impossible to differentiate exogenous from endogenous formaldehydeinduced DPCs. Lai et al. implemented the use of stable isotope labeled-formaldehyde in animal exposure experiments, which enabled the simultaneous quantification of endogenous and exogenous DPCs.10 In that study, the covalent linkage between the 2′-deoxyguanosine (dG) and cysteine (Cys), specifically Received: February 12, 2018 Published: April 13, 2018 350

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology

Scheme 1. Proposed Fragmentation Patterns of dG-Me-Cys and [15N5]-dG-[13CD2]-Me-Cys after HCD Fragmentationa

a The measured m/z and mass accuracy are shown in Table 1. The “N”, “C”, and “H” highlighted with red indicate the stable-isotope-labeled sites of [15N5]-dG-[13CD2]-Me-Cys.

affect the quantification accuracy, selectivity, and sensitivity due to the abundance of the co-eluting interferences in biological samples.23 Parallel reaction monitoring (PRM) is an emerging quantification method executed on a quadrupole Orbitrap mass spectrometer.24,25 PRM works by performing a similar selection of the target precursor ion on the quadrupole with an isolation window of 1−2 m/z, and then the full MS/MS spectrum is acquired at a high mass resolution. Following fragmentation, the Orbitrap mass analyzer provides the high mass accuracy. Post-acquisition extraction of the fragment ions with a tight tolerance greatly improves the selectivity and quantification by eliminating interferences from co-eluting species. Herein, we demonstrated the first study for structural detection and accurate quantification of dG-Me-Cys using Orbitrap mass spectrometry. The high mass resolution and mass accuracy provided by the Orbitrap mass analyzer showed the unbiased and accurate quantification of dG-Me-Cys. Our novel method using the advanced Orbitrap platform was further

termed dG-Me-Cys (Scheme 1), represents the formaldehydeinduced DPCs. Moreover, dG-Me-Cys linkage is one of the major covalent linkages derived from formaldehyde-induced DPCs. Specifically, the linkages between amino acids (cysteine, histidine, tryptophan, and lysine) and deoxynucleosides (2′deoxyguanosine, 2′-deoxyadenosine, and 2′-deoxycytidine) induced by formaldehyde have been identified by an in vitro experiment conducted by Lu et al.20 It is of note that traditional selected reaction monitoring (SRM) for targeted quantification was utilized in those three pioneering structural DPC studies. In traditional SRM performed on triple quadrupole (QqQ) mass spectrometry,21−23 the first and the third quadrupoles act as mass filters to specifically select predefined m/z values for a precursor ion. A specific fragment ion is generated from the fragmentation performed in the second quadrupole, which is used as a collision cell. A set of transitions (precursor/fragment ion pairs) are selectively monitored over time to reveal the chromatographic peak for quantification. The main drawback of using SRM for quantification is the low mass resolution of the quadrupole mass analyzer used for ion selection, which may 351

