Skip to main content
NCBI home page
Nucleic Acids Research logoLink to Nucleic Acids Research
. 2001 Nov 15;29(22):4716–4723. doi: 10.1093/nar/29.22.4716

Extrahelical cytosine bases in DNA duplexes containing d[GCC]n·d[GCC]n repeats: detection by a mechlorethamine crosslinking reaction

Pornchai Rojsitthisak 1, Rebecca M Romero 1, Ian S Haworth 1,a
PMCID: PMC92524  PMID: 11713322

Abstract

The cytosine–cytosine (C–C) pair is one of the least stable DNA mismatch pairs. The bases of the C–C mismatch are only weakly hydrogen bonded, and previous work has shown that, in certain sequence contexts, they can become unstacked from the core helix, and adopt an ‘extrahelical’ location. Here, using DNA duplexes with d[GCC]n·d[GCC]n fragments containing C–C mismatches in a 1,4 bp relationship, we show that cytosine bases of different formal mismatch pairs can be crosslinked by mechlorethamine. For example, in the duplex d[CTCTCGCCGCCGCCGTATC]·d[GATACGCCGCCGCCGAGAG], where underlined cytosine bases are present as the formal C–C mismatch pairs C7–C32, C10–C29 and C13–C26, we show that two mechlorethamine crosslinks form between C13 and C29 and between C10 and C32, in addition to crosslinks at C7–C32, C10–C29 and C13–C26 (we have reported previously the crosslinking of formal C–C pairs by mechlorethamine). We interpret the formation of the C13–C29 and C10–C32 crosslinks as evidence of an extrahelical location of the crosslinkable cytosines. Such extrahelical cytosine bases have been observed previously for a single C–C mismatch pair (in the so-called E-motif conformation). In the E-motif, the extrahelical cytosines are folded back towards the 5′-end of the duplex, consistent with our crosslinking data, and also consistent with the absence of C7–C29 and C10–C26 crosslinks in the current work. Hence, our data provide evidence for an extended E-motif DNA (eE-DNA) conformation in short d[GCC]n·d[GCC]n repeat fragments, and raise the possibility that such structures might occur in much longer d[GCC]n·d[GCC]n repeat tracts.

INTRODUCTION

Mismatched base pairs can be incorporated into DNA duplexes during replication and recombination. The repair of such potentially mutagenic lesions occurs with variable efficiency, depending on the nature of the mismatch pair (1–3). The cytosine–cytosine (C–C) pair is an example of a poorly repaired mismatch (1–3). The efficiency of repair can be correlated with the thermodynamic stability of the mismatch pair (4), and the C–C mismatch is amongst the least stable (5,6). Mismatch recognition, and therefore the structure of the mismatch pair, is a key element in the repair process. An early study of the C–C mismatch pair proposed an intrahelical structure that could be stabilized by protonation (7). More recent solution state nuclear magnetic resonance (NMR) data also suggested an intrahelical C–C mismatch pair (8). However, the bases of a C–C mismatch pair can also adopt ‘extrahelical’ locations in the minor groove of the duplex, in the so-called E-motif conformation (9). We return to this structure below.

In the past decade, mismatch pairs have also been shown to be of significance in the formation of unusual DNA conformations (10) associated with the so-called triplet repeat expansion diseases, or TREDS (11–13). For example, both strands of the d[CGG]n·d[CCG]n triplet repeat sequence associated with Fragile X syndrome (13) form hairpin structures which include mismatch pairs (14–23). The d[CCG]n hairpin has a C–C mismatch at every third base pair of the hairpin stem (19–23). These hairpins are conformationally flexible, and may contain C–C mismatch pairs in both intrahelical and extrahelical conformations. For further details, we refer the reader to two discussions of the d[CCG]n hairpin conformational flexibility (24,25).

The discovery of the E-motif structure by Gao et al. (9) provided the first evidence that the flexibility of mismatch pairs could lead to a major conformational change within a duplex sequence. The E-motif contains a central d[GCC]·d[GCC] fragment in which the mismatched cytosine bases are extrahelical. The thermodynamic driving force for the conformational change is provided by stacking of the guanine bases within the pseudo-dinucleotide d[GC]·d[GC] step that forms following unstacking of the mismatched bases (Fig. 1). Our observation of a distorted d[CCG]15 hairpin containing a d[GCC]n·d[GCC]n repeat stem (23) led us to propose that this hairpin might be partially in the form of an extended E-motif (23), in which the bases of multiple C–C mismatch pairs adopt extrahelical locations. The putative extended E-motif structure is shown schematically in Figure 1.

Figure 1.

Figure 1

Open in a new tab

Schematic representation of the possible conformers of a d[GCC]3·d[GCC]3 duplex fragment, showing molecules containing Watson–Crick pairs (hydrogen bonds represented by filled circles) and C–C mismatch pairs (open circles). The cytosine bases of the C–C mismatch pairs are labeled based on the numbering system of duplex III (Fig. 2C), which contains the d[GCC]3·d[GCC]3 fragment. (A) A duplex in which C7–C32, C10–C29 and C13–C26 are formal, intrahelical C–C mismatch pairs. (B) A duplex containing extrahelical cytosine bases, in which C10–C32 and C13–C29 form extrahelical pseudo-pairs.

