- Email: [email protected]
G-quadruplex formation by single-base mutation or deletion of mitochondrial DNA sequences
BBA - General Subjects 1863 (2019) 418–425
Contents lists available at ScienceDirect
BBA - General Subjects journal homepage: www.elsevier.com/locate/bbagen
G-quadruplex formation by single-base mutation or deletion of mitochondrial DNA sequences I-Te Chua, Chia-Chuan Wua,b, Ta-Chau Changa, a b
T
⁎
Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106, Taiwan, ROC Department of Agricultural Chemistry, National Taiwan University, Taipei 106, Taiwan, ROC
A R T I C LE I N FO
A B S T R A C T
Keywords: Mitochondrial DNA G-quadruplex Single-base mutation or deletion Structural diversity Structural variation Hairpin
Background: Mitochondrial DNA (mtDNA) mutations could lead to mitochondrial dysfunction, which plays a major role in aging, neurodegeneration, and cancer. Recently, we have highlighted G-quadruplex (G4) formation of putative G4-forming (PQF) mtDNA sequences in cells. Herein, we examine structural variation of G4 formation due to mutation of mtDNA sequences in vitro. Methods: The combined circular dichroism (CD), nuclear magnetic resonance (NMR), and polyacrylamide gel electrophoresis (PAGE) results provide complementary insights into the structural variation of the studied G-rich sequence and its mutants. Results: This study illustrates the structural diversity of mt10251, a G-rich mtDNA sequence with a 16-nt loop, ( GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG), including the coexistence of a hairpin structure and monomeric, dimeric, and tetrameric G4 structures of mt10251 in 20 mM K+ solution. Moreover, a single-base mutation of mt10251 can cause significant changes in terms of structural populations and polymorphism. In addition, singlebase mutations of near-but-not-PQF sequences can potentially change not-G4 to G4 structures. We further found 124 modified PQF sequences due to single-base mutations of near-but-not-PQF sequences in mtDNA. Conclusions: Single-base mutations of mt10251 could make significant changes in its structural variation and some single-base mutated sequences in mtDNA could form G4 structures in vitro. General significance: We illustrate the importance of single-base mutations of DNA sequences to the change of G4 formation in vitro. The use of single-base mutations by generating the fourth G-tract and followed by selection in shortening the longest loop size in the near-but-not-PQF sequences was conducted for the G4 formation.
1. Introduction G-quadruplexes (G4s) are unique four-stranded nucleic acid structures formed by guanine-rich (G-rich) sequences under physiological conditions. The importance of G4s comprises not only the protection of chromosome ends, but also the regulation of the expression of certain genes [1–3]. Thus, G4s have become potential targets for cancer therapeutics [2,4]. However, some G-rich sequences can adopt various G4 structures and possibly coexist in mixtures. For example, slight variations in human telomeric sequences can result in different G4 structures [5,6]. In addition, the sequence (TG4AG3)2TG4A2G2 in the c-myc gene promoter formed both intramolecular and intermolecular G4 structures in K+ solution [7]. Plavec et al. have recently studied the effect of single-base mutation in G-rich sequences of human papillomaviruses on the structural diversity of DNA [8]. To our knowledge, structural variation of G4 formation among monomers, dimers, and tetramers of a
⁎
native G-rich sequence caused by a single-base difference has not yet been reported. While considerable efforts have been made to demonstrate the biological role of nuclear G4 formation [9,10], the study of mitochondrial G4 formation is still in its nascent stages [11,12]. With reference to the large size of nuclear DNA (3.3 × 109 bps), human mitochondrial DNA (mtDNA) is extremely small (16,569 bps) [13]. Dong et al. [11] identified 9 PQF sequences with four tracts of three consecutive guanines and 178 PQF sequences with four tracts of two consecutive guanines for each length < 33 nt. We highlighted the existence of human mtDNA G4 structures in cells [14]. In addition, Lyonnais et al. [15] found that the human mitochondrial transcription factor A extracted from mitochondria could strongly bind to diverse G4 structures in vitro. It is noteworthy that mtDNA mutations, including point mutations, deletions, inversions, and copy number variations, can lead to mitochondrial dysfunction, which plays a major role in the pathogenesis
Corresponding author. E-mail address: [email protected] (T.-C. Chang).
https://doi.org/10.1016/j.bbagen.2018.11.009 Received 5 July 2018; Received in revised form 26 October 2018; Accepted 13 November 2018 Available online 26 November 2018 0304-4165/ © 2018 Published by Elsevier B.V.
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
2.3. Nuclear magnetic resonance (NMR) spectroscopy
of aging, neurodegeneration, and diseases such as cancer [16]. Previous studies have suggested that mtDNA contains numerous sequences with the potential to form the non-B DNA structures, which could be associated with mtDNA deletions [11,17]. In addition, the non-B DNA structures have been implicated in genetic instability and disease development [18]. Among non-B forms of DNA, G4 structures have recently attracted more attention [11,12,14,15]. Recently, we utilized CD to study 10 PQF mtDNA sequences each with a length < 33 nt [14]. Among them, only mt15653 (GGGTTAAT CGTGTGACCGCGGTGGCTGG) was unable to form G4 with a possible two G-quartets, while the others could form G4s in a 150 mM K+ solution. It is notable that mt15653 has a 16-nt-long loop. In contrast, mt10251 (GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG), located at positions 10251–10280 in reversed mtDNA of the NC_012920 reference sequence, can form G4s with a possible three G-quartets, even though it also contains a 16-nt-long loop. However, the G4 formation type of mt10251 in K+ solution is unclear. Previously, Mergny et al. reported possible G4 formation with longer loops comprising up to 30 nt of thymine [19]. Yang et al. demonstrated that the 1245 G-runs of BCL-2 sequence adopted a parallel G4 structure with a 13-nt loop in K+ solution [20]. In this work, the combination of NMR and PAGE has revealed the coexistence of monomeric, dimeric, and tetrameric G4 structures of mt10251 in 20 mM K+ solution. Moreover, a single-base mutation or deletion of mt10251 could cause significant changes in terms of structural populations and polymorphism. For example, substitution of G2 by T2 or G6 by T6 in mt10251 could change a G4 to a not-G4 structure, while the substitution of G25 by T25 or G29 by T29 in mt10251 could maintain tetrameric G4 formation. The former finding stimulated our interest to examine the possible G4 formations from non-PQF sequences with single-base mutations in mtDNA. Here, we name such non-PQF sequences with three G-tracts that can be converted to PQF sequences with four G-tracts by single-base mutations as near-but-not-PQF sequences. Since the point mutation is possible in mtDNA, we have screened mtDNA sequence and found 124 such near-but-not-PQF sequences. Data analyses of 20 PQF and 10 modified PQF sequences suggested that more than half of them can change not-G4 to G4 structures after single-base mutations in vitro. Notably, recent studies have suggested that two mutants of mt10251 with a single-base difference, mt10251-T24 by substitution of G24 with T24 and mt10251-d22 by deletion of A22, are potential risk factors in breast cancer [21].
All NMR experiments were performed on Bruker AVIII 500 MHz spectrometers equipped with a prodigy probehead at 25 °C. One-dimensional (1D) imino proton NMR spectra were recorded using a WATERGATE for water suppression. The strand concentrations of the NMR samples were typically 100 μM, containing 10% D2O in 10 mM Tris (pH 7.5) or 20 mM K+ conditions, with an internal reference of 0.01 mM DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid). 2.4. Polyacrylamide gel electrophoresis (PAGE) PAGE was performed using 16% polyacrylamide gel for samples without K+ and with the addition of 20 mM K+ in both the samples and gel solution. The DNA concentration for each sample was 100 μM, that is, the same concentration used for CD and NMR. PAGE was conducted at 150 V for 3.5 h at 4 °C. The gels were then photographed under ultraviolet (UV) light at 254 nm using a digital camera. 3. Results and discussion 3.1. Structural diversity of G4 formations in mt10251 The CD spectra of mt10251 showed a growth of the 265-nm band after the addition of 20 and 150 mM K+ solution (Fig. 1A). The detection of positive CD bands at 265 nm is a typical CD pattern for parallel G4 structures. NMR spectra of mt10251 showed imino proton signals in the region near 13 ppm in Tris buffer (Fig. 1B), indicating the presence of hairpin structure. After the addition of 20 and 150 mM K+, the detection of imino proton signals in the region of 10.5–12 ppm indicated the formation of G4 structures. Notably, the concomitance of imino proton signals near 13 ppm and 10.5–12 ppm in mt10251
A
C
(degxmol -1xcm 2 )
-5
8
mt10251
6 4 2 0
10 mM Tris + 20 mM K + overnight + 150 mM K+ overnight
-2 -4
2. Materials and methods
1 2 3
240 280 320 Wavelength (nm)
2.1. DNA preparation
B
DNA oligonucleotides were purchased from Bio Basic (Ontario, Canada) and dissolved in 10 mM Tris (pH 7.5). They were then subjected to heat-denaturation at 95 °C for 10 min and slowly annealed to room temperature at a rate of 1 °C/min. The annealed oligonucleotides were stored at 4 °C overnight until the further experiments. The DNA concentrations were determined using a UV–Vis absorption spectrometer (Implen, Germany).
mt10251 + 150 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
2.2. Circular dichroism (CD) spectroscopy
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
CD experiments were conducted using a spectropolarimeter (J-815, Jasco, Japan) with a bandwidth of 2 nm, at a scan speed of 50 nm/min and a step resolution of 0.2 nm over a spectral range of 210–350 nm. The DNA concentration in each sample was 100 μM (dissolved in 10 mM Tris (pH 7.5)), and a stock solution of 3 M KCl (Sigma-Aldrich, USA) was added to the DNA samples to attain a final K+ concentration of 20 mM. The observed signals were baseline subtracted.
