Possible Genetic Risks from Heat-Damaged DNA in Food (2024)

The consumption of foods prepared at high temperatureshas beenassociated with numerous health risks. To date, the chief identifiedsource of risk has been small molecules produced in trace levels bycooking and reacting with healthy DNA upon consumption. Here, we consideredwhether the DNA in food itself also presents a hazard. We hypothesizethat high-temperature cooking may cause significant damage to theDNA in food, and this damage might find its way into cellular DNAby metabolic salvage. We tested cooked and raw foods and found highlevels of hydrolytic and oxidative damage to all four DNA bases uponcooking. Exposing cultured cells to damaged 2′-deoxynucleosides(particularly pyrimidines) resulted in elevated DNA damage and repairresponses in the cells. Feeding a deaminated 2′-deoxynucleoside(2′-deoxyuridine), and DNA containing it, to mice resultedin substantial uptake into intestinal genomic DNA and promoted double-strandchromosomal breaks there. The results suggest the possibility of apreviously unrecognized pathway whereby high-temperature cooking maycontribute to genetic risks.

Food DNA Damage Hypothesis

Significantly, very littleresearch attention has been paid to the effect of elevated cookingtemperatures on the DNA in the food itself. DNA is one of three majorclasses of macromolecules in mammalian cells, accounting for 0.3%of cellular mass;⁶ this implies that theconsumption of a 500 g steak results in the ingestion of >1 g ofDNA(Table 1). Moreover,elevated temperatures have been shown to have adverse effects on DNAintegrity in DNA samples in vitro.^7,8 Thelack of studies on the effects of DNA damage in food may be due inpart to the perception that ingested DNA is not likely to be takenup in cells to influence cellular pathways.⁹ However, it has long been recognized that DNA, when fed orally tomammals, is rapidly fragmented and hydrolyzed, ultimately to 2′-deoxymononucleotides(chiefly, 5′-monophosphates) by nuclease enzymes present inpancreatic and intestinal juices.^9,10

In addition, 2′-deoxynucleoside 5′-monophosphatesare dephosphorylated by 5′-nucleotidase (intestinal phosphatase)activities in the cell membrane,¹⁰ andthe resulting free nucleosides (at least the canonical cases) canbe taken up into the intracellular environment and participate innucleotide salvage pathways (Supporting InformationFigure S1).⁹ Interestingly, althoughthe cellular nucleotide salvage pathway has been well studied withregard to canonical nucleosides/nucleotides, very little is knownabout the capability of damaged 2′-deoxynucleosides to be takenup into cells and incorporated into DNA there.¹¹ However, if damaged 2′-deoxynucleosides were indeedtaken up in salvage pathways, then this might present a significantrisk by the direct placement of damage in host DNA.

Taken together,these issues combine to present a potential mechanismwhereby the ingestion of damaged DNA from cooked food might resultin the incorporation of plant- or animal-derived damaged nucleosidesinto human DNA, resulting in genetic lesions and possible health risks.As a result, it could potentially be of significant health interestto determine to what degree high-temperature cooking can result indamage to the DNA in food sources and if damaged DNA can be digestedinto damaged nucleosides and indeed could have the capacity to enterhuman nucleotide salvage pathways and be incorporated into cellularDNA. We are aware of no previous studies of these issues.

Earlystudies of DNA stability in vitro have shownthat elevated temperature (milder than that employed in many cookingprocedures, Table S1) can accelerate thedeamination of 2′-deoxycytidine (dC) in DNA, resulting in 2′-deoxyuridine(dU),⁷ and also promotes the oxidationof guanine, resulting in the formation of 8-oxo-2′-deoxyguanosine(8-oxo-dG) along with other modified deoxynucleosides.⁸ 2′-Deoxyuridine, if incorporated into DNA by polymeraseenzymes, is a targeted substrate for base excision repair (BER),¹² and high levels of dU in cellular DNA can resultin elevated numbers of single-strand nicks and, if proximally localized,double-strand breaks (DSB), leading to genotoxicity and genomic rearrangements.¹³ Many damaged 2′-deoxynucleosides suchas 8-oxo-dG in DNA are highly mutagenic when incorporated and alsocan be genotoxic both in mitochondrial and nuclear DNA when subjectto DNA repair.¹⁴ Indeed, because damagednucleotides (when generated directly in cells) are potentially harmful,nucleotide pool sanitation enzymes exist to prevent their misincorporationinto DNA via inactivation of their 5′-triphosphate derivatives.¹⁵

We emphasize that this overall hypothesiscannot be proven in suchan initial study. Indeed, studies of small-molecule agents such asPAH and HCA in cooked foods have proceeded over decades, and risksto humans are seen only in large population studies. Thus, our goalis to test the individual parts of the food DNA hypothesis, whichmay lead to insights into its feasibility. To examine these hypothesizedissues, we addressed three chief questions regarding the potentialconnection of the cooking of food and DNA damage in human DNA: First,to what extent does cooking cause damage to DNA in food? Second, doescellular exposure to damaged 2′-deoxynucleosides evoke DNAdamage repair responses or chromosomal damage? Third, to what degreeare damaged DNAs digested and salvaged by cells and incorporated intocellular DNA?