As a corollary to these studies, we have also shown that the common nitrogen mustard, mechlorethamine (Fig. 2A), can form a DNA interstrand crosslink (Fig. 2B) at a C–C mismatch pair (25,26). Preliminary molecular modeling of the putative extended E-motif structure (Fig. 1) suggested that the N3 atoms of proximal, extrahelical cytosine bases (for example, C10 and C32 in Fig. 1) may be only ∼5 Å apart. Given this, we rationalized that pairs of extrahelical cytosine bases might be susceptible to mechlorethamine crosslinking. In this paper, we show that DNA duplexes containing two and three contiguous d[GCC]·d[GCC] helical fragments can be crosslinked by mechlorethamine at all the formal C–C mismatch pairs, and between specific pairs of cytosine bases that are not formally paired in the duplex structure. These results provide evidence for an extended E-motif DNA (eE-DNA) conformation.

Figure 2.

Figure 2

Open in a new tab

(A) Mechlorethamine. (B) A representation of the DNA interstrand crosslink formed by mechlorethamine at a C–C mismatch pair, showing the probable connectivities of the mechlorethamine crosslink through the cytosine N3 atoms. (C) Duplex sequences containing a single C–C mismatch pair (C10–C29, duplex I), two C–C mismatch pairs (C7–C32 and C10–C29, duplex II), three C–C mismatch pairs (C7–C32, C10–C29 and C13–C26, duplex III), and two C–C mismatch pairs (C7–C32 and C10–C29, duplex IV). Cytosine bases present as formal C–C mismatch pairs are underlined.

MATERIALS AND METHODS

Chemicals and reagents

Mechlorethamine and T4 polynucleotide kinase were purchased from Sigma. [γ-32P]ATP was purchased from ICN. Oligodeoxyribonucleotides were synthesized on an Applied Biosystems Model 394 automated synthesizer, deprotected, and purified with a COP cartridge. Other reagents were at least analytical grade.

32P 5′-end labeling of DNA

Column-purified synthetic DNA (2.5 µg, 0.5 nmol) was 5′-end labeled with [γ-32P]ATP (5 µl, 4500 Ci/mmol) by incubation in buffer (30 mM Tris pH 7.8, 10 mM MgCl2, 5 mM dithiothreitol) and 30 U of T4 polynucleotide kinase for 1 h at 37°C. The reaction was stopped by addition of 3 M sodium acetate (5.5 µl, pH 5.2) and pre-chilled 95% ethanol (150 µl). Unincorporated [γ-32P]ATP was removed by precipitation in pre-chilled 95% ethanol (200 µl) at –20°C overnight. The labeled DNA was lyophilized and resuspended in a 0.1 M NaCl solution.

Alkylation of DNA

The unlabeled complementary strand (2.5 µg) was added to a 0.1 M NaCl solution of the labeled oligodeoxyribonucleotide, heated to 65–70°C and then slowly cooled to room temperature. We note that the unlabeled strand is in excess, given the <100% recovery of the labeled DNA. Following annealing of the strands, a 1 µM duplex DNA solution containing 0.1 M NaCl and 10 mM Tris pH 7.5 was incubated at 37°C with 100 µM mechlorethamine in a total volume of 100 µl, and for the times indicated. For each experiment, a fresh solution of 100 mM mechlorethamine was prepared in dimethyl sulfoxide (DMSO), rapidly diluted to 10 mM and immediately added to the DNA solution. Following incubation, the reaction was terminated by addition of 3 M sodium acetate (5.5 µl, pH 5.2), tRNA (5 mg/ml, 5 µl) and pre-chilled 95% ethanol (150 µl) and precipitated in pre-chilled 95% ethanol (200 µl) at –20°C, washed and lyophilized. The DNA was then dissolved in distilled water (2 µl) and tracking dye (8 µl, 80% formamide, 1 mM EDTA, 0.025% bromophenol blue and xylene cyanol).

Detection and quantification of crosslinked DNA

The samples were loaded onto a 20% denaturing polyacrylamide gel [29:1 acrylamide:bisacrylamide, 8 M urea, 89 mM Tris–borate pH 8.5, 2 mM EDTA (TBE buffer), 0.4 mm thick, 38 × 31 cm, 2500 V, 45 W] until the xylene cyanol marker had migrated for a specific distance (duplex I 10 cm; duplex II and IV 22 cm; duplex III 28 cm). After the gel was exposed to X-ray film, the intensity of each band was quantified using Kodak Digital Science 1D software (Kodak Scientific Imaging Systems).

