Characterization and quantitation of busulfan DNA adducts in the blood of patients receiving busulfan therapy

Valeria Guidolin; Yupeng Li; Foster C Jacobs; Margaret L MacMillan; Peter W Villalta; Stephen S Hecht; Silvia Balbo

doi:10.1016/j.omto.2023.01.005

. 2023 Jan 20;28:197–210. doi: 10.1016/j.omto.2023.01.005

Characterization and quantitation of busulfan DNA adducts in the blood of patients receiving busulfan therapy

Valeria Guidolin ^1,^3,⁵, Yupeng Li ^1,^2,⁵, Foster C Jacobs ^1,³, Margaret L MacMillan ^1,⁴, Peter W Villalta ^1,², Stephen S Hecht ¹, Silvia Balbo ^1,^3,^∗

PMCID: PMC9938526 PMID: 36820303

Abstract

DNA alkylating drugs have been used as cancer chemotherapy with variable outcomes. The establishment of predictive biomarkers to identify patients who will effectively respond to treatment would allow for the development of personalized therapies. As the degree of interaction of alkylating drug with DNA plays a key role in their mechanism of action, our hypothesis is that the measurement of the DNA adducts formed by alkylating drugs could be used to inform patient stratification. Beginning with busulfan, we took advantage of our DNA adductomic approach to characterize DNA adducts formed by reacting busulfan with calf-thymus DNA. Samples collected from six patients undergoing busulfan-based chemotherapy prior to allogeneic hematopoietic cell transplantation were analyzed for the presence of busulfan-derived DNA adducts. Among the 15 adducts detected in vitro, 12 were observed in the patient blood confirming the presence of a large profile of DNA adducts in vivo. Two of the detected adducts were structurally confirmed by comparison with synthetic standards and quantified in patients. These data confirm our ability to comprehensively characterize busulfan-derived DNA damage and set the stage for the development of methods to support personalized chemotherapy.

Keywords: precision medicine, DNA adduct, biomarker, busulfan, mass spectrometry, alkylating agents

Graphical abstract

Introduction

Alkylating drug-based chemotherapy remains the first-line treatment choice for many types of cancer dating back to when nitrogen mustard was first introduced in 1940s.¹^,² To achieve their therapeutic effect, alkylating drugs must enter cancer cells and dysregulate cell cycles by modifying the DNA structure.³^,⁴ DNA damage is the desired effect on cancer cells as it leads to cell replication blockage and subsequent cell death, but it is also the cause of side effects when localized in healthy cells. Large variations in efficacy and toxicity of alkylating drugs have been widely observed in the clinic. Since an association between alkylating drug-based chemotherapy outcome and cancer cell DNA modifications has been established, it is reasonable to investigate the use of DNA damage products formed by alkylating drugs for patient stratification.⁵ Due to the development of new analytical technologies, it is now possible to characterize DNA damage and quantify the levels of specific types of DNA damage products. For example, in the blood DNA of patients receiving platinum drugs, the levels of DNA platinum adducts were shown to be closely associated with therapy outcome.⁵^,⁶^,⁷^,⁸ To better understand the profile of DNA damage caused by alkylating drugs, we have developed a DNA adductomic method that allows us to comprehensively study the DNA adduct formation by using high-resolution mass spectrometry (HRMS).⁹ On the basis of this method, we report here the first study of DNA adduct formation caused by the widely used front-line alkylating drug busulfan (BU) (1,4-butanediol dimethanesulfonate, 1; Figure 1).

Study rationale and sample workflow

The study rationale (A) and the work flow (B) of adductomic analysis of DNA samples. In the *in vitro* conditions with less complexity, DNA adducts are more readily identified due to their higher concentrations. This allows for a DNA adduct library generation that can be informative for *in vivo* evaluation of patient DNA samples.

BU is an alkylating agent historically used for the treatment of a variety of cancers.¹⁰ BU is nowadays frequently used in combination with other chemotherapeutic agents for pre-transplantation conditioning of patients undergoing allogeneic hematopoietic cell transplantation (HCT).¹¹^,¹² Similar to other alkylating drugs, BU-based therapy side effects include bone marrow suppression, myelosuppression, and hyperpigmentation.¹⁰ Although a clear benefit has been observed using a regimen of cyclophosphamide (CY) with high-dose oral BU in pre-transplantation conditioning,¹³^,¹⁴ high systemic exposure to BU has been associated with neurotoxicity and other conditions.¹⁵^,¹⁶^,¹⁷ Low systemic exposure to BU, however, may result in leukemic relapse and increased risks of graft rejection.¹⁸^,¹⁹ Altogether, these observations indicate a narrow therapeutic window for BU-based therapy in the clinic.²⁰

To inform the optimal dose of BU for cancer patients, efforts have been made to monitor its concentration in plasma.²¹^,²²^,²³ However, controversies regarding the advantage of monitoring systematic exposure of BU have been reported due to the presence of confounders.²⁰ Indeed, many retrospective studies evaluating the association of BU exposure to clinical outcomes were confounded by heterogeneity in the conditioning regimens and the baseline patient characteristics.²⁰

In this context, BU-derived DNA adducts can be considered as clinical outcome prediction biomarkers since they are the direct products resulting from BU therapy. The evaluation of their profile could provide more accurate information on the interaction of the drug with its target to be used to support predictive information for adjusting BU dose. This may be even more important when BU is used in combination with other alkylating drugs due to the observed BU-related drug-drug interactions.²⁴ However, to the best of our knowledge, these DNA adducts were only hypothesized to take part in BU-related cytotoxicity and had not been fully characterized before.²⁵ Using our above-mentioned adductomic approach,²⁶^,²⁷^,²⁸ we report here the first comprehensive study of BU-DNA adducts formed in vitro in BU-treated calf-thymus and bacterial DNA, along with the synthesis and characterization of two of the major identified BU-DNA adducts. Detection of the adducts in vivo in the blood DNA of cancer and Fanconi anemia patients receiving BU therapy was demonstrated, along with the establishment of a quantitative method for use in large sample population studies.

Results

Analysis of putative BU DNA adducts formed in vitro

A top-down DNA adductomic approach was utilized for the comprehensive characterization of BU-derived DNA adducts in calf-thymus DNA(CT-DNA), treated with the drug in vitro (Figure 1). The adductomic data-dependent constant neutral loss-MS³ (DDA-CNL/MS³) method can simultaneously screen for multiple DNA adducts by taking advantage of the common structural features of DNA adducts (the 2′-deoxyribose-nucleobase structure).⁹^,²⁶^,²⁷ In this study, the neutral loss of the four nucleobase moieties (G, 151.0494 amu; A, 135.0545 amu; T, 126.0429 amu; and C, 111.0433 amu) and the 2′-deoxyribose moiety (116.0474 amu) were monitored and coupled with subsequent MS³ fragmentation triggered upon neutral loss observation to gain additional structural characterization.

Using the DDA-CNL/MS³ method, DNA hydrolysates of BU-treated and untreated CT-DNA were analyzed and compared. Data analysis was performed as previously reported, and ions corresponding to DNA adducts present only in the exposed sample were considered.²⁶ Briefly, scrutiny of the MS² and MS³ spectra for each MS³-triggering ion confirmed that the fragments observed resulted from a DNA adduct and were used to exclude any MS³-triggering ions resulting from artifacts or false-positives. Specifically, for each ion, the MS³ spectrum was scrutinized to (1) confirm the presence of one of the nucleobases and/or its fragments as product ions, and (2) evaluate that the accurate mass corresponding to the modification accounts for a realistic chemical formula. Furthermore, the precursor extracted ion chromatogram (EIC) was evaluated to confirm its peak-like shape and a minimum of four sticks across the peak. Finally, the retention times of the full scan, MS² and MS³ spectra were evaluated to confirm that they coincided. Full-scan EICs for all candidate DNA adduct ions were generated for the exposed and unexposed samples, and only ions that were uniquely present in the exposed sample were annotated. Further confirmation was gained by evaluation of the presence of the adducts in both the ¹⁴N and ¹⁵N form when combining the analysis of ¹⁴N- and ¹⁵N-DNA exposed with BU.

A total of 20 compounds matching all the criteria were identified as putative BU-DNA adducts and listed in Table 1. The EIC for each of them is reported in Figure S2. As illustrated in Figure 2, a putative adduct with m/z 357.1530 shows a typical chromatogram with mass spectrometry (MS) traces of a clear full-scan peak aligning well with its MS² and MS³ signals. The neutral loss of guanine observed in the MS² fragmentation, together with the appearance of the guanine ion [Gua + H]⁺ in the MS³ fragmentation suggests that this putative DNA adduct is likely to be a cross-link product of the drug with two guanine bases. We thus tentatively assign the structure of this adduct as N7G-Bu-N7G, 6 (Figure S1).

Table 1.

DNA adducts putatively identified in BU-treated calf-thymus DNA and the blood DNA of patients receiving BU therapy

[M + H]⁺	tR (min)	NL	E. coli DNA (number of¹⁵N atoms)	Patient blood DNA^a
224.1135	8.6	N	5	X
257.0880	8.6	dR	5	X
262.0860	18.0	C	–	X
264.6485	8.3	G	5	X
290.1724	8.2	G	5	X
298.1430	28.9	A	5	X
300.1401	29.0	A	5	X
302.0920	16.0	G	5	–
306.1560	6.4	dR	5	X
316.1349	34.5	G	10	–
357.1531	9.6	G	10	X
358.1276	17.7	dR	5	X
378.1330	17.9	dR	–	X
418.1390	15.0	dR	5	X
432.1823	28.3	dR	7	–
441.2218	8.2	G	5	X
449.1726	15.6	dR	10	X
465.2080	49.7	T	–	–
469.1782	15.6	dR	–	–
482.2170	49.9	T	–	–

Open in a new tab

X indicates that adduct was detected in at least one patient.

MS traces of EIC of *m/z* 357.1530 and its MS² and MS³ fragmentation pattern in BU-treated CT-DNA

Confirmation of the putative BU DNA adducts using ¹⁵N-labeled DNA

¹⁵N-labeled Escherichia coli DNA is a useful tool to further confirm the putative BU-DNA adducts listed in Table 1.²⁹ A 1:1 mixture of unlabeled and ¹⁵N-labeled E. coli DNA exposed to BU was hydrolyzed and purified following the same approach described for the analysis of CT-DNA. The hydrolysate was analyzed by the same adductomic DDA-CNL/MS³ method, with the exception that ¹⁵N-labeled deoxyribonucleosides were included as monitored neutral losses. The acquired MS data were processed by the same approach described above. Only the peaks present in both ¹⁴N- and ¹⁵N-MS traces were considered as putative BU-DNA adducts, a representative example of which is shown in Figure 3. In this analysis, the presence of the two co-eluting peaks with m/z 357.1534 (unlabeled) and 367.1231 (¹⁵N-labeled) agreed with the putative BU cross-link adduct 6 observed in Table 1 and Figure 2. The mass difference of 9.9703 amu (indicating 10 ¹⁵N-labeled nitrogens) provides further evidence of forming the cross-link 6, since each guanine contains five nitrogens.

¹⁵N-DNA data for supporting DNA adducts characterization

(A) EIC of one detected adduct with *m/z* 357.1530, its ¹⁵N-version *m/z* 367.1234, and the corresponding MS² scan events. (B) MS² spectra of the ion *m/z* 357.1530 and the hypothesized structure. (C) MS² spectra of the ion *m/z* 367.1233 and the hypothesized structure (ion *m/z* 202.1071 is not related to the molecule of interest).