Fig. 1. Structural diversity of G4 formation in mt10251. (A) CD spectra and (B) NMR spectra of mt10251 in 10 mM Tris buffer, after the addition of 20 mM K+ and 150 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24 (T2AG3)4, HT48 (T2AG3)8, and HT96 (T2AG3)16 (lane 1), mt10251 in 20 mM K+ solution overnight (lane 2), and in 150 mM K+ solution overnight (lane 3). The same DNA concentration of 100 μM was used in the experiments of CD, NMR, and PAGE of this work. 419
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
suggested the coexistence of hairpin and G4 structures in K+ solution. PAGE was conducted to verify possible monomer, dimer, and tetramer conformations; PAGE results showed several discernible bands with dimeric and tetrameric G4s of mt10251 after the addition of 20 and 150 mM K+ (Fig. 1C). The use of an mt10251 concentration of 100 μM for PAGE and CD was simply to match the DNA concentration necessary for use in NMR experiments. Thus, significant intermolecular structures found at the mt10251 concentration of 100 μM were possibly due to high concentrations of DNA. Here, the gel assays for 20 μM and 100 μM mt10251 showed that the higher the mt10251 concentration, the more was the high-order population (Fig. S1). Nevertheless, we also detected the intermolecular G4 structures at 20 μM mt10251 even in 20 mM K+ solution. In addition, the gel assays of mt10251 showed that an increasing amount of the intermolecular population was detected at higher K+ concentrations (Fig. S1). Such finding was consistent with a previous study of the structural conversion of intramolecular and intermolecular G4 formations of bcl2mid [22]. Since we were interested in the intramolecular G4 formation for more biological relevance, the following studies were conducted using 20 mM K+ solution.
20 mM K+, whereas that of mt10251-T25 and mt10251-T29 showed two additional bands with tetrameric G4 formation (Fig. 2B). It appears that the first and second G-tracts play critical roles for G4 formation.
3.3. Search of single-base mutations in mt10251 to change multiple G4 structures to primarily monomeric G4 structures Given that the tetrameric G4 structure is unlikely to be formed in cells, we examined whether the mt10251-M1 sequence can favor the formation of monomeric G4 structures by substituting A23 and A27 with T23 and T27, and whether the mt10251-M2 sequence can favor the formation of monomeric G4 structures by substituting C15-C17 with T15T17. The CD spectra of mt10251-M1 and mt10251-M2 showed the growth of the 265-nm band after the addition of 20 mM K+ (Fig. S3A), suggesting parallel G4 formation. Their NMR spectra showed no imino proton signals at 13 ppm in Tris buffer (Fig. S3B), implying the absence of hairpin structure. After the addition of 20 mM K+, the NMR spectrum of mt10251-M2 showed fine imino proton signals in the 10.5–12 ppm region. The NMR spectrum of mt10251-M1 showed imino proton signals different from those of mt10251-M2, implying that they had formed different G4 structures. In addition, the PAGE results showed a major band of mt10251-M1 with dimeric G4 and a major band of mt10251-M2 with monomeric G4 (Fig. S3C). Since the major component of mt10251-M2 was a monomeric G4 structure, we were interested in examining the effect of a single C-base mutation of C15-C17 in mt10251 (substituting C15 with T15 in mt10251T15, G16 with T16 in mt10251-T16, and G17 with T17 in mt10251-T17, together with a single C-base deletion of C17 in mt10251-d17) on G4 formation. Their NMR spectra showed no imino proton signals near 13 ppm (Fig. 3A), indicating the absence of a hairpin structure. After the addition of 20 mM K+, imino proton signals in the 10.5–12 ppm region were similar for mt10251-T15, mt10251-T16, and mt10251-d17 and different for mt10251-T17 (Fig. 3A). Their CD spectra also showed substantial growth of the 265-nm band after the addition of 20 mM K+ (Fig. S4). The PAGE results showed a major dimer band and a weak monomer band for mt10251-T15, mt10251-T16, and mt10251-d17, and a major smeared monomer band for mt10251-T17 (Fig. 3B). To simulate the NMR spectrum of mt10251, we summed up the NMR spectra of mt10251-T29 for tetrameric, mt10251-T16 for dimeric, and mt10251-T17 for monomeric formations in a ratio of 5:3:1 based
3.2. Single G-base substitutions in mt10251 to determine hairpin formation To examine the possible pairings between the C-tract and each Gtract for hairpin formation, we studied numerous mt10251 mutants by substituting the middle G-base with a T-base, specifically, G2 with T2 in mt10251-T2, G6 with T6 in mt10251-T6, G25 with T25 in mt10251-T25, and G29 with T29 in mt10251-T29, to eliminate one G-tract in each sequence. The absence of imino proton signals near 13 ppm in the NMR spectrum of mt10251-T25 suggested that the G-tract of G24-G26 was involved in the hairpin formation of mt10251 (Fig. 2A). After addition of 20 mM K+, imino proton signals of mt10251-T2 and mt10251-T6 in the 10.5–12 ppm region were absent, indicating no G4 formation. Notably, the imino proton signals of mt10251-T25 and mt10251-T29 in this region were similar to those of mt10251 (Fig. 2A). Consistent with the NMR results, the CD spectra of mt10251-T2 and mt10251-T6 showed no appreciable CD band change, whereas those of mt10251T25 and mt10251-T29 showed a marked growth of the 265-nm band after the addition of 20 mM K+ (Fig. S2). PAGE analysis of mt10251-T2 and mt10251-T6 showed no additional bands after the addition of
A
B
mt10251-T2
mt10251-T6
+ 20 mM K+ overnight
+ 20 mM K+ overnight
1 2 3 4 5 10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
mt10251-T25
mt10251-T29
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 2. Single G-base substitution in mt10251. (A) NMR spectra of mt10251-T2, mt10251-T6, mt10251-T25, and mt10251-T29 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) PAGE assays of mt10251 (lane 1), mt10251-T2 (lane 2), mt10251-T6 (lane 3), mt10251-T25 (lane 4), and mt10251-T29 (lane 5) in 20 mM K+ solution overnight. 420
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
A mt10251-T15
mt10251-T16
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
mt10251-T17
mt10251-d17
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
B
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
C Mixed (5:3:1)
1 2 3 4 5
mt10251-T29 20 mM K+ mt10251-T16 20 mM K+ mt10251-T17 20 mM K+ mt10251 20 mM K+ 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 3. Single-base mutations in mt10251 to change multiple G4 structures to primarily monomeric G4 structures. (A) NMR spectra of mt10251-T15, mt10251-T16, mt10251-T17 and mt10251-d17 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) PAGE assays of mt10251 (lane 1), mt10251-T15 (lane 2), mt10251-T16 (lane 3), mt10251-T17 (lane 4), and mt10251-d17 (lane 5) in 20 mM K+ solution overnight. (C) Summing up NMR spectra of mt10251-T29 for tetramer, mt10251-T16 for dimer, and mt10251-T17 for monomer with the ratio of 5:3:1 based on the gel results to mimic the NMR spectrum of mt10251. Table 1 The sequences of mt10251 and its mutants studied in this work with weak (W), moderate (M), and strong (S) components based on the PAGE results. Name
Sequence
Monomer
Dimer
Tetramer
mt10251 mt10251-M1 mt10251-M2 mt10251-T2 mt10251-T6 mt10251-T25 mt10251-T29 mt10251-T15 mt10251-T16 mt10251-T17 mt10251-d17 mt10251-T24 mt10251-d22
GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTATGGGTGGG GGGTGGGAGTAGTTTTTTGCTAAGGGAGGG GTGTGGGAGTAGTTCCCTGCTAAGGGAGGG GGGTGTGAGTAGTTCCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTAAGTGAGGG GGGTGGGAGTAGTTCCCTGCTAAGGGAGTG GGGTGGGAGTAGTTTCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCTCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCTTGCTAAGGGAGGG GGGTGGGAGTAGTTCC↓TGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTAATGGAGGG GGGTGGGAGTAGTTCCCTGCT↓AGGGAGGG
W
W S
M
– –
– – S S
S – –
W W S W S
S S S S W
tetrameric intensity of each sequence obtained from the PAGE assays.