Cooking Results in High Levels of Damage to DNA in Food

We tested the in vitro thermostability of genomicDNA (gDNA) extracted from HeLa cells, focusing on the deaminationof cytosine, the most frequent form of heat-induced DNA damage in vitro.⁷ The extracted gDNAwas subjected to extended heating (95 °C) to accelerate the deaminationof cytosine to uracil in DNA (Figure ​Figure22a), and then the levels of uracil were measured withuracil-DNA glycosylase (UDG) and a fluorescence probe (UBER)¹⁶ specific to apyrimidinic/apurinic (AP) sitesin DNA (Figure ​Figure22b).The results show that heating DNA at this elevated temperature markedlyincreased the level of uracil in DNA over time, as a result of theaccelerated deamination reaction.

Open in a separate window

Figure 2

Measurements of specific forms of damagein DNA from food afterheating and cooking reveal elevated levels of damage. (a) Illustrationof deamination of cytosine affording uracil in DNA. (b) Uracil quantificationassay in gDNA extracted from HeLa cells, employing UDG and a fluorescentprobe for AP sites. (c) Procedure of DNA damage quantification withGC–MS/MS and LC–MS/MS in DNA extracted from food samples.(d–g) Levels of 10 types of DNA damage quantified with GC–MS/MSand LC–MS/MS in DNA extracted from raw (−) and cooked(B = boiled, R = roasted) food samples. n.d. = not determined. Cookedfoods were boiled (100 °C, 20 min) or roasted (220 °C, 15min) before DNA extraction. Uncertainties are standard deviations.

Cooking processes commonly involve temperaturesmuch higher than95 °C for minutes to hours (Supporting Information, Table S1), which suggests the possibility of significant DNA damagein food. Our initial observation of the elevated deamination of cytosinein extracted gDNA after heating prompted us to test the stabilityof DNA in food during cooking processes, focusing on multiple aspects:(i) To what extent is DNA in food damaged upon cooking and (ii) howdoes the type of food source and method of cooking affect damage?Ground beef (80% lean), ground pork (80% lean), and sliced potatoeswere cooked via boiling for 20 min or roasting for 15 min in an oven(220 °C), and then DNA was isolated from the heat-processed foodsas well as from uncooked controls (SupportingInformation Figure S2). We employed gas chromatography/tandemmass spectrometry (GC–MS/MS) and liquid chromatography/tandemmass spectrometry (LC–MS/MS) to identify and quantify distinctchemical forms of DNA base damage in the raw and cooked foods. Structuresof the DNA lesions measured are shown in SupportingInformation Figure S3 and in Figure ​Figure22d–g.

The analysis showed thatthe levels of all 10 DNA lesions testedwere significantly increased in the DNA extracted from heat-processedfoods compared to those in the raw foods (Figure ​Figure22d–g,). FapyAde, FapyGua, cis-ThyGly, and trans-ThyGly could not be detectedin DNA samples extracted from potatoes. For the meat sources, thehigher temperature of cooking (roasting) generated greater amountsof DNA damage than the lower-temperature cooking procedure (boiling).In absolute terms, the two most frequent forms of damage were dU (10-foldincrease after roasting) and 8-oxo-dG (3.5-fold increase after roasting).Relative to control levels (Supporting InformationFigure S3), dU and 8,5′-cyclopurine-2′-deoxynucleosides(8-fold increase in R-cdA after roasting) were increasedby the greatest factor. dU was found at levels of ∼300 basesper million nucleotides in meats after mild roasting (15 min) (Figure ​Figure22d). Given that heat-induceddeamination producing dU in isolated DNA continues to proceed overextended times (Figure ​Figure22b),¹⁸ hours of roasting or smoking couldpotentially result in higher levels of damage, although this was nottested here. For dU in briefly roasted beef, the amounts found herecorrespond to milligram quantities in a serving of cooked meat, asmuch as 1000 times greater than concentrations of HCA or PAH moleculesin cooked meats.¹⁹ The fact that the levelsof dU and 8-oxo-dG increased strongly and prominently implies thatboth the deamination and oxidation of DNA were strongly acceleratedduring the cooking of food, which was exposed both to heat and ambientoxygen. Moreover, many other DNA lesions were also increased substantiallyin the foods. For example, 8,5′-cyclopurine-2′-deoxynucleosideswere increased several-fold during roasting; these lesions are mutagenicand are documented to act as polymerase substrates in triphosphateform, although it is not yet known if phosphorylation occurs in cells.²⁰ Interestingly, we found that DNA damage aftercooking was considerably lower in potatoes than in pork and beef,suggesting that other components of plant tissues may confer substantialprotection.