Determination of the crosslinking site

The bands due to the crosslinked DNA were recovered from the gel using the crush-and-soak procedure. The DNA was then precipitated with ethanol, washed, lyophilized and resuspended in 10% aqueous piperidine in a total volume of 100 µl. To ensure complete cleavage of all alkylated bases, the crosslinked samples were heated for 1 h at 90°C. For the control Maxam–Gilbert G reaction, the DNA was cleaved using 10% aqueous piperidine, in a total volume of 100 µl, for 20 min at 90°C. Different incubation times for the control and mechlorethamine-treated DNA were required, because a 20 min incubation of the crosslinked DNA resulted in insufficient cleavage at the crosslinked sites. All samples were lyophilized overnight, resuspended in 2 µl distilled water and 8 µl tracking dye (80% formamide, 1 mM EDTA, 0.025% bromophenol blue and xylene cyanol), heated at 90°C for 2 min, chilled in an ice bath, and loaded onto a 20% denaturing polyacrylamide gel [29:1 acrylamide:bisacrylamide, 8 M urea, 89 mM Tris–borate pH 8.5, 2 mM EDTA (TBE buffer), 0.4 mm thick, 38 × 31 cm, 2500 V, 45 W] until the xylene cyanol marker had migrated 10 cm.

RESULTS

A DNA duplex containing two C–C mismatch pairs gives four crosslinked species with mechlorethamine

Duplex II (Fig. 2C), which contains two C–C mismatch pairs, was incubated with mechlorethamine and the products of the reaction were separated by electrophoresis on a 20% polyacrylamide denaturing gel (Fig. 3A). Separate reactions were carried out in which the duplex was 32P-labeled on the ‘top’ strand (Fig. 3A, lane 2) or the ‘bottom’ strand (Fig. 3A, lane 4). In each case, four crosslinked species were formed corresponding to bands a–d (Fig. 3A, lane 2) and bands e–h (Fig. 3A, lane 4), respectively. The DNA from each of these bands was extracted from the gel and subjected to Maxam–Gilbert sequencing. The results for each band are shown in Figure 3B and C. Band a (Fig. 3A, lane 2) corresponds to a duplex crosslinked through C10 (Fig. 3B) and band e (Fig. 3A, lane 4) corresponds to a duplex crosslinked through C29 (Fig. 3C). Hence, the slowest moving crosslinked species is a duplex crosslinked at the formal mismatch pair C10–C29 (Fig. 4A). Similarly, bands b and f (Fig. 3A–C) are due to a duplex crosslinked at the formal mismatch pair C7–C32 (Fig. 4B). The origins of the other bands in Figure 2A are discussed below. We note that for both pairs of bands (a and e, b and f, Fig. 2A) the electrophoretic mobility of the crosslinked duplex varies slightly depending on the position of the 32P-label (on the ‘top’ or ‘bottom’ strand). Control experiments were also performed in which either the ‘top’ or ‘bottom’ strands of the duplex were incubated separately with mechlorethamine. No crosslink bands were evident in any of the control experiments, proving that the bands in Figure 3A are due to interstrand crosslinks (data not shown).

Figure 3.

Figure 3

Open in a new tab

(A) Autoradiogram of a 20% DPAGE gel showing the products of incubation for 1 h of mechlorethamine with duplex II (Fig. 2C) carrying a 5′-end 32P-label on the ‘top’ strand (lane 2) and ‘bottom’ strand (lane 4). Lanes 1 and 3 are controls (no mechlorethamine). Bands are identified as X (crosslink), M (monoadduct) and S (unreacted single strands). Each individual band due to a mechlorethamine-crosslinked duplex is labeled from a to h, for identification purposes in the sequencing gels. (B) Autoradiogram of a 20% DPAGE gel showing the products of Maxam–Gilbert sequencing of the crosslink bands from (A) lane 2, which correspond to crosslinked duplexes carrying a 5′-end 32P-label on the ‘top’ strand. The lane labeled G shows the results of a Maxam–Gilbert guanine sequencing reaction, and identifies the positions on the gel of fragments cleaved at four guanine bases in the ‘top’ strand. Lanes a–d refer to the bands from (A), and correspond to cleavage at C10 (lanes a and d) or C7 (lanes b and c). (C) Autoradiogram of a 20% DPAGE gel showing the products of Maxam–Gilbert sequencing of the crosslink bands from (A) lane 4, which correspond to crosslinked duplexes carrying a 5′-end 32P-label on the ‘bottom’ strand. The lane labeled G shows the results of a Maxam–Gilbert guanine sequencing of the ‘bottom’ strand. Lanes e–h refer to the bands from (A), and correspond to cleavage at C29 (lanes e and g) or C32 (lanes f and h).

Figure 4.

Figure 4

Open in a new tab

Possible mechlorethamine-crosslinked species for duplex II (Fig. 2C). The central 9 bp of the duplex are shown, and ‘–M–’ indicates the location of a mechlorethamine crosslink. (A) C10–C29 intrahelical crosslink. (B) C7–C32 intrahelical crosslink. (C) C10–C29 and C7–C32 double intrahelical crosslink. (D) C10–C32 extrahelical crosslink in an eE-DNA conformation with 5′ foldback of the extrahelical cytosine bases. (E) C7–C29 extrahelical crosslink in an eE-DNA conformation with 3′ foldback of the extrahelical cytosine bases. This conformation was not observed under the conditions used.