We then manually excluded any peaks from the list of Table 1 (the column of unlabeled CT-DNA) that were not confirmed by the ¹⁴N/¹⁵N paired pattern in the BU-treated E. coli DNA. The number of putative BU-DNA adducts thus decreased to 15, as shown in Table 1. Insights on DNA adduct types were given by the mass difference between the ¹⁴N/¹⁵N paired peaks. As mentioned above, a mass difference of 4.9852 amu (corresponding to the adduct with five nitrogen atoms), indicates that the DNA adduct is a modification on a single Gua or Ade nucleobase (mono adduct). while a mass difference of 6.9890 or 9.9703 amu corresponds to a cross-link DNA adduct. Therefore, a total of four cross-links and 11 mono adducts are identified to be formed from BU treatment of DNA in vitro.

Targeted analysis of putative BU DNA adducts in the blood of patients

The presence of the putative BU-derived DNA adducts was investigated in the blood DNA of patients undergoing BU treatment. Blood samples were collected from six patients receiving BU therapy (Table 2). Samples were taken prior to the first administration (baseline sample) and within 24 h after the completion of the BU therapy (exposed sample). The paired blood DNA samples were analyzed using a targeted MS method for the 15 putative BU-DNA adducts (Table 1) observed in vitro. Twelve of the adducts were detected in at least one of the blood samples. Interestingly, two putative DNA adducts with m/z 357.1531 (tentatively assigned to be N7G-Bu-N7G, 6) and 441.2218 (unknown) were present in all the six blood DNA samples of patients receiving BU therapy (Table S1). Some adducts, such as that with m/z 224.1135 (tentatively assigned to be N7G-Bu-OH, 7) and 449.1726 (unknown), were also detected in five out of six patients. These results identified the BU adduct profile to be used to define the interaction of the drug with its molecular target in patients. These efforts led to the development of a quantitative/profiling method pairing the quantitation of the most common BU-derived DNA modifications, adducts 6 and 7, with the profiling of all 15 detected putative DNA adducts.

Table 2.

Diagnosis, sex, and levels of DNA adducts 6 and 7 in patients receiving BU-based chemotherapy

Patient	Diagnosis	Sex	N7G-Bu-N7G (6) Adducts/10⁶ nucleosides	N7G-Bu-OH (7) Adducts/10⁶ nucleosides
1	Acute myeloid leukemia	F	0.62	12.8
2	Acute myeloid leukemia	F	1.54	26.7
3	Fanconi anemia	F	0.38	ND
4	Multiple sulfatase deficiency	F	2.02	13.7
5	Acute myeloid leukemia	F	0.62	28.2
6	Fanconi anemia	M	1.60	26.3

Open in a new tab

DNA modifications derived from the alkylating drug BU were profiled with our novel adductomics approach in vitro. The detected adducts were investigated in patients undergoing BU-based chemotherapy, two of which were chemically characterized and quantified. This study sets the stage for the evaluation of BU adducts for therapy personalization.

Chemical synthesis of BU-derived DNA adducts

The synthesis of adducts 6 and 7 is illustrated in Figure S1. The same scheme was followed for the synthesis of their corresponding isotopically labeled analogs. The synthesis started with a common step in which the N²-NH₂ of [¹⁵N₅]2′-deoxyguanosine (dGuo) was first protected with the dimethylformamidine group to improve the regioselectivity of the alkylation reaction by 1,4-dibromobutane with 2. This alkylation reaction requires nearly 2 weeks to complete (Figure S3) and room temperature stirring to achieve a better outcome. Then the intermediate 3 was reacted with excess 2 to yield the cross-link precursor 4, which was easily converted to the final compound 6 via the NH₄OH-catalyzed deprotection reaction. During the cross-linking step, the bromine atom of 3 also underwent hydrolysis and formed the mono adduct precursor 5. This precursor was readily converted to the mono adduct 7 by treatment with NH₄OH in MeOH. The same route was followed starting from [¹⁵N₅]dGuo to obtain the corresponding isotopically labeled analogs of 6 and 7.

Characterization of BU-derived DNA adducts

One critical aspect of the characterization of BU-derived DNA adducts 6 and 7 is to confirm the regioselective attack of the Gua-N7 position in the alkylation steps of Figure S1. During the first alkylation step, compound 3 was formed as the major product (Figure S3). As shown by Figure S4, the heteronuclear multiple bond correlations (HMBCs) between the methylene group (H1′) and the 5- and 8-carbons of the guanine base clearly demonstrate that the alkylation reaction is Gua-N7 selective under the mild conditions used. No such correlation is observed between the same methylene group and the 4-carbon of the guanine base. The full NMR characterization of compound 3 is also presented in Figure S4. The chemical structure of 7 is thus the Gua-N7 alkylation product since the hydrolysis step and the deprotection step will not change the alkylation position.

For the structural assignment of the cross-link adduct 6, additional HMBC evidence was needed for the second alkylation reaction by the intermediate 3 with 2. As shown in Figure S5, correlations similar to those described for compound 3, between the methylene group (H1′) and the Gua C5 and C8, were clearly observed. No signal was observed between the same methylene group and Gua C4 further supporting the structure of intermediate 4. The full NMR characterization of this intermediate can also be found in the Figure S5. Similarly, the regioselectivity of the final cross-link 6 remains the same after the deprotection reaction from compound 4.

The MS study of the synthesized standards provided convincing evidence for their structural assignments. MS fragmentation patterns of 6 and 7 both agreed with likely pathways. As shown in Figure 4, the major product ion of the cross-link adduct 6 is m/z 206.1033, which is the fragment ion resulting from the loss of a guanine; the major product ion of the mono adduct 7 is m/z 224.1138, which is the fragment ion resulting from the loss of the alkyl modification. The isotope-labeled chemical standards [¹⁵N₅]6 and [¹⁵N₅]7 eluted at the same high-pressure liquid chromatography (HPLC) retention time compared with the unlabeled adducts 6 and 7 in the BU-exposed CT-DNA (Figure 5). Their MS² fragmentation patterns were also identical to the unlabeled standards, with the most abundant product ions being m/z 211.0876 and 157.0412, respectively.

Mass spectra of synthetic standards

HRMS analysis of synthetic standards of cross-link 6 (A) and mono adduct 7 (B).

Chromatograms resulting from co-injection of standards with the CT-DNA sample exposed to BU for analyte confirmation

(A) EIC of *m/z* 224.1135 and 229.0986. (B and C) Relative MS² spectra. (D) EIC of *m/z* 357.1531 and 362.1382. (E and F) Relative MS² spectra.

Quantification of N7G-Bu-N7G and N7G-Bu-OH in blood DNA of patients undergoing BU-based chemotherapy

The targeted HPLC-NSI-HRMS/MS method was used for the quantitation of the cross-link adduct 6 and the mono adduct 7 in the blood DNA of BU-exposed patients (Figure 6). The limit of detection (LOD) and limit of quantitation (LOQ) for adduct 6 were 0.015 and 0.05 fmol on-column, respectively; the LOD and LOQ for adduct 7 were 0.025 and 0.25 fmol on-column, respectively. Standard curves for both compounds were established by analyzing solutions containing a fixed amount of isotope-labeled adducts (6, 7.5 fmol; 7, 5 fmol) mixed with increasing amounts of the unlabeled standards, followed by regression analysis of the calculated and experimentally determined relative response ratios (Figure 6). Solvent blanks were periodically injected to detect potential analyte carry-over. The concentration ranges for the calibration curves and validation experiments were chosen to cover the range of the levels of adducts found in human DNA samples. The calibration curves showed good linearity within the concentration ranges (R² = 0.99 for 6 and 0.98 for 7). The assay accuracy was determined as a percentage of the added amount of adducts to 20 μg of CT-DNA. The adducts were detected with excellent accuracies of 94% and 101.48% (n = 6) for 6 and 7, respectively. Good linearity was also observed across the tested concentration ranges in the accuracy study. The precision of this assay was also satisfactory, with 6.9% for 6 and 7.6% for 7 by calculating the average of the coefficient of variation (CV) at each concentration of the curve. The total recoveries averaged 81% and 64% for 6 and 7 respectively.

Example of DNA adduct detection in patients and calibration curves

(A) Representative EIC of respectively adduct 6, its internal standard and 7, and relative internal standard in patient DNA. (B and C) Measured and spiked amount of adducts 6 and 7 for method validation. Error bars for measures repeated in triplicates (n=3).

As shown by Table 2, both the cross-link 6 and the mono adduct 7 were readily detected in most patients. The mono adduct 7 occurred in much higher abundance than the cross-link adduct 6. The concentration of 7 was 6.8–20.6 times higher than that of 6 in each patient except for patient 4. A high interindividual variation was observed in the levels of adducts 6 and 7. This may be due to the difference of total BU dose given to the patients, to interindividual difference in forming and repairing DNA adducts, or to a combination of both.

Discussion

In this study, we characterized for the first time BU-derived DNA adducts in the blood of HCT patients receiving this therapy. Twelve out of 15 putative DNA adducts observed in vitro were detected in patients. Two of them were characterized to be the cross-link DNA adduct 6 and the mono DNA adduct 7. A quantitative method for the analysis of these two adducts was developed to be combined with the screening analysis for all other BU-derived DNA modifications identified here. This study provides a promising approach to develop predictive biomarkers for future use to monitor the effects of BU and support personalized therapies.

The mechanism of action for DNA alkylating drugs such as BU is their ability to damage DNA. However, this interaction can also result in therapy toxicities, especially when high-dosage treatment is required.⁵ On the other hand, low doses of alkylating drugs result in insufficient cancer cell DNA damage, which has been associated with drug resistance and therapy failure.⁵ A fine balance needs to be found to administer these drugs at appropriate concentrations. Furthermore, patient-specific aspects, such as DNA repair mechanisms or drug metabolisms, are variables that may dramatically influence the generation of DNA damage and therefore the efficacy of the treatment. A profile of BU-derived DNA adducts provides integrated information on the drug activity and individual variables that may influence it and therefore is an excellent candidate as a predictive biomarker in precision oncology.⁵

There are limited studies showing the addition products of BU with glutathione.³⁰^,³¹^,³² Scian and coworkers studied the role of BU-related metabolite γ-glutamyldehydroalanylglycine (EdAG) and the action of BU on intact proteins, highlighting novel mechanisms of BU toxicity.³³^,³⁴^,³⁵ However, to the best of our knowledge, no study has characterized DNA adducts formed by BU. Iwamoto and coworkers demonstrated that BU caused DNA damage at 5′-GA-3′ and 5′-GG-3′ sequences in vitro with detection of two structurally unknown adducts.²⁵ They concluded that the generation of cross-links may be a strong mechanism of BU cytotoxicity. Souliotis and colleagues highlighted the relevance of mono adducts in therapeutic effect, supporting a two-step mechanism of generation of cross-linked adducts. Indeed, cross-links may be generated by a delayed binding to a second site in DNA.³⁶ Therefore, repair of mono adducts may interfere with the generation of cross-linked adducts and module the drug cytotoxicity.³⁶ Interstrand lesions are thought to play a central role in the cytotoxic response because they prevent DNA strand separation, which is required for DNA replication and transcription.³⁷ Overall, these observations support the importance of monitoring the full profile of adducts rather than focusing only on few major ones. Traditionally, the interactions between alkylating drugs and nucleic acids have been studied by targeting a few specific analytes resulting from the main chemical modifications occurring according to the known reactivity of the drug with DNA’s reactive sites. Our novel DNA adductomics approach, instead, allows the simultaneous screening of all DNA covalent modifications (not only the major ones), including both known and previously unidentified ones. This more comprehensive method may also include those resulting from metabolic modifications or activation pathways that may be unexpected. The possibility to screen how the drug, directly or through its metabolites, interacts with DNA gives a full picture of all the possible interactions of the drug and the metabolites with their target. In addition to the direct action of drug with DNA, this allows us to understand the role of all the metabolic processes involved. Furthermore, different DNA modifications may be affected by DNA repair mechanisms in various ways and at different rates, thus screening for multiple DNA adducts allows us to potentially monitor the influence of repairing enzymes on the action of the drug.