Evidence That Damaged 2′-Deoxynucleosides Are Salvagedby Cells in Culture and Evoke DNA Damage and Repair Responses

Following up on our findings that cooking markedly damages DNA infood, we next asked whether damaged DNA components pose risks to cellsby acting as substrates for nucleotide salvage. Canonical DNA in foodis ultimately digested into nucleosides by the gastrointestinal digestionsystem and then absorbed in the small intestine and transported intocells and circulation.^9,10 Enzymes that are responsiblefor the nucleotide salvage pathway are known to exhibit imperfectselectivity, enabling the DNA uptake of modified nucleosides suchas 5-bromo-2′-deoxyuridine (BrdU) and 5-ethynyl-2′-deoxyuridine(EdU) which are employed as markers of cellular DNA synthesis.²¹ We hypothesized that cooking-damaged DNA, digestedinto damaged 2′-deoxynucleosides upon consumption, might alsobe taken up into cellular DNA in a similar fashion (Figure ​Figure33a). The challenge in measuringthe levels of damaged 2′-deoxynuclosides, if any, incorporatedinto cellular DNA is that the presence of such a lesion in genomicor mitochondrial DNA will be difficult to quantify directly, as itis being actively removed by repair pathways before it can be measured.

See Also

Vegetarian hot dogs found to contain meat, human DNA in disturbing sausage study

Open in a separate window

Figure 3

CellularDNA damage responses to incubation with damaged 2′-deoxynucleosides.(a) Illustration of the pathway of damaged DNA in food to be incorporatedinto cellular DNA. (b) Flow cytometry results showing the relativefluorescence intensity of UBER in cells incubated with 200 μmol/Lof damaged 2′-deoxynucleosides for 24 h, reflecting the BERactivity of mitochondrial DNA. (c) Fluorescence intensity of UBERmeasured with flow cytometry in HeLa cells incubated with 200 μmol/LdU in the presence of a varied concentration of TAS-114 for 2 days.(d) Immuno-fluorescence intensity of γ-H2AX (a biomarker ofDSB) in cells incubated with 200 μmol/L of damaged 2′-deoxynucleosidesfor 24 h. (e) Immunofluorescence images of γ-H2AX in cells incubatedwith dU and/or TAS-114, showing evidence of elevated DSB in the cells.(f) Cytotoxicity of damaged nucleosides in CHO cells measured by acolony formation assay (N = 4, *p ≤ 0.05, **p ≤ 0.01, ***p < 0.001 by Dunnett’s multiple comparisons test). (g) Chromosomalaberrations are elevated in CHO cells after incubation with 200 μmol/Ldamaged nucleosides for 24 h. *p ≤ 0.05. (h)Representative images of a chromatid break (gap), resulting from double-strandedDNA damage after exposure to 200 μmol/L dU (additional imagesin Supporting Information, Figure S7).(i) Evidence for the mutagenicity of damaged nucleosides after exposureto 200 μmol/L damaged nucleosides for 24 h, as measured by theHPRT mutation assay in CHO cells. N = 8, *p ≤ 0.05, **p ≤ 0.01, and***p < 0.001 by Dunnett’s multiple comparisonstest; uncertainties are standard deviations.

To bypass this issue, we initially measured theBER activity ofcellular DNA evoked by the addition of damaged 2′-deoxynucleosides.As the forms of damaged 2′-deoxynucleosides studied here (exceptfor the 8,5′-cyclopurine-2′-deoxynucleosides which arerepaired by the nucleotide excision repair pathway) are known to besubstrates for BER, the appearance of elevated BER activity impliesthe direct incorporation of damaged 2′-deoxynucleosides inthe DNA.²⁰

We employed a fluorescentprobe specific to BER activity in cells(UBER) to gain evidence of cellular salvage and triphosphorylation,which are necessary for the incorporation of damaged 2′-deoxynucleosidesinto DNA. UBER binds covalently to AP sites in mitochondrial DNA (mtDNA)in intact cells and has been utilized for measuring mitochondrialBER responses to reactive oxygen species (ROS).²² While mitochondrial DNA lesions do not pose the directcancer risks that those in genomic DNA do, the incorporation of damagednucleosides into mtDNA would provide evidence for successful intracellularsalvage and polymerase incorporation. The experiments included 10different damaged 2′-deoxynucleosides (structures are shownin Supporting Information, Figure S4) and4 cell lines (HeLa, MCF-7, HEK293, and SW620). We found that mitochondrialBER activity in cells increased in the presence of several of thedamaged 2′-deoxynucleosides tested (Figure ​Figure33b), apparently as a result of defensive responsesto increases in lesions in mtDNA.²³ Tofurther investigate the relationship between the enhanced DNA repairactivity and salvage pathways that enable the incorporation of damagednucleoside into DNA, we tested the effect of a chemical inhibitorof a nucleotide sanitization enzyme for one of the damaged components(dU).

TAS-114 is an inhibitor of dUTPase, which hydrolyzes dUTPintodUMP to prevent the misincorporation of dU into DNA.²⁴ Inhibitor treatment in HeLa cells resulted in the furtherenhancement of BER signals in response to the incubation with dU inthe cell culture medium (Figure ​Figure33c). This adds support to the notion that dU from externalsources can be taken up via salvage and is iteratively phosphorylatedto form the triphosphate analogue, enabling its incorporation intocellular DNA. Prior studies of dUTPase have suggested that it existsto address the hydrolysis of cytidine nucleotides that occurs directlyin cells,²⁵ while the new findings suggestthat it can also prevent damage imported from external sources. Thus,the BER data for mitochondrial DNA support the notion of cellularsalvage and uptake of damaged nucleosides, but further data were neededto assess any effects in chromosomal DNA.

