Replication Studies of Alkyl Phosphotriester Lesions in Human Cells

Abstract

Alkyl phosphotriester (alkyl-PTE) lesions in DNA are shown to be poorly repaired; however, little is known about how these lesions impact DNA replication in human cells. Here, we investigated how the S_pand R_p diastereomers of four alkyl-PTE lesions (alkyl = Me, Et, nPr, or nBu) at the TT site perturb DNA replication in HEK293T cells. We found that these lesions moderately impede DNA replication and that their replicative bypass is accurate. Moreover, CRISPR-Cas9-mediated depletion of Pol η or Pol ζ resulted in significantly attenuated bypass efficiencies for both diastereomers of nPr- and nBu-PTE adducts, and the S_P diastereomer of Et-PTE. Diminished bypass efficiencies were also detected for the R_p diastereomer of nPr- and nBu-PTE lesions upon ablation of Pol κ. Together, our study uncovered the impact of the alkyl-PTE lesions on DNA replication in human cells and revealed the roles of individual translesion synthesis DNA polymerases in bypassing these lesions.

Graphical Abstract

Endogenous and exogenous genotoxic agents constantly challenge the integrity of the genome.¹ Among them, various alkylating agents are known to react with DNA to yield a battery of DNA lesions with alkyl groups being conjugated to different positions of nucleobases.^2,3 The ensuing DNA adducts may inhibit DNA replication and transcription and induce mutations in these important DNA metabolic processes.³ Apart from nucleobase modifications, alkylating agents can attack directly one of the two noncarbon-bound oxygen atoms of the internucleotide phosphodiester linkage, leading to the formation of alkyl phosphotriester (alkyl-PTE) lesions in two diastereomeric configurations, i.e., S_p and R_P (Figure 1).^4–6

Figure 1.

Structures of the two diastereomers of alkyl-PTE lesions employed in the present study.

Several studies have been conducted to examine the occurrence of alkyl-PTE lesions. For instance, it was found that approximately 1% and 15% of the total alkylation adducts were alkyl-PTE lesions upon treatment of DNA with methylmethanesulfonate and ethylmethanesulfonate, respectively.⁷ In addition, Me- and Et-PTE lesions were shown to be highly persistent in mammalian tissues, suggesting the poor repair of these lesions in vivo.^8,9 Recent studies also documented that exposure of rodents to tobacco-specific N-nitrosamines gives rise to the accrual of substantial levels of alkyl-PTE lesions in tissue DNA.^10–12 The persistence of the alkyl-PTE lesions in tissues suggests that they may be appropriate biomarkers for assessing cumulative exposure to genotoxic alkylating agents and that they are likely encountered by cellular DNA replication and transcription machineries.

Some studies have been conducted to examine the repair and biological end points of alkyl-PTE lesions. Along this line, the S_P-Me-PTE could be repaired by the Ada methyltransferase in Escherichia coli,^13,14 though no such methyltransferase activity has yet been detected in human cells.^4,15 Additionally, the activities of SF2 helicases, some of which are involved in DNA repair, were strongly inhibited by iPr-PTE in vitro,¹⁶ and Et-PTE was found to inhibit T4 DNA polymerase.¹⁷ Recently, Wu and Wang^18–20 conducted comprehensive investigations about the replication of a series of alkyl-PTE lesions (with the alkyl group being Me, Et, nPr, nBu, and pyridyloxobutyl) in R_P or S_P configuration at the TT dinucleotide site in E. coli cells. Interestingly, it was found that only the replication across the S_P-Me-PTE is mutagenic, and the mutagenic bypass of this lesion entails Ada protein.^18–20 In addition, Tan et al.²¹ showed that the S_P diastereomer of the Me- and nPr-PTEs lesions did not appreciably perturb transcription efficiency in cultured human cells, whereas the R_P diastereomer of the two adducts moderately and strongly blocked transcription, respectively. No replication study, however, has been conducted for alkyl-PTE lesions in human cells.

