Quantification of Chemical and Mechanical Effects on the Formation of

Subscriber access provided by UNIV OF NEWCASTLE

Article

Quantification of Chemical and Mechanical Effects on the Formation of G-quadruplex and i-Motif in Duplex DNA Sangeetha Selvam, Shankar Mandal, and Hanbin Mao Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.7b00279 • Publication Date (Web): 24 Jul 2017 Downloaded from http://pubs.acs.org on July 25, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biochemistry is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Quantification of Chemical and Mechanical Effects on the Formation of G-quadruplex and i-Motif in Duplex DNA Sangeetha Selvam,1 Shankar Mandal,1 Hanbin Mao1,* 1

Department of Chemistry and Biochemistry, Kent State University, Kent, Ohio, 44242, USA *

Corresponding author: HM, [email protected] (330-672-9380)

1 Environment ACS Paragon Plus

Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Abstract Formation of biologically significant tetraplex DNA species, such as G-quadruplexes and i-motifs, is affected by chemical (ions and pH) and mechanical (superhelicity [σ] and molecular crowding) factors.

Due to the extremely challenging experimental conditions, relative im-

portance of these factors on tetraplex folding is unknown. In this work, we quantitatively evaluated the chemical and mechanical effects on the population dynamics of DNA tetraplexes in Insulin Linked Polymorphic Region using magneto-optical tweezers. By mechanically unfolding individual tetraplexes, we found that ions and pH have the largest effects on the formation of Gquadruplex and i-motif, respectively. Interestingly, superhelicity has the second largest effect followed by molecular crowding condition. While chemical effects are specific to tetraplex species, mechanical factors show generic influences. The predominant effect of chemical factors can be attributed to the fact that they directly change the stability of a specific tetraplex whereas the mechanical factors, superhelicity in particular, reduce the stability of the competing species by changing the kinetics of the melting and annealing of the duplex DNA template in a nonspecific manner. The substantial dependence of tetraplexes on superhelicity provides strong support that DNA tetraplexes can serve as topological sensors to modulate fundamental cellular processes such as transcription.


Page 2 of 28

Page 3 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

Introduction G-quadruplex and i-motif are tetraplex DNA structures formed in guanine (G) and cytosine (C) rich regions, respectively. While G-quadruplex contains a stack of G-quartets, each of which is made of four guanine residues orchestrated into a planar structure by Hoogsteen bonds, i-motif consists of a stack of hemiprotonated cytosine-cytosine pairs (C-CH+) (see Figure 1). Recently, it has been demonstrated that either G-quadruplex or i-motif can regulate transcription of oncogenes such as bcl-2 (1-3) and c-Myc (4). Formation of either G-quadruplex or i-motif requires appropriate chemical conditions. While G-quadruplex prefers potassium cations, i-motif forms at slightly acidic conditions (pH~5.5). Since cells maintain a homeostasis with pH = 7.4 and 100 mM K+, the G-quadruplex structure is more likely to form than the i-motif. Formation of either structure, however, is only possible after duplex DNA disintegrates into single-stranded DNA. It has been found that the molecular crowding due to the presence of ~40% macrobiomolecules inside cells can destabilize duplex DNA (5, 6). Another widely occurring cellular condition, negative DNA superhelicity, has shown a similar property to weaken duplex DNA (7). As a result, the G-quadruplex population has been found to increase under physiologically relevant superhelical conditions with superhelicity (σ) ranging from σ = -0.05 to -0.1 (8). By the same token, it is expected that i-motif formation should also be facilitated under negative template superhelicity, although this prediction is yet to be verified. To test this prediction, it is necessary to quantify the formation of tetraplexes in a duplex DNA template under various chemical and mechanical conditions. However, such quantification is extremely challenging. First, there is a lack of reliable ensemble approach to providing a DNA template with desirable superhelicity. In fact, a mixture of superhelicities is often produced under bulk conditions (9). In addition, due to the presence of mechanical and chemical factors, as well as the implication of both G-quadruplex and i-motif in complementary DNA strands, it is of high complexity in experimental design and results interpretation.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. (A) Schematic of mechanical unfolding in ILPR DNA constructs using magneto-optical tweezers. The sequence of interest from ILPR was sandwiched between two double-strand (ds) DNA handles which was tethered between the two optical traps. (B) and (C) depict the formation of G-quadruplex (red) and i-motif (blue) in wild-type and mutant ILPR DNA constructs, respectively. Chemical structures of the main components in the G-quadruplex and i-motif are shown. M+ depicts monovalent cation in (B). (D) Twist-extension curves of the wild-type (WT) dsDNA construct containing the ILPR sequence in a 10 mM MES buffer (pH 5.5) with 100 mM Li+. In this work, we quantified for the first time the molecular population dynamics of DNA tetraplexes (G-quadruplex and i-motif) from the promoter sequence of Insulin Linked Polymorphic Region (ILPR) under chemical (ions and pH) and mechanical (template superhelicity and