In the present study, we first investigated the formation of BU-DNA adducts in our in vitro samples with BU-exposed CT-DNA and ¹⁴N-/¹⁵N-DNA mixture from E. coli, and identified four putative cross-links including adduct 6 and 11 mono adducts including adduct 7. Since the synthesis and characterization of all adducts would require enormous effort, we focused on the development of a quantitative method for the two major adducts 6 and 7 in the profile, namely those that were present in most samples. In the future, these two adducts will serve as standards for the development of analytical methods aimed at profiling all DNA adducts and evaluating their relationship with patient outcome.

The chemical synthesis of 6 and 7 was pursued for absolute characterization and development of a quantitative method to be used in parallel to the profiling. The synthetic strategy for synthesizing the cross-link adduct 6 utilized two sequential alkylation reactions instead of the classical method of treating dGuo with the alkylating agent (such as nitrogen mustard³ and bis(2-chloroethyl)ethylamine³⁸) in a one-step reaction. The advantages of our method are (1) allowing a full NMR characterization of the Gua-N7-alkyl product obtained from each alkylation reaction, and (2) yielding cleaner reaction mixtures for a better separation of the desired product. We failed to identify desired products from the reaction mixture of dGuo with BU or 1,4-dibromobutane due to the complexity of the HPLC traces.

The success of the first-step alkylation (Figure S1) is critical. This reaction requires a long reaction time (14 days) with mild stirring at room temperature to achieve completion (Figure S3). Increasing the reaction temperatures (e.g., 60°C) or use of different solvents (e.g., DMSO) and added bases (e.g., Et₃N, DBU, K₂CO₃) failed to facilitate the reaction; all the reactions yielded much more complex products with no improved yields of the desired product 3. For the second-step alkylation, the reaction yield for compound 4 is very low even with excess starting material 2. The product 5 is considered the hydrolysis product from 3 during the second-step alkylation reaction in a similar low yield. The mild condition from the first-step alkylation reaction did not work for the second-step alkylation, even after up to 14 days. One important observation is that the excess amount of NH₄OH used for the deprotection reaction of 4 could degrade the product 6 but did not appear to affect compound 7.

Using the synthesized standards, we were able for the first time to quantify BU-derived DNA adducts in patients. The concentrations of 7 ranged from 12.8 to 28.2 adducts/10⁶ nucleotides (except for patient 3), significantly higher than those of 6, which ranged from 0.38 to 2.02 adducts/10⁶ nucleotides. Out of the 15 putative DNA adducts detected in vitro, 12 were detected in at least one patient undergoing BU-based chemotherapy. Interestingly, several DNA adducts are in common among patients underlining a general reaction mechanism between BU and DNA. Chemical modifications to guanine and adenine are observed to be the most frequent ones as commonly reported in past investigations of alkylating drugs.³^,³⁶^,³⁸^,³⁹ An isotope dilution method was developed and validated for quantifying the cross-link adduct 6 and the mono adduct 7. The cross-link adduct 6 and the mono adduct 7 were formed at high levels, ranging from 0.3 to 2 and from 1 to 263 adducts every 10⁶ nucleosides, respectively. The higher amount of the mono adduct is consistent with the observations of Soulioti et al., Palom et al., and Hemminki in the cases of cyclophosphamide and mitomycin exposure.³⁶^,³⁹^,⁴⁰ However, due to the small number of patients in the present study with different doses and imprecise therapy time courses, the evaluation of the effects of adducts 6 and 7 on therapeutic outcome was not possible but can be done in the future.

In conclusion, we explored for the first time the BU-DNA adductome using a DNA adductomic approach. This resulted in a list of 15 putative DNA adducts formed by BU treatment in CT-DNA. Twelve of the 15 adducts were detected in the blood DNA of six patients undergoing BU treatment, and two of these were characterized to be the cross-link DNA adduct 6 and the mono DNA adduct 7 by comparison with their corresponding synthesized standards. An isotope dilution HPLC-NSI-HRMS/MS method was developed to quantify the two adducts in the blood DNA. Their formation occurred in relatively high concentrations, suggesting their potential as promising biomarkers for the investigation of interindividual differences in BU-induced DNA adducts for patient stratification.

Materials and methods

Chemicals and supplies

Methanol (MeOH, liquid chromatography [LC]-MS grade), acetonitrile (ACN, LC-MS grade), isopropanol (IPA), and formic acid (FA, 98% _v/v) were purchased from Fluka (St. Louis, MO, USA). Water was purified by a Milli-Q system (Milford, MA, USA). Cell Lysis Solution, Proteinase K, and RNase-A were purchased from Qiagen (Germantown, MD, USA). DNase I recombinant expressed by Pichia pastoris (R-DNase, 10,000 U/mg), phosphodiesterase-1 extracted from Crotalus adamanteus (PDE-1, 0.4 U/mg), recombinant alkaline phosphatase expressed by P. pastoris (R-ALP, 7,000 U/mg), and calf-thymus DNA (CT-DNA) were purchased from Roche (St. Louis, MO, USA). Single-membrane filtration devices Microcon (10-kDa cutoff, 0.5 mL) were purchased from Amicon (Billerica, MA, USA). Silanized vials (0.3, 1.2, and 4 mL) were purchased from ChromTech (Apple Valley, MN, USA). [¹⁵N₅]2′-deoxyguanosine (dGuo) was purchased from Cambridge Isotope Laboratories. All other chemicals and supplies were purchased from Sigma-Aldrich or Fisher Scientific.

Synthesis of chemical standards

N'-(9-((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-6-oxo-6,9-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (2, N²-dmf-dGuo)

N,N-dimethylacetamide dimethyl acetal (2.0 mL) was added to a solution of dGuo (3.5 mmol, 1.0 g) in MeOH (9.5 mL). The reaction mixture was stirred at room temperature for 3 days. After reaction, the precipitated white solid was filtered and washed with cold MeOH. The solid was dried and used directly without further purification (1.05 g, 93%). ¹H NMR (500 MHz, DMSO-d₆) δ 11.31 (s, 1H, H1), 8.55 (s, 1H, (CH₃)₂NCH = N-), 8.03 (s, 1H, H8), 6.25 (dd, J = 7.9, 6.1 Hz, 1H, H1′), 5.29 (d, J = 3.9 Hz, 1H, 3′-OH), 4.93 (t, J = 5.5 Hz, 1H, 5′-OH), 4.37 (dq, J = 6.1, 3.0 Hz, 1H, H3′), 3.83 (td, J = 4.6, 2.5 Hz, 1H, H4′), 3.57 (dt, J = 11.7, 5.1 Hz, 1H, H5′a), 3.51 (ddd, J = 11.6, 5.8, 4.4 Hz, 1H, H5′b), 3.16 (s, 3H, -CH₃), 3.03 (s, 3H, -CH₃), 2.59 (ddd, J = 13.5, 8.0, 5.7 Hz, 1H, H2′a), and 2.23 (ddd, J = 13.2, 6.1, 2.9 Hz, 1H, H2′b). For the synthesis of the isotopically labeled compound [¹⁵N₅]2, N,N-dimethylacetamide dimethyl acetal (1.0 mL) was added to a solution of [¹⁵N₅]dGuo (5.0 mg, 0.018 mmol) in MeOH (1.0 mL). After stirring at room temperature for 3 days, the reaction mixture was dried to remove most of the solvent and diluted with H₂O (0.5 mL) before performing reverse-phase HPLC purification using Waters Associates (Milford, MA) systems equipped with a Shimadzu SPD-10A 0.2 mm Prep UV-vis detector (254 nm). A Luna 5-μm C18(2) 100 Å 250 × 10-mm column purchased from Phenomenex (Torrance, CA) was used for the separation. A 50-min program was used with a flow rate of 4 mL/min and a gradient starting from 5% MeOH in H₂O for 10 min, then increasing linearly to 90% MeOH in H₂O over 25 min. After holding at 90% MeOH in H₂O for 3 min, the gradient was returned to the initial condition of 5% MeOH in H₂O over 5 min. The system was equilibrated for 7 min before the next injection. The desired product [¹⁵N₅]2 was collected at the retention time of 22 min ¹H NMR (500 MHz, MeOD) δ 8.64 (d, J = 2.3 Hz, 1H, (CH₃)₂NCH = ¹⁵N-), 8.06 (dd, J = 11.3, 7.8 Hz, 1H, H8), 6.37 (ddd, J = 7.7, 6.1, 1.4 Hz, 1H, H1′), 4.55 (dt, J = 6.1, 3.0 Hz, 1H, H3′), 4.01 (q, J = 3.5 Hz, 1H, H4′), 3.79 (dd, J = 12.0, 3.6 Hz, 1H, H5′a), 3.72 (dd, J = 12.0, 3.9 Hz, 1H, H5′b), 3.21 (s, 3H, -CH₃), 3.12 (s, 3H, -CH₃), 2.73 (dddd, J = 13.6, 8.3, 6.1, 2.7 Hz, 1H, H2′a), and 2.39 (dddd, J = 13.5, 6.2, 3.2, 1.3 Hz, 1H, H2′b).

N'-(7-(4-bromobutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (3, N²-dmf-Gua-N7-butylbromide)

A solution of 7 (0.5 mmol, 161 mg) and 1,4-dibromobutane (2.5 mmol, 300 μL) in N,N-dimethylformamide (DMF, 2.0 mL) was stirred at room temperature for 14 days. The reaction mixture turned from a white suspension to a clear yellow solution after 8 days. After completion, the reaction mixture was subjected directly to reverse-phase HPLC for purification using the same conditions described for the synthesis of [¹⁵N₅]2. The desired product 3 eluted at 30.4 min as the major peak (Figure S3). After drying under a stream of N₂, compound 3 was collected as a white solid (101 mg, 60%). The full NMR spectra are presented in Figure S4. ¹H NMR (500 MHz, DMSO-d₆) δ 11.34 (s, 1H, H1), 8.60 (s, 1H, (CH₃)₂NCH = N-), 8.01 (s, 1H, H8), 4.26 (t, J = 6.7 Hz, 2H, -CH₂CH₂CH₂CH₂Br), 3.53 (td, J = 6.7, 2.2 Hz, 2H, -CH₂CH₂CH₂CH₂Br), 3.14 (s, 3H, -CH₃), 3.01 (s, 3H, -CH₃), 1.96–1.86 (m, 2H, -CH₂CH₂CH₂CH₂Br), 1.79–1.66 (m, 2H, -CH₂CH₂CH₂CH₂Br). ¹³C NMR (126 MHz, DMSO) δ 158.90 (C4), 157.64 ((CH₃)₂NCH = N-), 156.45 (C2), 155.27 (C6), 143.40 (C8), 110.70 (C5), 45.12 (-CH₂CH₂CH₂CH₂Br), 40.46 (-CH₃), 34.51 (-CH₃), 34.30 (-CH₂CH₂CH₂CH₂Br), 29.24 (-CH₂CH₂CH₂CH₂Br), and 29.06 (-CH₂CH₂CH₂CH₂Br). HRMS (Orbitrap): [M + H]⁺ calculated (calc’d) 341.0720, 343.0700; 286.0298, 288.0278; found 341.0665, 343.0641; 286.0329, 288.0310. For the synthesis of the isotopically labeled compound [¹⁵N₅]3, 1,4-dibromobutane (20 μL) was added to a solution of [¹⁵N₅]2 in DMF (0.5 mL). The reaction mixture was stirred at room temperature for 14 days. After reaction, the reaction mixture was subjected directly to reverse-phase HPLC, and the desired product [¹⁵N₅]3 was collected at the retention time of 30.3 min under the same conditions described for the synthesis of [¹⁵N₅]2. HRMS (Orbitrap): [M + H]⁺ calc’d 346.0572, 348.0552; 291.0150, 293.0130; found 346.0578, 348.0648; 291.0343, 293.0314.

