English

• Nucleic Acid methylation represents an important type of nucleic acid modifications that may play crucial roles in diverse biological processes in living organisms [1, 2]. Of them, 5-methylcytosine and its oxidative derivates (e.g., 5-hydroxymethylcytosine) are well-established epigenetic marks dynamically occuring in DNA and RNA [3, 4]. In addition, N⁶-methyladenine modification in RNA has been acknowledged as a novel epigenetic modification involved in regulation of various biological activities such as tissue development, cellular DNA damage response, and stem cell fate determination [5, 6]. Methylation at N⁶ position of deoxyadenosine (dA) has also been characterized as a dominant internal modification in bacteria since the discovery of N⁶-methyldeoxyadenosine (6 mdA) five decades ago (Fig. 1) [7]. With the recent development of highly sensitive detection techniques, 6 mdA has also been observed in several eukaryotes, such as Chlamydomonas [8], C. elegans [9], Drosophila [10, 11], and human [12-15]. Increasing evidence has indicated the possible involvement of 6 mdA methylation in development, stress response and diseases including neurodegenerative disorders and cancers [11, 14-18].

  Figure 1

  

  Figure 1.  Chemical structures of deoxyadenosine (dA) and N⁶-methyldeoxyadenosine (6 mdA).

  DownLoad: Full-Size Img PowerPoint

  Understanding the biological significance of 6 mdA necessitates the investigation about how this modified nucleoside affects the flow of genetic information during DNA replication. It has been reported that 6 mdA reduced the rate of DNA replication mediated by the Klenow fragment of Escherichia coli (E. coli) DNA polymerase I and the large fragment of Bst DNA polymerase in vitro [19]. 6 mdA was also found to partially inhibit in-vitro DNA replication by human DNA polymerase (Pol) η and Pol ι, which play important roles in the translesion synthesis (TLS) of various types of DNA adducts [20, 21]. In addition, since the 6 mdA methyltransferase METTL4 can modulate the mitochondrial DNA copy number, the presence of 6 mdA has been hypothesized to affect mitochondrial DNA replication in mammals [14]. To date, the biological effects of 6 mdA during genomic DNA replication in mammalian cells remain elusive.

  We have previously reported the use of shuttle vector technology, together with liquid chromatography-tandem mass spectrometry (LC-MS/MS)-based analysis, for successful evaluation of the biological consequences and repair of multiple types of DNA adducts during replication in mammalian cells [22, 23]. It has also been reported that the next-generation sequencing (NGS) methodology is capable of determining the replicative bypass and mutagenic properties of DNA adducts in vitro and in E. coli cells, which are consistent with those obtained from LC-MS/MS and other traditional approaches but at a much higher throughput [24-26]. Herein, we combined our traditional shuttle vector technology with NGS to investigate the biological consequences of 6 mdA methylation during replication in human cells (Figs. 2A and B). We also investigated the potential roles of the major TLS DNA polymerases, including Pol κ, Pol η, Pol ι and Pol ζ, in the replication bypass of 6 mdA in human cells.

  Figure 2

  

  Figure 2.  Experimental outline. (A) A schematic diagram illustrating the procedures for the construction of the plasmid harboring a site-specifically incorporated 6 mdA. (B) NGS-based strategy for assessing the impact of the 6 mdA on DNA replication. 'X' indicates 6 mdA, and the C/C mismatch site is underlined. 'M' represents the nucleobase formed at the initial 6 mdA site after replication, and 'N' designates the paired nucleobase of 'M' in the complementary strand.

  DownLoad: Full-Size Img PowerPoint

  Using the gapped vector-based method [25, 26], we first constructed the double-stranded pTGFP-Hha10 shuttle vector housing a site-specifically incorporated 6 mdA (Fig. 2A). We also introduced C/C mismatched bases near the modified nucleoside site so as to differentiate the replication products of 6 mdA-bearing strand from that of the unmodified complementary strand, which was used as an internal control as described elsewhere [23, 27]. In this vein, introduction of mismatched base pair has been accepted as an efficient strategy to distinguish the replication products of the modified nucleoside-bearing and unmodified control strands, which may be replicated at distinct efficiencies in mammalian cells [23, 27].