A nucleotide gap ingDNA generated during the BER repair processis filled in by DNA polymerase β, and the resulting nick issealed by ligase IIIα.²⁶ However,high levels of lesions in DNA can accumulate, and multiple base excisionsin clustered DNA lesions can result in proximal nucleotide gaps andnicks in both strands. This results in DSB, a serious form of DNAdamage that can cause both chromosomal rearrangements and indel mutations.^26,27 Thus, we assessed the levels of nuclear DSB after the incubationof damaged nucleosides with cells, employing phosphorylated histoneH2AX (γ-H2AX) as a biomarker of cellular responses to DSB.²⁸ The immunofluorescence assay results showedthat levels of γ-H2AX were increased significantly in HeLa cellsafter incubation with 200 μmol/L of damaged nucleoside dU, 5-OH-dU,5-OH-dC, DH-dT, or dTg for 24 h (Figure ​Figure33d). This result supports the hypothesis thatdamaged nucleosides were taken up into gDNA and subjected to highlevels of BER there, ultimately resulting in DSB. To further confirmthat DSB resulted from the metabolic activation of damaged nucleosides,the nucleotide sanitization pool was inhibited with TAS-114 in thepresence of dU. Upregulated misincorporation of dU with TAS-114 exhibitedmuch higher responses to DSB (Figure ​Figure33e), further supporting the involvement of the salvagepathway leading to BER and DSB when this form of damage is exposedto cells.

Further Assessment of Genetic Damage after Exposure to HighLevels of Damaged 2′-Deoxynucleosides

Given the abovedata documenting signals of mitochondrial base excision repair andchromosomal double-strand breaks in the DNA of cells incubated withdamaged nucleosides, particularly for certain pyrimidines (Figure ​Figure33b,d), we tested thecytotoxicity of exposing damaged nucleosides to Chinese hamster ovary(CHO) cells using the colony formation assay (Figure ​Figure33f). Incubating the cells with 30–200μM damaged pyrimidines dU, 5-OH-dU, and 5-OHdC for 8 days revealedsignificant cytotoxicity, while exposure to 8-oxo-dG did not. We performedfurther experiments to explicitly measure specific types of damagethat may occur in chromosomal DNA upon exposure to high levels ofdamaged nucleosides. CHO cells were incubated for 24 h with 200 μmol/LdU, 8-oxo-dG, 5-OH-dU, and 5-OH-dC and then analyzed for chromosomalaberrations (after arresting the cells in the metaphase) comparedto controls without added nucleoside. The data show an average ∼3-foldincrease in chromosomal aberrations in the presence of the damagedpyrimidines dU, 5-OH-dU, and 5-OH-dC, including chromatid gaps, chromatidexchanges, and chromosomal rearrangements, while 8-oxo-dG showed littleor no significant increase (Figure ​Figure33g,h and Supporting InformationFigure S6). For chromosomal aberrations excluding gaps, thepyrimidines induced a yet larger 4-fold increase (Supporting Information Figure S6). We also evaluated possiblemutagenic effects of the damaged nucleosides using a hypoxanthinephosphoribosyl transferase (HPRT) mutagenesis assay,²⁹ which is commonly used to measure mutagenicity in mammaliancells, and two of the damaged nucleosides (dU and 8-oxo-dG) were foundto induce statistically elevated levels (1.8- and 2.0-fold) of mutations(p = 0.0145 and 0.0062, Figure ​Figure33i), while 5-hydroxypyrimidines also showedaverage increases (∼1.7-fold) in mutagenicity but did not reach p > 0.05 (p = 0.085).

Consumption of a Damaged 2′-Deoxynucleoside Contributesto DNA Damage in the Small Intestines of Rodents

Given ourobservations that the cellular uptake of damaged 2′-deoxynucleosidescan induce mitochondrial and genomic DNA damage in the cell culture,we pursued an animal model of this pathway to test whether damagednucleosides that are orally consumed survive the digestive systemand find their way into the DNA of tissues. As with animal studiesof mutagenic small-molecule food species such as HCA and PAH, we employedhigh concentrations to observe maximal responses in a short span.We note that the concentration of damaged DNA used in these feedingexperiments is in the same range of those used in prior PAH metabolitestudies, while the amount of damaged DNA in food is calculated tobe 3 to 4 order of magnitude higher than the metabolites.³⁰ 2′-Deoxyuridine, the most abundant formof DNA damage caused by the cooking processes, was fed to mice (2mg dU in 200 μL of PBS buffer daily) for a week through oralgavage (Figure ​Figure44a).After oral administration of dU, intestinal tissues (the site of absorptionof canonical nucleosides) were examined for levels of damage in genomicDNA. From the tissue hom*ogenates, gDNA was extracted and the levelsof dU and 8-oxo-dG as a control were quantified with LC–MS/MS.

Open in a separate window

Figure 4

Adversegenetic effects of feeding high levels of a damaged nucleosideto mice. (a) Schematic illustration of oral feeding of dU to miceand analysis of intestinal tissue. LC–MS/MS quantificationresults of (b) dU and (c) 8-oxo-dG in gDNA extracted from the intestinesof control (−) and dU-fed mice (+), showing 2.5-fold to 15-foldincreases in levels of dU in the genomic DNA from these tissues. (d)Immunostaining of γ-H2AX in villi in the small intestine, showingenhanced DNA double-strand break (DSB) signals in response to dU feeding.Also shown are images of crypts in the large intestine. Tissues werecostained with Hoechst 33343 (5 μg/mL) to highlight nuclearDNA. (e) Quantified intensities of γ-H2AX from red channelsin panel d. Uncertainties are standard deviations (****p ≤ 0.0001) by the unpaired t test.