N',N''-(butane-1,4-diylbis(6-oxo-6,7-dihydro-1H-purine-7,2-diyl))bis(N,N-dimethylformimidamide) (4, bis(N²-dmf-Gua-N7)butyl, and N'-(7-(4-hydroxybutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (5, N²-dmf-Gua-N7-butanol)

To a solution of 3 (0.22 mmol, 101 mg) in DMF (1.5 mL), compound 2 (0.29 mmol, 94 mg) was added. The reaction mixture was stirred at 60°C for 2 days and then analyzed by reverse-phase HPLC directly under the same conditions described for the synthesis of [¹⁵N₅]7. The desired products 4 and 5 were collected at 25.6 and 26.0 min, respectively, in very low yields (<1%) (Figure S6). The isotopically labeled compounds [¹⁵N₅]4 and [¹⁵N₅]5 were synthesized in the same way as 4 and 5. A solution of 2 (16.2 mg, 0.05 mmol) was added to a DMF (0.5 mL) solution of [¹⁵N₅]3. The reaction mixture was stirred at 60°C for 2 days. After reaction, the resulting mixture was subjected to reverse-phase HPLC directly using the same separation conditions as described above. The desired compounds [¹⁵N₅]4 and [¹⁵N₅]5 were collected at the retention times of 25.6 and 26.0 min, respectively.

Compound 4 was fully characterized by one- and two-dimensional NMR and HRMS. The NMR spectra are presented in Figure S5. ¹H NMR (500 MHz, MeOD) δ 8.59 (s, 2H, (CH₃)₂NCH = N-), 7.99 (s, 2H, H8), 4.48–4.30 (m, 4H, N-CH₂CH₂-), 3.18 (s, 6H, -CH₃), 3.10 (s, 6H, -CH₃), 2.18–1.69 (m, 4H, N-CH₂CH₂-). ¹³C NMR (126 MHz, MeOD) δ 160.50 (C4), 159.44 (CH₃)₂NCH = N-), 158.85 (C2), 144.92 (C8), 112.03 (C5), 47.45 (N-CH₂CH₂-), 41.29 (-CH₃), 35.16 (-CH₃), and 28.84 (N-CH₂CH₂-). HRMS (Orbitrap): [M + H]⁺ calc’d 467.2374; found 467.2359. However, parent is in lower abundance than its fragment m/z 266.1310. The structure of [¹⁵N₅]4 was confirmed by HRMS. HRMS (Orbitrap): [M + H]⁺ calc’d 472.2226, 266.1310, 261.1458; found 266.0663 and 261.2174 (no parent mass was observed, as above noted due to in-source fragmentation and spontaneous loss of Boc group in the source).

Compound 5 was characterized by ¹H NMR (Figure S7) and HRMS. ¹H NMR (500 MHz, MeOD) δ 8.61 (s, 1H, (CH₃)₂NCH = N-), 8.07 (s, 1H, H1), 8.01 (s, 1H, H8), 4.37 (t, J = 7.1 Hz, 2H, -CH₂CH₂CH₂CH₂OH), 4.18 (t, J = 6.4 Hz, 2H, -CH₂CH₂CH₂CH₂OH), 3.19 (s, 3H, -CH₃), 3.11 (s, 3H, -CH₃), 1.99 (p, J = 7.3 Hz, 2H, -CH₂CH₂CH₂CH₂OH), and 1.68 (dt, J = 14.1, 6.6 Hz, 2H, -CH₂CH₂CH₂CH₂OH). HRMS (Orbitrap): [M + H]⁺ calc’d 279.1564; found 279.1573. The structure of [¹⁵N₅]5 was confirmed by HRMS. HRMS (Orbitrap): calc’d [M + H]⁺ 284.1416; found 284.1414.

7,7'-(Butane-1,4-diyl)bis(2-amino-1,7-dihydro-6H-purin-6-one) (6, N7G-Bu-N7G)

To a solution of 4 (∼1 mg) in MeOH (0.5 mL) was added 28%–30% NH₄OH solution (0.1 mL). The reaction mixture was stirred at room temperature overnight. After completion, the reaction mixture was dried by a stream of N₂ and the residue was reconstituted in H₂O before subjecting it to reverse-phase HPLC for purification using the same conditions described for the synthesis of [¹⁵N₅]2. The desired product 6 was eluted at 21.1 min (Figure S8). The ¹H NMR spectra are presented in Figure S8. ¹H NMR (500 MHz, MeOD) δ 7.69 (s, 2H, H8), 4.31 (t, J = 6.8 Hz, 4H, N-CH₂CH₂-), 4.08 (t, J = 6.9 Hz, 4H, N-CH₂CH₂-). ¹H NMR (500 MHz, DMSO-d₆) δ 7.63 (s, 2H, H8), 4.20 (t, J = 6.4 Hz, 4H, N-CH₂CH₂-), and 3.92 (t, J = 6.8 Hz, 4H, N-CH₂CH₂-). HRMS (Orbitrap): [M + H]⁺ calc’d 357.1530; found 357.1525. The isotopically labeled compound [¹⁵N₅]6 was synthesized using the same method used for 6. To a solution of [¹⁵N₅]4 in MeOH (0.5 mL) was added 28%–30% NH₄OH solution (0.1 mL), and the mixture was stirred at room temperature overnight. After reaction, the solvent was removed by drying with a stream of N₂ and the residue was reconstituted in H₂O before subjecting it to reverse-phase HPLC for purification using the same conditions described for the synthesis of [¹⁵N₅]2. The desired product [¹⁵N₅]6 was collected at 21.2 min. The structure of [¹⁵N₅]6 was confirmed by HRMS. HRMS (Orbitrap): [M + H]⁺ calc’d 362.1382; found 362.1382.

2-Amino-7-(4-hydroxybutyl)-1,7-dihydro-6H-purin-6-one (7, N7G-Bu-OH)

Compound 7 and its isotopically labeled analog [¹⁵N₅]7 were synthesized similarly as compound 6. The desired product 7 eluted from HPLC at 20.1 min (Figure S9) under the same conditions described for the synthesis of [¹⁵N₅]2. The desired product [¹⁵N₅]7 was collected at 20.0 min. The structure of compound 7 was confirmed by ¹H NMR (Figure S9) and HRMS. ¹H NMR (500 MHz, DMSO) δ 10.70 (s, 1H, H1), 7.89 (s, 1H, H8), 6.07 (s, 2H, N²-NH₂), 4.41 (t, J = 5.2 Hz, 1H, -OH), 4.17 (t, J = 6.9 Hz, 2H, -CH₂CH₂CH₂CH₂OH), 1.77 (p, J = 7.1 Hz, 2H, -CH₂CH₂CH₂CH₂OH), and 1.32 (dt, J = 14.1, 6.6 Hz, 2H, -CH₂CH₂CH₂CH₂OH). HRMS (Orbitrap): [M + H]⁺ calc’d 224.1142; found 224.1141. The structure of [¹⁵N₅]7 was confirmed by HRMS. HRMS (Orbitrap): [M + H]⁺ calc’d 229.0994; found 229.0993.

DNA isolation and purification

¹⁵N-DNA generation

E. coli (MG1655 strain) was cultured in ¹⁵N-labeled minimal medium (5 mL) at 37°C overnight for three generations to obtain a uniformly labeled ¹⁵N-strain. The optical density 600 (OD₆₀₀) was used to monitor bacterial growth until stationary phase. The bacteria culture was centrifuged at 3,000 × g for 10 min at 24°C. The bacterial pellet was re-suspended in 50% glycerol in bacterial medium solution, frozen, and stored at −80°C until use. ¹⁵N-labeled minimal medium (1 L) was prepared using 200 mL of M9 salts (Na₂HPO₄·7H₂O, 64 g KH₂PO₄, 15 g, NaCl, 2.5 g, and ¹⁵NH₄Cl, 5.0 g in deionized H₂O, 1 L), and added to 20 mL of glucose (20%; Sigma-Aldrich), 2 mL of MgSO₄ (1 M; Fisher Scientific), and 100 μL of CaCl₂ (1 M; Fisher Scientific). A starter culture was prepared by inoculating 10 μL of ¹⁵N- and ¹⁴N-stock bacteria into 5 mL of medium and grown overnight. Fifty microliters of cells were then inoculated in 1 L of medium and incubated overnight. Cells were centrifuged and the pellet stored at −80°C until use (Figure S10).

Bacterial DNA isolation

Cells were re-suspended in 25 mL of Cell Lysis Solution (Qiagen) and treated with 150 μL each of Proteinase K (24 h, 24°C) and RNase-A (2 h, 24°C). Proteins were precipitated by adding 7.5 mL of Protein Precipitation Solution (Qiagen). The pellet was discarded and the supernatant was transferred into a new tube containing an equal amount of cold isopropanol (IPA) to precipitate the DNA. The mixture was centrifuged, the supernatant was discarded, and the DNA pellet was sequentially washed with 70% (v/v) IPA in H₂O and 100% IPA. The DNA pellet was dried, re-suspended in buffer (20 mM Tris, 2 mM MgCl₂ pH 7.4), and stored at −20°C. The yield and purity of the DNA were assessed using a nanodrop UV-visible (UV-vis) spectrophotometer monitoring the 260- and 280-nm wavelengths.

Sample collection from patients and whole-blood DNA extraction

This study was approved by the University of Minnesota Institutional Review Board (IRB) (no. 1506M74263). All patients signed IRB-approved informed consent in accordance with the Declaration of Helsinki. Blood (∼3 mL) was obtained from four cancer and two Fanconi Anemia patients undergoing allogeneic HCT receiving BU as part of their preparative regimens. Patients received a 2-h infusion of intravenous BU prior to their HCT. Whole blood was obtained from a central venous catheter immediately prior to the first dose of BU (0 h) and within 24 h after the completion of the BU infusion protocol for a total of six baseline samples to be compared with the matching six samples post treatment. Whole-blood DNA extraction was performed immediately after the sample collection. Whole-blood DNA extraction was performed using Qiagen DNA extraction kits (Qiagen, Valencia, CA) by the manufacturer’s instructions with slight modifications as previously reported.⁴¹ DNA amounts were estimated by UV and subsequently determined by HPLC analysis of dGuo in enzymatic hydrolysates as described below. DNA samples (10–50 μg in 200 μL of Tris buffer) were spiked with 30 fmol of [¹⁵N₅]N7G-Bu-OH and 50 fmol of [¹⁵N₅]N7G-Bu-N7G as internal standards and subjected to enzymatic hydrolysis as described below.

DNA exposure to BU in vitro

CT-DNA (100 μg) or ¹⁴N- and ¹⁵N-bacterial DNA (210 μg each; 1:1 ratio) was dissolved in Tris buffer (20 mM Tris, 2 mM MgCl₂, pH 7.4). BU (in DMSO) was added to the DNA solutions separately to reach a final concentration of 100 μM (246 ng/mL) in each batch. This is close to the levels measured in patient plasma of previous studies.⁴² The total volume of the corresponding sample was 200 μL. The resulting mixture was incubated at 37°C overnight. DNA exposed to DMSO (1% v/v) was used as a negative control. DNA isolation was performed by IPA precipitation. Briefly, 1 mL of cold IPA was added to each sample vial. The precipitated DNA was isolated, washed sequentially with 1 mL each of 70% IPA and 100% IPA twice, and dried under a stream of N₂. All steps of the protocol were performed using silanized glass vials.