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology

of 10 μL of 30% acetic acid. After adding 8 fmol of internal standard ([15N5]-dG-[13CD2]-Me-Cys), the final reaction mixture was subjected to centrifugation at 12 000 × g at 4 °C for 40 min in a Nanosep Centrifugal Device to remove enzymes prior to HPLC purification. HPLC Purification and Fractionation of DPCs. The target analytes, endogenous and exogenous dG-Me-Cys, and their internal standard were purified from the filtrate using an Agilent 1200 Series UV HPLC System with two C18 reverse-phase columns connected in series (Waters Atlantis T3, 3 μm, 15 cm × 4.6 mm i.d.).10 The detection wavelength and column temperature were set at 254 nm and 15 °C, respectively. The mobile phases consisted of 0.05% acetic acid in water (A) and CH3CN (B). The flow rate was 0.45 mL/min, and the elution gradient conditions were set as follows: 0 min, 2% B; 3 min, 2% B; 42 min, 4.2% B; 43 min, 4.2% B; 43.5 min, 80% B; 46 min, 80% B; 47.5 min, 2% B; and 55 min, 2% B. The target analytes were eluted at a retention time range between 37 and 41 min. The fractions containing target compounds were combined and concentrated to approximately 20 μL using a vacuum concentrator before LC-MS/MS analysis. The amount of digested dG in each sample was quantitated by the UV peak area (λ = 254 nm) based on each freshly prepared calibration curve to estimate the dG amount in each sample loaded on the column. Nano-LC/ESI/MS/MS Analysis. LC-MS/MS analysis was conducted using an UltiMate 3000 RSLCnano system coupled to a Q Exactive HF Hybrid Quadrupole-Orbitrap mass spectrometer through an EASY-Spray ion source for nanoelectrospray (Thermo Fisher Scientific). DPCs were separated on a PepMap C18 analytical column (2 μm particle, 50 cm × 75 μm i.d., catalog number ES503, Thermo Fisher Scientific). At first, the DPCs were loaded into a C18 trapping column (5 μm particle, 0.5 cm × 300 μm i.d., catalog number 160454, Thermo Fisher Scientific) at a flow rate of 5 μL/min for 3 min using 0.1% formic acid in ddH2O as a loading solvent. A binary solvent system consisting of 0.1% formic acid in ddH2O (solvent A) and 0.1% formic acid in CH3CN (solvent B) was used for LC separation at a flow rate of 250 nL/min. LC separation was performed using the following gradient setting: held at 1% B for 3 min (trapping time), 1− 20% B in 16 min, 20−90% B in 0.1 min, held at 90% B for 9.9 min, 90−1% B in 0.1 min, and held at 4% B for 13.9 min for re-equilibrating the column. MS and MS/MS data were both acquired in the profile mode. One full scan coupled with the targeted PRM mode with an inclusion list comprised of m/z 401.12 (endogenous dG-Me-Cys), 404.14 (exogenous dG-[13CD2]-Me-Cys), and 409.13 (internal standard [15N5]-dG-[13CD2]-Me-Cys). The Orbitrap resolution for the full MS (380−420 m/z scan range) and MS/MS spectra was set to 240 000 and 60 000 at m/z 200, respectively. The automatic gain control (AGC) target for the full scan and MS/MS scan was 1 × 106 and 5 × 105 with a maximum fill time of 100 and 250 ms, respectively. Precursors in the targeted PRM mode were isolated with a window of 1.4 m/z and fragmented with HCD fragmentation (higher-energy collisional dissociation, normalized collision energy of 25). Calibration Curve. Standard curves were established by plotting the peak area ratios of solutions containing a fixed concentration of [15N5]-dG-[13CD2]-Me-Cys (internal standard) at 0.25 nM and increasing the concentrations of dG-Me-Cys (analytical standard) from 0.00390625 to 4 nM. The 2-fold serial dilution was performed from the highest stock solution of 4 nM of analytical standard using internal standard (0.25 nM) in 0.05% acetic acid as a diluent. The linear calibration curve was based on the loading amount ratio versus the integrated peak area ratio between the analytical and internal standards. The method validation was performed for consecutive intraday and interday evaluations. Data Analysis. The LC-MS/MS raw data generated from the Q Exactive HF instrument were analyzed by Xcalibur software (Thermo Fisher Scientific). The quantification analysis was performed using Skyline v3.7.0.11317,27 and the chromatographic peak area ratios of endogenous dG-Me-Cys (m/z 401.12377→m/z 164.05668) and exogenous dG-[13CD2]-Me-Cys (m/z 404.13948→m/z 167.07238) over that of internal standard [15N5]-dG-[13CD2]-Me-Cys (m/z

applied to detect and quantify both endogenous and exogenous formaldehyde-induced DPCs in human cell lines.