Extrahelical cytosine bases can be crosslinked by mechlorethamine: evidence for an eE-DNA conformation

In the E-motif DNA structure (9) the extrahelical cytosine bases in the d[GCC]·d[GCC] duplex fragment are positioned in the DNA minor groove, and folded back towards the 5′-end of their respective strands. If duplex II contained an eE-DNA structure with similar properties, a crosslink could conceivably form between C10 and C32 (Fig. 4D). On the other hand, if the extrahelical cytosine bases fold back towards the 3′-end of their respective strands, then a C7–C29 crosslink might be formed (Fig. 4E). Band d (Fig. 3A, lane 2) results from a crosslink formed through C10 (Fig. 3B). Similarly, band h (Fig. 3A, lane 4) results from a crosslink formed through C32 (Fig. 3C). This result is consistent with crosslinking of the duplex at C10–C32 (Fig. 4D). A similar analysis of the origin of band c (Fig. 3A, lane 2) and band g (Fig. 3A, lane 4) suggests the presence of a duplex crosslinked through C7–C29 (Fig. 4E). However, it is also apparent that a duplex containing intrahelical C–C mismatch pairs and crosslinks at both C7–C32 and C10–C29 (Fig. 4C) would give identical sequencing data (through cleavage of the alkylated cytosine bases proximal to the 32P-label on each strand, C7 and C29). The identity of the species giving bands c and g is resolved in the following section.

A kinetic analysis suggests that multiple mechlorethamine C–C crosslink formation can occur in a single duplex

The origin of bands c and g (Fig. 3A) for duplex II was determined by observing the formation of the crosslink bands as a function of time (Fig. 5). Crosslink formation in duplex II at the formal mismatch C–C pairs, C10–C29 and C7–C32 (bands a and b, respectively, Fig. 5A), reaches a maximum of 14% (C10–C29) and 12% (C7–C32) of the total DNA after 80 min (Fig. 5B). For time points beyond 80 min, there is apparent elimination of the species carrying either the C10–C29 or the C7–C32 crosslink. We conclude that this is due to the conversion of species with single crosslinks into a species having two crosslinks (at both C10–C29 and C7–C32; Fig. 4C). Hence, band c (Figs 3A and 5A) and band g (Fig. 3A) are due to a double crosslinked species (Fig. 4C). The rate of formation (Fig. 5B) of the species giving band c is consistent with this conclusion, and the amount of this species continues to rise as the single crosslink species are eliminated (Fig. 5B). To confirm that the elimination of the single crosslink species in duplex II was due to their conversion to double crosslinked duplexes, and not just to degradation of the C10–C29 and C7–C32 crosslinks, we performed a kinetic analysis of crosslinking of duplex I, which contains only one C–C mismatch pair (Fig. 2C). Formation of the C10–C29 intrahelical crosslink in duplex I reaches ∼28% of the total DNA after 100 min, beyond which no change is observed (Fig. 6).

Figure 5.

Figure 5

Figure 5

Open in a new tab

Kinetics of mechlorethamine interstrand crosslink formation with duplex II (Fig. 2C) for reaction times up to 5 h. (A) Autoradiogram of a 20% DPAGE gel following incubation of 5′-end ‘top’ strand 32P-labeled duplex II with 100 µM mechlorethamine. Lane 0 is a control (no mechlorethamine) and all other lanes show the products of incubation for the time (in minutes) indicated above the lane. Bands are identified as X (crosslink), M (monoadduct) and S (unreacted single strands), and individual crosslink bands are labeled a–d, consistent with Figure 3A (lane 2). (B) Quantification of the autoradiogram showing the time course of crosslink formation for band a (C10–C29, intrahelical) (filled squares), band b (C7–C32, intrahelical) (filled triangles), band c (C7–C32 and C10–C29, double intrahelical) (open circles) and band d (C10–C32, extrahelical) (open squares).

Figure 6.

Figure 6

Figure 6

Open in a new tab

Kinetics of mechlorethamine interstrand crosslink formation with duplex I (Fig. 2C) for reaction times up to 5 h. (A) Autoradiogram of a 20% DPAGE gel following incubation of 5′-end ‘top’ strand 32P-labeled duplex I with 100 µM mechlorethamine. Lane 0 is a control (no mechlorethamine) and all other lanes show the products of incubation for the time (in minutes) indicated above the lane. Bands are identified as X (crosslink), M (monoadduct) and S (unreacted single strands). (B) Quantification of the autoradiogram showing the time course of crosslink formation of the C10–C29 intrahelical crosslink.