Page 4 of 28

Page 5 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

molecular crowding) conditions by single-molecule mechanical unfolding methods and innovative data analyses. Using magneto-optical tweezers instrument (7), we first established that DNA superhelicity facilitated the formation of i-motif in the double-stranded Insulin Linked Polymorphic Region. Like G-quadruplex, the promotion effect is due to increased unwinding rate of duplex DNA and reduced reannealing rate of the two complementary DNA strands. By comparison between mutant (MT) and wild-type (WT) torsionally constrained ILPR constructs, next, we deconvoluted G-quadruplex and i-motif species under 32 sets of different chemical and mechanical conditions. Data were then presented with an intuitive layout, triangle binary plot (or TBP), which ranks the contribution of each factor to promote either G-quadruplex or i-motif species. By comparing the population probability in any given condition with a variation in only one factor, we were able to scrutinize the effect of each factor on the folding of the DNA tetraplexes. Through this method, we found that while potassium ion and acidity predominate the formations of G-quadruplex and i-motif, respectively, negative superhelicity has the second most potent effect, followed by molecular crowding. As negative DNA superhelicity showed a more substantial effect on the tetraplex population than other cellular conditions such as molecular crowding, our results support the role of tetraplexes as a dynamic topology sensor during transcription or replication processes where superhelicity of DNA template is changed transiently. Materials and Methods Materials DNA fragments were purchased from Integrated DNA Technologies (Coralville, Iowa). Before use, the DNA fragments were purified by 8% polyacrylamide gel electrophoresis. All chemicals were purchased from VWR (West Chester, PA) with purity >95% and used directly without further purification. Restriction enzymes were purchased from New England Biolabs Inc. (Ipswich, MA). Streptavidin coated magnetic particles (1.8 µm diameter), digoxigeninantibody coated polystyrene particles (2.1 µm diameter), and 130 nm polystyrene marker particles were purchased from Spherotech Inc. (Lake Forest, IL). Synthesis of torsionally constrained DNA construct A

five-segment

DNA

construct

that

contains

the

C-rich

region

(5′-

TGT[CCCCACACCCCTGT]2) from ILPR was synthesized according to literature (7) (Figure


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

S1). Briefly, the C-rich region of ILPR region along with the complementary strand (G-rich DNA, refer Table 1 for sequence) was ligated to two long DNA duplex handles, which were prepared by PCR using λ-DNA template (position 27493–30485 and 33496–35561, see reference 7). The ligated product was subject to a nested PCR for further amplification. The amplified DNA construct was digested sequentially with EcoRI and HindIII restriction enzymes to yield restriction sites on either end of the DNA. After each digestion, the resulting product was consecutively ligated to two short duplex DNA fragments, which were labeled with multiple copies of either digoxigenin-11-dUTP or biotin-11-dUTP in both strands by PCR. The resultant DNA construct was used for single-molecule force ramping experiments. Table 1. Sequence of G-rich and C-rich DNA strands used in this study G-rich sequence 5’- ACAGGGGTGTGGGGACAGGGGTGTGGGGACA (wild type construct) G-rich sequence 5’- ACAGGGGTGTGTGTACAGTGGTGTGTGTACA (mutant construct) C-rich sequence 5’- TGTCCCCACACCCCTGTCCCCACACCCCTGT (wild type and mutant constructs)