DNA hydrolysis and sample enrichment

DNA enzymatic hydrolysis

DNA obtained above was dissolved in Tris buffer and incubated with R-DNase (0.5 U/μg DNA) at room temperature overnight. The following day, an additional amount of R-DNase (0.5 U/μg DNA) and R-ALP-1 (0.4 U/μg DNA) and PDE-1 (0.02 mU/μg DNA) were added and the mixture was incubated at 37°C for 70 min. The mixture was further incubated at room temperature overnight. After hydrolysis, enzymes were removed via filtration (Microcon, 10-kDa cutoff) by centrifuging at 12,000 × g for 60 min. The filtrate was collected for the analyte analysis and dGuo quantitation.

dGuo quantitation by HPLC

Quantitation of dGuo was carried out using a Dionex UltiMate 3000 RSLCnano System (Thermo Scientific, Waltham, MA) with a UV detector set at 254 nm. A 300-μm internal diameter (ID) × 15-cm C18 column (2 μm, 100 Å) (Thermo Scientific, Waltham, MA) was used with (A) H₂O and (B) MeOH as the mobile phase. The injection volume was 1 μL and the flow rate was 15 μL/min. Starting from 5% B for 2 min, a linear gradient was applied increasing from 5% to 25% B in 10 min. The gradient increased to 95% B in 3 min and was held for 5 min before returning back to the initial condition in 2 min. The instrument was equilibrated for 3 min (25 min total run time) before the next injection. A calibration curve for dGuo (0.0625–50 ng/μL in H₂O) was run in triplicate and used to calculate the dGuo content in each sample.⁴³

Sample enrichment and purification

DNA hydrolysates were partially purified by solid-phase extraction using Strata-X cartridges (30 μm, Phenomenex, Torrance, CA) that were activated with 3 mL of MeOH and preconditioned with 1 mL of H₂O. The hydrolysates were loaded on the cartridges and washed sequentially with 1 mL of H₂O and 1 mL of 5% MeOH in H₂O, and the analytes were eluted with 1 mL of 100% MeOH and 1 mL of MeOH containing 2% formic acid, and both fractions were evaporated to dryness and reconstituted in 2% MeOH in H₂O (LC-MS grade, Fluka) to a final volume of 20 μL prior to MS analysis.

Adductomic DDA-CNL/MS³ method and data analysis

LC conditions

An aliquot of 1 μL of DNA hydrolysate was injected in a nanoflow Ultra Performance Liquid Chromatography (UPLC) system (Ultimate 3000 RSLCnano UPLC, Thermo Scientific, Waltham, MA). The UPLC system was equipped with a 5-μL autosampler injection loop. Chromatographic separation was achieved using a hand-packed commercially available fused-silica emitter (230 × 0.075-mm ID, 15-μm orifice, New Objective, Woburn MA) with C18 stationary phase (5 μm, 100, Luna Phenomenex, Torrance, CA). The mobile phase consisted of (A) 0.05% (v/v) formic acid in H₂O and (B) CH₃CN. The eluent was held at 2% B for 2 min, brought to 20% B in 24 min, then to 60% B in 10 min, to 98% B in 1 min, and then maintained at 98% for 4 min. The column was re-equilibrated for 4 min. The injection valve position was switched at 6 min to take the injection loop out of the flow path.

MS

All the MS-based analyses were conducted using a hybrid high-field Orbitrap mass spectrometer (Fusion, Thermo Scientific, Waltham, MA). The LC system was interfaced to the mass spectrometer using a Nanoflex ESI ion source (Nanoflex Thermo Scientific, Waltham, MA), which operated in positive ion mode at 2.5 kV. The ion transfer tube temperature was 300°C, and the radio frequency (RF) lens (%) setting was 60. The DDA-CNL/MS³ method consisted of a full scan with data-dependent MS² and a neutral loss MS³ acquisition (NL-MS³). The full scan (m/z 150–750) was performed with quadrupole filtering, maximum injection time of 50-ms, automatic gain control target of 50%, and a resolution setting of 120,000. Data-dependent MS² parameters were dynamic exclusion of 30 s, mass tolerance of ±5 ppm, repeat count of 1, minimum intensity of 2 × 10⁴, and a cycle time of 3 s. An MS² exclusion list of 18 masses, consisting of unmodified deoxyribonucleosides and their electrostatically bound dimer ions, was included in the method with a mass tolerance of ±5 ppm.²⁹ The MS² scan events were triggered on the basis of an intensity threshold of 5 × 10³. The MS² fragmentation was performed with a quadrupole isolation width of m/z 1.5, higher-energy C-trap dissociation (HCD) collision energy 30%, automatic gain control (AGC) value of 400%, maximum injection time of 54 ms, and Orbitrap detection at a resolution setting of 30,000. For the NL-MS³ data acquisitions, the MS² product ions were isolated by the ion trap with an isolation window of m/z 2.0, and the MS³ fragmentation was triggered upon observation of the neutral loss of the 2′-deoxyribose (dR) or the base moieties (-dR, 116.0474; -G, 151.0494; -A, 135.0545; -T, 126.0429; -C, 111.0433; all within a mass tolerance of ±5 ppm). Masses of the respective ¹⁵N-nucleosides were also included. The MS parameters for the NL-MS³ scan were 30% HCD fragmentation, Orbitrap detection resolution of 15,000, AGC value of 200%, and a maximum injection time of 200 ms.

Data analysis

For DNA adduct screening data analysis, RawConverter (http://fields.scripps.edu/rawconv/) was used to convert the Thermo Scientific raw data files to mzXML files, which were imported into an SQL database, where data filtration on the basis of MS³ triggering ion m/z and retention time was performed. MS³ triggering ions present in the exposed samples with both m/z (±5 ppm) and retention times (±30 s) different from those in the control samples were selected and reported as putative BU-DNA adducts. The absence of the putative BU-DNA adducts in the control samples was further confirmed by generating EICs for all the putative DNA adduct precursor masses at 5 ppm mass tolerance using the Qualbrowser component of the Xcalibur 3.0 software package (Thermo Scientific, Waltham, MA). The MS² and MS³ spectra of each putative DNA adduct were subsequently evaluated for structural information, and the peak areas of the precursor mass were determined.

The presence of the resulting adducts was further confirmed using ¹⁴N- and ¹⁵N-DNA exposed to the drug, processed as described above, and mixed in a 1:1 ratio prior to LC-MS analysis (Figure S2). In this resulting sample, the presence of co-eluting peaks and MS³ events corresponding to ¹⁴N- and ¹⁵N-labeled version of the same adduct supports adduct identification. Briefly, the mass difference between ¹⁴N and ¹⁵N corresponding to the presence of two, three, four, five, six, seven, eight, nine, or 10 ¹⁵N in the molecule was added to each DNA adduct uniquely found in the exposed CT-DNA sample. The resulting masses were manually searched in QualBrowser in the ¹⁵N-DNA sample, and MS² and MS³ spectra were investigated and compared with that of the corresponding putative adducts found in CT-DNA. The same retention time (±0.1 min), neutral loss, and fragmentation patterns were expected. Furthermore, the presence of a co-eluting peak and MS³ event corresponding to the ¹⁴N version of the putative DNA adduct was verified. DNA adducts uniquely present in the exposed sample and having the corresponding ¹⁴N and ¹⁵N peaks were annotated.

Targeted MS method for investigating and quantitating BU-DNA adducts in human blood

LC conditions

The DNA hydrolysates were reconstituted in 20 μL of 2% MeOH in H₂O (LC-MS grade, Fluka) and 1 μL was injected with the same UPLC system and column conditions as described in the adductomic DDA-CNL/MS³ method. The mobile phase consisted of (A) 5 mM NH₄OAc in H₂O and (B) CH₃CN. The elution program included an isocratic step (2% of B for 5.5 min at 0.9 μL/min), followed by a two-step linear gradient of B (3%/min for 19 min and 19%/min for 2 min, both at a flow rate of 0.3 μL/min) and it concluded with a washing isocratic step, performed at 98% of B for 2 min at 0.3 μL/min. At the end of the elution program, the LC system was equilibrated for 3 min at isocratic conditions (2% of B, 0.9 μL/min). The injection valve position was switched at 6 min to take the injection loop out of the flow path.

Mass spectrometry

The targeted DNA adduct analysis was performed using the targeted MS² analysis of 15 BU-derived DNA adducts (Table 2), with a quadrupole isolation window width of m/z 1.5, maximum injection time of 54 ms, AGC of 1,000%, and resolution setting of 30,000. MS fragmentation was performed using HCD with a stepped collision energy (25%). The fragmentation of the DNA adduct precursor ions [MH]⁺ results in neutral loss of the dR, G, A, T, or C moiety to produce the corresponding [MH-dR]⁺, [MH-G]⁺, [MH-A]⁺, [MH-T]⁺, and [MH-C]⁺ product ions whose masses were extracted for detection of the adducts.

BU-derived DNA adduct quantitation

LODs for adducts N7G-Bu-N7G (6, Figure 1) and N7G-Bu-OH (7) were determined using their H₂O solutions. LOQ, accuracy, and precision for the MS quantitation method were determined by analyzing CT-DNA enriched with different amounts of synthesized standards 6 and 7 (0, 5, 10, 50, 100, and 200 fmol for adduct 6 and a constant amount of 10 fmol of [¹⁵N₅]6; 0, 1, 5, 10, 30, and 100 fmol for adduct 7 and a constant amount of 5 fmol of [¹⁵N₅]7). Each sample was prepared in triplicate. The LOQ was defined as the lowest amount added to CT-DNA that produced a CV lower than 20%. Accuracy was determined by comparing added and measured amounts of the adducts at each level. Precision was determined by the intraday CV in the triplicate samples. Quantitation of the adducts was done comparing the sample peak area with a calibration curve (0.05, 0.1, 0.3, 0.5, 1, 3, 10, 100, and 200 fmol for adducts 6 and 7).

Acknowledgments

MS was carried out in the Analytical Biochemistry Shared Resource of the Masonic Cancer Center, University of Minnesota, supported in part by Cancer Center Support Grant CA-077598. Salary support for P.W.V., head of the Analytical Biochemistry Shared Resource, was provided by the National Cancer Institute Grant CA-211256. The authors would like to thank Dr. Laura Maertens and Carsten Spry for their help with initiating and coordinating the sample collection for this study. We also wish to thank the patients for their participation in this study. Additionally, we acknowledge the hard work and dedication of the nurses and other medical personnel of the M Health Fairview University of Minnesota Medical Center for their help with collecting the samples. We thank Robert Carlson for editorial assistance.

Author contributions

V.G., S.B., and Y.L. conceived the idea of this manuscript and designed the experiments. Y.L. and V.G. performed the synthesis and characterization of adducts. V.G. and F.J. collected samples from clinics and isolated DNA from blood. V.G., Y.L., and F.J. performed standard purification. V.G. and F.J. performed DNA sample preparation, HRMS, and data analysis. M.L.M. supported sample collection in the clinic together with providing inputs on the project and a clinical point of view. P.W.V. contributed to the optimization of LC-MS-based methods in addition to useful inputs on the project. S.B. and S.S.H. supervised the project. V.G. and Y.L. drafted the manuscript. All authors have read and agreed to the published version of the manuscript.

Declaration of interests

The authors declare no competing interests.