A duplex containing three C–C mismatch pairs gives three intrahelical and two extrahelical C–C crosslinked species with mechlorethamine

To confirm the conclusions drawn from duplex II, we examined the mechlorethamine crosslinking reactions of duplex III, which contains three C–C mismatch pairs (Fig. 2C). Separate reactions were carried out in which duplex III was 32P-labeled on the ‘top’ strand (Fig. 7A, lane 2) or the ‘bottom’ strand (Fig. 7A, lane 4). In each experiment, five crosslinked species were apparent on the denaturing gel, corresponding to bands a–e (Fig. 7A, lane 2) and bands f–j (Fig. 7A, lane 4), respectively. The sequencing of each of these bands is shown in Figure 7B and C, and the structures of the five crosslinked species that were identified are shown in Figure 8. An analysis of the sequencing gels shows that the crosslinked species corresponding to bands a and f (Fig. 7A) is that containing a C10–C29 intrahelical crosslink (Fig. 8A). Similarly, bands b and g are due to a C13–C26 intrahelical crosslink (Fig. 8B) and bands c and h correspond to a C7–C32 intrahelical crosslink (Fig. 8C). The two pairs of faster moving bands (d and i, e and j, Fig. 7A) are due to the crosslinking of eE-DNA conformers containing extrahelical cytosine bases folded towards the 5′-end. Hence, bands d and i result from a C10–C32 extrahelical crosslink (Fig. 8D) and bands e and j correspond to a C13–C29 extrahelical crosslink (Fig. 8E).

Figure 7.

Figure 7

Open in a new tab

(A) Autoradiogram of a 20% DPAGE gel showing the products of incubation for 1 h of mechlorethamine with duplex III (Fig. 2C) carrying a 5′-end 32P-label on the ‘top’ strand (lane 2) and ‘bottom’ strand (lane 4). Lanes 1 and 3 are controls (no mechlorethamine). Bands are identified as X (crosslink), M (monoadduct) and S (unreacted single strands). Each individual band due to a mechlorethamine-crosslinked duplex is labeled from a to j, for identification purposes in the sequencing gels. (B) Autoradiogram of a 20% DPAGE gel showing the products of Maxam–Gilbert sequencing of the crosslink bands from (A) lane 2, which correspond to crosslinked duplexes carrying a 5′-end 32P-label on the ‘top’ strand. The lane labeled G shows the results of a Maxam–Gilbert guanine sequencing reaction, and identifies the positions of fragments cleaved at four guanine bases in the ‘top’ strand. Lanes a–e refer to bands from (A), and correspond to cleavage at C7 (lane c), C10 (lanes a and d) or C13 (lanes b and e). (C) Autoradiogram of a 20% DPAGE gel showing the products of Maxam–Gilbert sequencing of the crosslink bands from (A) lane 4, which correspond to crosslinked duplexes carrying a 5′-end 32P-label on the ‘bottom’ strand. The lane labeled G shows the results of a Maxam–Gilbert guanine sequencing of the ‘bottom’ strand. Lanes f–j refer to bands from (A), and correspond to cleavage at C26 (lane g), C29 (lanes f and j) or C32 (lanes h and i).

Figure 8.

Figure 8

Open in a new tab

Mechlorethamine-crosslinked species for duplex III (Fig. 2C). The central 9 bp of the duplex are shown, and ‘–M–’ indicates the location of a mechlorethamine crosslink. (A) C10–C29 intrahelical crosslink. (B) C13–C26 intrahelical crosslink. (C) C7–C32 intrahelical crosslink. (D) C10–C32 extrahelical crosslink. (E) C13–C29 extrahelical crosslink.

Double crosslinks can also form in a duplex containing three C–C mismatch pairs

To confirm the results for duplex III, we performed a similar kinetic analysis to those described for duplexes I and II. The data are shown in Figure 9. In addition to the five crosslinked species identified above, bands due to further species develop after ∼30 min (bands labeled XX in Fig. 9A). We have not attempted specific identification of these bands, but presumably they correspond to the three different ways of forming two intrahelical C–C crosslinks in duplex III. We note that the decrease in the intensity of the single crosslink bands, as these species are converted to the double crosslink species, is not as pronounced for duplex III (Fig. 9B) compared with duplex II (Fig. 5B). This is due probably to the greater number of bands present for duplex III, and the correspondingly lower intensity. However, there does appear to be a small, but consistent, decrease in the intensity of the single crosslink species for duplex III after 1 h (Fig. 9B).

Figure 9.

Figure 9

Figure 9

Open in a new tab

Kinetics of mechlorethamine interstrand crosslink formation with duplex III (Fig. 2C) for reaction times up to 4 h. (A) Autoradiogram of a 20% DPAGE gel following incubation of 5′-end ‘top’ strand 32P-labeled duplex III with mechlorethamine. Lane 0 is a control (no mechlorethamine) and all other lanes show the products of incubation for the time (in minutes) indicated above the lane. Bands are identified as X (single crosslinks), XX (double crosslinks), M (monoadduct) and S (unreacted single strands), and individual crosslink bands are labeled a–e, consistent with Fig. 7A (lane 2). (B) Quantification of the autoradiogram showing the time course of crosslink formation for band a (C10–C29, intrahelical) (filled circles), band b (C13–C26, intrahelical) (filled squares), band c (C7–C32, intrahelical) (filled triangles), band d (C10–C32, extrahelical) (open squares), band e (C13–C29, extrahelical, data obscured by the band d data) (open triangles), and bands representing three possible double, intrahelical crosslink species (crosses).