on the PAGE results (Fig. 3C). The agreement between the simulated and experimental results supported the unusual complexity of conformational changes from a hairpin structure to multiple monomeric, dimeric, tetrameric G4 structures of mt10251 after the addition of 20 mM K+. Importantly, a single-base mutation of mt10251 can cause significant changes in terms of structural populations and polymorphism. Table 1 listed the sequences of mt10251 and its mutants studied in this work and the relative monomeric, dimeric, and
3.4. Conversion of near-but-not-PQF sequences to PQF sequences by singlebase mutations Because a single-base mutation can change a G4 to a not-G4 structure, we were interested in examining the possibility of G4 formation due to a single-base mutation of a near-but-not-PQF sequence, which 421
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
C
Molar ellipticity [ ] x10 -5 (degxmol -1xcm2 )
A 4 3 2 1 0 -1 -2 10 mM Tris + 20 mM K+ overnight -3
mt5027
2 1 0 -1
240 280 320 Wavelength (nm)
mt5027-G18
4
mt5027-d18
3
1 2 3 4
2 1 0 -1 10 mM Tris + 20 mM K+overnight
10 mM Tris + 20 mM K+overnight -2
240 280 320 Wavelength (nm)
240 280 320 Wavelength (nm)
B mt5027
mt5027-G18
mt5027-d18
+ 20 mM K+ overnight
10 mM Tris 12.5 12.0 11.5 11.0 10.5 ppm
12.5 12.0 11.5 11.0 10.5 ppm
12.5 12.0 11.5 11.0 10.5 ppm
Fig. 4. Single-base mutations of near-but-not-PQF mt5027 sequence. (A) CD spectra and (B) NMR spectra of mt5027, mt5027-G18 and mt5027-d18 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24, HT48, and HT96 (lane 1), mt5027 (lane 2), mt5027-G18 (lane 3), and mt5027-d18 (lane 4) in 20 mM K+ solution overnight.
gives rise to a PQF sequence. We first studied a native mtDNA sequence of mt5027 and two mutants with a single-base substitution of T18 by G18 for mt5027-G18 and a single-base deletion of T18 for mt5027-d18. The CD spectra showed no appreciable difference for mt5027, but a significant difference for mt5027-G18 and mt5027-d18 after the addition of 20 mM K+ (Fig. 4A). The detection of two positive CD bands near 240 and 295 nm, together with a negative CD band near 265 nm, represents a typical pattern for antiparallel G4 structures. Consistent with the CD results, imino proton signals in the 10.5–12.5 ppm region due to G4 structures were not detected in the NMR spectrum of mt5027, but were detected in the spectra of mt5027-G18 and mt5027-d18 after the addition of 20 mM K+ (Fig. 4B). Notably, more imino proton signals were detected for mt5027-G18 than for mt5027-d18, implying that mt5027-G18 could have formed at least two different G4 structures. The PAGE results for mt5027, mt5027-G18, and mt5027-d18 showed a single monomer band (Fig. 4C). Although PAGE was not able to distinguish the multiple G4 structures of mt5027-G18, it indicated the absence of intermolecular G4 formation. Indeed, single-base mutations or deletions of near-but-not-PQF sequences could benefit the G4 formation in vitro. We further investigated 9 additional near-but-not-PQF sequences together with their single-base mutations to characterize their possible G4 formations in 20 mM K+ solution (Table 2 and Fig. S5). Indeed, NMR spectra showed no imino proton signals in the 10.5–12.0 ppm region for most of these near-but-not-PQF sequences, but a number of imino proton signals in this region for most of the single-base mutated sequences. Exceptions include the presence of a number of fine peaks of imino proton signals in the 10.5–12.0 ppm region form mt1363 sequence and the absence of imino proton signals in this region from mt10550-G5 mutant, which deserves further study. Of interest was that such imino proton signals in this region were not detected in mt2080G7 with substitution of A7 by G7, but a number of fine peaks in this region were detected in mt2080-G14 with substitution of T14 by G14. It is not clear whether this difference is due to an 8-nt loop existed in mt2080-G7 but a 6-nt loop existed in mt2080-G14 or different loop sequences. Nevertheless, this finding suggested that generating the fourth G-tract and shortening the longest loop size in a near-but-notPQF sequence by a single-base mutation could benefit the G4
Table 2 Near-but-not-PQF sequences together with their single-base mutations and 4 PQF sequences studied in this work. Name
Sequence
mt5027 mt5027-G18 mt5027-d18 mt2080 mt2080-G7 mt2080-G14 mt1363 mt1363-G13 mt2971 mt2971-G5 mt5054 mt5054-G13 mt6797 mt6797-G11 mt7908 mt7908-G12 mt10550 mt10550-G5 mt12077 mt12077-G2 mt14229 mt14229-G7 mt1067 mt1563 mt10213 mt13056
GGAAGGGGTAGGCTATGTG GGAAGGGGTAGGCTATGGG GGAAGGGGTAGGCTATG↓G GGAATGATGGTTGTCTTTGG GGAATGGTGGTTGTCTTTGG GGAATGATGGTTGGCTTTGG GGGATGGCGGATAG GGGATGGCGGATGG GGCGCTTGTCAGGGAGG GGCGGTTGTCAGGGAGG GGGGGTTGAGAATGAGTGTGAGG GGGGGTTGAGAAGGAGTGTGAGG GGAAATGGTGAAGGG GGAAATGGTGGAGGG GGGTGGTGATTAGTCGG GGGTGGTGATTGGTCGG GGCTCGAATAAGGAGG GGCTGGAATAAGGAGG GACGGGTTGGGCCAGGGG GGCGGGTTGGGCCAGGGG GGCTTATGCGGAGG GGCTTAGGCGGAGG GGTCGCCTAGGAGGTCTGG GGCGCCATTGGCGTGAAGGTAGCGG GGTGTAAGGAGAAGATGGTTAGG GGGTGATGGTAGATGTGGCGGG
formation.
3.5. Estimation of the number of near-but-not-PQF sequences for potential G4 formation in mtDNA To estimate the number of near-but-not-PQF sequences in mtDNA, we first studied four native PQF sequences with each containing four tracts of GG and a loop size of 7 nt. The NMR results showed that mt1067 and mt1563 did not form G4 structure, while mt10213 and mt13056 could form G4 structures (Table 2 and Fig. 5A). For simplicity, 422
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
A mt1067
mt1563
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5ppm
mt10213
mt13056
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
B The number of the longest loop base (nmax )
10 G4 (In this work) Not G4 (In this work) G4 (Reference 24) Not G4 (Reference 24)
8 6 4 2 0 0.0
0.5
1.0
1.5 2.0 G4Hscore
2.5
3.0
Fig. 5. NMR results used in characterizing the G4 formation. (A) NMR spectra of mt1067, mt1563, mt10213 and mt13056 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) The plot of 30 PQF sequences by using the G4Hscore and the number of the longest loop bases to analyze the possible G4 formation with two G-quartets.
of the longest loop bases (nmax) to analyze the possible G4 formation with two G-quartets (Fig. 5B). The plot comprised 10 sequences obtained from ref. 24 and 20 sequences studied in this work. The results suggested that those PQF sequences characterized with either nmax ≤ 4 or G4Hscore ≥ 1.2 have very high possibility to form G4 structures, while those sequences characterized with nmax > 4 and G4Hscore < 1.2 have low possibility to form G4 structures. Accordingly, there are 82 out of 118 modified mtDNA sequences which are likely to form G4 structures with two G-quartets in vitro.
we used the loop size between 1 nt and 7 nt in our estimation of PQF sequences with four tracts of GG. Using the QGRS program to analyze the PQF sequence in mtDNA (NC_012920) [23], it was found that there were 9 PQF sequences with four tracts of GGG constrained by loop sizes between 1 and 16 nt, and 79 PQF sequences with four tracts of GG constrained by loop sizes between 1 and 7 nt (Table S1). Considering the possible overlapping PQF sequences, we only counted one sequence. We further found 124 near-but-not-PQF sequences in mtDNA (NC_012920), which contain 6 modified sequences with possible three G-quartets and 118 modified sequences with possible two G-quartets, as listed in Table S2. Recently, a G4Hunter algorithm that took into account the G-richness and G-skewness of a given sequence was developed to evaluate the G4 propensity of 209 sequences from the human mtDNA in vitro. These mtDNA sequences were further classified as stable G4 (71), unstable G4 (63), and not-G4 (75) based on G4Hscores and biophysical results [24]. Here we used the G4Hunter program to calculate the G4Hscore for each sequence and then made a plot by using the G4Hscore and the number
3.6. G4 structures of specific mutants of mt10251: mt10251-d22 and mt10251-T24 We have studied the G4 formations of mt10251-T24 and mt10251d22 sequences because they are possible risk factors in breast cancer [21]. The CD spectra of mt10251-T24 and mt10251-d22 showed growth of the 265-nm band after the addition of 20 mM K+ (Fig. 6A), implying the G4 formation. Their NMR spectra showed no imino proton 423
BBA - General Subjects 1863 (2019) 418–425
4
mt10251-T24
3 2 1 0
-1
10 mM Tris + 20 mM K+overnight
-2 240
Molar ellipticity [ ] x10-5 (degxmol-1xcm 2 )
A
Molar ellipticity [ ] x10-5 (degxmol-1xcm 2)
I.-T. Chu et al.