  The 6 mdA-bearing pTGFP-Hha10 plasmid was transfected into HEK293T cells as well as the isogenic cells deficient in Pol κ, Pol η, Pol ι or Pol ζ for the replication study. After 24 h of incubation, the progenies of 6 mdA-bearing plasmid were isolated from human cells, and residual unreplicated plasmid DNA was removed by a combined treatment of DpnI and exonuclease Ⅲ [22, 23]. The DNA region of interest was PCR amplified and indexed with sample-specific barcodes, which designed the respective five host cell lines and the three biological replications. The 15 sets of tagged PCR products were combined and adenylated at the 3′-end, and then ligated to PE adapters. The ligation products were subsequently amplified with PE PCR primers, and the resulting PCR product was purified and subjected to NGS analysis (Fig. 2B, Table S1 in Supporting information).

  We obtained a total of ~2 million valid NGS reads for the replication products of 6 mdA-bearing plasmid from triplicate replication experiments in five host cell lines (Tables S2 and S3 in Supporting information). The effect of 6 mdA on replication efficiency can be determined by the 'bypass efficiency', which was calculated by dividing the total number of sequencing reads from 6 mdA-carrying strand with that obtained from its unmodified complementary strand. The quantification data showed that the bypass efficiencies of 6 mdA varied from ~87.3% to 99.0% in HEK293T cells as well as the isogenic cells deficient in Pol κ, Pol η, Pol ι or Pol ζ (Fig. 3), which suggested that the presence of 6 mdA did not substantially inhibit DNA replication in human cells.

  Figure 3

  

  Figure 3.  Bypass efficiencies of 6 mdA in wild-type HEK293T cells and the relevant TLS polymerase-knockout (KO) cells. The data represent the means and standard deviations of results from three independent experiments.

  DownLoad: Full-Size Img PowerPoint

  Using the NGS-based assay, the effect of 6 mdA on replication fidelity can be determined by the 'base substitution frequency', that is, the relative distribution of nucleobase (A, T, C or G) frequencies opposite the original modified site from 6 mdA-carrying strand. It turned out that the mutation frequencies of A → C, A → G or A → T arising from the replication of the 6 mdA-carrying strand in wild-type and TLS polymerase-deficient HEK293T cells were generally occurring at frequencies of 0.01%–0.03% (Fig. 4A), which were similar to the background error rates (0.01%–0.07%) identified from the unmodified complementary strand (Fig. 4B). These results indicated that 6 mdA did not compromise the fidelity of DNA replication in human cells.

  Figure 4

  

  Figure 4.  Mutation frequencies of 6 mdA (A) and control (B) in wild-type HEK293T cells and the relevant TLS polymerase-knockout (KO) cells. The data represent the means and standard deviations of results from three independent experiments.

  DownLoad: Full-Size Img PowerPoint

  As discussed above, DNA 6 mdA modification has recently attracted substantial research interests regarding their genomic distribution patterns and potential biological functions in eukaryotes [7, 28]. In this study, we have demonstrated that 6 mdA did not inhibit DNA replication or induce mutations in human HEK293T cells using a next-generation sequencing-based cellular replication assay. In addition, replication bypass of 6 mdA is also highly efficient and accurate in human cells that are deficient in any of the four TLS polymerases examined in our study. These results are inconsistent with the observation that 6 mdA can impede considerably in vitro DNA replication mediated by several distinct DNA polymerases including the Klenow fragment of E. coli DNA polymerase I as well as human Pol η and Pol ι [19-21]. These differences could be, at least in part, attributed to the fact that the in vitro replication studies were carried out with a single-type DNA polymerase, whereas the cellular DNA replication often involved the collaborative participation of different DNA polymerases and their auxiliary proteins [26, 29].