A duplex containing two contiguous d[CCG]·d[CCG] repeats does not undergo crosslinking of cytosine bases that are not formally paired

Duplex IV contains two d[CCG]·d[CCG] repeat duplex fragments, and two formal C–C mismatch pairs positioned similarly to those in duplex II (Fig. 2C). Incubation of duplex IV with mechlorethamine gave only two crosslinked species for reaction times up to 6 h (Fig. 10). These bands correspond to intrahelical crosslinks at C10–C29 and C7–C32 (data not shown). No evidence of crosslinking between extrahelical cytosine bases was obtained, suggesting that the eE-DNA conformation cannot form in duplex IV.

Figure 10.

Figure 10

Figure 10

Open in a new tab

Kinetics of mechlorethamine interstrand crosslink formation with duplex IV (Fig. 2C) for reaction times up to 6 h. (A) Autoradiogram of a 20% DPAGE gel following incubation of 5′-end ‘top’ strand 32P-labeled duplex IV with mechlorethamine. Lane 0 is a control (no mechlorethamine) and all other lanes show the products of incubation for the time (in minutes) indicated above the lane. Bands are identified as X (crosslink), M (monoadduct) and S (unreacted single strands). (B) Quantification of the autoradiogram showing the time course of crosslink formation for the C10–C29 intrahelical crosslink (filled squares) and the C7–C32 intrahelical crosslink (filled circles).

Quantification of the mechlorethamine C–C crosslink reactions

Quantitative data derived for mechlorethamine crosslinking of duplexes I, II, III and IV are shown in Table 1. The data are shown as the amount of crosslinked DNA, expressed as a percentage of the total DNA, after a 4 h reaction time (by which time the mechlorethamine has either reacted with the DNA, or been hydrolyzed). For duplex I, the 28% crosslink formation for the intrahelical crosslink is consistent with our previous reports of this reaction (25,26). For duplex II, 31% of the DNA is crosslinked through intrahelical C–C mismatch pairs (including duplexes that carry two crosslinks). A further 8% of the total DNA is crosslinked through the extrahelical C10–C32 pair, giving an approximate ratio of intrahelical:extrahelical crosslinks of 4:1. A similar analysis of the duplex III crosslinks gives a ratio of intrahelical:extrahelical crosslinks close to 2:1 (Table 1).

Table 1. Quantification of mechlorethamine crosslinking of duplexes containing intrahelical and extrahelical C–C crosslink sites.

Duplex Crosslink position Crosslink location Amount of crosslinka Total crosslinkb Intra:extra ratioc
I C10–C29 Intrahelical 28 28 –
II C10–C29 Intrahelical 11 39 31:8
  C7–C32 Intrahelical 10    
  C10–C29 and C7–C32 Double 10    
  C10–C32 Extrahelical 8    
III C10–C29 Intrahelical 5 33 23:10
  C13–C26 Intrahelical 6    
  C7–C32 Intrahelical 6    
  C10–C32 Extrahelical 5    
  C13–C29 Extrahelical 5    
  Three speciesd Double 3 × 2    
IV C10–C29 Intrahelical 18 34 34:0
  C7–C32 Intrahelical 16    
Open in a new tab

aThe amount of crosslinked DNA for each species, expressed as a percentage of the total DNA after 4 h reaction time.

bThe total amount of crosslinked DNA for each duplex after 4 h reaction time.

cThe ratio of intrahelical:extrahelical crosslinked species after 4 h of reaction time, expressed as a percentage of the total DNA.

dSpecies containing two intrahelical crosslinks (Fig. 9A) were not sequenced for duplex III.

DISCUSSION

We have proposed previously that multiple extrahelical cytosine bases might occur in d[GCC]n·d[GCC]n repeat duplex fragments (23), but we were unable to obtain direct evidence of an eE-motif DNA conformation. Here, we have shown that pairs of cytosine bases in different formal C–C mismatch pairs, and located in a 1,4 interstrand relationship, can be crosslinked by mechlorethamine. In a B-DNA duplex, the N3 atoms (through which we believe the crosslink forms) of these cytosines are ∼12 Å apart. Hence, if the cytosine bases remain intrahelical, this distance appears to be too large to allow mechlorethamine crosslinking, even if DNA bending occurs, as in the 1,3 G–G crosslink formed by mechlorethamine (27). It is possible that initial mechlorethamine monoadduct formation could induce an extrahelical cytosine conformation. However, this would still leave a significant distance between the monoadduct and the target (intrahelical) cytosine for completion of the crosslink, and we believe that such a mechanism of crosslink formation is unlikely. We have also considered the possibility of slippage of the DNA to bring different cytosine bases into proximity, but this seems unlikely, given the required disruption at the Watson–Crick paired duplex termini. We also note that, under the reaction conditions used, we would not expect a significant amount of the 1,3 G–G crosslinked species (25), although it is possible that some of the weaker bands visible at longer incubation times may be due to such species.

