The PX-Motif of DNA Binds Specifically to E. coli DNA Polymerase I

Subscriber access provided by TULANE UNIVERSITY

Article

The PX-Motif of DNA Binds Specifically to E. coli DNA Polymerase I Xiang Gao, Matthew Gethers, Si-ping Han, William A. Goddard, Ruojie Sha, Richard Cunningham, and Nadrian C. Seeman Biochemistry, Just Accepted Manuscript • DOI: 10.1021/acs.biochem.8b01148 • Publication Date (Web): 17 Dec 2018 Downloaded from http://pubs.acs.org on December 19, 2018

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 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 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.

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

The PX-Motif of DNA Binds Specifically to E. coli DNA Polymerase I Xiang Gao1, Matthew Gethers2, Si-ping Han2, William A. Goddard III2, Ruojie Sha1, Richard P. Cunningham3* and Nadrian C. Seeman1* 1Department

of Chemistry New York University New York, NY 10003, USA 2Materials

and Process Simulation Center, MC139-74 California Institute of Technology Pasadena, CA 91125, USA

3Department

of Biological Sciences State University of New York at Albany Albany, NY 12222, USA

Correspondence to Nadrian C. Seeman at [email protected] or to Richard P. Cunningham at [email protected]

ACS Paragon Plus Environment

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 2 of 28

ABSTRACT The PX-motif of DNA is a four-stranded structure wherein two parallel juxtaposed double helical domains are fused by crossovers at every point where the strands approach each other. Consequently, its twist and writhe are about half those of conventional DNA. This property has been shown to relax supercoiled plasmid DNA in circumstances where head-to-head homology exists within the plasmid; the homology can be either complete homology or every-other-half-turn homology, known as PX-homology. It is clearly of interest to establish whether the cell contains proteins that interact with this unusual and possibly functional motif. We have examined E. coli extracts to seek such a protein. We find by gel mobility studies that the PX motif is apparently bound by a cellular component: Fractionation of this binding activity reveals that the component is DNA polymerase I (pol I). Although the PX motif binds to pol I, we find that PX-DNA is not able to serve as a substrate for the extension of a shortened strand. We cannot say at this time whether the binding is coincidence or whether it represents an activity of pol I that is currently unknown.

We have modeled the

interaction of pol I and PX-DNA using symmetry considerations and molecular dynamics.


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

Numerous DNA motifs other than the traditional Watson-Crick double helix have been discovered, designed and modeled in experimental systems. Prominent among these structures are the Holliday junction,1 multi-arm junction motifs,2 the DX motif and its congeners,3 the 6-helix bundle4, DNA origami,5 the tensegrity triangle,6 and paranemic crossover PX-DNA.7 Many of these motifs have been used in the construction of nanoscale objects, lattices and devices.8 The paranemic crossover motif, PX-DNA, has been used for over a decade as a component of nanomechanical devices. From a biological perspective, the PX-DNA motif appears to be a vehicle for relaxing supercoiled DNA in the presence of head-to-head homology in two portions of a plasmid.9 This phenomenon suggests that PX-DNA might be involved in homology recognition within the cell,

indicating

its

possible

involvement

in

processes

like

recombination and repair, where homology features prominently. The structure of the PX molecule is shown in Figure 1, where it is compared with the structures of two B-DNA double helices, one drawn in red and one in blue. Those helices can form PX-DNA by wrapping around each other to yield the 4-stranded structure shown on the right.

One can produce this arrangement of strands by

selecting sequences that pair in the fashion indicated.7 However, it has also been shown that if the blue and red helices are homologous components of a larger supercoiled plasmid, the PX-DNA structure


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 4 of 28

can be generated, because it relaxes the supercoiling: The pitches of the B-DNA double helices and of the PX-DNA molecule are indicated in Figure 1; since the pitch of the PX molecule is roughly twice that of the B-DNA molecules, its overall twist is about half that of B-DNA. Consequently, from the standpoint of twist, the formation of PX-DNA can relax a supercoiled molecule that contains it. A related argument applies to the writhe.9

Each helix of the PX

molecule in Figure 1 contains four half-turns in its repeat. Traveling down from the top of the left helix, there is a red-red half-turn, a redblue half-turn, a blue-blue half-turn and a blue-red half-turn. Note that the 'half-turns' are not the same size, because some are major groove spacings and some are minor groove spacings. Homology is only necessary in the mixed half-turns, where red pairs with blue, so it is possible to form PX-DNA in super-coiled DNA with an interrupted homology, every other half-turn.9 Thus, only the half-turns labeled 'H' need to be homologous, and those labeled 'U' do not require homology.