C
8
mt10251-d22
6
1 2 3 4
4 2 0 -2
10 mM Tris + 20 mM K+ overnight
-4
280 320 Wavelength (nm)
240 280 320 Wavelength (nm)
B mt10251-T24
mt10251-d22
+ 20 mM K overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
+
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 6. Single-base mutation or deletion in mt10251. (A) CD spectra and (B) NMR spectra of mt10251-T24 and mt1025-d22 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24, HT48, and HT96 (lane 1), mt10251 (lane 2), mt10251-T24 (lane 3), mt10251d22 (lane 4) in 20 mM K+ solution overnight.
formation, which may happen in mtDNA. Although we have not found such mutated sequences in mitochondria at present, we are not able to totally eliminate the possible association of such mutations with mitochondrial dysfunction.
signals at 13 ppm in Tris buffer (Fig. 6B), implying the absence of hairpin structure. After the addition of 20 mM K+, the NMR spectrum of mt10251-T24 was very similar to that of mt10251-T25, while the NMR spectrum of mt10251-d22 showed different imino proton signals from those of mt10251-T24, implying that they formed different G4 structures. In addition, the PAGE results showed two bands of mt10251-T24 with tetrameric G4 s and a major band of mt10251-d22 with monomeric G4, together with a weak tetrameric G4 band (Fig. 6C). Although the tetrameric G4 formation of mt10251-T24 is unlikely prevalent in mitochondria, the monomeric G4 formation of mt10251-d22 may be accessible in mitochondria. At present, it is not clear whether the potential risk of mt10251-d22 in breast cancer is correlated with its monomeric G4 formation. The possible correlation of mtDNA mutations to G4-related diseases deserves further study. With reference to a recently developed method [15,25], it is possible to selectively pull down G4 structures from isolated mtDNA for future study.
Conflict of interest The authors declare no conflict of interest. Acknowledgements This work was supported by the Ministry of Science and Technology [MOST-106-2119-M-001-030-MY3] and Academia Sinica of the Republic of China. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbagen.2018.11.009.
4. Conclusions The major findings of this study involve the demonstration of the unusually complex structural diversity of mt10251 and the significant changes in terms of structural populations and polymorphism by a single-base mutation or deletion of mt10251. For example, mt10251 forms only a small amount of monomeric G4 structures. However, mt10251-T17 and mt10251-d22 form primarily monomeric G4 structures, which may be of more biological relevance. The underlying mechanisms of different types of G4 formation from different singlebase mutations of mt10251 deserve further study. In addition, single-base mutations can potentially change not-G4 to G4 structures. In this work, we found 124 near-but-not-PQF sequences in reversed mtDNA of the NC_012920 reference sequence. Further studies predicted that 82 single-base mutated sequences are very likely to form G4 structures with two G-quartets in vitro. Given that mtDNA has been reported to mutate evolutionarily 10 to 17 times faster than nuclear DNA and that oxidative damage can increase point mutations of mtDNA [26], it is likely that additional PQF sequences due to singlebase mutations of near-but-not-PQF sequences could benefit the G4
References [1] M. Tarsounas, M. Tijsterman, Genomes and G-quadruplexes: for better or for worse, J. Mol. Biol. 425 (2013) 4782–4789. [2] D. Rhodes, H.J. Lipps, G-quadruplexes and their regulatory roles in biology, Nucleic Acids Res. 43 (2015) 8627–8637. [3] R. Hänsel-Hertsch, M. Di Antonio, S. Balasubramanian, DNA G-quadruplexes in the human genome: detection, functions and therapeutic potential, Nat. Rev. Mol. Cell Biol. 18 (2017) 279–284. [4] S. Balasubramanian, L.H. Hurley, S. Neidle, Targeting G-quadruplexes in gene promoters: a novel anticancer strategy? Nat. Rev. Drug Discov. 10 (2011) 261–275. [5] J. Dai, M. Carvera, D. Yang, Polymorphism of human telomeric quadruplex structures, Biochimie 90 (2008) 1172–1183. [6] A.T. Phan, Human telomeric G-quadruplex: structures of DNA and RNA sequences, FEBS J. 277 (2010) 1107–1117. [7] A. Siddiqui-Jain, C.L. Grand, D.J. Bearss, L.H. Hurley, Direct evidence for a Gquadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 11593–11598. [8] M. Marušiča, L. Hošnjakb, P. Krafčikovac, M. Poljakb, V. Viglaskyc, J. Plavec, The effect of single nucleotide polymorphisms in G-rich regions of high-risk human papillomaviruses on structural diversity of DNA, Biochim. Biophys. Acta 1861 ( (2017) 1229–1236.
424
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
[17] J. Damas, J. Carneiro, J. Gonçalves, J.B. Stewart, D.C. Samuels, A. Amorim, F. Pereira, Mitochondrial DNA deletions are associated with non-B DNA conformations, Nucleic Acids Res. 40 (2012) 7606–7621. [18] G. Wang, K.M. Vasquez, Effects of replication and transcription in DNA structurerelated genetic instability, Genes 8 (2017) 17. [19] A. Guedin, J. Gros, P. Alberti, J.L. Mergny, How long is too long? Effects of loop size on G-quadruplex stability, Nucleic Acids Res. 38 (2010) 7858–7868. [20] P. Agrawal, C. Lin, R.I. Mathad, M. Carver, D. Yang, The major G-quadruplex formed in the human BCL-2 proximal promoter adopts a parallel structure with a 13-nt loop in K+ solution, J. Am. Chem. Soc. 136 (2014) 1750–1753. [21] L. Li, L. Chen, J. Li, W. Zhang, Y. Liao, J. Chen, Z. Sun, Correlational study on mitochondrial DNA mutations as potential risk factors in breast cancer, Oncotarget 7 (2016) 31270–31283. [22] C.-T. Lin, T.-Y. Tseng, Z.-F. Wang, T.-C. Chang, Structural conversion of intramolecular and intermolecular G-quadruplexes of bcl2mid: the effect of potassium concentration and ion exchange, J. Phys. Chem. B 115 (2011) 2360–2370. [23] O. Kikin, L. D'Antonio, P.S. Bagga, QGRS Mapper: a web-based server for predicting G-quadruplexes in nucleotide sequences, Nucleic Acids Res. 34 (2006) W676–W682. [24] A. Bedrat, L. Lacroix, J.L. Mergny, Re-evaluation of G-quadruplex propensity with G4Hunter, Nucleic Acids Res. 44 (2016) 1746–1759. [25] N. Busto, P. Calvo, J. Santolaya, J.M. Leal, A. Guédin, G. Barone, T. Torroba, J.L. Mergny, B. García, Fishing G-quadruplex in solution with a perylene diimide derivative labeled with biotins, Chem. Eu. J. 24 (2018) 11292–11296. [26] H.A.L. Tuppen, E.L. Blakely, D.M. Turnbull, R.W. Taylor, Mitochondrial DNA mutations and human disease, Biochim. Biophys. Acta 1797 (2010) 113–128.
[9] S. De, F. Micho, DNA secondary structures and epigenetic determinants of cancer genome evolution, Nat. Struct. Mol. Biol. 18 (2011) 950–955. [10] R. Hänsel-Hertsch, D. Beraldi, S.V. Lensing, G. Marsico, K. Zyner, A. Parry, M. Di Antonio, J. Pike, H. Kimura, M. Narita, D. Tannahill, S. Balasubramanian, Gquadruplex structures mark human regulatory chromatin, Nat. Genet. 48 (2016) 1267–1272. [11] D.W. Dong, F. Pereira, S.P. Barrett, J.E. Kolesar, K. Cao, J. Damas, L.A. Yatsunyk, F.B. Johnson, B.A. Kaufman, Association of G-quadruplex forming sequences with human mtDNA deletion breakpoints, BMC Genomics 15 (2014) 677. [12] S.K. Bharti, J.A. Sommers, J. Zhou, D.L. Kaplan, J.N. Spelbrink, J.L. Mergny, R.M. Brosh Jr., DNA sequences proximal to human mitochondrial DNA deletion breakpoints prevalent in human disease form G-quadruplexes, a class of DNA structures inefficiently unwound by the mitochondrial replicative Twinkle helicase, J. Biol. Chem. 289 (2014) 29975–29993. [13] R.W. Taylor, D.M. Turnbull, Mitochondrial DNA mutations in human disease, Nat. Rev. Genet. 6 (2005) 389–402. [14] W.-C. Huang, T.-Y. Tseng, Y.-T. Chen, C.-C. Chang, Z.-F. Wang, C.-L. Wang, T.N. Hsu, P.-T. Li, C.-T. Chen, J.-J. Lin, P.-J. Lou, T.-C. Chang, Direct evidence of mitochondrial G-quadruplex DNA by using fluorescent anti-cancer agents, Nucleic Acids Res. 43 (2015) 10102–10113. [15] S. Lyonnais, A. Tarrés-Solé, A. Rubio-Cosials, A. Cuppari, R. Brito, J. Jaumot, R. Gargallo, M. Vilaseca, C. Silva, A. Granzhan, M.P. Teulade-Fichou, R. Eritja, M. Solà, The human mitochondrial transcription factor A is a versatile G-quadruplex binding protein, Sci. Rep. 7 (2017) 43992. [16] P.F. Chinnery, Mitochondrial disease in adults: what's old and what's new? EMBO Mol. Med. 7 (2015) 1503–1512.