Synthesis of Mutant DNA construct The mutant construct contains five segments similar to the wild-type. The synthesis strategy is split into three parts (Figure S2). In the mutant construct, the sequence of interest (SOI) is the duplex strand obtained from annealing mutated G-rich sequence with wild-type C-rich sequence (Table 1). Part I of the synthesis is to connect the SOI to the right dsDNA handle to form a threesegment DNA product (SOI-H1-H2). The right dsDNA handle contains Handle 1 (H1) and Handle 2 (H2), which were ligated to SOI sequentially. Part II of the synthesis is to synthesize the left dsDNA handle by ligating Handle 4 (H4) with Handle 3 (H3). All the four handles (H1, H2, H3 and H4) used in this strategy were amplified from λ-DNA templates through PCR, which was followed by digestion with specific restriction enzymes to produce sticky ends for respective ligations. Handles 2 and 4 contain multiple copies of digoxigenin-11-dUTP and biotin-11-dUTP respectively. These modified dUTP copies were incorporated during the PCR amplification of


Page 6 of 28

Page 7 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

the corresponding λ-DNA templates. In Part III of the synthesis, the final mutant construct consisting of five DNA segments (H4-H3-SOI-H1-H2) was synthesized by ligating the DNA products from Part-I (SOI-H1-H2) and Part-II (H4-H3). Magneto-optical tweezers Detailed description of the instrument has been reported elsewhere (7). Briefly, a dualtrap optical tweezers instrument (10) was used. The rotation of an optically trapped superparamagnetic bead (diameter 1.8 µm, Spherotech Inc.) was controlled by a pair of magnets (Neodymium magnet (NdFeB), CMS Magnetics, TX, USA) with a servomotor (Lego®, Mindstorm) placed 2.0 ± 0.5 cm above the trapped magnetic bead. This distance ensured minimal interference of magnetic force to the optical force generated by laser traps. The DNA construct between two optically trapped beads (see below) was kept in a force range of 0.3 pN – 2.5 pN during rotations, followed by equilibration for two minutes at 0 pN before the force-ramping experiment at each DNA superhelicity (σ = 0.1 to -0.2). Single-molecule force ramping experiments As discussed above, the dsDNA construct has multi-labelling of either biotin or digoxigenin on each end of both DNA strands. This design enables the attachment of the two DNA helices to the two micrometer-sized particles in a torsionally constrained manner (see Figure S3). The DNA construct (0.016 nM) was incubated with anti-digoxigenin antibody coated polystyrene beads (2.1 µm, Spherotech Inc., IL) for 2 hours. Following incubation, these beads were dispersed in 1 mL of 10 mM MES buffer (pH 5.5) with 100 mM Li+ and injected into the bottom channel of a three-channel microfluidic chamber (11) (see SI Figure S3). The three-channel microfluidic chamber was prepared by placing a laser-cut (Universal Laser Systems Inc, AZ,USA) 127-µm-thick parafilm (VWR, West Chester, PA, USA) pattern (as shown in Figure S3) between two coverslips, which were bonded using thermal heating at 85 °C. The top channel of the chamber was flown with streptavidin coated superparamagnetic beads (1.8 µm, Spherotech Inc., IL), which were also dispersed in a 10 mM MES buffer at pH 5.5 with 100 mM Li+. Notice, 100 mM Li+ was added in buffers to avoid the formation of G-quadruplex. It is known that Li+ has no effect on the i-motif formation (12).