■

MATERIALS AND METHODS

Chemicals and Materials. Unless otherwise specified, all reagents and chemicals used in this study were purchased from Sigma-Aldrich (St. Louis, MO). Acetonitrile (CH3CN) solvent and formaldehyde solution (37%, w/w) were purchased from Thermo Fisher Scientific (Rockford, IL). [15N5]-2′-Deoxyguanosine and [13CD2]-formaldehyde solution (20% w/w in D2O) were obtained from Cambridge Isotope Laboratories, Inc. (Tewksbury, MA). DNAzol and Turbo DNase were obtained from Lift Technologies. Proteinase K was obtained from FivePrime (San Francisco, CA). Nanosep Centrifugal Devices (MWCO 3K) were purchased from Pall Lift Sciences. All of the solvents used were at least HPLC-grade. Synthesis of dG-Me-Cys and [15N5]-dG-[13CD2]-Me-Cys Standards. The synthesized standards were prepared by enzymatic digestion of dG-Me-GSH (glutathione) according to previous publications.10,26 Briefly, GSH was incubated with formaldehyde in a sodium phosphate buffer at 37 °C for 3 h. Then, 2′-deoxyguanosine (dG) was added for a cross-linking reaction at 37 °C for 14 h to produce dG-Me-GSH. The dG-Me-GSH was digested by carboxypeptidase Y and leucine aminopeptidase M in sodium phosphate buffer with the presence of MgCl2 and CaCl2 at room temperature for 15 h in order to release dG-Me-Cys. The same method was applied to synthesize [15N5]-dG-[13CD2]-Me-Cys using [15N5]-dG and [13CD2]formaldehyde as starting materials. The dG-Me-Cys and [15N5]-dG[13CD2]-Me-Cys were further purified and quantified as described previously.10,26 The dG-Me-Cys and [15N5]-dG-[13CD2]-Me-Cys were used as the analytical and internal standard to construct the calibration curve for the accurate quantification of the endogenous and exogenous DPCs (dG-Me-Cys). Cell Culture and Cell Treatment with [13CD2]-Formaldehyde. HEK293T cells were obtained from the American Type Culture Collection (Camden, NJ). The cells were maintained as exponentially growing monolayer cultures in Dulbecco’s modified Eagle’s medium (DMEM) supplemented with 10% fetal bovine serum (FBS) and 1% penicillin-streptomycin in a humidified incubator at 37 °C with 5% CO2. HEK293T cells were subcultured 1 day before the [13CD2]formaldehyde treatment. The cells were washed with phosphatebuffered saline (PBS) to remove the FBS just prior to treatment. The washed cells were treated with DMEM containing 0, 100, or 250 μM [13CD2]-formaldehyde for 2 h at 37 °C. After treatment, the cells were washed with PBS to remove the [13CD2]-formaldehyde before further isolation of the DNA−protein cross-links. DNA−Protein Cross-Link Isolation and Enzymatic Digestion. The DNA−protein cross-links (DPCs) were isolated from the harvested cells using DNAzol (Thermo Fisher Scientific). The cells were dissolved in 1 mL of DNAzol reagent by pipetting before adding proteinase K for overnight digestion at room temperature. Further DPCs isolation and enzymatic digestion were previously described with some minor modifications.10 Briefly, DPCs were precipitated by 100% ethanol at −20 °C for 2 h, and then they were centrifuged at 14 000 × g at 4 °C for 5 min. The pellets (DPCs) were washed with 75% ethanol before being reconstituted with 0.5 mL of predigestion buffer (40 mmol/L ammonium acetate (pH 6.0), CaCl2 (10 mmol/L), and pronase (1.4 U/mL)) for overnight digestion at room temperature. After digestion, the supernatant was collected by centrifuged at 12 000 × g at 4 °C for 5 min. Then, DPCs were precipitated at −20 °C for 2 h using 10% volume sodium acetate (3 M) and 2.5-fold volume of 100% ethanol. DPCs were pelleted by centrifugation at 12 000 × g at 4 °C for 5 min, and then they were washed with 75% ethanol. Pellets were reconstituted with 0.45 mL of digestion buffer (40 mmol/L ammonium acetate (pH 6.0), MgCl2 (10 mmol/L), CaCl2 (10 mmol/L), DNase I (44.4 U/mL), alkaline phosphatase (3.3 U/mL), phosphodiesterase I (0.0067 U/mL), prolidase (3.3 U/mL), carboxypeptidase Y (0.83 U/mL), and aminopeptidase M (0.083 U/mL)) for overnight digestion at room temperature. The enzymatic reaction was terminated with the addition 352

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology 409.12465→m/z 172.05756) were directly reported from Skyline after manually checking the peak integration.

shows the extracted ion chromatograms (XIC) of dG-Me-Cys (Figure 1A) and [15N5]-dG-[13CD2]-Me-Cys (Figure 1B). Because of the high mass accuracy obtained from the Orbitrap, the XIC was executed using the exact mass of the target product ion of m/z 164.05723 and m/z 172.05811 for dG-MeCys and [15N5]-dG-[13CD2]-Me-Cys, respectively, within 5 ppm mass accuracy. In addition, the precursor ions of target analytes were revealed by the high mass resolution and accuracy in the full scan spectra (Figure 1, left inserts). The product ion spectrum (MS/MS) after HCD fragmentation was further used to confirm the identities and structures of target analytes (Figure 1, right inserts). The fragmentation patterns under HCD fragmentation were proposed in Scheme 1. The peaks of m/z 164.05668 and m/z 172.05756 for dG-MeCys and [15N5]-dG-[13CD2]-Me-Cys, respectively, were the most intense product ions (“product ion 5”, Table 1) after fragmentation on two chemical bonds as indicated by a red dashed line. After fragmentation, all five of the proposed product ions were detected, and the measured mass accuracies for the precursor and product ions were less than 3 ppm compared to their exact mass (Table 1). These data demonstrate the confident identification and detectability of synthetic dG-Me-Cys by high-resolution, accurate-mass Orbitrap mass spectrometry. Linearity of Calibration Curve for Quantification. For quantification of dG-Me-Cys by Q Exactive HF, we implemented the targeted PRM mode.28 Briefly, this mode sequentially isolates target precursor ions for HCD fragmentation and collects the full product ion spectrum. Quantification was based on the post-acquisition processing to reconstruct the chromatographic peaks using a narrow tolerance ( m/z 164.05668), exogenous dG-[13CD2]-Me-Cys (m/z 404.14 > m/ z 167.07238), and internal standard [15N5]-dG-[13CD2]-MeCys (m/z 409.13 > m/z 172.05756). Figure 2 shows the wide dynamic range of the calibration curve for quantification of dG-Me-Cys. Plotting amount ratio versus the integrated peak area ratio over internal standard indicates the high quantification capability (at least 3 orders) using the PRM mode on Orbitrap MS. In this tested calibration curve, the lowest calibrator was 15.625 amol of dG-Me-Cys loaded onto the LC column. Intra- and interday precisions and