  It has been suggested that DNA 6 mdA modification may act as a new epigenetic mark that is associated with the regulation of gene expression [30]. In this respect, the enrichment of 6 mdA around the transcription start sites or transposon regions was found to be correlated with increased transcriptional activity [8, 10, 31]. It was also reported that 6 mdA can impact transcription by interfering with RNA polymerase elongation, changing chromatin structure, or modifying transcription factor binding affinities [32, 33]. Since epigenetic DNA modifications generally do not compromise the flow of genetic information during replication [34], our findings that 6 mdA has a negligible effect on DNA replication in human cells is in keeping with its potential role in epigenetic regulation of gene expression.

  Similar as 6 mdA, it was reported that the replication of 5-methyldeoxycytidine (5mdC) and its oxidation products, including 5-hydroxymethyl-dC, 5-formyl-dC and 5-carboxyl-dC, are highly efficient and accurate in human cells, consistent with the notion that these oxidized cytidine derivatives may serve as novel epigenetic marks. These epigenetic DNA modifications do not perturb Watson-Crick base pairing, which may account for their non-cytotoxic effects in human cells [4, 34]. On the other hand, the N1 and N3 positions of dA are also among the major methylation sites in DNA, and the resulting DNA adducts 1 mdA and 3 mdA are highly cytotoxic because they can impair Watson-Crick interactions and strongly block DNA replication in human cells [35, 36].

  In conclusion, we have employed NGS, together with shuttle vector technology, to quantatively assess how 6 mdA affects the fidelity and efficiency of DNA replication in human cells. Our results showed that 6 mdA has a negligible impact on DNA replication in human cells, which is consistent with its potential role as an epigenetic mark. It is worth noting that we only evaluated the effect of 6 mdA on DNA replication in a single sequence context, and it is possible that the biological consequences of 6 mdA during replication may be infuenced by sequence contexts. In this vein, previous studies have shown that DNA sequence contexts could affect the repair and replicative bypass of several DNA lesions such as the ultraviolet light-induced 6-4 photoproduct [37, 38]. In addition, it has been recently reported that 6 mdA can be modified by ALKBH1 protein to give N⁶-hydroxymethyl-dA in mammalian genomic DNA [39], and it would be interesting to investigate how this new type of chemical modification modulates the process of DNA replication.in mammalian cells. A more complete understanding of the biological consequences of 6 mdA also requires a further assessment of the modulation activities of cellular repair proteins such as 6 mdA demethylases in mammalian cells. The shuttle vector- and NGS-based strategy described in this study should be applicable for quantitatively assessing the potential effects of sequence contexts and cellular repair activities on the replication of 6 mdA as well as other DNA adducts such as N⁶-hydroxymethyl-dA and the oxidation products of 5mdC in the future.

  Declaration of competing interest

  The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

  Acknowledgments

  We thank Prof. Yinsheng Wang for kindly providing the initial pTGFPHha10 vector and TLS polymerase knockout cell lines used in the present study. This work was supported by the National Natural Science Foundation of China (Nos. 21807030, 21907028), the Science and Technology Innovation Program of Hunan Province (No. 2019RS2020), Natural Science Foundation of Hunan Province (No. 2020JJ5046), and the Fundamental Research Funds for the Central Universities (Nos. 531118010061, 531118010259).

  Supplementary materials

  Supplementary material associated with this article can be found, in the online version, at doi:10.1016/j.cclet.2021.08.066.