Given the above, we conclude that the mechlorethamine crosslink formed between cytosine bases that are not formally paired occurs because the bases are extrahelical. We further conclude that these bases are positioned similarly to those observed in E-motif DNA (9). The extrahelical cytosine bases (of a single C–C pair) in the E-motif are folded towards the 5′-end of their respective strands (9). This positions the bases adjacent to the preceding d[CG] dinucleotide step in a deepened minor groove. Our crosslinking data suggest that a similar arrangement occurs in eE-motif DNA, and that two extrahelical cytosine bases can form a ‘pseudo-pair’ in the minor groove. Crosslinks form between cytosine bases that would be proximal in 5′-end foldback conformations (C10–C32 in duplex II, and C10–C32 and C13–C29 in duplex III). They are not observed for the equivalent pseudo-pairs (C7–C29 in duplex II, and C7–C29 and C10–C26 in duplex III) that might be expected in a 3′-end foldback conformation.

Duplexes II and IV contain identically positioned formal C–C pairs, but the eE-DNA conformation is only adopted by duplex II. This may be because the transition to the extrahelical conformation in duplex II results in the formation of a pseudo-d[GC]·d[GC] step, which is energetically favorable because of effective stacking of the guanine bases (9). The equivalent transition in duplex IV would result in formation of a pseudo-d[CG]·d[CG] step. Presumably the stacking interaction in this step is insufficient to drive the transition, and the eE-DNA conformation cannot form. We also note that there is no double, intrahelical crosslink species formed for duplex IV. This may be due to subtle differences in the duplex geometry following the initial crosslink formation.

NMR spectra of d[CCG]n hairpins suggest a dynamic equilibrium between conformers containing intrahelical and extrahelical cytosine bases (22). A similar equilibrium appears to be present in the d[GCC]n·d[GCC]n duplex fragments, and it is apparent that there is a significant amount of eE-DNA present. The ratio of crosslinks (Table 1) does not necessarily reflect the ratio of intrahelical:extrahelical cytosine bases prior to crosslinking, because the rate of formation of the crosslink species may be different. However, it is of note that the intrahelical:extrahelical crosslink ratio is of the order of 4:1 in the d[GCC]2·d[GCC]2 fragment and ∼2:1 in the d[GCC]3·d[GCC]3 fragment. This perhaps suggests greater stability of the extrahelical conformation with an increased number of repeats. It will be of interest to determine if extended runs of d[GCC]n·d[GCC]n repeats, and perhaps other sequences with appropriately spaced C–C mismatch pairs, adopt even more stable eE-DNA conformations.

Acknowledgments

ACKNOWLEDGEMENTS

This work was supported by NIH (NIA) Program Project Grant AG17179 (I.S.H.) and a grant from Chulalongkorn University, Bangkok, Thailand (P.R.).