Here, we generated double-stranded plasmids containing a site-specifically inserted alkyl-PTE lesion (the alkyl group is a Me, Et, nPr or nBu) in two diastereomeric configurations, R_P and S_P (Figure 1), and examined how these lesions block DNA replication and elicit mutations in cultured human cells that are proficient in translesion synthesis (TLS), or deficient in TLS polymerases, Pol η, ι, κ, or ζ.²² We employed a previously established strand-specific PCR-competitive replication and adduct bypass (SSPCR-CRAB) assay.²³ To this end, we incorporated a single stereochemically defined alkyl-PTE lesion at a specific site in a double-stranded shuttle vector carrying an SV40 replication origin and prepared the corresponding lesion-free control vector (Figure 2A). We also included a C/C mismatch two nucleotides away from the lesion site to differentiate the replication products originated from the lesion-situated strand and the complementary lesion-free strand.²³ We premixed the lesion-carrying or control plasmids individually with a lesion-free competitor vector at fixed molar ratios and transfected them concurrently into HEK293T cells. Relative to the control vector, the competitor vector harbored three additional nucleotides in the region sandwiched with the recognition sites of the two restriction enzymes used for the cleavage of the PCR products (i.e., NcoI and SfaNI).

Figure 2.

Schematic diagrams showing the procedures for the preparation of the lesion-bearing plasmid (A) and the SSPCR-CRAB assay (B). TT in red denotes the site of alkyl-PTE lesions. The C/C mismatch site is underlined. “P1” designates one of the PCR primers, and it contains a G as the terminal 3′-nucleotide, corresponding to the C/C mismatch site of the lesion-bearing genome. It also contains a C/A mismatch two nucleotides away from its 3′-terminus for improving the specificity of PCR. “M” and “X” denote the nucleobase generated at the two nucleosides flanking the alkyl-PTE lesion after replication, and “N” and “Y” represent the paired nucleobase of M and X, respectively, in the complementary strand.

The progeny genomes were extracted from human cells 24 h following the transfection, and the residual unreplicated plasmids were removed by treatment with DpnI and exonuclease III. The progeny plasmids were then PCR-amplified using a pair of primers flanking the site where the lesion was initially installed. In this vein, one of the primers (P1) contained a G at the 3′-terminus, which corresponded to the C/C mismatch locus (Figure 2B). This allowed for the selective amplification of the progeny genomes emanating from the replication of the lesion-containing bottom strand. Furthermore, we placed a C/A mismatch in the P1 primer two nucleotides from its 3′ terminus to further improve the specificity of strand-specific PCR, as documented elsewhere.²⁴ The ensuing PCR products were digested by NcoI and SfaNI (Figure 2B), and the restriction fragments were analyzed by native PAGE and LC-MS/MS (Figures 3 and 4 and Figures S1–S3). The quantification data from these analyses were then employed to calculate the bypass efficiencies and mutation frequencies, as described in the Materials and Methods.

Figure 3.

Restriction digestion and postlabeling method for determining the bypass efficiencies and mutation frequencies of the PTE lesions in HEK293T cells. (A) Sample processing for restriction digestion using NcoI and SfaNI and postlabeling assay (p* indicates a ³²P-labeled phosphate group). The recognition sequences for restriction enzymes are indicated with arrows. (B) Representative gel images showing the NcoI/SfaNI-produced restriction fragments of interest. Synthetic ODN with the same sequence as the restriction fragment arising from the competitor vector, i.e., d-(CATGGCGATATGCTGT), is designated as “16-mer Comp”; “13-mer GT”, “13-mer GG”, “13-mer AA”, and “13-mer TT” denote the standard synthetic ODNs d(CATGGCXMGCTGT), where “XM” is GT, GG, AA, and TT, respectively.

Figure 4.

Bypass efficiencies of alkyl-PTE lesions. The data represent the mean and standard deviation of results from three independent cellular replication experiments. The p values referred to the differences between the wild-type (WT) HEK293T cells and the isogenic polymerase-deficient cells, and they were calculated using two-tailed, unpaired t test. *0.01 < p < 0.05; **0.001 < p < 0.01.

Our results showed that the Me- and Et-PTE lesions were more readily bypassed than the nPr- and nBu-PTE adducts. The bypass efficiencies for the S_p and R_p diastereomers of Me-PTE were ~43% and 58%, respectively, and the corresponding values were 52% and 38% for Et-PTE lesions, 30% and 32% for nPr-PTE adducts, and 28% and 25% for nBu-PTE lesions (Figure 4, Table S1).