  
  
• 1. [1]

     S. Horvath, K. Raj, Nat. Rev. Genet. 19 (2018) 371-384 doi: 10.1038/s41576-018-0004-3

  2. [2]

     S. Zaccara, R.J. Ries, S.R. Jaffrey, Nat. Rev. Mol. Cell Biol. 20 (2019) 608-624 doi: 10.1038/s41580-019-0168-5

  3. [3]

     J. Song, C. Yi, ACS Chem. Biol. 12 (2017) 316-325 doi: 10.1021/acschembio.6b00960

  4. [4]

     L. Shen, C.X. Song, C. He, Y. Zhang, Annu. Rev. Biochem. 83 (2014) 585-614 doi: 10.1146/annurev-biochem-060713-035513

  5. [5]

     B.S. Zhao, I.A. Roundtree, C. He, Nat. Rev. Mol. Cell Biol. 18 (2017) 31-42 doi: 10.1038/nrm.2016.132

  6. [6]

     X. Deng, R. Su, H. Weng, et al., Cell Res. 28 (2018) 507-517 doi: 10.1038/s41422-018-0034-6

  7. [7]

     H. Heyn, M. Esteller, Cell 161 (2015) 710-713 doi: 10.1016/j.cell.2015.04.021

  8. [8]

     Y. Fu, G.Z. Luo, K. Chen, et al., Cell 161 (2015) 879-892 doi: 10.1016/j.cell.2015.04.010

  9. [9]

     E.L. Greer, M.A. Blanco, L. Gu, et al., Cell 161 (2015) 868-878 doi: 10.1016/j.cell.2015.04.005

  10. [10]

      G. Zhang, H. Huang, D. Liu, et al., Cell 161 (2015) 893-906 doi: 10.1016/j.cell.2015.04.018

  11. [11]

      S. He, G. Zhang, J. Wang, et al., Nat. Commun. 10 (2019) 2219 doi: 10.1038/s41467-019-10202-3

  12. [12]

      C.L. Xiao, S. Zhu, M.H. He, et al. e307, Mol. Cell. 71 (2018) 306–318. doi: 10.1016/j.molcel.2018.06.015

  13. [13]

      M.J. Koziol, C.R. Bradshaw, G.E. Allen, et al., Nat. Struct. Mol. Biol. 23 (2016) 24-30 doi: 10.1038/nsmb.3145

  14. [14]

      Z. Hao, T. Wu, X. Cui, et al., Mol. Cell 78 (2020) 382–395 doi: 10.1016/j.molcel.2020.02.018

  15. [15]

      C.W.Q. Koh, Y.T. Goh, J.D.W. Toh, et al., Nucleic Acids Res. 46 (2018) 11659-11670 doi: 10.1093/nar/gky1104

  16. [16]

      C. Ma, R. Niu, T. Huang, et al., Nat. Cell Biol. 21 (2019) 319-327 doi: 10.1038/s41556-018-0238-5

  17. [17]

      T.P. Wu, T. Wang, M.G. Seetin, et al., Nature 532 (2016) 329-333 doi: 10.1038/nature17640

  18. [18]

      B. Yao, Y. Cheng, Z. Wang, et al., Nat. Commun. 8 (2017) 1122 doi: 10.1038/s41467-017-01195-y

  19. [19]

      S. Wang, J. Wang, X. Zhang, et al., Chem. Sci. 7 (2016) 1440-1446 doi: 10.1039/C5SC02902C

  20. [20]

      K. Du, X. Zhang, Z. Zou, et al., DNA Repair 78 (2019) 81-90 doi: 10.1016/j.dnarep.2019.03.015

  21. [21]

      S. Zhang, B. Li, K. Du, et al., Biochimie 168 (2020) 134-143 doi: 10.1016/j.biochi.2019.10.018

  22. [22]

      C. You, A.L. Swanson, X. Dai, et al., J. Biol. Chem. 288 (2013) 28548-28556 doi: 10.1074/jbc.M113.480459

  23. [23]

      B. Yuan, C. You, N. Andersen, et al, J. Biol. Chem. 286 (2011) 17503-17511 doi: 10.1074/jbc.M111.232835

  24. [24]

      S.C. Chang, B.I. Fedeles, J. Wu, et al., Nucleic Acids Res. 43 (2015) 5489-5500 doi: 10.1093/nar/gkv243

  25. [25]

      D.J. Taggart, T.L. Camerlengo, J.K. Harrison, et al., Nucleic Acids Res. 41 (2013) e96 doi: 10.1093/nar/gkt141