Footnotes

Supplemental information can be found online at https://doi.org/10.1016/j.omto.2023.01.005.

Supplemental information

Document S1. Figures S1–S10 and Table S1

mmc1.pdf^{(2.2MB, pdf)}

Document S2. Article plus supplemental information

mmc2.pdf^{(4.1MB, pdf)}

Data availability

The data that support the findings of this study are available from the authors on reasonable request; see author contributions for specific datasets.

References

1.Diethelm-Varela B., Ai Y., Liang D., Xue F. Nitrogen mustards as anticancer chemotherapies: historic perspective, current developments and future trends. Curr. Top. Med. Chem. 2019;19:691–712. doi: 10.2174/1568026619666190401100519. [DOI] [PubMed] [Google Scholar]
2.Goodman L.S., Wintrobe M.M. Nitrogen mustard therapy; use of methyl-bis (beta-chloroethyl) amine hydrochloride and tris (beta-chloroethyl) amine hydrochloride for Hodgkin's disease, lymphosarcoma, leukemia and certain allied and miscellaneous disorders. J. Am. Med. Assoc. 1946;132:126–132. doi: 10.1001/jama.1946.02870380008004. [DOI] [PubMed] [Google Scholar]
3.Malayappan B., Johnson L., Nie B., Panchal D., Matter B., Jacobson P., Tretyakova N. Quantitative high-performance liquid chromatography-electrospray ionization tandem mass spectrometry analysis of bis-N7-guanine DNA-DNA cross-links in white blood cells of cancer patients receiving cyclophosphamide therapy. Anal. Chem. 2010;82:3650–3658. doi: 10.1021/ac902923s. [DOI] [PubMed] [Google Scholar]
4.Johnson L.A., Malayappan B., Tretyakova N., Campbell C., MacMillan M.L., Wagner J.E., Jacobson P.A. Formation of cyclophosphamide specific DNA adducts in hematological diseases. Pediatr. Blood Cancer. 2012;58:708–714. doi: 10.1002/pbc.23254. [DOI] [PMC free article] [PubMed] [Google Scholar]
5.Stornetta A., Zimmermann M., Cimino G.D., Henderson P.T., Sturla S.J. DNA adducts from anticancer drugs as candidate predictive markers for precision medicine. Chem. Res. Toxicol. 2017;30:388–409. doi: 10.1021/acs.chemrestox.6b00380. [DOI] [PMC free article] [PubMed] [Google Scholar]
6.Wang S.S., Zimmermann M., Zhang H., Lin T.Y., Malfatti M., Haack K., Turteltaub K.W., Cimino G.D., de Vere White R., Pan C.X., Henderson P.T. A diagnostic microdosing approach to investigate platinum sensitivity in non-small cell lung cancer. Int. J. Cancer. 2017;141:604–613. doi: 10.1002/ijc.30747. [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Wang L.E., Yin M., Dong Q., Stewart D.J., Merriman K.W., Amos C.I., Spitz M.R., Wei Q. DNA repair capacity in peripheral lymphocytes predicts survival of patients with non-small-cell lung cancer treated with first-line platinum-based chemotherapy. J. Clin. Oncol. 2011;29:4121–4128. doi: 10.1200/JCO.2010.34.3616. [DOI] [PMC free article] [PubMed] [Google Scholar]
8.Zimmermann M., Wang S.S., Zhang H., Lin T.Y., Malfatti M., Haack K., Ognibene T., Yang H., Airhart S., Turteltaub K.W., et al. Microdose-induced drug-DNA adducts as biomarkers of chemotherapy resistance in humans and mice. Mol. Cancer Ther. 2017;16:376–387. doi: 10.1158/1535-7163.MCT-16-0381. [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Balbo S., Hecht S.S., Upadhyaya P., Villalta P.W. Application of a high-resolution mass-spectrometry-based DNA adductomics approach for identification of DNA adducts in complex mixtures. Anal. Chem. 2014;86:1744–1752. doi: 10.1021/ac403565m. [DOI] [PMC free article] [PubMed] [Google Scholar]
10.U.S. Food and Drug Administration Prescribing information of Myleran (busulfan) 2-mg scored tablets. 2003. https://www.accessdata.fda.gov/drugsatfda_docs/label/2003/09386slr023_myleran_lbl.pdf
11.Ciurea S.O., Andersson B.S. Busulfan in hematopoietic stem cell transplantation. Biol. Blood Marrow Transpl. 2009;15:523–536. doi: 10.1016/j.bbmt.2008.12.489. [DOI] [PMC free article] [PubMed] [Google Scholar]
12.McCune J.S., Holmberg L.A. Busulfan in hematopoietic stem cell transplant setting. Expert Opin. Drug Metab. Toxicol. 2009;5:957–969. doi: 10.1517/17425250903107764. [DOI] [PubMed] [Google Scholar]
13.Clift R.A., Buckner C.D., Thomas E.D., Bensinger W.I., Bowden R., Bryant E., Deeg H.J., Doney K.C., Fisher L.D., Hansen J.A., et al. Marrow transplantation for chronic myeloid leukemia: a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood. 1994;84:2036–2043. [PubMed] [Google Scholar]
14.Clift R.A., Radich J., Appelbaum F.R., Martin P., Flowers M.E., Deeg H.J., Storb R., Thomas E.D. Long-term follow-up of a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide for patients receiving allogenic marrow transplants during chronic phase of chronic myeloid leukemia. Blood. 1999;94:3960–3962. [PubMed] [Google Scholar]
15.Vassal G., Deroussent A., Hartmann O., Challine D., Benhamou E., Valteau-Couanet D., Brugières L., Kalifa C., Gouyette A., Lemerle J. Dose-dependent neurotoxicity of high-dose busulfan in children: a clinical and pharmacological study. Cancer Res. 1990;50:6203–6207. [PubMed] [Google Scholar]
16.McDonald G.B., Sandmaier B.M., Mielcarek M., Sorror M., Pergam S.A., Cheng G.S., Hingorani S., Boeckh M., Flowers M.D., Lee S.J., et al. Survival, nonrelapse mortality, and relapse-related mortality after allogeneic hematopoietic cell transplantation: comparing 2003-2007 versus 2013-2017 cohorts. Ann. Intern. Med. 2020;172:229–239. doi: 10.7326/M19-2936. [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Eberly A.L., Anderson G.D., Bubalo J.S., McCune J.S. Optimal prevention of seizures induced by high-dose busulfan. Pharmacotherapy. 2008;28:1502–1510. doi: 10.1592/phco.28.12.1502. [DOI] [PubMed] [Google Scholar]
18.Slattery J.T., Sanders J.E., Buckner C.D., Schaffer R.L., Lambert K.W., Langer F.P., Anasetti C., Bensinger W.I., Fisher L.D., Appelbaum F.R., et al. Graft-rejection and toxicity following bone marrow transplantation in relation to busulfan pharmacokinetics. Bone Marrow Transpl. 1995;16:31–42. [PubMed] [Google Scholar]
19.Slattery J.T., Clift R.A., Buckner C.D., Radich J., Storer B., Bensinger W.I., Soll E., Anasetti C., Bowden R., Bryant E., et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood. 1997;89:3055–3060. [PubMed] [Google Scholar]
20.Palmer J., McCune J.S., Perales M.A., Marks D., Bubalo J., Mohty M., Wingard J.R., Paci A., Hassan M., Bredeson C., et al. Personalizing busulfan-based conditioning: considerations from the American society for blood and marrow transplantation practice guidelines committee. Biol. Blood Marrow Transpl. 2016;22:1915–1925. doi: 10.1016/j.bbmt.2016.07.013. [DOI] [PubMed] [Google Scholar]
21.Andersson B.S., Thall P.F., Madden T., Couriel D., Wang X., Tran H.T., Anderlini P., de Lima M., Gajewski J., Champlin R.E. Busulfan systemic exposure relative to regimen-related toxicity and acute graft-versus-host disease: defining a therapeutic window for i.v. BuCy2 in chronic myelogenous leukemia. Biol. Blood Marrow Transpl. 2002;8:477–485. doi: 10.1053/bbmt.2002.v8.pm12374452. [DOI] [PubMed] [Google Scholar]
22.Schuler U., Schroer S., Kühnle A., Blanz J., Mewes K., Kumbier I., Proksch B., Zeller K.P., Ehninger G. Busulfan pharmacokinetics in bone marrow transplant patients: is drug monitoring warranted? Bone Marrow Transpl. 1994;14:759–765. [PubMed] [Google Scholar]
23.Bartelink I.H., Lalmohamed A., van Reij E.M.L., Dvorak C.C., Savic R.M., Zwaveling J., Bredius R.G.M., Egberts A.C.G., Bierings M., Kletzel M., et al. Association of busulfan exposure with survival and toxicity after haemopoietic cell transplantation in children and young adults: a multicentre, retrospective cohort analysis. Lancet Haematol. 2016;3:e526–e536. doi: 10.1016/s2352-3026(16)30114-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
24.Myers A.L., Kawedia J.D., Champlin R.E., Kramer M.A., Nieto Y., Ghose R., Andersson B.S. Clarifying busulfan metabolism and drug interactions to support new therapeutic drug monitoring strategies: a comprehensive review. Expert Opin. Drug Metab. Toxicol. 2017;13:901–923. doi: 10.1080/17425255.2017.1360277. [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Iwamoto T., Hiraku Y., Oikawa S., Mizutani H., Kojima M., Kawanishi S. DNA intrastrand cross-link at the 5'-GA-3' sequence formed by busulfan and its role in the cytotoxic effect. Cancer Sci. 2004;95:454–458. doi: 10.1111/j.1349-7006.2004.tb03231.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
26.Guidolin V., Carlson E.S., Carrà A., Villalta P.W., Maertens L.A., Hecht S.S., Balbo S. Identification of new markers of alcohol-derived DNA damage in humans. Biomolecules. 2021;11:366. doi: 10.3390/biom11030366. [DOI] [PMC free article] [PubMed] [Google Scholar]
27.Stornetta A., Villalta P.W., Hecht S.S., Sturla S.J., Balbo S. Screening for DNA alkylation mono and cross-linked adducts with a comprehensive LC-MS3 adductomic approach. Anal. Chem. 2015;87:11706–11713. doi: 10.1021/acs.analchem.5b02759. [DOI] [PMC free article] [PubMed] [Google Scholar]
28.Carrà A., Guidolin V., Dator R.P., Upadhyaya P., Kassie F., Villalta P.W., Balbo S. Targeted high resolution LC/MS3 adductomics method for the characterization of endogenous DNA damage. Front. Chem. 2019;7:658. doi: 10.3389/fchem.2019.00658. [DOI] [PMC free article] [PubMed] [Google Scholar]
29.Alexander E.M., Kreitler D.F., Guidolin V., Hurben A.K., Drake E., Villalta P.W., Balbo S., Gulick A.M., Aldrich C.C. Biosynthesis, mechanism of action, and inhibition of the enterotoxin tilimycin produced by the opportunistic pathogen Klebsiella oxytoca. ACS Infect. Dis. 2020;6:1976–1997. doi: 10.1021/acsinfecdis.0c00326. [DOI] [PMC free article] [PubMed] [Google Scholar]
30.Hassan M., Ehrsson H. Urinary metabolites of busulfan in the rat. Drug Metab. Dispos. 1987;15:399–402. [PubMed] [Google Scholar]
31.Hassan M., Oberg G., Ehrsson H., Ehrnebo M., Wallin I., Smedmyr B., Tötterman T., Eksborg S., Simonsson B. Pharmacokinetic and metabolic studies of high-dose busulphan in adults. Eur. J. Clin. Pharmacol. 1989;36:525–530. doi: 10.1007/BF00558081. [DOI] [PubMed] [Google Scholar]
32.Bouligand J., Deroussent A., Simonnard N., Opolon P., Morizet J., Connault E., Daudigeos E., Re M., Paci A., Vassal G. Induction of glutathione synthesis explains pharmacodynamics of high-dose busulfan in mice and highlights putative mechanisms of drug interaction. Drug Metab. Dispos. 2007;35:306–314. doi: 10.1124/dmd.106.012880. [DOI] [PubMed] [Google Scholar]
33.Scian M., Atkins W.M. The busulfan metabolite EdAG irreversibly glutathionylates glutaredoxins. Arch. Biochem. Biophys. 2015;583:96–104. doi: 10.1016/j.abb.2015.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Scian M., Atkins W.M. Supporting data for characterization of the busulfan metabolite EdAG and the Glutaredoxins that it adducts. Data Brief. 2015;5:161–170. doi: 10.1016/j.dib.2015.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
35.Scian M., Guttman M., Bouldin S.D., Outten C.E., Atkins W.M. The myeloablative drug busulfan converts cysteine to dehydroalanine and lanthionine in redoxins. Biochemistry. 2016;55:4720–4730. doi: 10.1021/acs.biochem.6b00622. [DOI] [PMC free article] [PubMed] [Google Scholar]
36.Souliotis V.L., Dimopoulos M.A., Sfikakis P.P. Gene-specific formation and repair of DNA monoadducts and interstrand cross-links after therapeutic exposure to nitrogen mustards. Clin. Cancer Res. 2003;9:4465–4474. [PubMed] [Google Scholar]
37.Stornetta A., Villalta P.W., Gossner F., Wilson W.R., Balbo S., Sturla S.J. DNA adduct profiles predict in vitro cell viability after treatment with the experimental anticancer prodrug PR104A. Chem. Res. Toxicol. 2017;30:830–839. doi: 10.1021/acs.chemrestox.6b00412. [DOI] [PMC free article] [PubMed] [Google Scholar]
38.Gruppi F., Hejazi L., Christov P.P., Krishnamachari S., Turesky R.J., Rizzo C.J. Characterization of nitrogen mustard formamidopyrimidine adduct formation of bis(2-chloroethyl)ethylamine with calf thymus DNA and a human mammary cancer cell line. Chem. Res. Toxicol. 2015;28:1850–1860. doi: 10.1021/acs.chemrestox.5b00297. [DOI] [PMC free article] [PubMed] [Google Scholar]
39.Hemminki K. Binding of metabolites of cyclophosphamide to DNA in a rat liver microsomal system and in vivo in mice. Cancer Res. 1985;45:4237–4243. [PubMed] [Google Scholar]
40.Palom Y., Suresh Kumar G., Tang L.Q., Paz M.M., Musser S.M., Rockwell S., Tomasz M. Relative toxicities of DNA cross-links and monoadducts: new insights from studies of decarbamoyl mitomycin C and mitomycin C. Chem. Res. Toxicol. 2002;15:1398–1406. doi: 10.1021/tx020044g. [DOI] [PubMed] [Google Scholar]
41.Walters K., Stornetta A., Jacobs F., Villalta P.W., Razzoli M., Grant M., Zordoky B., Bartolomucci A., Borgatti A., Balbo S. Identification of new candidate biomarkers to support doxorubicin treatments in canine cancer patients. BMC Vet. Res. 2021;17:378. doi: 10.1186/s12917-021-03062-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
42.Moon S.Y., Lim M.K., Hong S., Jeon Y., Han M., Song S.H., Lim K.S., Yu K.S., Jang I.J., Lee J.W., et al. Quantification of human plasma-busulfan concentration by liquid chromatography-tandem mass spectrometry. Ann. Lab. Med. 2014;34:7–14. doi: 10.3343/alm.2014.34.1.7. [DOI] [PMC free article] [PubMed] [Google Scholar]
43.Li Y., Ma B., Cao Q., Balbo S., Zhao L., Upadhyaya P., Hecht S.S. Mass spectrometric quantitation of pyridyloxobutyl DNA phosphate adducts in rats chronically treated with N'-nitrosonornicotine. Chem. Res. Toxicol. 2019;32:773–783. doi: 10.1021/acs.chemrestox.9b00007. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Document S1. Figures S1–S10 and Table S1