Replication experiments conducted in HEK293T cells with Pol η, Pol ι, Pol κ, and Pol ζ being individually ablated by CRISPR-Cas9²² unveiled the functions of these polymerases in supporting the replicative bypass of alkyl-PTE lesions. In particular, we found that the bypass efficiencies for S_P-Et-PTE and both diastereomers of the nPr- and nBu-PTE lesions were significantly attenuated in HEK293T cells depleted of Pol η or Pol ζ, supporting the roles of these two polymerases in bypassing these lesions (Figure 4, Table S1). Furthermore, removal of Pol κ resulted in significantly diminished bypass efficiencies for the R_p diastereomer of nPr- and nBu-PTE lesions. Knockout of Pol ι resulted in moderate diminutions in bypass efficiencies for both diastereomers of the nPr- and nBu-PTE products (Figure 4, Table S1).

The results from PAGE and LC-MS/MS analyses of restriction fragments of PCR products from the progeny genomes also allowed us to assess the mutagenic properties of the alkyl-PTE lesions. It turned out that none of the alkyl-PTE adducts were mutagenic in parental HEK293T cells, or the isogenic cells deficient in any of the four TLS polymerases (Figures S1–S3).

Exposure to alkylating agents is known to induce appreciable levels of alkyl-PTE lesions in DNA. Although a number of studies have been conducted for assessing the occurrence and persistence of alkyl-PTE lesions, their biological consequences in mammalian cells are largely ignored.^4,21 In the present study, we aim to attain a comprehensive understanding about the biological end points of the alkyl-PTE lesions by examining how these lesions impede DNA replication and induce mutations in human cells and how replication across these lesions is modulated by TLS DNA polymerases. Hence, this is the first study of the effects of any alkyl-PTE lesions on DNA replication in human cells.

Our results showed that most alkyl-PTE lesions studied herein constitute moderate impediments to DNA replication in human cells, with no pronounced differences being observed for the S_p and R_P diastereomers. The latter finding differs from the observations made in E. coli, where these lesions in the R_P configuration exert stronger blockage effects on DNA replication than those in the S_p configuration.²⁵ While the exact reason behind this difference is unclear, the alkyl group in the R_P and S_P diastereomers projects into the major groove and extends out perpendicularly from double-stranded DNA,⁴ respectively; such differences in structural perturbation to duplex DNA elicited by the two diastereomers may be differentially recognized by bacterial DNA replication machinery, but not its mammalian counterpart. In addition, the replication blockage effects of nPr- and nBu-PTE lesions are exacerbated upon individual depletion of Pol η, Pol ι, Pol κ, or Pol ζ, underscoring the important roles of these polymerases in bypassing these lesions (Figure 4, Table S1). Together, our results support that the bypass efficiencies of the alkyl-PTE lesions are influenced by both the length of the alkyl chain in the lesions and TLS DNA polymerases in host cells. The exact mechanisms underlying the involvement of these TLS polymerases in supporting the replicative bypass of the alkyl-PTE lesions remain unclear, and future biochemical and structural biology studies may offer some insights into these mechanisms.

No mutagenic products were detectable for either diastereomer of any of the four alkyl-PTE lesions in HEK293T cells or the isogenic cells depleted of any of the four TLS polymerases (Figures S2 and S3). This perhaps can be attributed to the lack of alterations in the Watson–Crick hydrogen bonding properties of nucleobases imposed by alkylation of the backbone phosphate group. This finding also indicates that the presence of the alkyl-PTE lesions does not perturb the DNA polymerases’ recognition of the hydrogen bonding face of the nucleobases flanking the lesions. This finding differs from the observations made in E. coli cells in the respect that the S_P-Me-PTE at TT dinucleotide site could direct substantial frequencies of nucleotide misincorporations at the two nucleosides flanking the damage site, where Ada protein is required for the mutagenic bypass of the lesion.¹⁸ The difference is not surprising viewing that Ada recognizes only S_p-Me-PTE^13,14 and no Ada-like activity has been detected in human cells.^4,15 It is of note that, in the present study, we only considered the alkyl-PTE lesions at the TT dinucleotide site, and it will be interesting to examine how these lesions flanked by other nucleotides are recognized by the DNA replication machinery in human cells.

In summary, we investigated comprehensively the influence of alkyl-PTE adducts on the efficiency and fidelity of DNA replication in cultured human cells and revealed the roles of four major TLS polymerases (Pol η, κ, ι, and ζ) in bypassing these lesions, which afforded novel insights into the biological consequences of alkyl-PTE lesions.