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

During the single-molecule experiments, the magnetic beads were trapped in the top optical trap and the DNA-digoxigenin-antibody coated polystyrene particles (DNA-bead conjugate) were trapped in the bottom optical trap. The DNA construct was tethered between the two beads by moving the top bead close to the DNA coated polystyrene bead in the bottom trap via the biotin-streptavidin linkage. The tethered DNA molecule was subject to torsional stress by rotating the magnet placed above the chamber (see Figure 1A and S3). It is noteworthy that while rotating the magnets, only the magnetic bead in the top trap rotated while the polystyrene bead at the bottom trap did not rotate (7). This introduced superhelicity in the DNA that was tethered between the two beads in a torsionally constrained fashion. For clear visualization of the rotation of the magnetic bead alone, 130 nm polystyrene particles (Spherotech Inc., IL) were flown in both top and bottom channel and were attached to the two trapped beads as markers (see Figure 1A and S3). The force-ramping experiments were performed by moving the top laser trap (magnetic bead) away from the bottom (polystyrene bead) with a load rate of 5.5 pN/s and the tension in the DNA tether was recorded in LabVIEW program (National Instruments) at 1000 Hz, which was Savitzky-Golay filtered to 100 Hz by Matlab (The MathWorks, Inc., MA) programs. Data analyses were carried out in IGOR Pro program (WaveMetrics, Portland, OR). The DNA tethered between the two trapped beads with and without the pair of magnets displayed identical tensions, verifying minimal interference of magnetic force on the optical force measurement in the instrument (7). Calculation of change in contour length (∆L) Change in extension (∆x) of the unfolding event at certain force (F) was converted to change in contour length (∆L) using the worm-like-chain model given below (13, 14), ∆ ∆

= 1 −

+ …………………………………………………………………… (1)

where kB is the Boltzmann constant, T is absolute temperature, P is the persistence length of ds DNA (50.8nm) (15), and S is the stretching modulus (1243 pN) (15). Expected ∆L during the unfolding of i-motif or partially folded structure was calculated using the following equation,


Page 8 of 28

Page 9 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

∆ = ∗ − ………………………………………………………………………… (2) where N is the number of nucleotides, Lbp is the contour length per base pair in double-stranded DNA (0.34 nm), and x is the end-to-end distance of a folded structure, which is estimated as 0.7 nm for i-motif and 2.7 nm for a partially folded structure presumably a triplex conformation (12, 16). This calculation gave ∆L of 7.8 nm and 3.4 nm for i-motif (N = 25 nt) and partially folded (N = 18 nt) structures, respectively. Deconvolution of i-motif and partially folded structures To deconvolute the i-motif and partially folded structure in the ∆L histograms, the ∆L histograms were first fitted with two Gaussian functions (see Figure 2A and B (iii)). The overlapping region between the two populations were randomly assigned to either the i-motif or partially folded structure according to a ratio determined by the two Gaussian functions at a particular ∆L range within the region (12). Deconvolution of the rupture force was based on the fact that each data point in the ∆L histogram corresponds to a rupture force value. Calculations of the ∆Gunfold and kunfold were based on deconvoluted work histogram (Figure S4) and rupture force histogram (Figure S5) respectively. Calculation of change in free energy Change in the unfolding free energy of i-motif (∆Gunfold) was calculated using Jarzynski equation (17): ∆

!"