mmc1.pdf^{(2.2MB, pdf)}

Document S2. Article plus supplemental information

mmc2.pdf^{(4.1MB, pdf)}

Data Availability Statement

The data that support the findings of this study are available from the authors on reasonable request; see author contributions for specific datasets.

[bib1] 1.Diethelm-Varela B., Ai Y., Liang D., Xue F. Nitrogen mustards as anticancer chemotherapies: historic perspective, current developments and future trends. Curr. Top. Med. Chem. 2019;19:691–712. doi: 10.2174/1568026619666190401100519. [DOI] [PubMed] [Google Scholar]

[bib2] 2.Goodman L.S., Wintrobe M.M. Nitrogen mustard therapy; use of methyl-bis (beta-chloroethyl) amine hydrochloride and tris (beta-chloroethyl) amine hydrochloride for Hodgkin's disease, lymphosarcoma, leukemia and certain allied and miscellaneous disorders. J. Am. Med. Assoc. 1946;132:126–132. doi: 10.1001/jama.1946.02870380008004. [DOI] [PubMed] [Google Scholar]

[bib3] 3.Malayappan B., Johnson L., Nie B., Panchal D., Matter B., Jacobson P., Tretyakova N. Quantitative high-performance liquid chromatography-electrospray ionization tandem mass spectrometry analysis of bis-N7-guanine DNA-DNA cross-links in white blood cells of cancer patients receiving cyclophosphamide therapy. Anal. Chem. 2010;82:3650–3658. doi: 10.1021/ac902923s. [DOI] [PubMed] [Google Scholar]

[bib4] 4.Johnson L.A., Malayappan B., Tretyakova N., Campbell C., MacMillan M.L., Wagner J.E., Jacobson P.A. Formation of cyclophosphamide specific DNA adducts in hematological diseases. Pediatr. Blood Cancer. 2012;58:708–714. doi: 10.1002/pbc.23254. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib5] 5.Stornetta A., Zimmermann M., Cimino G.D., Henderson P.T., Sturla S.J. DNA adducts from anticancer drugs as candidate predictive markers for precision medicine. Chem. Res. Toxicol. 2017;30:388–409. doi: 10.1021/acs.chemrestox.6b00380. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib6] 6.Wang S.S., Zimmermann M., Zhang H., Lin T.Y., Malfatti M., Haack K., Turteltaub K.W., Cimino G.D., de Vere White R., Pan C.X., Henderson P.T. A diagnostic microdosing approach to investigate platinum sensitivity in non-small cell lung cancer. Int. J. Cancer. 2017;141:604–613. doi: 10.1002/ijc.30747. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib7] 7.Wang L.E., Yin M., Dong Q., Stewart D.J., Merriman K.W., Amos C.I., Spitz M.R., Wei Q. DNA repair capacity in peripheral lymphocytes predicts survival of patients with non-small-cell lung cancer treated with first-line platinum-based chemotherapy. J. Clin. Oncol. 2011;29:4121–4128. doi: 10.1200/JCO.2010.34.3616. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib8] 8.Zimmermann M., Wang S.S., Zhang H., Lin T.Y., Malfatti M., Haack K., Ognibene T., Yang H., Airhart S., Turteltaub K.W., et al. Microdose-induced drug-DNA adducts as biomarkers of chemotherapy resistance in humans and mice. Mol. Cancer Ther. 2017;16:376–387. doi: 10.1158/1535-7163.MCT-16-0381. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib9] 9.Balbo S., Hecht S.S., Upadhyaya P., Villalta P.W. Application of a high-resolution mass-spectrometry-based DNA adductomics approach for identification of DNA adducts in complex mixtures. Anal. Chem. 2014;86:1744–1752. doi: 10.1021/ac403565m. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib10] 10.U.S. Food and Drug Administration Prescribing information of Myleran (busulfan) 2-mg scored tablets. 2003. https://www.accessdata.fda.gov/drugsatfda_docs/label/2003/09386slr023_myleran_lbl.pdf

[bib11] 11.Ciurea S.O., Andersson B.S. Busulfan in hematopoietic stem cell transplantation. Biol. Blood Marrow Transpl. 2009;15:523–536. doi: 10.1016/j.bbmt.2008.12.489. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib12] 12.McCune J.S., Holmberg L.A. Busulfan in hematopoietic stem cell transplant setting. Expert Opin. Drug Metab. Toxicol. 2009;5:957–969. doi: 10.1517/17425250903107764. [DOI] [PubMed] [Google Scholar]

[bib13] 13.Clift R.A., Buckner C.D., Thomas E.D., Bensinger W.I., Bowden R., Bryant E., Deeg H.J., Doney K.C., Fisher L.D., Hansen J.A., et al. Marrow transplantation for chronic myeloid leukemia: a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide. Blood. 1994;84:2036–2043. [PubMed] [Google Scholar]

[bib14] 14.Clift R.A., Radich J., Appelbaum F.R., Martin P., Flowers M.E., Deeg H.J., Storb R., Thomas E.D. Long-term follow-up of a randomized study comparing cyclophosphamide and total body irradiation with busulfan and cyclophosphamide for patients receiving allogenic marrow transplants during chronic phase of chronic myeloid leukemia. Blood. 1999;94:3960–3962. [PubMed] [Google Scholar]

[bib15] 15.Vassal G., Deroussent A., Hartmann O., Challine D., Benhamou E., Valteau-Couanet D., Brugières L., Kalifa C., Gouyette A., Lemerle J. Dose-dependent neurotoxicity of high-dose busulfan in children: a clinical and pharmacological study. Cancer Res. 1990;50:6203–6207. [PubMed] [Google Scholar]

[bib16] 16.McDonald G.B., Sandmaier B.M., Mielcarek M., Sorror M., Pergam S.A., Cheng G.S., Hingorani S., Boeckh M., Flowers M.D., Lee S.J., et al. Survival, nonrelapse mortality, and relapse-related mortality after allogeneic hematopoietic cell transplantation: comparing 2003-2007 versus 2013-2017 cohorts. Ann. Intern. Med. 2020;172:229–239. doi: 10.7326/M19-2936. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib17] 17.Eberly A.L., Anderson G.D., Bubalo J.S., McCune J.S. Optimal prevention of seizures induced by high-dose busulfan. Pharmacotherapy. 2008;28:1502–1510. doi: 10.1592/phco.28.12.1502. [DOI] [PubMed] [Google Scholar]

[bib18] 18.Slattery J.T., Sanders J.E., Buckner C.D., Schaffer R.L., Lambert K.W., Langer F.P., Anasetti C., Bensinger W.I., Fisher L.D., Appelbaum F.R., et al. Graft-rejection and toxicity following bone marrow transplantation in relation to busulfan pharmacokinetics. Bone Marrow Transpl. 1995;16:31–42. [PubMed] [Google Scholar]

[bib19] 19.Slattery J.T., Clift R.A., Buckner C.D., Radich J., Storer B., Bensinger W.I., Soll E., Anasetti C., Bowden R., Bryant E., et al. Marrow transplantation for chronic myeloid leukemia: the influence of plasma busulfan levels on the outcome of transplantation. Blood. 1997;89:3055–3060. [PubMed] [Google Scholar]

[bib20] 20.Palmer J., McCune J.S., Perales M.A., Marks D., Bubalo J., Mohty M., Wingard J.R., Paci A., Hassan M., Bredeson C., et al. Personalizing busulfan-based conditioning: considerations from the American society for blood and marrow transplantation practice guidelines committee. Biol. Blood Marrow Transpl. 2016;22:1915–1925. doi: 10.1016/j.bbmt.2016.07.013. [DOI] [PubMed] [Google Scholar]