425
Contents lists available at ScienceDirect
BBA - General Subjects journal homepage: www.elsevier.com/locate/bbagen
G-quadruplex formation by single-base mutation or deletion of mitochondrial DNA sequences I-Te Chua, Chia-Chuan Wua,b, Ta-Chau Changa, a b
T
⁎
Institute of Atomic and Molecular Sciences, Academia Sinica, Taipei 106, Taiwan, ROC Department of Agricultural Chemistry, National Taiwan University, Taipei 106, Taiwan, ROC
A R T I C LE I N FO
A B S T R A C T
Keywords: Mitochondrial DNA G-quadruplex Single-base mutation or deletion Structural diversity Structural variation Hairpin
Background: Mitochondrial DNA (mtDNA) mutations could lead to mitochondrial dysfunction, which plays a major role in aging, neurodegeneration, and cancer. Recently, we have highlighted G-quadruplex (G4) formation of putative G4-forming (PQF) mtDNA sequences in cells. Herein, we examine structural variation of G4 formation due to mutation of mtDNA sequences in vitro. Methods: The combined circular dichroism (CD), nuclear magnetic resonance (NMR), and polyacrylamide gel electrophoresis (PAGE) results provide complementary insights into the structural variation of the studied G-rich sequence and its mutants. Results: This study illustrates the structural diversity of mt10251, a G-rich mtDNA sequence with a 16-nt loop, ( GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG), including the coexistence of a hairpin structure and monomeric, dimeric, and tetrameric G4 structures of mt10251 in 20 mM K+ solution. Moreover, a single-base mutation of mt10251 can cause significant changes in terms of structural populations and polymorphism. In addition, singlebase mutations of near-but-not-PQF sequences can potentially change not-G4 to G4 structures. We further found 124 modified PQF sequences due to single-base mutations of near-but-not-PQF sequences in mtDNA. Conclusions: Single-base mutations of mt10251 could make significant changes in its structural variation and some single-base mutated sequences in mtDNA could form G4 structures in vitro. General significance: We illustrate the importance of single-base mutations of DNA sequences to the change of G4 formation in vitro. The use of single-base mutations by generating the fourth G-tract and followed by selection in shortening the longest loop size in the near-but-not-PQF sequences was conducted for the G4 formation.
1. Introduction G-quadruplexes (G4s) are unique four-stranded nucleic acid structures formed by guanine-rich (G-rich) sequences under physiological conditions. The importance of G4s comprises not only the protection of chromosome ends, but also the regulation of the expression of certain genes [1–3]. Thus, G4s have become potential targets for cancer therapeutics [2,4]. However, some G-rich sequences can adopt various G4 structures and possibly coexist in mixtures. For example, slight variations in human telomeric sequences can result in different G4 structures [5,6]. In addition, the sequence (TG4AG3)2TG4A2G2 in the c-myc gene promoter formed both intramolecular and intermolecular G4 structures in K+ solution [7]. Plavec et al. have recently studied the effect of single-base mutation in G-rich sequences of human papillomaviruses on the structural diversity of DNA [8]. To our knowledge, structural variation of G4 formation among monomers, dimers, and tetramers of a
⁎
native G-rich sequence caused by a single-base difference has not yet been reported. While considerable efforts have been made to demonstrate the biological role of nuclear G4 formation [9,10], the study of mitochondrial G4 formation is still in its nascent stages [11,12]. With reference to the large size of nuclear DNA (3.3 × 109 bps), human mitochondrial DNA (mtDNA) is extremely small (16,569 bps) [13]. Dong et al. [11] identified 9 PQF sequences with four tracts of three consecutive guanines and 178 PQF sequences with four tracts of two consecutive guanines for each length < 33 nt. We highlighted the existence of human mtDNA G4 structures in cells [14]. In addition, Lyonnais et al. [15] found that the human mitochondrial transcription factor A extracted from mitochondria could strongly bind to diverse G4 structures in vitro. It is noteworthy that mtDNA mutations, including point mutations, deletions, inversions, and copy number variations, can lead to mitochondrial dysfunction, which plays a major role in the pathogenesis
Corresponding author. E-mail address: [email protected] (T.-C. Chang).
https://doi.org/10.1016/j.bbagen.2018.11.009 Received 5 July 2018; Received in revised form 26 October 2018; Accepted 13 November 2018 Available online 26 November 2018 0304-4165/ © 2018 Published by Elsevier B.V.
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
2.3. Nuclear magnetic resonance (NMR) spectroscopy
of aging, neurodegeneration, and diseases such as cancer [16]. Previous studies have suggested that mtDNA contains numerous sequences with the potential to form the non-B DNA structures, which could be associated with mtDNA deletions [11,17]. In addition, the non-B DNA structures have been implicated in genetic instability and disease development [18]. Among non-B forms of DNA, G4 structures have recently attracted more attention [11,12,14,15]. Recently, we utilized CD to study 10 PQF mtDNA sequences each with a length < 33 nt [14]. Among them, only mt15653 (GGGTTAAT CGTGTGACCGCGGTGGCTGG) was unable to form G4 with a possible two G-quartets, while the others could form G4s in a 150 mM K+ solution. It is notable that mt15653 has a 16-nt-long loop. In contrast, mt10251 (GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG), located at positions 10251–10280 in reversed mtDNA of the NC_012920 reference sequence, can form G4s with a possible three G-quartets, even though it also contains a 16-nt-long loop. However, the G4 formation type of mt10251 in K+ solution is unclear. Previously, Mergny et al. reported possible G4 formation with longer loops comprising up to 30 nt of thymine [19]. Yang et al. demonstrated that the 1245 G-runs of BCL-2 sequence adopted a parallel G4 structure with a 13-nt loop in K+ solution [20]. In this work, the combination of NMR and PAGE has revealed the coexistence of monomeric, dimeric, and tetrameric G4 structures of mt10251 in 20 mM K+ solution. Moreover, a single-base mutation or deletion of mt10251 could cause significant changes in terms of structural populations and polymorphism. For example, substitution of G2 by T2 or G6 by T6 in mt10251 could change a G4 to a not-G4 structure, while the substitution of G25 by T25 or G29 by T29 in mt10251 could maintain tetrameric G4 formation. The former finding stimulated our interest to examine the possible G4 formations from non-PQF sequences with single-base mutations in mtDNA. Here, we name such non-PQF sequences with three G-tracts that can be converted to PQF sequences with four G-tracts by single-base mutations as near-but-not-PQF sequences. Since the point mutation is possible in mtDNA, we have screened mtDNA sequence and found 124 such near-but-not-PQF sequences. Data analyses of 20 PQF and 10 modified PQF sequences suggested that more than half of them can change not-G4 to G4 structures after single-base mutations in vitro. Notably, recent studies have suggested that two mutants of mt10251 with a single-base difference, mt10251-T24 by substitution of G24 with T24 and mt10251-d22 by deletion of A22, are potential risk factors in breast cancer [21].
All NMR experiments were performed on Bruker AVIII 500 MHz spectrometers equipped with a prodigy probehead at 25 °C. One-dimensional (1D) imino proton NMR spectra were recorded using a WATERGATE for water suppression. The strand concentrations of the NMR samples were typically 100 μM, containing 10% D2O in 10 mM Tris (pH 7.5) or 20 mM K+ conditions, with an internal reference of 0.01 mM DSS (4,4-dimethyl-4-silapentane-1-sulfonic acid). 2.4. Polyacrylamide gel electrophoresis (PAGE) PAGE was performed using 16% polyacrylamide gel for samples without K+ and with the addition of 20 mM K+ in both the samples and gel solution. The DNA concentration for each sample was 100 μM, that is, the same concentration used for CD and NMR. PAGE was conducted at 150 V for 3.5 h at 4 °C. The gels were then photographed under ultraviolet (UV) light at 254 nm using a digital camera. 3. Results and discussion 3.1. Structural diversity of G4 formations in mt10251 The CD spectra of mt10251 showed a growth of the 265-nm band after the addition of 20 and 150 mM K+ solution (Fig. 1A). The detection of positive CD bands at 265 nm is a typical CD pattern for parallel G4 structures. NMR spectra of mt10251 showed imino proton signals in the region near 13 ppm in Tris buffer (Fig. 1B), indicating the presence of hairpin structure. After the addition of 20 and 150 mM K+, the detection of imino proton signals in the region of 10.5–12 ppm indicated the formation of G4 structures. Notably, the concomitance of imino proton signals near 13 ppm and 10.5–12 ppm in mt10251
A
C
(degxmol -1xcm 2 )
-5
8
mt10251
6 4 2 0
10 mM Tris + 20 mM K + overnight + 150 mM K+ overnight
-2 -4
2. Materials and methods
1 2 3
240 280 320 Wavelength (nm)
2.1. DNA preparation
B
DNA oligonucleotides were purchased from Bio Basic (Ontario, Canada) and dissolved in 10 mM Tris (pH 7.5). They were then subjected to heat-denaturation at 95 °C for 10 min and slowly annealed to room temperature at a rate of 1 °C/min. The annealed oligonucleotides were stored at 4 °C overnight until the further experiments. The DNA concentrations were determined using a UV–Vis absorption spectrometer (Implen, Germany).
mt10251 + 150 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
2.2. Circular dichroism (CD) spectroscopy
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
CD experiments were conducted using a spectropolarimeter (J-815, Jasco, Japan) with a bandwidth of 2 nm, at a scan speed of 50 nm/min and a step resolution of 0.2 nm over a spectral range of 210–350 nm. The DNA concentration in each sample was 100 μM (dissolved in 10 mM Tris (pH 7.5)), and a stock solution of 3 M KCl (Sigma-Aldrich, USA) was added to the DNA samples to attain a final K+ concentration of 20 mM. The observed signals were baseline subtracted.