The results showed that dU was present at significantlyhigherlevels in gDNA from the small intestines of dU-fed mice compared tothat in control mice (Figure ​Figure44b). Increases were substantial, with an increase of up to2000 dU per million gDNA bases in the duodenum and jejunum, correspondingto 15-fold and 3.5-fold increases, respectively. Note that these elevatedlevels were observed in the presence of presumably intact DNA repairpathways in the mice and thus are likely lower than actual initialuptake. DNA incorporation levels of dU in the ileum were significant(2.5-fold increase), albeit smaller than in the earlier digestivetract and nonexistent in colon tissue, consistent with prior studiesshowing that canonical 2′-deoxynucleosides are primarily absorbedearlier in the digestive tract.³¹ It seemspossible that the absence of villi in the colon resulted in a negligibleincorporation of dU. Given that colorectal cancer is more frequentthan small intestinal cancer in the clinic, further studies are neededregarding this localization. For control experiments, the level of8-oxo-dG (not fed in the experiment) was measured in mouse intestinalDNA, showing no significant difference between the two, confirmingthat dU feeding caused no oxidation of dG (Figure ​Figure44c). The results confirm that the increasedlevel of dU in gDNA of dU-fed mice resulted from direct DNA incorporationof the damaged component rather than by indirectly inducing ROS (whichmay also increase the deamination of dC).³²

We further employed the γ-H2AX immunostaining assayto measureDSB in mouse intestinal tissues after a week of oral administrationof dU. Microscopic images of the stained intestinal tissue showedthat the level of γ-H2AX was significantly higher in epithelialcells of villi in small intestines from dU-fed mice than that of controlmice (Figure ​Figure44d,e).In contrast, mice fed with 2 mg of dC, the canonical 2′-deoxynucleosideprecursor of dU, daily for a week showed no observable enhancementin DSB levels, implying that the imbalance of the nucleotide poolis not a chief cause of these signals (SupportingInformation, Figure S8). Taken together, our results suggestthat dU in the diet may be taken up in enterocytes of villi of thesmall intestine after consumption and then enters the intracellularsalvage pathway followed by incorporation into cellular DNA.

At least at the high concentrations tested here, this results inelevated DNA repair responses leading to increased DSB. The resultsdocument that an orally ingested damaged 2′-deoxynucleoside,generated at high levels in food DNA during cooking, potentially survivesthe digestive system and find its way into cellular DNA, leading toa serious form of DNA damage in intestinal tissues.

Damaged DNA Can Be Digested and Incorporated into Cellular DNAupon Consumption

The above experiments evaluated the effectsof feeding a deaminated nucleoside monomer to rodents; however, ourhypothesis requires that the ingestion of DNA or DNA fragments containingsuch damage is followed by processing by nuclease and phosphataseactivities in the stomach, gut, and cells to the nucleotide/nucleosideform. Although this has been established for canonical DNA, it hasnot yet been tested with damaged DNA to our knowledge. To test this,we synthesized an oligodeoxynucleotide (5′-d(UUUUC)-3′)containing four dU residues, and in vitro digestionof the oligodeoxynucleotide by a commercial digestive enzyme mix wasanalyzed with HPLC (Figure ​Figure55a). We found that the oligodeoxynucleotide was digested completelyinto free 2′-deoxynucleotides (dU and dC). As a second testwith native digestive enzymes, lysates were prepared from the hom*ogenizedstomach and small intestines of mice, including gastric and intestinaljuices. HPLC analysis clearly showed that the damaged oligodeoxynucleotidewas digested, releasing the corresponding 2′-deoxynucleosidesin the presence of both gastric and intestinal lysates. Thus, thedamaged DNA base (even with consecutive substitution) does not preventthe digestion of DNA containing it.

Open in a separate window

Figure 5

(a) HPLC analysis of the in vitro digestion of10 μg of a damaged oligodeoxynucleotide (5′-d(UUUUC)-3′)with a digestive enzyme mix, gastric lysate, or gastrointestinal lysateat 37 °C for 24 h. (b) Levels of dU and (c) control 8-oxo-dGin gDNA extracted from the intestines of control (−) and damagedDNA-fed mice (+), showing >10-fold increases in levels of dU inthegenomic DNA from later small intestinal tissue. (d) Immunostainingof γ-H2AX in villi of the small intestine (ileum), showing enhancedDNA double-strand break (DSB) signals (in red) in response to damagedoligonucleotide feeding for 1 week. (e) Quantification results ofred fluorescence (level of γ-H2AX) measured from epithelialcells in panel d. ****p ≤ 0.0001 by the unpaired t test. Tissues were co-stained with blue Hoechst 33343(5 μg/mL) to highlight nuclear DNA. Uncertainties are standarddeviations.