■

RESULTS Detection of dG-Me-Cys by Q Exactive HF. In our previous study, Lai et al. demonstrated the detection and quantification of dG-Me-Cys using low mass resolution triple quadrupole mass spectrometry with the SRM mode used to detect and quantify dG-Me-Cys.10 In the current study, for the first time, we have evaluated the detectability for dG-Me-Cys using a high-resolution, accurate-mass Orbitrap mass spectrometry. At first, the synthetic standards, dG-Me-Cys and the stable isotope labeled [15N5]-dG-[13CD2]-Me-Cys, were individually analyzed by LC-MS/MS Orbitrap mass spectrometry. Figure 1

Figure 1. XIC of synthetic standards for (A) dG-Me-Cys (m/z 401.12377→m/z 164.05668) and (B) [15N5]-dG-[13CD2]-Me-Cys (m/z 409.12465→m/z 172.05756). Inserts show the isotope envelope patterns and corresponding MS/MS spectra. R indicates the mass resolution.

Table 1. Mass Accuracy of dG-Me-Cys and [15N5]-dG-[13CD2]-Me-Cys Measured in Q Exactive HF Mass Spectrometry [15N5]-dG-[13CD2]-Me-Cys

dG-Me-Cys

a

ion typea

measured m/z

theoretical m/z

mass accuracy (ppm)

measured m/z

theoretical m/z

mass accuracy (ppm)

precursor ion product ion 1 product ion 2 product ion 3 product ion 4 product ion 5

401.12344 117.05489 122.02722 280.10388 285.07626 164.05653

401.12377 117.05462 122.02702 280.10403 285.07643 164.05668

−0.83 2.28 1.60 −0.54 −0.60 −0.94

409.12460 117.05486 122.02721 288.10489 293.07727 172.05757

409.12465 117.05462 122.02702 288.10491 293.07731 172.05756

−0.13 2.03 1.52 −0.07 −0.13 0.06

The chemical structures are shown in Scheme 1. 353

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology

Table 3. Formaldehyde-Induced Exogenous dG-Me-Cys in HEK293T Cell Lines Treated by 0, 100, and 250 μM [13CD2]-Formaldehyde dG-Me-Cys (cross-link/108 dG)

a

formaldehyde treated concentration (μM)

endogenous

exogenous

0 100 250

0.66 ± 0.06 (n = 4) 0.51 ± 0.11 (n = 4) 0.67 ± 0.08 (n = 4)

NDa 0.93 ± 0.04 12.41 ± 0.69

ND, not detected.

Clearly, the exogenous DPCs increased with increased formaldehyde concentration (Figure 3, right panel). In addition, the mass accuracy of those product ions was less than 1 ppm. Chromatographic Peak Integration Using High-Resolution, Accurate-Mass Orbitrap Mass Spectrometry. The Orbitrap mass analyzer provides the benefits of high mass accuracy ( 10.

Figure 2. The calibration curve by plotting the amount ratio versus area ratio. The x-axis is the added dG-Me-Cys (analytical standard, AS) concentrations in the fixed concentration of [15N5]-dG-[13CD2]Me-Cys (internal standard, IS) at 0.25 nM, and the final amount ratios are from 0.015625 to 16. The y-axis is the integrated peak area ratios between AS and IS. The inset shows a zoomed in region of the amount ratio from 0.015625 to 0.5.

accuracies for dG-Me-Cys quantification were also evaluated, and variations were less than 10% when measuring the known dG-Me-Cys over the IS amount ratio, providing accurate and precise quantification results (Table 2). Measurement of Endogenous and Exogenous dG-MeCys in HEK293T Cells Treated with Isotope-Labeled Formaldehyde. Next, we applied this quantification strategy to measure endogenous dG-Me-Cys (m/z 401.12 > m/z 164.05668) and exogenous dG-[13CD2]-Me-Cys (m/z 404.14 > m/z 167.07238) in HEK293T cells treated with [13CD2]formaldehyde at 0, 100, and 250 μM (Table 3). While no exogenous DPCs were detected in the control group, exogenous dG-[13CD2]-Me-Cys was 0.94 ± 0.04 and 12.41 ± 0.69 cross-links/108 dG after being treated with 100 and 250 μM, respectively. In this in vivo experiment, the cells were treated with isotope-labeled formaldehyde for only 2 h, resulting in a rapid exogenous DPC generation; however, the amount of endogenous DPCs was not affected under this treatment. This observation indicated that the exposureinduced stress did not alter the amount of the endogenous formaldehyde-induced DPCs. The chromatographic peaks for those endogenous and exogenous DPCs, as well as internal standards integrated by Skyline software, are shown in Figure 3.