Fig. 1. Structural diversity of G4 formation in mt10251. (A) CD spectra and (B) NMR spectra of mt10251 in 10 mM Tris buffer, after the addition of 20 mM K+ and 150 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24 (T2AG3)4, HT48 (T2AG3)8, and HT96 (T2AG3)16 (lane 1), mt10251 in 20 mM K+ solution overnight (lane 2), and in 150 mM K+ solution overnight (lane 3). The same DNA concentration of 100 μM was used in the experiments of CD, NMR, and PAGE of this work. 419
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
suggested the coexistence of hairpin and G4 structures in K+ solution. PAGE was conducted to verify possible monomer, dimer, and tetramer conformations; PAGE results showed several discernible bands with dimeric and tetrameric G4s of mt10251 after the addition of 20 and 150 mM K+ (Fig. 1C). The use of an mt10251 concentration of 100 μM for PAGE and CD was simply to match the DNA concentration necessary for use in NMR experiments. Thus, significant intermolecular structures found at the mt10251 concentration of 100 μM were possibly due to high concentrations of DNA. Here, the gel assays for 20 μM and 100 μM mt10251 showed that the higher the mt10251 concentration, the more was the high-order population (Fig. S1). Nevertheless, we also detected the intermolecular G4 structures at 20 μM mt10251 even in 20 mM K+ solution. In addition, the gel assays of mt10251 showed that an increasing amount of the intermolecular population was detected at higher K+ concentrations (Fig. S1). Such finding was consistent with a previous study of the structural conversion of intramolecular and intermolecular G4 formations of bcl2mid [22]. Since we were interested in the intramolecular G4 formation for more biological relevance, the following studies were conducted using 20 mM K+ solution.
20 mM K+, whereas that of mt10251-T25 and mt10251-T29 showed two additional bands with tetrameric G4 formation (Fig. 2B). It appears that the first and second G-tracts play critical roles for G4 formation.
3.3. Search of single-base mutations in mt10251 to change multiple G4 structures to primarily monomeric G4 structures Given that the tetrameric G4 structure is unlikely to be formed in cells, we examined whether the mt10251-M1 sequence can favor the formation of monomeric G4 structures by substituting A23 and A27 with T23 and T27, and whether the mt10251-M2 sequence can favor the formation of monomeric G4 structures by substituting C15-C17 with T15T17. The CD spectra of mt10251-M1 and mt10251-M2 showed the growth of the 265-nm band after the addition of 20 mM K+ (Fig. S3A), suggesting parallel G4 formation. Their NMR spectra showed no imino proton signals at 13 ppm in Tris buffer (Fig. S3B), implying the absence of hairpin structure. After the addition of 20 mM K+, the NMR spectrum of mt10251-M2 showed fine imino proton signals in the 10.5–12 ppm region. The NMR spectrum of mt10251-M1 showed imino proton signals different from those of mt10251-M2, implying that they had formed different G4 structures. In addition, the PAGE results showed a major band of mt10251-M1 with dimeric G4 and a major band of mt10251-M2 with monomeric G4 (Fig. S3C). Since the major component of mt10251-M2 was a monomeric G4 structure, we were interested in examining the effect of a single C-base mutation of C15-C17 in mt10251 (substituting C15 with T15 in mt10251T15, G16 with T16 in mt10251-T16, and G17 with T17 in mt10251-T17, together with a single C-base deletion of C17 in mt10251-d17) on G4 formation. Their NMR spectra showed no imino proton signals near 13 ppm (Fig. 3A), indicating the absence of a hairpin structure. After the addition of 20 mM K+, imino proton signals in the 10.5–12 ppm region were similar for mt10251-T15, mt10251-T16, and mt10251-d17 and different for mt10251-T17 (Fig. 3A). Their CD spectra also showed substantial growth of the 265-nm band after the addition of 20 mM K+ (Fig. S4). The PAGE results showed a major dimer band and a weak monomer band for mt10251-T15, mt10251-T16, and mt10251-d17, and a major smeared monomer band for mt10251-T17 (Fig. 3B). To simulate the NMR spectrum of mt10251, we summed up the NMR spectra of mt10251-T29 for tetrameric, mt10251-T16 for dimeric, and mt10251-T17 for monomeric formations in a ratio of 5:3:1 based
3.2. Single G-base substitutions in mt10251 to determine hairpin formation To examine the possible pairings between the C-tract and each Gtract for hairpin formation, we studied numerous mt10251 mutants by substituting the middle G-base with a T-base, specifically, G2 with T2 in mt10251-T2, G6 with T6 in mt10251-T6, G25 with T25 in mt10251-T25, and G29 with T29 in mt10251-T29, to eliminate one G-tract in each sequence. The absence of imino proton signals near 13 ppm in the NMR spectrum of mt10251-T25 suggested that the G-tract of G24-G26 was involved in the hairpin formation of mt10251 (Fig. 2A). After addition of 20 mM K+, imino proton signals of mt10251-T2 and mt10251-T6 in the 10.5–12 ppm region were absent, indicating no G4 formation. Notably, the imino proton signals of mt10251-T25 and mt10251-T29 in this region were similar to those of mt10251 (Fig. 2A). Consistent with the NMR results, the CD spectra of mt10251-T2 and mt10251-T6 showed no appreciable CD band change, whereas those of mt10251T25 and mt10251-T29 showed a marked growth of the 265-nm band after the addition of 20 mM K+ (Fig. S2). PAGE analysis of mt10251-T2 and mt10251-T6 showed no additional bands after the addition of
A
B
mt10251-T2
mt10251-T6
+ 20 mM K+ overnight
+ 20 mM K+ overnight
1 2 3 4 5 10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
mt10251-T25
mt10251-T29
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 2. Single G-base substitution in mt10251. (A) NMR spectra of mt10251-T2, mt10251-T6, mt10251-T25, and mt10251-T29 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) PAGE assays of mt10251 (lane 1), mt10251-T2 (lane 2), mt10251-T6 (lane 3), mt10251-T25 (lane 4), and mt10251-T29 (lane 5) in 20 mM K+ solution overnight. 420
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
A mt10251-T15
mt10251-T16
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
mt10251-T17
mt10251-d17
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
B
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
C Mixed (5:3:1)
1 2 3 4 5
mt10251-T29 20 mM K+ mt10251-T16 20 mM K+ mt10251-T17 20 mM K+ mt10251 20 mM K+ 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 3. Single-base mutations in mt10251 to change multiple G4 structures to primarily monomeric G4 structures. (A) NMR spectra of mt10251-T15, mt10251-T16, mt10251-T17 and mt10251-d17 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) PAGE assays of mt10251 (lane 1), mt10251-T15 (lane 2), mt10251-T16 (lane 3), mt10251-T17 (lane 4), and mt10251-d17 (lane 5) in 20 mM K+ solution overnight. (C) Summing up NMR spectra of mt10251-T29 for tetramer, mt10251-T16 for dimer, and mt10251-T17 for monomer with the ratio of 5:3:1 based on the gel results to mimic the NMR spectrum of mt10251. Table 1 The sequences of mt10251 and its mutants studied in this work with weak (W), moderate (M), and strong (S) components based on the PAGE results. Name
Sequence
Monomer
Dimer
Tetramer
mt10251 mt10251-M1 mt10251-M2 mt10251-T2 mt10251-T6 mt10251-T25 mt10251-T29 mt10251-T15 mt10251-T16 mt10251-T17 mt10251-d17 mt10251-T24 mt10251-d22
GGGTGGGAGTAGTTCCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTATGGGTGGG GGGTGGGAGTAGTTTTTTGCTAAGGGAGGG GTGTGGGAGTAGTTCCCTGCTAAGGGAGGG GGGTGTGAGTAGTTCCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTAAGTGAGGG GGGTGGGAGTAGTTCCCTGCTAAGGGAGTG GGGTGGGAGTAGTTTCCTGCTAAGGGAGGG GGGTGGGAGTAGTTCTCTGCTAAGGGAGGG GGGTGGGAGTAGTTCCTTGCTAAGGGAGGG GGGTGGGAGTAGTTCC↓TGCTAAGGGAGGG GGGTGGGAGTAGTTCCCTGCTAATGGAGGG GGGTGGGAGTAGTTCCCTGCT↓AGGGAGGG
W
W S
M
– –
– – S S
S – –
W W S W S
S S S S W
tetrameric intensity of each sequence obtained from the PAGE assays.