[bib21] 21.Andersson B.S., Thall P.F., Madden T., Couriel D., Wang X., Tran H.T., Anderlini P., de Lima M., Gajewski J., Champlin R.E. Busulfan systemic exposure relative to regimen-related toxicity and acute graft-versus-host disease: defining a therapeutic window for i.v. BuCy2 in chronic myelogenous leukemia. Biol. Blood Marrow Transpl. 2002;8:477–485. doi: 10.1053/bbmt.2002.v8.pm12374452. [DOI] [PubMed] [Google Scholar]

[bib22] 22.Schuler U., Schroer S., Kühnle A., Blanz J., Mewes K., Kumbier I., Proksch B., Zeller K.P., Ehninger G. Busulfan pharmacokinetics in bone marrow transplant patients: is drug monitoring warranted? Bone Marrow Transpl. 1994;14:759–765. [PubMed] [Google Scholar]

[bib23] 23.Bartelink I.H., Lalmohamed A., van Reij E.M.L., Dvorak C.C., Savic R.M., Zwaveling J., Bredius R.G.M., Egberts A.C.G., Bierings M., Kletzel M., et al. Association of busulfan exposure with survival and toxicity after haemopoietic cell transplantation in children and young adults: a multicentre, retrospective cohort analysis. Lancet Haematol. 2016;3:e526–e536. doi: 10.1016/s2352-3026(16)30114-4. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib24] 24.Myers A.L., Kawedia J.D., Champlin R.E., Kramer M.A., Nieto Y., Ghose R., Andersson B.S. Clarifying busulfan metabolism and drug interactions to support new therapeutic drug monitoring strategies: a comprehensive review. Expert Opin. Drug Metab. Toxicol. 2017;13:901–923. doi: 10.1080/17425255.2017.1360277. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib25] 25.Iwamoto T., Hiraku Y., Oikawa S., Mizutani H., Kojima M., Kawanishi S. DNA intrastrand cross-link at the 5'-GA-3' sequence formed by busulfan and its role in the cytotoxic effect. Cancer Sci. 2004;95:454–458. doi: 10.1111/j.1349-7006.2004.tb03231.x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib26] 26.Guidolin V., Carlson E.S., Carrà A., Villalta P.W., Maertens L.A., Hecht S.S., Balbo S. Identification of new markers of alcohol-derived DNA damage in humans. Biomolecules. 2021;11:366. doi: 10.3390/biom11030366. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib27] 27.Stornetta A., Villalta P.W., Hecht S.S., Sturla S.J., Balbo S. Screening for DNA alkylation mono and cross-linked adducts with a comprehensive LC-MS3 adductomic approach. Anal. Chem. 2015;87:11706–11713. doi: 10.1021/acs.analchem.5b02759. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib28] 28.Carrà A., Guidolin V., Dator R.P., Upadhyaya P., Kassie F., Villalta P.W., Balbo S. Targeted high resolution LC/MS3 adductomics method for the characterization of endogenous DNA damage. Front. Chem. 2019;7:658. doi: 10.3389/fchem.2019.00658. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib29] 29.Alexander E.M., Kreitler D.F., Guidolin V., Hurben A.K., Drake E., Villalta P.W., Balbo S., Gulick A.M., Aldrich C.C. Biosynthesis, mechanism of action, and inhibition of the enterotoxin tilimycin produced by the opportunistic pathogen Klebsiella oxytoca. ACS Infect. Dis. 2020;6:1976–1997. doi: 10.1021/acsinfecdis.0c00326. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib30] 30.Hassan M., Ehrsson H. Urinary metabolites of busulfan in the rat. Drug Metab. Dispos. 1987;15:399–402. [PubMed] [Google Scholar]

[bib31] 31.Hassan M., Oberg G., Ehrsson H., Ehrnebo M., Wallin I., Smedmyr B., Tötterman T., Eksborg S., Simonsson B. Pharmacokinetic and metabolic studies of high-dose busulphan in adults. Eur. J. Clin. Pharmacol. 1989;36:525–530. doi: 10.1007/BF00558081. [DOI] [PubMed] [Google Scholar]

[bib32] 32.Bouligand J., Deroussent A., Simonnard N., Opolon P., Morizet J., Connault E., Daudigeos E., Re M., Paci A., Vassal G. Induction of glutathione synthesis explains pharmacodynamics of high-dose busulfan in mice and highlights putative mechanisms of drug interaction. Drug Metab. Dispos. 2007;35:306–314. doi: 10.1124/dmd.106.012880. [DOI] [PubMed] [Google Scholar]

[bib33] 33.Scian M., Atkins W.M. The busulfan metabolite EdAG irreversibly glutathionylates glutaredoxins. Arch. Biochem. Biophys. 2015;583:96–104. doi: 10.1016/j.abb.2015.08.005. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib34] 34.Scian M., Atkins W.M. Supporting data for characterization of the busulfan metabolite EdAG and the Glutaredoxins that it adducts. Data Brief. 2015;5:161–170. doi: 10.1016/j.dib.2015.09.002. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib35] 35.Scian M., Guttman M., Bouldin S.D., Outten C.E., Atkins W.M. The myeloablative drug busulfan converts cysteine to dehydroalanine and lanthionine in redoxins. Biochemistry. 2016;55:4720–4730. doi: 10.1021/acs.biochem.6b00622. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib36] 36.Souliotis V.L., Dimopoulos M.A., Sfikakis P.P. Gene-specific formation and repair of DNA monoadducts and interstrand cross-links after therapeutic exposure to nitrogen mustards. Clin. Cancer Res. 2003;9:4465–4474. [PubMed] [Google Scholar]

[bib37] 37.Stornetta A., Villalta P.W., Gossner F., Wilson W.R., Balbo S., Sturla S.J. DNA adduct profiles predict in vitro cell viability after treatment with the experimental anticancer prodrug PR104A. Chem. Res. Toxicol. 2017;30:830–839. doi: 10.1021/acs.chemrestox.6b00412. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib38] 38.Gruppi F., Hejazi L., Christov P.P., Krishnamachari S., Turesky R.J., Rizzo C.J. Characterization of nitrogen mustard formamidopyrimidine adduct formation of bis(2-chloroethyl)ethylamine with calf thymus DNA and a human mammary cancer cell line. Chem. Res. Toxicol. 2015;28:1850–1860. doi: 10.1021/acs.chemrestox.5b00297. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib39] 39.Hemminki K. Binding of metabolites of cyclophosphamide to DNA in a rat liver microsomal system and in vivo in mice. Cancer Res. 1985;45:4237–4243. [PubMed] [Google Scholar]

[bib40] 40.Palom Y., Suresh Kumar G., Tang L.Q., Paz M.M., Musser S.M., Rockwell S., Tomasz M. Relative toxicities of DNA cross-links and monoadducts: new insights from studies of decarbamoyl mitomycin C and mitomycin C. Chem. Res. Toxicol. 2002;15:1398–1406. doi: 10.1021/tx020044g. [DOI] [PubMed] [Google Scholar]

[bib41] 41.Walters K., Stornetta A., Jacobs F., Villalta P.W., Razzoli M., Grant M., Zordoky B., Bartolomucci A., Borgatti A., Balbo S. Identification of new candidate biomarkers to support doxorubicin treatments in canine cancer patients. BMC Vet. Res. 2021;17:378. doi: 10.1186/s12917-021-03062-x. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib42] 42.Moon S.Y., Lim M.K., Hong S., Jeon Y., Han M., Song S.H., Lim K.S., Yu K.S., Jang I.J., Lee J.W., et al. Quantification of human plasma-busulfan concentration by liquid chromatography-tandem mass spectrometry. Ann. Lab. Med. 2014;34:7–14. doi: 10.3343/alm.2014.34.1.7. [DOI] [PMC free article] [PubMed] [Google Scholar]

[bib43] 43.Li Y., Ma B., Cao Q., Balbo S., Zhao L., Upadhyaya P., Hecht S.S. Mass spectrometric quantitation of pyridyloxobutyl DNA phosphate adducts in rats chronically treated with N'-nitrosonornicotine. Chem. Res. Toxicol. 2019;32:773–783. doi: 10.1021/acs.chemrestox.9b00007. [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Characterization and quantitation of busulfan DNA adducts in the blood of patients receiving busulfan therapy

Valeria Guidolin

Yupeng Li

Foster C Jacobs

Margaret L MacMillan

Peter W Villalta

Stephen S Hecht

Silvia Balbo

Abstract

Graphical abstract

Introduction

Figure 1.

Results

Analysis of putative BU DNA adducts formed in vitro

Table 1.

Figure 2.

Confirmation of the putative BU DNA adducts using 15N-labeled DNA

Figure 3.

Targeted analysis of putative BU DNA adducts in the blood of patients

Table 2.

Chemical synthesis of BU-derived DNA adducts

Characterization of BU-derived DNA adducts

Figure 4.

Figure 5.

Quantification of N7G-Bu-N7G and N7G-Bu-OH in blood DNA of patients undergoing BU-based chemotherapy

Figure 6.

Discussion

Materials and methods

Chemicals and supplies

Synthesis of chemical standards

N'-(9-((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-6-oxo-6,9-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (2, N2-dmf-dGuo)

N'-(7-(4-bromobutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (3, N2-dmf-Gua-N7-butylbromide)

N',N''-(butane-1,4-diylbis(6-oxo-6,7-dihydro-1H-purine-7,2-diyl))bis(N,N-dimethylformimidamide) (4, bis(N2-dmf-Gua-N7)butyl, and N'-(7-(4-hydroxybutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (5, N2-dmf-Gua-N7-butanol)

7,7'-(Butane-1,4-diyl)bis(2-amino-1,7-dihydro-6H-purin-6-one) (6, N7G-Bu-N7G)

2-Amino-7-(4-hydroxybutyl)-1,7-dihydro-6H-purin-6-one (7, N7G-Bu-OH)

DNA isolation and purification

15N-DNA generation

Bacterial DNA isolation

Sample collection from patients and whole-blood DNA extraction

DNA exposure to BU in vitro

DNA hydrolysis and sample enrichment

DNA enzymatic hydrolysis

dGuo quantitation by HPLC

Sample enrichment and purification

Adductomic DDA-CNL/MS3 method and data analysis

LC conditions

MS

Data analysis

Targeted MS method for investigating and quantitating BU-DNA adducts in human blood

LC conditions

Mass spectrometry

BU-derived DNA adduct quantitation

Acknowledgments

Author contributions

Declaration of interests

Footnotes

Supplemental information

Data availability

References

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases

Confirmation of the putative BU DNA adducts using ¹⁵N-labeled DNA

N'-(9-((2R,4S,5R)-4-hydroxy-5-(hydroxymethyl)tetrahydrofuran-2-yl)-6-oxo-6,9-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (2, N²-dmf-dGuo)

N'-(7-(4-bromobutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (3, N²-dmf-Gua-N7-butylbromide)

N',N''-(butane-1,4-diylbis(6-oxo-6,7-dihydro-1H-purine-7,2-diyl))bis(N,N-dimethylformimidamide) (4, bis(N²-dmf-Gua-N7)butyl, and N'-(7-(4-hydroxybutyl)-6-oxo-6,7-dihydro-1H-purin-2-yl)-N,N-dimethylformimidamide (5, N²-dmf-Gua-N7-butanol)

¹⁵N-DNA generation

Adductomic DDA-CNL/MS³ method and data analysis