■

DISCUSSION It is well-known that endogenous formaldehyde is an important metabolic intermediate from multiple pathways, such as onecarbon pool metabolism, amino acid metabolism, lipid peroxidation, and various demethylation processes.29,30 Because

Table 2. Intra- and Interday Accuracy and Precision Evaluation for dG-Me-Cys (Analytical Standard, AS) by a Spiked in Fixed Amount of [15N5]-dG-[13CD2]-Me-Cys (Internal Standard, IS) Determined by LC-MS/MS (n = 3) linear regression equation

R2

added AS/IS amount ratio

measured AS/IS amount ratio

RSD (%)

accuracy (%)

day 1

y = 1.0538x − 0.0027

0.9994

day 2

y = 1.0735x − 0.005

0.9999

day 3

y = 1.0599x − 0.0113

0.9984

0.03125 0.5 0.03125 0.5 0.03125 0.5 0.03125 0.5

0.02863 0.49927 0.03152 0.48266 0.03403 0.49524 0.03139 0.49239

6.54 1.09 8.87 0.84 4.17 1.33 8.61 1.76

91.6 99.9 100.9 96.5 108.9 99.0 100.4 98.5

intra/interday

interday

354

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology

Figure 3. Chromatographic peak integrations for the quantification of endogenous (m/z 401.12377 > m/z 164.05668, left panel) and exogenous (m/z 404.13948 > m/z 167.07238, right panel) dG-Me-Cys in HEK293T cells treated with 0 (A), 100 (B), and 250 μM [13CD2]formaldehyde (C) for 2 h. The endogenous and exogenous dG-MeCys measured were marked with red and orange lines, respectively. The internal standard, [15N5]-dG-[13CD2]-Me-Cys (m/z 409.12465 → m/z 172.05756), was marked with a blue line. Mass accuracy of the integrated peaks reported by Skyline was also provided in the unit of ppm. The ratio of the peak area was calculated by the integrated peak area of dG-Me-Cys over that of the internal standard.

of the high reactivity of formaldehyde and its common presence in all tissues and body fluids,9,10,26 formaldehyde is known as an important source of endogenous DNA damages, and the repair of formaldehyde-induced DPCs is important.9,15 Our group previously implemented stable isotope-labeled formaldehyde and utilized its corresponding synthetic internal standard for dG-Me-Cys to accurately and structurally measure, for the first time, both endogenous and exogenous DPCs in exposed animal tissues.10 In this study, we further advanced this methodology for DPC quantification using a state-of-the-art Orbitrap MS with high mass resolution and accuracy. The quantification method presented in this study is based on the targeted acquisition mode, PRM, executed on an Orbitrap instrument. PRM has been widely used for large-scale targeted peptide quantification.25,31 Unlike the traditional SRM method performed on triple quadrupole mass spectrometry, the PRM mode collects the full product ion spectrum using high mass resolution and accuracy. The PRM quantification is achieved by the post-acquisition extraction of the target product ions with tight mass tolerance for reconstructing the chromatographic peak. Moreover, owing to the product ion spectrum obtained using high mass resolution and accuracy, the selectivity of the quantification was significantly improved using PRM over SRM.28 As expected, PRM presents the background signal-free quantification by extracting the exact mass of the target product ion(s) with a tight mass tolerance, drastically reducing the background noise signal (Figure 4A). Conversely, the low mass resolution of SRM acquisition results

Figure 4. XIC of exogenous dG-Me-Cys in HEK293T cells treated with 250 μM [13CD2]-formaldehyde for 2 h in mass tolerance of 5 ppm (A) and 2095 ppm (B) using Orbitrap MS. (C) LC-SRM spectrum for exogenous dG-Me-Cys in rat nasal tissue exposed to 15 ppm [13CD2]-formaldehyde for 6 h per day for 4 days using triple quadruple MS, and corresponding animal exposure experiments, sample preparation, and LC-MS/MS methods were described previously by Lai et al.10 It is of note that some m/z regions were zoomed in for demonstrating the background signals.