on the PAGE results (Fig. 3C). The agreement between the simulated and experimental results supported the unusual complexity of conformational changes from a hairpin structure to multiple monomeric, dimeric, tetrameric G4 structures of mt10251 after the addition of 20 mM K+. Importantly, a single-base mutation of mt10251 can cause significant changes in terms of structural populations and polymorphism. Table 1 listed the sequences of mt10251 and its mutants studied in this work and the relative monomeric, dimeric, and
3.4. Conversion of near-but-not-PQF sequences to PQF sequences by singlebase mutations Because a single-base mutation can change a G4 to a not-G4 structure, we were interested in examining the possibility of G4 formation due to a single-base mutation of a near-but-not-PQF sequence, which 421
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
C
Molar ellipticity [ ] x10 -5 (degxmol -1xcm2 )
A 4 3 2 1 0 -1 -2 10 mM Tris + 20 mM K+ overnight -3
mt5027
2 1 0 -1
240 280 320 Wavelength (nm)
mt5027-G18
4
mt5027-d18
3
1 2 3 4
2 1 0 -1 10 mM Tris + 20 mM K+overnight
10 mM Tris + 20 mM K+overnight -2
240 280 320 Wavelength (nm)
240 280 320 Wavelength (nm)
B mt5027
mt5027-G18
mt5027-d18
+ 20 mM K+ overnight
10 mM Tris 12.5 12.0 11.5 11.0 10.5 ppm
12.5 12.0 11.5 11.0 10.5 ppm
12.5 12.0 11.5 11.0 10.5 ppm
Fig. 4. Single-base mutations of near-but-not-PQF mt5027 sequence. (A) CD spectra and (B) NMR spectra of mt5027, mt5027-G18 and mt5027-d18 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24, HT48, and HT96 (lane 1), mt5027 (lane 2), mt5027-G18 (lane 3), and mt5027-d18 (lane 4) in 20 mM K+ solution overnight.
gives rise to a PQF sequence. We first studied a native mtDNA sequence of mt5027 and two mutants with a single-base substitution of T18 by G18 for mt5027-G18 and a single-base deletion of T18 for mt5027-d18. The CD spectra showed no appreciable difference for mt5027, but a significant difference for mt5027-G18 and mt5027-d18 after the addition of 20 mM K+ (Fig. 4A). The detection of two positive CD bands near 240 and 295 nm, together with a negative CD band near 265 nm, represents a typical pattern for antiparallel G4 structures. Consistent with the CD results, imino proton signals in the 10.5–12.5 ppm region due to G4 structures were not detected in the NMR spectrum of mt5027, but were detected in the spectra of mt5027-G18 and mt5027-d18 after the addition of 20 mM K+ (Fig. 4B). Notably, more imino proton signals were detected for mt5027-G18 than for mt5027-d18, implying that mt5027-G18 could have formed at least two different G4 structures. The PAGE results for mt5027, mt5027-G18, and mt5027-d18 showed a single monomer band (Fig. 4C). Although PAGE was not able to distinguish the multiple G4 structures of mt5027-G18, it indicated the absence of intermolecular G4 formation. Indeed, single-base mutations or deletions of near-but-not-PQF sequences could benefit the G4 formation in vitro. We further investigated 9 additional near-but-not-PQF sequences together with their single-base mutations to characterize their possible G4 formations in 20 mM K+ solution (Table 2 and Fig. S5). Indeed, NMR spectra showed no imino proton signals in the 10.5–12.0 ppm region for most of these near-but-not-PQF sequences, but a number of imino proton signals in this region for most of the single-base mutated sequences. Exceptions include the presence of a number of fine peaks of imino proton signals in the 10.5–12.0 ppm region form mt1363 sequence and the absence of imino proton signals in this region from mt10550-G5 mutant, which deserves further study. Of interest was that such imino proton signals in this region were not detected in mt2080G7 with substitution of A7 by G7, but a number of fine peaks in this region were detected in mt2080-G14 with substitution of T14 by G14. It is not clear whether this difference is due to an 8-nt loop existed in mt2080-G7 but a 6-nt loop existed in mt2080-G14 or different loop sequences. Nevertheless, this finding suggested that generating the fourth G-tract and shortening the longest loop size in a near-but-notPQF sequence by a single-base mutation could benefit the G4
Table 2 Near-but-not-PQF sequences together with their single-base mutations and 4 PQF sequences studied in this work. Name
Sequence
mt5027 mt5027-G18 mt5027-d18 mt2080 mt2080-G7 mt2080-G14 mt1363 mt1363-G13 mt2971 mt2971-G5 mt5054 mt5054-G13 mt6797 mt6797-G11 mt7908 mt7908-G12 mt10550 mt10550-G5 mt12077 mt12077-G2 mt14229 mt14229-G7 mt1067 mt1563 mt10213 mt13056
GGAAGGGGTAGGCTATGTG GGAAGGGGTAGGCTATGGG GGAAGGGGTAGGCTATG↓G GGAATGATGGTTGTCTTTGG GGAATGGTGGTTGTCTTTGG GGAATGATGGTTGGCTTTGG GGGATGGCGGATAG GGGATGGCGGATGG GGCGCTTGTCAGGGAGG GGCGGTTGTCAGGGAGG GGGGGTTGAGAATGAGTGTGAGG GGGGGTTGAGAAGGAGTGTGAGG GGAAATGGTGAAGGG GGAAATGGTGGAGGG GGGTGGTGATTAGTCGG GGGTGGTGATTGGTCGG GGCTCGAATAAGGAGG GGCTGGAATAAGGAGG GACGGGTTGGGCCAGGGG GGCGGGTTGGGCCAGGGG GGCTTATGCGGAGG GGCTTAGGCGGAGG GGTCGCCTAGGAGGTCTGG GGCGCCATTGGCGTGAAGGTAGCGG GGTGTAAGGAGAAGATGGTTAGG GGGTGATGGTAGATGTGGCGGG
formation.
3.5. Estimation of the number of near-but-not-PQF sequences for potential G4 formation in mtDNA To estimate the number of near-but-not-PQF sequences in mtDNA, we first studied four native PQF sequences with each containing four tracts of GG and a loop size of 7 nt. The NMR results showed that mt1067 and mt1563 did not form G4 structure, while mt10213 and mt13056 could form G4 structures (Table 2 and Fig. 5A). For simplicity, 422
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
A mt1067
mt1563
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5ppm
mt10213
mt13056
+ 20 mM K+ overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris 14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
B The number of the longest loop base (nmax )
10 G4 (In this work) Not G4 (In this work) G4 (Reference 24) Not G4 (Reference 24)
8 6 4 2 0 0.0
0.5
1.0
1.5 2.0 G4Hscore
2.5
3.0
Fig. 5. NMR results used in characterizing the G4 formation. (A) NMR spectra of mt1067, mt1563, mt10213 and mt13056 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (B) The plot of 30 PQF sequences by using the G4Hscore and the number of the longest loop bases to analyze the possible G4 formation with two G-quartets.
of the longest loop bases (nmax) to analyze the possible G4 formation with two G-quartets (Fig. 5B). The plot comprised 10 sequences obtained from ref. 24 and 20 sequences studied in this work. The results suggested that those PQF sequences characterized with either nmax ≤ 4 or G4Hscore ≥ 1.2 have very high possibility to form G4 structures, while those sequences characterized with nmax > 4 and G4Hscore < 1.2 have low possibility to form G4 structures. Accordingly, there are 82 out of 118 modified mtDNA sequences which are likely to form G4 structures with two G-quartets in vitro.
we used the loop size between 1 nt and 7 nt in our estimation of PQF sequences with four tracts of GG. Using the QGRS program to analyze the PQF sequence in mtDNA (NC_012920) [23], it was found that there were 9 PQF sequences with four tracts of GGG constrained by loop sizes between 1 and 16 nt, and 79 PQF sequences with four tracts of GG constrained by loop sizes between 1 and 7 nt (Table S1). Considering the possible overlapping PQF sequences, we only counted one sequence. We further found 124 near-but-not-PQF sequences in mtDNA (NC_012920), which contain 6 modified sequences with possible three G-quartets and 118 modified sequences with possible two G-quartets, as listed in Table S2. Recently, a G4Hunter algorithm that took into account the G-richness and G-skewness of a given sequence was developed to evaluate the G4 propensity of 209 sequences from the human mtDNA in vitro. These mtDNA sequences were further classified as stable G4 (71), unstable G4 (63), and not-G4 (75) based on G4Hscores and biophysical results [24]. Here we used the G4Hunter program to calculate the G4Hscore for each sequence and then made a plot by using the G4Hscore and the number
3.6. G4 structures of specific mutants of mt10251: mt10251-d22 and mt10251-T24 We have studied the G4 formations of mt10251-T24 and mt10251d22 sequences because they are possible risk factors in breast cancer [21]. The CD spectra of mt10251-T24 and mt10251-d22 showed growth of the 265-nm band after the addition of 20 mM K+ (Fig. 6A), implying the G4 formation. Their NMR spectra showed no imino proton 423
BBA - General Subjects 1863 (2019) 418–425
4
mt10251-T24
3 2 1 0
-1
10 mM Tris + 20 mM K+overnight
-2 240
Molar ellipticity [ ] x10-5 (degxmol-1xcm 2 )
A
Molar ellipticity [ ] x10-5 (degxmol-1xcm 2)
I.-T. Chu et al.