REFERENCES

  • 1.Kramer B., Kramer,W. and Fritz,H.-J. (1984) Different base/base mismatches are corrected with different efficiencies by the methyl-directed DNA mismatch-repair system of E.coli. Cell, 38, 879–887. [DOI] [PubMed] [Google Scholar]
  • 2.Su S.-S., Lahue,R.S., Au,K.G. and Modrich,P. (1988) Mispair specificity of methyl-directed DNA mismatch correction in vitro. J. Biol. Chem., 263, 6829–6835. [PubMed] [Google Scholar]
  • 3.Gasc A.-M., Sicard,A.-M. and Claverys,J.-P. (1989) Repair of single- and multiple-substitution mismatches during recombination in Streptococcus pneumoniae. Genetics, 120, 29–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Werntges H., Steger,G., Riesner,D. and Fritz,H.-J. (1986) Mismatches in DNA double strands: thermodynamic parameters and their correlation to repair efficiencies. Nucleic Acids Res., 14, 3773–3790. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Aboul-ela F., Koh,D. and Tinoco,I.,Jr (1985) Base-base mismatches. Thermodynamics of double helix formation for dCA3XA3G + dCT3YT3G (X,Y = A,C,G,T). Nucleic Acids Res., 13, 4811–4824. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Peyret N., Ananda Seneviratne,P., Allawi,H.T. and SantaLucia,J.,Jr (1999) Nearest-neighbor thermodynamics and NMR of DNA sequences with internal A.A,C.C,G.G and T.T mismatches. Biochemistry, 38, 3468–3477. [DOI] [PubMed] [Google Scholar]
  • 7.Brown T., Leonard,G.A., Booth,E.D. and Kneale,G. (1990) Influence of pH on the conformation and stability of mismatch base-pairs in DNA. J. Mol. Biol., 212, 437–440. [DOI] [PubMed] [Google Scholar]
  • 8.Boulard Y., Cognet,J.A.H. and Fazakerley,G.V. (1997) Solution structure as a function of pH of two central mismatches,C.T and C.C, in the 29 to 39 K-ras gene sequence, by nuclear magnetic resonance and molecular dynamics. J. Mol. Biol., 268, 331–347. [DOI] [PubMed] [Google Scholar]
  • 9.Gao X., Huang,X., Smith,K.G., Zheng,M. and Liu,H. (1995) New antiparallel duplex motif of DNA CCG repeats that is stabilized by extrahelical bases symmetrically located in the minor groove. J. Am. Chem. Soc., 117, 8883–8884. [Google Scholar]
  • 10.Sinden R.R. (1999) Biological implications of the DNA structures associated with disease-causing triplet repeats. Am. J. Hum. Genet., 64, 346–353. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Bowater R.P. and Wells,R.D. (2001) The intrinsically unstable life of DNA triplet repeats associated with human hereditary disorders. Prog. Nucleic Acid Res. Mol. Biol., 66, 159–202. [DOI] [PubMed] [Google Scholar]
  • 12.Timchenko L.T. and Caskey,C.T. (1999) Triplet repeat disorders: discussion of molecular mechanisms. Cell. Mol. Life Sci., 55, 1432–1447. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Jin P. and Warren,S.T. (2000) Understanding the molecular basis of fragile X syndrome. Hum. Mol. Genet., 9, 901–908. [DOI] [PubMed] [Google Scholar]
  • 14.Gacy A.M., Goellner,G., Jurani,N., Macura,S. and McMurray,C.T. (1995) Trinucleotide repeats that expand in human disease form hairpin structures in vitro. Cell, 81, 533–540. [DOI] [PubMed] [Google Scholar]
  • 15.Chen X., Mariappan,S.V.S., Catasti,P., Ratliff,R., Moyzis,R.K., Laayoun,A., Smith,S.S., Bradbury,E.M. and Gupta,G. (1995) Hairpins are formed by the single DNA strands of the fragile X triplet repeats: structure and biological implications. Proc. Natl Acad. Sci. USA, 92, 5199–5203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Mitas M., Yu,A., Dill,J. and Haworth,I.S. (1995) The trinucleotide repeat sequence d(CGG)15 forms a heat-stable hairpin containing Gsyn.Ganti base pairs. Biochemistry, 34, 12803–12811. [DOI] [PubMed] [Google Scholar]
  • 17.Nadel Y., Weisman-Shomer,P. and Fry,M. (1995) The fragile X syndrome single strand d(CGG)n nucleotide repeats readily fold back to form unimolecular hairpin structures. J. Biol. Chem., 270, 28970–28977. [DOI] [PubMed] [Google Scholar]
  • 18.Usdin K. (1998) NGG-triplet repeats form similar intrastrand structures: implications for the triplet expansion diseases. Nucleic Acids Res., 26, 4078–4085. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Mariappan S.V.S., Catasti,P., Chen,X., Ratliff,R., Moyzis,R.K., Bradbury,E.M. and Gupta,G. (1996) Solution structures of the individual single strands of the fragile X DNA triplets (GCC)n.(GGC)n. Nucleic Acids Res., 24, 784–792. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Darlow J.M. and Leach,D.R.F. (1998) Evidence for two preferred hairpin folding patterns in d(CGG).d(CCG) repeat tracts in vivo. J. Mol. Biol., 275, 17–23. [DOI] [PubMed] [Google Scholar]
  • 21.Mariappan S.V., Silks,III,L.A., Bradbury,E.M. and Gupta,G. (1998) Fragile X DNA triplet repeats, (GCC)n, form hairpins with single hydrogen-bonded cytosine.cytosine mispairs at the CpG sites: isotope-edited nuclear magnetic resonance spectroscopy on (GCC)n with selective 15N4-labeled cytosine bases. J. Mol. Biol., 283, 111–120. [DOI] [PubMed] [Google Scholar]
  • 22.Zheng M., Huang,X., Smith,G.K., Yang,X. and Gao,X. (1996) Genetically unstable CXG repeats are structurally dynamic and have a high propensity for folding. An NMR and UV spectroscopic study. J. Mol. Biol., 264, 323–336. [DOI] [PubMed] [Google Scholar]
  • 23.Yu A., Barron,M.D., Romero,R.M., Christy,M., Gold,B., Dai,J., Gray,D.M., Haworth,I.S. and Mitas,M. (1997) At physiological pH, d(CCG)15 forms a hairpin containing protonated cytosines and a distorted helix. Biochemistry, 36, 3687–3699. [DOI] [PubMed] [Google Scholar]
  • 24.Darlow J.M. and Leach,D.R.F. (1998) Secondary structures in d(CGG) and d(CCG) repeat tracts. J. Mol. Biol., 275, 3–16. [DOI] [PubMed] [Google Scholar]
  • 25.Romero R.M., Rojsitthisak,P. and Haworth,I.S. (2001) DNA Interstrand crosslink formation by mechlorethamine at a cytosine-cytosine mismatch pair: kinetics and sequence dependence. Arch. Biochem. Biophys., 386, 143–153. [DOI] [PubMed] [Google Scholar]
  • 26.Romero R.M., Mitas,M. and Haworth,I.S. (1999) Anomalous cross-linking by mechlorethamine of DNA duplexes containing C-C mismatch pairs. Biochemistry, 38, 3641–3648. [DOI] [PubMed] [Google Scholar]
  • 27.Rink S.M., Solomon,M.S., Taylor,M.J., Raja,S.B., McLaughlin,L.W. and Hopkins,P.B. (1993) Covalent structure of a nitrogen mustard-induced DNA interstrand cross-link: an N7-to-N7 linkage of deoxyguanosine residues at the duplex sequence 5′-d(GNC). J. Am. Chem. Soc., 115, 2551–2557. [Google Scholar]

Articles from Nucleic Acids Research are provided here courtesy of Oxford University Press

RESOURCES