in background signals from the co-eluting compounds that may interfer with the target ion(s) even after significantly reducing the sample complexity by HPLC purification (Figure 4C). The old version of QqQ MS was previously used to detect dG-Me-Cys by Lai et al.,10 and the limit of quantification was 37.5 amol on the column based on the S/N ratio >10. The lowest calibrator tested in the current study was 15.625 amol loaded onto the column, with a good signal response and linearity covering 3 orders of magnitude (Figure 2). In addition, the precision and accuracy of this new methodology showed a significant improvement for dG-Me-Cys quantification (Table 2). Moreover, the scan rate of the instrument is an important 355

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology

*Phone: 919 966 6139; E-mail: [email protected].

factor for the targeted quantification. The scan rate on the Orbitrap MS is based on the settings between the maximum injection time and the Orbitrap resolution, because the instrument enables the accumulation and measurement of ion population in parallel.32 Collecting the basic criterion of at least 8−10 data points over a chromatographic peak ensures precise quantification.31 Only a limited number of precursors were monitored in this PRM method, therefore the maximum injection time for each precursor can be increased to 250 ms in order to pursue high detection sensitivity and still have enough data points for quantification (Figure 3). To our knowledge, the amount of formaldehyde-DPCs (dGMe-Cys) from any cell lines has not been reported to date. In this study, the dG-Me-Cys level in a human epithelial cell line was analyzed by applying our new Orbitrap-based quantification method (Table 3). The amount of exogenous DPCs in the cells significantly increased from 0.93 ± 0.04 to 12.41 ± 0.69 cross-link/10 8 dG when being treated with a higher concentration of labeled formaldehyde (100 and 250 μM, respectively). In contrast to the significant changes observed in exogenous DPCs levels in HEK293T cells after treatment with labeled formaldehyde, no significant difference was detected in the amount of endogenous DPCs. Clearly, the exogenous formaldehyde exposure stress did not affect the endogenous DPCs formation based on the exposure data from the cell line and animal models. This observation is in agreement with our previous study using animal models exposed to stable isotopelabeled formaldehyde, showing the steady-state concentrations of endogenous DPCs.10 Furthermore, the endogenous DPCs level in HEK293T cells (0.66 ± 0.06 cross-link/108 dG) was at least 2 and up to 23-fold lower than those in tissues analyzed in our previous study, including peripheral blood mononuclear cells (1.34 ± 0.25), bone marrow (2.30 ± 0.30), nasal epithelium (3.59 ± 1.01), and liver (15.46 ± 1.98 cross-link/ 108 dG).10 The endogenous formaldehyde concentration in the blood is ∼100 μM, and much higher concentrations were observed in the rat liver and nasal mucosa.33 The significantly lower formaldehyde concentrations in cell lines have been reported by Kato et al.34 The concentration differences of endogenous formaldehyde may account for the lower endogenous DPCs level in the cell line compared to tissues, especially in the liver. To detect this low level of DPCs in cell lines requires a more sensitive detection platform demonstrated in this study. Moreover, the methodology demonstrated herein can be applied to detect other linkages of formaldehydeinduced DPCs20 and different forms of DNA damages, such as interstrand DNA−DNA cross-links.7,8 In summary, we demonstrated a sensitive method for DPC quantification by an Orbitrap-based MS platform, which provides benefits of high mass resolution and accuracy for unambiguous analysis. The high mass resolution and accuracy of the Orbitrap mass analyzer generated an interference-free XIC spectrum for accurate quantification. This quantification method was further applied to detect endogenous and exogenous DPC levels in HEK293T cells treated with labeled formaldehyde. In addition, the capability of this method to detect low levels of endogenous DPCs will significantly facilitate the understanding of the molecular mechanisms for DPCs repair.

■

ORCID

Chih-Wei Liu: 0000-0002-0823-0252 Xu Tian: 0000-0001-8174-5669 Hadley J. Hartwell: 0000-0003-4453-3266 Kun Lu: 0000-0002-8125-2394 Funding

This work was partially supported by the NIH/NIEHS Grant (R03ES024147) and a grant from the American Chemistry Council. Notes

The authors declare no competing financial interest.

■

ABBREVIATIONS

■

REFERENCES

Cys, cysteine; dG, 2′-deoxyguanosine; DPCs, DNA−protein cross-links; DMEM, Dulbecco’s modified Eagle’s medium; FBS, fetal bovine serum; HCD, higher-energy collisional dissociation; MS/MS, tandem mass spectrometry; PBS, phosphatebuffered saline; PRM, parallel reaction monitoring; QqQ, triple quadrupole; SRM, selected reaction monitoring; XIC, extracted ion chromatogram