C
8
mt10251-d22
6
1 2 3 4
4 2 0 -2
10 mM Tris + 20 mM K+ overnight
-4
280 320 Wavelength (nm)
240 280 320 Wavelength (nm)
B mt10251-T24
mt10251-d22
+ 20 mM K overnight
+ 20 mM K+ overnight
10 mM Tris
10 mM Tris
+
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
14.5 14.0 13.5 13.0 12.5 12.0 11.5 11.0 10.5 ppm
Fig. 6. Single-base mutation or deletion in mt10251. (A) CD spectra and (B) NMR spectra of mt10251-T24 and mt1025-d22 in 10 mM Tris buffer and after the addition of 20 mM K+ overnight at 25 °C. (C) PAGE assays of marker bands of HT24, HT48, and HT96 (lane 1), mt10251 (lane 2), mt10251-T24 (lane 3), mt10251d22 (lane 4) in 20 mM K+ solution overnight.
formation, which may happen in mtDNA. Although we have not found such mutated sequences in mitochondria at present, we are not able to totally eliminate the possible association of such mutations with mitochondrial dysfunction.
signals at 13 ppm in Tris buffer (Fig. 6B), implying the absence of hairpin structure. After the addition of 20 mM K+, the NMR spectrum of mt10251-T24 was very similar to that of mt10251-T25, while the NMR spectrum of mt10251-d22 showed different imino proton signals from those of mt10251-T24, implying that they formed different G4 structures. In addition, the PAGE results showed two bands of mt10251-T24 with tetrameric G4 s and a major band of mt10251-d22 with monomeric G4, together with a weak tetrameric G4 band (Fig. 6C). Although the tetrameric G4 formation of mt10251-T24 is unlikely prevalent in mitochondria, the monomeric G4 formation of mt10251-d22 may be accessible in mitochondria. At present, it is not clear whether the potential risk of mt10251-d22 in breast cancer is correlated with its monomeric G4 formation. The possible correlation of mtDNA mutations to G4-related diseases deserves further study. With reference to a recently developed method [15,25], it is possible to selectively pull down G4 structures from isolated mtDNA for future study.
Conflict of interest The authors declare no conflict of interest. Acknowledgements This work was supported by the Ministry of Science and Technology [MOST-106-2119-M-001-030-MY3] and Academia Sinica of the Republic of China. Appendix A. Supplementary data Supplementary data to this article can be found online at https:// doi.org/10.1016/j.bbagen.2018.11.009.
4. Conclusions The major findings of this study involve the demonstration of the unusually complex structural diversity of mt10251 and the significant changes in terms of structural populations and polymorphism by a single-base mutation or deletion of mt10251. For example, mt10251 forms only a small amount of monomeric G4 structures. However, mt10251-T17 and mt10251-d22 form primarily monomeric G4 structures, which may be of more biological relevance. The underlying mechanisms of different types of G4 formation from different singlebase mutations of mt10251 deserve further study. In addition, single-base mutations can potentially change not-G4 to G4 structures. In this work, we found 124 near-but-not-PQF sequences in reversed mtDNA of the NC_012920 reference sequence. Further studies predicted that 82 single-base mutated sequences are very likely to form G4 structures with two G-quartets in vitro. Given that mtDNA has been reported to mutate evolutionarily 10 to 17 times faster than nuclear DNA and that oxidative damage can increase point mutations of mtDNA [26], it is likely that additional PQF sequences due to singlebase mutations of near-but-not-PQF sequences could benefit the G4
References [1] M. Tarsounas, M. Tijsterman, Genomes and G-quadruplexes: for better or for worse, J. Mol. Biol. 425 (2013) 4782–4789. [2] D. Rhodes, H.J. Lipps, G-quadruplexes and their regulatory roles in biology, Nucleic Acids Res. 43 (2015) 8627–8637. [3] R. Hänsel-Hertsch, M. Di Antonio, S. Balasubramanian, DNA G-quadruplexes in the human genome: detection, functions and therapeutic potential, Nat. Rev. Mol. Cell Biol. 18 (2017) 279–284. [4] S. Balasubramanian, L.H. Hurley, S. Neidle, Targeting G-quadruplexes in gene promoters: a novel anticancer strategy? Nat. Rev. Drug Discov. 10 (2011) 261–275. [5] J. Dai, M. Carvera, D. Yang, Polymorphism of human telomeric quadruplex structures, Biochimie 90 (2008) 1172–1183. [6] A.T. Phan, Human telomeric G-quadruplex: structures of DNA and RNA sequences, FEBS J. 277 (2010) 1107–1117. [7] A. Siddiqui-Jain, C.L. Grand, D.J. Bearss, L.H. Hurley, Direct evidence for a Gquadruplex in a promoter region and its targeting with a small molecule to repress c-MYC transcription, Proc. Natl. Acad. Sci. U. S. A. 99 (2002) 11593–11598. [8] M. Marušiča, L. Hošnjakb, P. Krafčikovac, M. Poljakb, V. Viglaskyc, J. Plavec, The effect of single nucleotide polymorphisms in G-rich regions of high-risk human papillomaviruses on structural diversity of DNA, Biochim. Biophys. Acta 1861 ( (2017) 1229–1236.
424
BBA - General Subjects 1863 (2019) 418–425
I.-T. Chu et al.
[17] J. Damas, J. Carneiro, J. Gonçalves, J.B. Stewart, D.C. Samuels, A. Amorim, F. Pereira, Mitochondrial DNA deletions are associated with non-B DNA conformations, Nucleic Acids Res. 40 (2012) 7606–7621. [18] G. Wang, K.M. Vasquez, Effects of replication and transcription in DNA structurerelated genetic instability, Genes 8 (2017) 17. [19] A. Guedin, J. Gros, P. Alberti, J.L. Mergny, How long is too long? Effects of loop size on G-quadruplex stability, Nucleic Acids Res. 38 (2010) 7858–7868. [20] P. Agrawal, C. Lin, R.I. Mathad, M. Carver, D. Yang, The major G-quadruplex formed in the human BCL-2 proximal promoter adopts a parallel structure with a 13-nt loop in K+ solution, J. Am. Chem. Soc. 136 (2014) 1750–1753. [21] L. Li, L. Chen, J. Li, W. Zhang, Y. Liao, J. Chen, Z. Sun, Correlational study on mitochondrial DNA mutations as potential risk factors in breast cancer, Oncotarget 7 (2016) 31270–31283. [22] C.-T. Lin, T.-Y. Tseng, Z.-F. Wang, T.-C. Chang, Structural conversion of intramolecular and intermolecular G-quadruplexes of bcl2mid: the effect of potassium concentration and ion exchange, J. Phys. Chem. B 115 (2011) 2360–2370. [23] O. Kikin, L. D'Antonio, P.S. Bagga, QGRS Mapper: a web-based server for predicting G-quadruplexes in nucleotide sequences, Nucleic Acids Res. 34 (2006) W676–W682. [24] A. Bedrat, L. Lacroix, J.L. Mergny, Re-evaluation of G-quadruplex propensity with G4Hunter, Nucleic Acids Res. 44 (2016) 1746–1759. [25] N. Busto, P. Calvo, J. Santolaya, J.M. Leal, A. Guédin, G. Barone, T. Torroba, J.L. Mergny, B. García, Fishing G-quadruplex in solution with a perylene diimide derivative labeled with biotins, Chem. Eu. J. 24 (2018) 11292–11296. [26] H.A.L. Tuppen, E.L. Blakely, D.M. Turnbull, R.W. Taylor, Mitochondrial DNA mutations and human disease, Biochim. Biophys. Acta 1797 (2010) 113–128.
[9] S. De, F. Micho, DNA secondary structures and epigenetic determinants of cancer genome evolution, Nat. Struct. Mol. Biol. 18 (2011) 950–955. [10] R. Hänsel-Hertsch, D. Beraldi, S.V. Lensing, G. Marsico, K. Zyner, A. Parry, M. Di Antonio, J. Pike, H. Kimura, M. Narita, D. Tannahill, S. Balasubramanian, Gquadruplex structures mark human regulatory chromatin, Nat. Genet. 48 (2016) 1267–1272. [11] D.W. Dong, F. Pereira, S.P. Barrett, J.E. Kolesar, K. Cao, J. Damas, L.A. Yatsunyk, F.B. Johnson, B.A. Kaufman, Association of G-quadruplex forming sequences with human mtDNA deletion breakpoints, BMC Genomics 15 (2014) 677. [12] S.K. Bharti, J.A. Sommers, J. Zhou, D.L. Kaplan, J.N. Spelbrink, J.L. Mergny, R.M. Brosh Jr., DNA sequences proximal to human mitochondrial DNA deletion breakpoints prevalent in human disease form G-quadruplexes, a class of DNA structures inefficiently unwound by the mitochondrial replicative Twinkle helicase, J. Biol. Chem. 289 (2014) 29975–29993. [13] R.W. Taylor, D.M. Turnbull, Mitochondrial DNA mutations in human disease, Nat. Rev. Genet. 6 (2005) 389–402. [14] W.-C. Huang, T.-Y. Tseng, Y.-T. Chen, C.-C. Chang, Z.-F. Wang, C.-L. Wang, T.N. Hsu, P.-T. Li, C.-T. Chen, J.-J. Lin, P.-J. Lou, T.-C. Chang, Direct evidence of mitochondrial G-quadruplex DNA by using fluorescent anti-cancer agents, Nucleic Acids Res. 43 (2015) 10102–10113. [15] S. Lyonnais, A. Tarrés-Solé, A. Rubio-Cosials, A. Cuppari, R. Brito, J. Jaumot, R. Gargallo, M. Vilaseca, C. Silva, A. Granzhan, M.P. Teulade-Fichou, R. Eritja, M. Solà, The human mitochondrial transcription factor A is a versatile G-quadruplex binding protein, Sci. Rep. 7 (2017) 43992. [16] P.F. Chinnery, Mitochondrial disease in adults: what's old and what's new? EMBO Mol. Med. 7 (2015) 1503–1512.
425