Finally, we tested whether directly feeding DNAcontaining deaminatedbases would lead to the observable incorporation of damage into intestinaltissue. Mice were fed the above synthetic oligodeoxynucleotide (2mg in 200 μL of PBS daily) for 7 days, and the DNA extractedfrom intestinal tissue was then analyzed for damage content. The resultsshowed that the level of dU in the extracted mouse gDNA was significantlyincreased (>10-fold) in the later part of the small intestine uponthe consumption of the damaged oligodeoxynucleotide, while the levelof 8-oxo-dG as a control remained unchanged (Figure ​Figure55b,c). Considering that the digestion of thedamaged oligodeoxynucleotide requires digestive enzymes in the smallintestine (Figure ​Figure55a), lesser absorption/incorporation into the early part of the smallintestine may plausibly reflect the requirement for complete digestionduring the transit of the intestine. Also, consistent with the aboveexperiments, small intestine tissues showed increased level of DSBafter the feeding of the damaged DNA (Figure ​Figure55d,e).

• Jaegerstad M.; Skog K.Genotoxicity of Heat-Processed Foods. Mutat.Res.2005, 574, 156–172. 10.1016/j.mrfmmm.2005.01.030. [PubMed] [CrossRef] [Google Scholar]
• Cross A. J.; Sinha R.Meat-Related Mutagens/Carcinogens in the Etiology of Colorectal Cancer. Environ. Mol. Mutagen.2004, 44, 44–55. 10.1002/em.20030. [PubMed] [CrossRef] [Google Scholar]
  Anderson K.E.; Sinha R.; Kulldorff M.; Gross M.; Lang N. P.; Barber C.; Harnack L.; DiMagno E.; Bliss R.; Kadlubar F. F.Meat Intakeand Cooking Techniques: Associations with Pancreatic Cancer. Mutat. Res.2002, 506, 225–231. 10.1016/S0027-5107(02)00169-0. [PubMed] [CrossRef] [Google Scholar]
  Liu G.; Zong G.; Wu K.; Hu Y.; Li Y.; Willett W. C.; Eisenberg D. M.; Hu F. B.; Sun Q.Meat Cooking Methods and Risk ofType 2 Diabetes: Results from Three Prospective Cohort Studies. Diabetes Care2018, 41, 1049–1060. 10.2337/dc17-1992. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Ramirez-Anaya J. d. P.; Samaniego-Sanchez C.; Castaneda-Saucedo M. C.; Villalon-Mir M.; de la Serrana H. L.-G.Phenols and the Antioxidant Capacityof Mediterranean Vegetables Prepared with Extra Virgin Olive Oil UsingDifferent Domestic Cooking Techniques. Foodchemistry2015, 188, 430–438. 10.1016/j.foodchem.2015.04.124. [PubMed] [CrossRef] [Google Scholar]
• Shaughnessy D. T.; Gangarosa L. M.; Schliebe B.; Umbach D. M.; Xu Z.; MacIntosh B.; Knize M. G.; Matthews P. P.; Swank A. E.; Sandler R. S.Inhibition of Fried Meat-Induced Colorectal DNA Damageand Altered Systemic Genotoxicity in Humans by Crucifera, Chlorophyllin,and Yogurt. PloS one2011, 6, e18707. 10.1371/journal.pone.0018707. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Colvin M. E.; Hatch F. T.; Felton J. S.Chemical and BiologicalFactors AffectingMutagen Potency. Mutat. Res.1998, 400, 479–492. 10.1016/S0027-5107(98)00073-6. [PubMed] [CrossRef] [Google Scholar]
• Wu J.; Xiao J.; Zhang Z.; Wang X.; Hu S.; Yu J.Ribogenomics: the Scienceand Knowledge of RNA. Genomics, proteomics &bioinformatics2014, 12, 57–63. 10.1016/j.gpb.2014.04.002. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Lindahl T.; Nyberg B.Heat-Induced Deamination of CytosineResidues in DeoxyribonucleicAcid. Biochemistry1974, 13, 3405–3410. 10.1021/bi00713a035. [PubMed] [CrossRef] [Google Scholar]
• Bruskov V. I.; Malakhova L. V.; Masalimov Z. K.; Chernikov A. V.Heat-InducedFormation of Reactive Oxygen Species and 8-Oxoguanine, a Biomarkerof Damage to DNA. Nucleic Acids Res.2002, 30, 1354–1363. 10.1093/nar/30.6.1354. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Jonas D.; Elmadfa I.; Engel K.-H.; Heller K.; Kozianowski G.; König A.; Müller D.; Narbonne J.; Wackernagel W.; Kleiner J.Safety Considerations of DNA in Food. Ann. Nutr. Metab.2001, 45, 235–254. 10.1159/000046734. [PubMed] [CrossRef] [Google Scholar]
• Carver J. D.; Walker W. A.The Role of Nucleotides in Human Nutrition. J. Nutr. Biochem1995, 6, 58–72. 10.1016/0955-2863(94)00019-I. [CrossRef] [Google Scholar]