(1) Vaz, B., Popovic, M., Newman, J. A., Fielden, J., Aitkenhead, H., Halder, S., Singh, A. N., Vendrell, I., Fischer, R., Torrecilla, I., Drobnitzky, N., Freire, R., Amor, D. J., Lockhart, P. J., Kessler, B. M., McKenna, G. W., Gileadi, O., and Ramadan, K. (2016) Metalloprotease SPRTN/DVC1 Orchestrates Replication-Coupled DNAProtein Crosslink Repair. Mol. Cell 64, 704−719. (2) Sczepanski, J. T., Wong, R. S., McKnight, J. N., Bowman, G. D., and Greenberg, M. M. (2010) Rapid DNA-protein cross-linking and strand scission by an abasic site in a nucleosome core particle. Proc. Natl. Acad. Sci. U. S. A. 107, 22475−22480. (3) Garaycoechea, J. I., Crossan, G. P., Langevin, F., Daly, M., Arends, M. J., and Patel, K. J. (2012) Genotoxic consequences of endogenous aldehydes on mouse haematopoietic stem cell function. Nature 489, 571−575. (4) Jackson, S. P., and Bartek, J. (2009) The DNA-damage response in human biology and disease. Nature 461, 1071−1078. (5) Tubbs, A., and Nussenzweig, A. (2017) Endogenous DNA Damage as a Source of Genomic Instability in Cancer. Cell 168, 644− 656. (6) Nakamura, J., and Swenberg, J. A. (1999) Endogenous apurinic/ apyrimidinic sites in genomic DNA of mammalian tissues. Cancer research 59, 2522−2526. (7) Johnson, K. M., Price, N. E., Wang, J., Fekry, M. I., Dutta, S., Seiner, D. R., Wang, Y., and Gates, K. S. (2013) On the formation and properties of interstrand DNA-DNA cross-links forged by reaction of an abasic site with the opposing guanine residue of 5′-CAp sequences in duplex DNA. J. Am. Chem. Soc. 135, 1015−1025. (8) Price, N. E., Johnson, K. M., Wang, J., Fekry, M. I., Wang, Y., and Gates, K. S. (2014) Interstrand DNA-DNA cross-link formation between adenine residues and abasic sites in duplex DNA. J. Am. Chem. Soc. 136, 3483−3490. (9) Pontel, L. B., Rosado, I. V., Burgos-Barragan, G., Garaycoechea, J. I., Yu, R., Arends, M. J., Chandrasekaran, G., Broecker, V., Wei, W., Liu, L., Swenberg, J. A., Crossan, G. P., and Patel, K. J. (2015) Endogenous Formaldehyde Is a Hematopoietic Stem Cell Genotoxin and Metabolic Carcinogen. Mol. Cell 60, 177−188. (10) Lai, Y., Yu, R., Hartwell, H. J., Moeller, B. C., Bodnar, W. M., and Swenberg, J. A. (2016) Measurement of Endogenous versus Exogenous Formaldehyde-Induced DNA-Protein Crosslinks in Animal Tissues by Stable Isotope Labeling and Ultrasensitive Mass Spectrometry. Cancer Res. 76, 2652−2661.