,-

. = −#$ %&' ∑) /0 ) *+ …………………………………………………….. (3)

where N is the number of features observed in the experiment and W is the amount of work done to unfold an i-motif, which is equal to the hysteresis area between stretching and relaxing forceextension curves (18). Previous investigation on a single-stranded ILPR G-quadruplex have verified that ∆Gunfold obtained with current loading rate of 5.5 pN/s is identical with that calculated from melting experiments.(18) Accuracy of this calculation is further confirmed here by identical ∆Gunfold values obtained from a nicked ILPR construct with loading rates of 5.5 pN/s and 1.0 pN/s.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 28

Calculation of unfolding kinetics of i-motif The unfolding kinetics of the i-motif was estimated using the Dudko model (19), 1

kunfold k ( F ) k (F ) x ‡ F 1− v p( F ) ∝ exp[ ‡ − ‡ (1 − v) ] ……………………………………. (4) r xr xr ∆G‡ 1

1

x‡ F v −1 x‡ F v ‡ v ∆ G − − ) exp{ [ 1 ( 1 v) ]}, where k ( F ) = kunfold (1 − ∆G‡ ∆G‡ kunfold is the unfolding rate constant at zero force, ∆G ‡ is the activation energy of unfolding, x ‡ is the activation distance from the folded state to the transition state, r is the loading rate (5.5 pN/s), and ν is the factor describing the shape of the energy barrier (ν = 1/2 for a sharp, cusp-like shape barrier and ν = 2/3 for a softer, cubic shape). Here we fit the force histograms of i-motif in mutant construct (MT, see text) using either ν = 1/2 or 2/3. The values reported in Table S2 are the average of the results from the two fittings using ν = 1/2 and 2/3. From the kunfold and the ∆Gunfold obtained from equation 3, the kfold was calculated by the equation ∆Gunfold = -RT ln K, where K = kunfold/ kfold. Calculation of twisting energy Twisting energy (Etwist) during rotation of a torsionally constrained template is calculated by(20), 123/42 = 5 6 78 , where τ is the torque and dθ is an infinitely small twisting angle introduced to the template. In the B-form dsDNA, torque (τ) is related to the added twisting angle (θ) 9

by the function (21), 6 = 8, here C is the torsional modulus (88 ± 4 nm kBT for 4.2 kbp DNA)(22), L is the contour length of the DNA (1425 nm). Twisting energy for clockwise rotation of B DNA towards its melting can therefore be calculated by, :

; 123/42 = 5:0 @

0 ? 9

9

8 78 = ×

B ?CD

……………………… (5)

where θmelt=2πn is the added twisted angle in radians until melting occurs, here n=6.4 is the number of clockwise rotations before melting occurs for the DNA construct under investigation (4191 bp).


Page 11 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

During the melting, torque becomes constant (τmelt = -10 pN.nm) (21). In this case, twisting energy is given by, 123/42 = 6EF!2 × 2H'2 ………………………………………………………… (6) where nt = 33.6 is the clockwise turns from the melting torque (σ = -0.016) to reach σ = -0.1 for the 4191 bp DNA construct. By summing up equations 5 and 6, the total twisting energy is expressed by, 12

2I! 23/42

9

= ×

B ?CD

+ 6EF!2 × 2H'2 ……………………………………… (7)

The first and second terms of equation 7 are equal to the yellow and grey regions in Figure S6, respectively. The total twisting energy for the construct was calculated to be 332.2 kcal/mol, which is equal to 0.079 kcal/mol for every base pair. Circular Dichroism (CD) measurements The circular dichroic spectra of the DNA samples were collected in JASCO J-810 CD spectrometer under nitrogen atmosphere. Each spectrum was measured with a scan speed of 50 nm/min by averaging three scans. All the DNA samples were heated to 55 oC for 10 minutes and rapidly cooled to 4 oC prior to CD measurements. The 285 nm CD signal, which is characteristic of i-motif structure, is higher for MT dsDNA than the wild-type dsDNA (WT, see text). To retrieve i-motif population in MT construct from that observed in the WT construct, the differences between the CD spectra of WT and MT under different buffer conditions are plotted (Figure S7). The ratio of the CD signal at 285 nm in this differential spectrum (see Figure S7A black) (CDover-expressed iM = CDMT-CDWT) over the signal from single-stranded C-rich DNA (not shown in the Figure) (12) in a 10 mM MES buffer (pH 5.5) with 100 mM LiCl was calculated as 0.14. CD experiments were also carried out in a 10 mM MES buffer (pH 5.5) with 100 mM LiCl and 40% BSA, which gave a similar ratio of 0.15 (compare Figure S7A green and red traces). Similarly, CD spectra in the buffer of 10 mM Tris (pH 7.4) with 100 mM KCl also gave the correction factor of 0.14 (compare Figure S7B blue and pink traces). Therefore, we used a factor of 1/(1-0.14) to estimate the i-motif population in the WT construct from that observed in the MT construct.