• Henderson P. T.; Evans M. D.; Cooke M. S.Salvage of OxidizedGuanine Derivativesin the (2′-deoxy) Ribonucleotide Pool as Source of Mutationsin DNA. Mutat. Res. Genet. Toxicol. Environ.Mutagen.2010, 703, 11–17. 10.1016/j.mrgentox.2010.08.021. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Kavli B.; Slupphaug G.; Krokan H.. Genomic Uracilin Biology, Immunity and Cancer. DNA Damage,DNA Repair and Disease; Royal Society ofChemistry, 2020; Vol. 1, pp 220–248. [Google Scholar]
• Krokan H. E.; Drabløs F.; Slupphaug G.Uracil in DNA–Occurrence,Consequences and Repair. Oncogene2002, 21, 8935–8948. 10.1038/sj.onc.1205996. [PubMed] [CrossRef] [Google Scholar]
• Mundt J. M.; Hah S. S.; Sumbad R. A.; Schramm V.; Henderson P. T.Incorporationof Extracellular 8-OxodG into DNA and RNA Requires Purine NucleosidePhosphorylase in MCF-7 Cells. Nucleic AcidsRes.2007, 36, 228–236. 10.1093/nar/gkm1032. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Hizi A.; Herzig E.dUTPase: the Frequently OverlookedEnzyme Encoded by Many Retroviruses. Retrovirology2015, 12, 70. 10.1186/s12977-015-0198-9. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
  Gad H.; Koolmeister T.; Jemth A.-S.; Eshtad S.; Jacques S. A.; Ström C. E.; Svensson L. M.; Schultz N.; Lundbäck T.; Einarsdottir B. O.MTH1 Inhibition Eradicates Cancer by Preventing Sanitationof the dNTP Pool. Nature2014, 508, 215–221. 10.1038/nature13181. [PubMed] [CrossRef] [Google Scholar]
• Wilson D. L.; Kool E. T.Ultrafast OximeFormation Enables Efficient FluorescenceLight-Up Measurement of DNA Base Excision. J.Am. Chem. Soc.2019, 141, 19379–19388. 10.1021/jacs.9b09812. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Herbel W.; Montag A.Nucleo-Compounds in Protein-RichFood. Unters. Forsch.1987, 185, 119–122. 10.1007/BF01850090. [PubMed] [CrossRef] [Google Scholar]
  Lassek E.; Montag A.Nucleic Acid Components in Carbohydrate-Rich Food. Unters. Forsch.1990, 190, 17–21. 10.1007/BF01188257. [PubMed] [CrossRef] [Google Scholar]
• Lewis C. A.; Crayle J.; Zhou S.; Swanstrom R.; Wolfenden R.Cytosine Deamination and the Precipitous Decline ofSpontaneous Mutation during Earth’s History. Proc. Natl. Acad. Sci. U. S. A.2016, 113, 8194–8199. 10.1073/pnas.1607580113. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Sahin S.; Ulusoy H. I.; Alemdar S.; Erdogan S.; Agaoglu S.The Presence of Polycyclic AromaticHydrocarbons (PAHs)in Grilled Beef, Chicken and Fish by Considering Dietary Exposureand Risk Assessment. Food Sci. Anim. Resour.2020, 40, 675–688. 10.5851/kosfa.2020.e43. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
  Sinha R.; Rothman N.; Salmon C.; Knize M.; Brown E.; Swanson C.; Rhodes D.; Rossi S.; Felton J.; Levander O.Heterocyclic Amine Content in Beef Cooked by DifferentMethods to Varying Degrees of Doneness and Gravy Made from Meat Drippings. Food Chem. Toxicol.1998, 36, 279–287. 10.1016/S0278-6915(97)00162-2. [PubMed] [CrossRef] [Google Scholar]
• Chatgilialoglu C.; Ferreri C.; Geacintov N. E.; Krokidis M. G.; Liu Y.; Masi A.; Shafirovich V.; Terzidis M. A.; Tsegay P. S.5′,8-Cyclopurine Lesions in DNA Damage: Chemical, Analytical, Biological,and Diagnostic Significance. Cells2019, 8, 513. 10.3390/cells8060513. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Zauri M.; Berridge G.; Thézénas M.-L.; Pugh K. M.; Goldin R.; Kessler B. M.; Kriaucionis S.CDA DirectsMetabolism of Epigenetic Nucleosides Revealing a Therapeutic Windowin Cancer. Nature2015, 524, 114–118. 10.1038/nature14948. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
  Gratzner H. G.MonoclonalAntibody to 5-Bromo-and 5-Iododeoxyuridine: A New Reagent for Detectionof DNA Replication. Science1982, 218, 474–475. 10.1126/science.7123245. [PubMed] [CrossRef] [Google Scholar]