AUTHOR INFORMATION

Corresponding Authors

*Phone: 919 966 7337; E-mail: [email protected]. 356

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357

Article

Chemical Research in Toxicology (11) Barker, S., Weinfeld, M., and Murray, D. (2005) DNA-protein crosslinks: their induction, repair, and biological consequences. Mutat. Res., Rev. Mutat. Res. 589, 111−135. (12) Nakano, T., Xu, X., Salem, A. M. H., Shoulkamy, M. I., and Ide, H. (2017) Radiation-induced DNA-protein cross-links: Mechanisms and biological significance. Free Radical Biol. Med. 107, 136−145. (13) Groehler, A. t., Villalta, P. W., Campbell, C., and Tretyakova, N. (2016) Covalent DNA-Protein Cross-Linking by Phosphoramide Mustard and Nornitrogen Mustard in Human Cells. Chem. Res. Toxicol. 29, 190−202. (14) Stingele, J., Bellelli, R., and Boulton, S. J. (2017) Mechanisms of DNA-protein crosslink repair. Nat. Rev. Mol. Cell Biol. 18, 563−573. (15) Stingele, J., Schwarz, M. S., Bloemeke, N., Wolf, P. G., and Jentsch, S. (2014) A DNA-dependent protease involved in DNAprotein crosslink repair. Cell 158, 327−338. (16) Gherezghiher, T. B., Ming, X., Villalta, P. W., Campbell, C., and Tretyakova, N. Y. (2013) 1,2,3,4-Diepoxybutane-induced DNAprotein cross-linking in human fibrosarcoma (HT1080) cells. J. Proteome Res. 12, 2151−2164. (17) de Graaf, B., Clore, A., and McCullough, A. K. (2009) Cellular pathways for DNA repair and damage tolerance of formaldehydeinduced DNA-protein crosslinks. DNA Repair 8, 1207−1214. (18) Ye, X., Ji, Z., Wei, C., McHale, C. M., Ding, S., Thomas, R., Yang, X., and Zhang, L. (2013) Inhaled formaldehyde induces DNAprotein crosslinks and oxidative stress in bone marrow and other distant organs of exposed mice. Environmental and molecular mutagenesis 54, 705−718. (19) Swenberg, J. A., Lu, K., Moeller, B. C., Gao, L., Upton, P. B., Nakamura, J., and Starr, T. B. (2011) Endogenous versus exogenous DNA adducts: their role in carcinogenesis, epidemiology, and risk assessment. Toxicol. Sci. 120, S130−S145. (20) Lu, K., Ye, W., Zhou, L., Collins, L. B., Chen, X., Gold, A., Ball, L. M., and Swenberg, J. A. (2010) Structural characterization of formaldehyde-induced cross-links between amino acids and deoxynucleosides and their oligomers. J. Am. Chem. Soc. 132, 3388−3399. (21) Picotti, P., and Aebersold, R. (2012) Selected reaction monitoring-based proteomics: workflows, potential, pitfalls and future directions. Nat. Methods 9, 555−566. (22) Shi, T., Su, D., Liu, T., Tang, K., Camp, D. G., 2nd, Qian, W. J., and Smith, R. D. (2012) Advancing the sensitivity of selected reaction monitoring-based targeted quantitative proteomics. Proteomics 12, 1074−1092. (23) Shi, T., Song, E., Nie, S., Rodland, K. D., Liu, T., Qian, W. J., and Smith, R. D. (2016) Advances in targeted proteomics and applications to biomedical research. Proteomics 16, 2160−2182. (24) Lesur, A., and Domon, B. (2015) Advances in high-resolution accurate mass spectrometry application to targeted proteomics. Proteomics 15, 880−890. (25) Bourmaud, A., Gallien, S., and Domon, B. (2016) Parallel reaction monitoring using quadrupole-Orbitrap mass spectrometer: Principle and applications. Proteomics 16, 2146−2159. (26) Yu, R., Lai, Y., Hartwell, H. J., Moeller, B. C., Doyle-Eisele, M., Kracko, D., Bodnar, W. M., Starr, T. B., and Swenberg, J. A. (2015) Formation, Accumulation, and Hydrolysis of Endogenous and Exogenous Formaldehyde-Induced DNA Damage. Toxicol. Sci. 146, 170−182. (27) MacLean, B., Tomazela, D. M., Shulman, N., Chambers, M., Finney, G. L., Frewen, B., Kern, R., Tabb, D. L., Liebler, D. C., and MacCoss, M. J. (2010) Skyline: an open source document editor for creating and analyzing targeted proteomics experiments. Bioinformatics 26, 966−968. (28) Peterson, A. C., Russell, J. D., Bailey, D. J., Westphall, M. S., and Coon, J. J. (2012) Parallel reaction monitoring for high resolution and high mass accuracy quantitative, targeted proteomics. Mol. Cell. Proteomics 11, 1475−1488. (29) Dhareshwar, S. S., and Stella, V. J. (2008) Your prodrug releases formaldehyde: should you be concerned? No! J. Pharm. Sci. 97, 4184− 4193.

(30) Shi, Y., Lan, F., Matson, C., Mulligan, P., Whetstine, J. R., Cole, P. A., Casero, R. A., and Shi, Y. (2004) Histone demethylation mediated by the nuclear amine oxidase homolog LSD1. Cell 119, 941− 953. (31) Gallien, S., Kim, S. Y., and Domon, B. (2015) Large-Scale Targeted Proteomics Using Internal Standard Triggered-Parallel Reaction Monitoring (IS-PRM). Mol. Cell. Proteomics 14, 1630−1644. (32) Scheltema, R. A., Hauschild, J. P., Lange, O., Hornburg, D., Denisov, E., Damoc, E., Kuehn, A., Makarov, A., and Mann, M. (2014) The Q Exactive HF, a Benchtop mass spectrometer with a pre-filter, high-performance quadrupole and an ultra-high-field Orbitrap analyzer. Mol. Cell. Proteomics 13, 3698−3708. (33) Heck, H., and Casanova, M. (2004) The implausibility of leukemia induction by formaldehyde: a critical review of the biological evidence on distant-site toxicity. Regul. Toxicol. Pharmacol. 40, 92−106. (34) Kato, S., Burke, P. J., Koch, T. H., and Bierbaum, V. M. (2001) Formaldehyde in human cancer cells: detection by preconcentrationchemical ionization mass spectrometry. Anal. Chem. 73, 2992−2997.

357

DOI: 10.1021/acs.chemrestox.8b00040 Chem. Res. Toxicol. 2018, 31, 350−357