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 28

Results and Discussion Formation of i-motif increases with decreasing superhelicity in torsionally constrained ILPR DNA To perform the superhelicity experiment in the magneto-optical tweezers, we first trapped two particles by two separate laser foci. One particle (2.1 µm in diameter) is made of polystyrene while the other superparamagnetic material iron oxide (1.8 µm, Figure 1A). A double-stranded DNA

(dsDNA)

construct

containing

the

ILPR

i-motif

forming

sequence,

5′-

(TGTCCCCACACCCC)2TGT, was then tethered between these two particles. To ensure that the DNA was torsionally constrained between the two trapped particles by affinity attachments, the two strands of the DNA construct were labelled with multiple biotin molecules on one end and multiple digoxigenin molecules on the other end (see Figure 1, Figure S3). By rotating a pair of magnets located above the magnetic bead either clockwise or counter-clockwise (viewed from above), negative or positive superhelicity can be respectively introduced according to the equation, Lk = Tw + Wr (23), where Lk is the link number, Tw is the physical twist between the two DNA strands, and Wr is the supercoiling along the dsDNA axis. Before each experiment, we confirmed the DNA tethered between two optically trapped particles was torsionally constrained by two observations. First, we observed a characteristic plateau of 110 pN, instead of 65 pN, during force-ramping experiments (Figure S8) (24, 25). Second, we followed the extension of the tethered DNA with superhelicity at a particular force (Figure 1D). At 4 pN, we observed that the extension of the DNA construct remained constant, independent of template superhelicity. When the force was reduced to 1.5 pN, the extension remained constant under negative superhelicities while it was reduced with positive superhelicities. Further reduction of the force to 0.3 pN caused the extension to decrease with either positive or negative superhelicities. All these observations indicated a torsionally constrained DNA in which the decrease in extension at low force reflects the unsymmetrical formation of plectoneme structures due to the right-handed chirality in the B-DNA (7, 22, 26-28). With the reliable means to introduce torsionally constrained DNA template in our magneto-optical tweezers, we varied template superhelicity to evaluate its effect on possible Gquadruplex or i-motif structures in the ILPR DNA duplex. Since the behavior of G-quadruplex in


Page 13 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

torsionally constrained DNA has been probed previously (7), here we first evaluated i-motif structures in the duplex DNA.