  26. [26]

      B. Yuan, J. Wang, H. Cao, R. Sun, Y. Wang, Nucleic Acids Res. 39 (2011) 5945-5954 doi: 10.1093/nar/gkr159

  27. [27]

      J.H. Yoon, G. Bhatia, S. Prakash, L. Prakash, Proc. Natl. Acad. Sci. USA 107 (2010) 14116-14121 doi: 10.1073/pnas.1007795107

  28. [28]

      G.Z. Luo, C. He, Nat. Struct. Mol. Biol. 24 (2017) 503-506 doi: 10.1038/nsmb.3412

  29. [29]

      G. Maga, G. Villani, E. Crespan, et al., Nature 447 (2007) 606-608 doi: 10.1038/nature05843

  30. [30]

      K.J. Wu, Cancer Lett. 494 (2020) 40-46 doi: 10.1016/j.canlet.2020.08.025

  31. [31]

      S.J. Mondo, R.O. Dannebaum, R.C. Kuo, et al., Nat. Genet. 49 (2017) 964-968 doi: 10.1038/ng.3859

  32. [32]

      W. Wang, L. Xu, L. Hu, et al., J. Am. Chem. Soc. 139 (2017) 14436-14442 doi: 10.1021/jacs.7b06381

  33. [33]

      K. Sugimoto, S. Takeda, H. Hirochika, Plant J. 36 (2003) 550-564 doi: 10.1046/j.1365-313X.2003.01899.x

  34. [34]

      D. Ji, C. You, P. Wang, Y. Wang, Chem. Res. Toxicol. 27 (2014) 1304-1309 doi: 10.1021/tx500169u

  35. [35]

      J. Conde, J.H. Yoon, J.R. Choudhury, L. Prakash, S. Prakash, J. Biol. Chem. 290 (2015) 29794-29800 doi: 10.1074/jbc.M115.693010

  36. [36]

      J.H. Yoon, J.R. Choudhury, J. Park, S. Prakash, L. Prakash, J. Biol. Chem. 292 (2017) 18682-18688 doi: 10.1074/jbc.M117.808659

  37. [37]

      X. Huang, A. Kolbanovskiy, X. Wu, Biochemistry 42 (2003) 2456-2466 doi: 10.1021/bi026912q

  38. [38]

      P. Shriber, Y. Leitner-Dagan, N. Geacintov, T. Paz-Elizur, Z. Livneh, Mutat. Res. 780 (2015) 71-76 doi: 10.1016/j.mrfmmm.2015.08.002

  39. [39]

      J. Xiong, T.T. Ye, C.J. Ma, et al., Nucleic Acids Res. 47 (2019) 1268-1277 doi: 10.1093/nar/gky1218

• 

  Figure 1  Chemical structures of deoxyadenosine (dA) and N⁶-methyldeoxyadenosine (6 mdA).

  下载: 全尺寸图片 幻灯片
  

  Figure 2  Experimental outline. (A) A schematic diagram illustrating the procedures for the construction of the plasmid harboring a site-specifically incorporated 6 mdA. (B) NGS-based strategy for assessing the impact of the 6 mdA on DNA replication. 'X' indicates 6 mdA, and the C/C mismatch site is underlined. 'M' represents the nucleobase formed at the initial 6 mdA site after replication, and 'N' designates the paired nucleobase of 'M' in the complementary strand.

  下载: 全尺寸图片 幻灯片
  

  Figure 3  Bypass efficiencies of 6 mdA in wild-type HEK293T cells and the relevant TLS polymerase-knockout (KO) cells. The data represent the means and standard deviations of results from three independent experiments.

  下载: 全尺寸图片 幻灯片
  

  Figure 4  Mutation frequencies of 6 mdA (A) and control (B) in wild-type HEK293T cells and the relevant TLS polymerase-knockout (KO) cells. The data represent the means and standard deviations of results from three independent experiments.

  下载: 全尺寸图片 幻灯片
•