  Salic A.; Mitchison T. J.A Chemical Method for Fast and Sensitive Detectionof DNA Synthesis in vivo. Proc. Natl. Acad.Sci. U. S. A.2008, 105, 2415–2420. 10.1073/pnas.0712168105. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Jun Y. W.; Albarran E.; Wilson D. L.; Ding J.; Kool E. T.FluorescenceImaging of Mitochondrial DNA Base Excision Repair Reveals Dynamicsof Oxidative Stress Responses. Angew. Chem.,Int. Ed.2022, 61, e202111829. 10.1002/anie.202111829. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Bonda E.; Rahav G.; Kaya A.; Bakhanashvili M.p53 in theMitochondria, as a Trans-Acting Protein, Provides Error-CorrectionActivities during the Incorporation of Non-Canonical dUTP into DNA. Oncotarget2016, 7, 73323. 10.18632/oncotarget.12331. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Yano W.; Yokogawa T.; Wakasa T.; Yamamura K.; Fujioka A.; Yoshisue K.; Matsushima E.; Miyahara S.; Miyakoshi H.; Taguchi J.TAS-114, a First-in-ClassDual dUTPase/DPD Inhibitor,Demonstrates Potential to Improve Therapeutic Efficacy of Fluoropyrimidine-BasedChemotherapy. Mol. Cancer Ther.2018, 17, 1683–1693. 10.1158/1535-7163.MCT-17-0911. [PubMed] [CrossRef] [Google Scholar]
• Ladner R. D.The Roleof dUTPase and Uracil-DNA Repair in Cancer Chemotherapy. Curr. Protein Pept. Sci.2001, 2, 361–370. 10.2174/1389203013380991. [PubMed] [CrossRef] [Google Scholar]
• Cannan W. J.; Pederson D. S.Mechanisms and Consequencesof Double-Strand DNA BreakFormation in Chromatin. J. Cell. Physiol.2016, 231, 3–14. 10.1002/jcp.25048. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Eccles L. J.; O’Neill P.; Lomax M. E.Delayed Repair ofRadiation Induced Clustered DNA Damage: Friend or Foe?. Mutat. Res. - Fundam. Mol. Mec.2011, 711, 134–141. 10.1016/j.mrfmmm.2010.11.003. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
  Thompson P. S.; Cortez D.New Insights into Abasic Site Repair and Tolerance. DNA repair2020, 90, 102866. 10.1016/j.dnarep.2020.102866. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Mah L.; El-Osta A.; Karagiannis T.γH2AX: A Sensitive MolecularMarker of DNA Damage and Repair. Leukemia2010, 24, 679–686. 10.1038/leu.2010.6. [PubMed] [CrossRef] [Google Scholar]
• Haskins J. S.; Su C.; Maeda J.; Walsh K. D.; Haskins A. H.; Allum A. J.; Froning C. E.; Kato T. A.Evaluating the Genotoxic and CytotoxicEffects of Thymidine Analogs, 5-Ethynyl-2′-Deoxyuridine and5-Bromo-2′-Deoxyurdine to Mammalian Cells. Int. J. Mol. Sci.2020, 21, 6631. 10.3390/ijms21186631. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• LeBlanc D. P.; Meier M.; Lo F. Y.; Schmidt E.; Valentine C.; Williams A.; Salk J. J.; Yauk C. L.; Marchetti F.Duplex SequencingIdentifies Genomic Features that Determine Susceptibility to Benzo[a]pyrene-Inducedin vivo Mutations. BMC Genomics2022, 23, 542. 10.1186/s12864-022-08752-w. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Theisinger A.; Grenacher B.; Rech K.; Scharrer E.Nucleosides are EfficientlyAbsorbed by Na-Dependent Transport Across the Intestinal Brush BorderMembrane in Veal Calves. J. Dairy Sci.2002, 85, 2308–2314. 10.3168/jds.S0022-0302(02)74311-7. [PubMed] [CrossRef] [Google Scholar]
• Cadet J.; Wagner J. R.DNA Base Damageby Reactive Oxygen Species, OxidizingAgents, and UV Radiation. Cold Spring Harb.Perspect. Biol.2013, 5, a012559. 10.1101/cshperspect.a012559. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Chen Y.; Liu X.; Sun X.; Zhang J.; Mi Y.; Li Q.; Guo Z.Synthesis and Antioxidant Activityof Cationic 1, 2, 3-Triazole Functionalized Starch Derivatives. Polymers2020, 12, 112. 10.3390/polym12010112. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
  Yang S.; Guo Z.; Miao F.; Xue Q.; Qin S.The Hydroxyl RadicalScavenging Activity of Chitosan, Hyaluronan, Starch and Their O-CarboxymethylatedDerivatives. Carbohydr. Polym.2010, 82, 1043–1045. 10.1016/j.carbpol.2010.06.014. [CrossRef] [Google Scholar]
• Watling C. Z.; Schmidt J. A.; Dunneram Y.; Tong T. Y.; Kelly R. K.; Knuppel A.; Travis R. C.; Key T. J.; Perez-Cornago A.Risk of Cancerin Regular and Low Meat-Eaters, Fish-Eaters, and Vegetarians: a ProspectiveAnalysis of UK Biobank Participants. BMC Medicine2022, 20, 73. 10.1186/s12916-022-02256-w. [PMC free article] [PubMed] [CrossRef] [Google Scholar]
• Nagy G. N.; Leveles I.; Vértessy B. G.PreventiveDNA Repair by Sanitizingthe Cellular (Deoxy) Nucleoside Triphosphate Pool. FEBS journal2014, 281, 4207–4223. 10.1111/febs.12941. [PubMed] [CrossRef] [Google Scholar]
• Kiwerska K.; Szyfter K.DNA Repair in Cancer Initiation, Progression, and Therapy—aDouble-Edged Sword. J. Appl. Genet.2019, 60, 329–334. 10.1007/s13353-019-00516-9. [PMC free article] [PubMed] [CrossRef] [Google Scholar]