It is well known that Li+ ions hinder G-quadruplex formation

while they have little effect on the i-motif in single-stranded DNA (29). In duplex DNA, however, negative superhelicity can also promote G-quadruplex formation (7). Therefore, it is possible that G-quadruplex may be responsible for the unfolding feature observed in negatively supercoiled torsionally constrained wild-type ILPR DNA. To probe the superhelicity effect on the ILPR i-motif in torsionally constrained DNA, we used a mutant construct (MT, Figure 1C) in which five guanine residues in the G-rich strand are mutated to thymines while the C-rich strand remains unchanged. CD signatures confirmed that G-quadruplex does not form in the mutant single-stranded G-rich DNA (Figure S9). Force-ramping experiments for the MT construct at 5.5 pN/s loading rate in a 10 mM 2(N-morpholino)ethanesulfonic acid (MES) buffer with 100 mM LiCl at pH 5.5 (see Materials and Methods) revealed two different unfolding features with change in contour length (∆L) centered at 3.4 (±0.6) nm and 7.3 (±0.5) nm, respectively (Figure 2A (i)). While the 7.3 nm population matches with that of unfolding ILPR i-motif structures in the duplex DNA (see Materials and Methods for calculation), the 3.4 nm ∆L species agrees well with a partially folded structure that involves three C-rich tracts (30). Additional evidence for this assignment comes from the experiments with nicked DNA, which contains two populations with similar ∆L values (Figure 2A (iii)). Previously, in a nicked DNA duplex that contains the same ILPR sequence, we have assigned the two populations with the similar ∆L values as ILPR i-motif and partially folded structures (12). With the assignment of i-motif and partially folded structures, we deconvoluted the two species under different superhelicities (7) for the mutant template (see Materials and Methods for details). We found percentage formation of fully folded i-motif increases with decreasing superhelicity (Figure 3A). At positive superhelicity (σ = 0.1), there is 9.5% population of fully folded i-motif in the mutant construct. This number increases slowly with decreasing superhelicity until around the melting torque (σ = -0.016) (21, 22, 31). Beyond the melting torque, the population increases rapidly to ~30%. Such a result suggests a correlation between the melting of DNA and the formation of i-motif. At σ = -0.10, the i-motif population reaches a plateau with ~33% population, suggesting equilibrated i-motif and duplex DNA populations. Similar observations have


Biochemistry

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 28

been found for the formation of G-quadruplex in the same ILPR sequence (7). It is noteworthy that the population of the partially folded structure remains constant with superhelicity in the mutant construct (~3%, Figure 3A). Since i-motif population increases with negative superhelicity, it suggests partially folded structure has decreased population with respect to the i-motif.

Figure 2. Mechanical property of ILPR secondary structures in a 10 mM MES buffer (pH 5.5) with 100 mM Li+ for (A) Mutant and (B) Wild-type constructs. (i) Typical F-X traces observed during force-ramping experiments. Red and blue traces contain unfolding events for small and big structures, respectively (see corresponding blown-up boxes for details). Histograms of un-


Page 15 of 28

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biochemistry

folding force (ii) and change in contour length (iii) of ILPR secondary structures in nicked and torsionally constrained molecules at σ = 0 and σ = -0.1. Black solid curves represent Gaussian fittings. See Figures S8 and S9 for the histograms of other superhelicities in the mutant and wild-type constructs respectively and Table S1 for the number of events investigated. Understanding the superhelicity effect from thermodynamic and kinetic perspectives As reported previously for ILPR G-quadruplex (7), the level of i-motif population at different template superhelicity could be a result of competing processes such as denaturation of dsDNA, reannealing to dsDNA, and formation of tetraplex. To fully understand the effect of superhelicity on the i-motif formation, we wish to investigate the equilibrated processes from thermodynamic and kinetic perspectives. To this end, we first analyzed rupture force histograms for the ILPR i-motif under different superhelicities of the mutant construct in the same MES buffer with 100 mM LiCl as described above. We found rupture force of i-motif does not vary significantly with template superhelicity (~ 32 pN, Figures 2A(ii) and S10). The average unfolding force for either i-motif or partially folded structure is 32 pN at the loading rate of 5.5 pN/s. Since the trap stiffness of the magneto-optical tweezers is around 0.201 nm/pN, the pulling rate is equivalent to 27.5 nm/s (or 81 bp/s), which is close to the average transcription speed observed in bacteria (32). Therefore, the unfolding force at this pulling rate serves as a physiologically relevant estimation for the mechanical stability of the i-motif. When RNA polymerase (RNAP) encounters i-motif, the tug-of-war between the stall force at which RNAP stops transcription and mechanical stability of i-motif determines whether RNAP can continue its action via disassembling i-motif. Given that the known RNA polymerases have stall forces

Quantification of Chemical and Mechanical Effects on the Formation of

Recommend Documents