This type of interrupted homology is termed 'PX-

homology.'9 Interrupted homology has recently been reported in the RIP system.10


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

Figure 1. The PX Structure Compared with Two Double Helices. Two B-DNA double helices are shown at left, one red and one blue. The pitch is shown as the length of a helical repeat. The PX-DNA structure is shown at right as though it were formed from those two helices. There is an apparent dyad axis down the middle, related two helices connected by crossovers that exist wherever possible. In this representation, the left helix of the PX-DNA molecule consists of a red-red half-turn at top, followed by a red/blue half-turn, a blue-blue half turn and a blue red half turn, a sequence that then repeats. The right helix is similar, except that it starts with a blue-blue half-turn at top. To form this structure from the B-DNA helices at left, only the mixed half-turns (indicated H) require homology, whereas the others (U) do not. Major grooves are indicated by W (wide) and minor grooves are indicated by N (narrow).


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 6 of 28

Methods and Materials: Design of DNA Sequences: The DNA sequences for PX-DNA (termed PXC) were designed by applying the principles of sequence symmetry minimization using SEQUIN.14 Selected strands were labeled with fluorescein at the 5´terminus. To make sure that the protein binding detected was structure-specific and not sequence-specific, we designed duplexes to act as competitors. Duplex competitor 1 (DC1) and duplex competitor 2 (DC2) were designed from the PX-DNA complex and contain the same sequences as in the two helices of the PXC molecule.

Synthesis and Purification of DNA Strands:

Unmodified DNA strands used in this study were ordered from IDT; modified DNA strand labeled with Fluorescein was synthesized on an Applied Biosystems 380B automatic DNA synthesizer and then removed from the support, and deprotected using routine phosphoramidite procedures. To increase purity, all DNA strands were purified by denaturing gel electrophoresis. Denaturing gels contained

8.3

M

urea,

10-20%

acrylamide

(19:1,

acrylamide/bisacrylamide) and were run at 55 °C. The running buffer consists of 100 mM Tris-HCl (pH 8.3), 89 mM boric acid, 2 mM EDTA (1X TBE). The sample buffer consists of 10 mM NaOH, 1 mM EDTA, containing 0.1% xylene cyanol FF tracking dye. Bands were cut out


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

of denaturing gels and eluted in 500 mM ammonium acetate, 10 mM magnesium acetate, and 1 mM EDTA (termed elution buffer). And then n-butanol was used to extract DNA and get rid of excess ethidium, followed by ethanol precipitation.

Preparation of DNA Complexes:

Strand concentrations were determined using the absorbance recorded at OD260. The four strands for making a PX DNA molecule were mixed at a ratio of 1:1:1:1 in 40 mM Tris-HCl, 20 mM acetic acid, 2 mM EDTA, and 12.5 mM calcium acetate (TAE/Ca2+). Duplex competitors were mixed at a ratio of 1:1 using the same buffer. Those mixtures were heated to 90 °C for 5 min and cooled by the following protocol: 20 min at 65 °C, 20 min at 45 °C, 30 min at 37 °C, 30 min at room temperature. Structural integrity of the constructs was checked on a non-denaturing gel containing 8% acrylamide (19:1, acrylamide: bisacrylamide).

Polymerization Reaction:

DNA substrates (PX or duplex), three different kinds of Pol I (complete Pol I, Klenow Fragment or Exo-) and 1 µL 1 mM dNTP were incubated in 50 mM NaCl, 10 mM Tris-HCl, 10 mM MgCl2, 1 mM DTT at 37 °C for 20 minutes. After that, 10 µL loading buffer (deionized formamide/ 10x TBE= 9:1) was added and incubated at 90 °C for 5 minutes. The sample was then loaded onto a nondenaturing

gel

composing

of

15%

acrylamide


(19:1,

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

acrylamide/bisacrylamide), 100 mM Tris-HCl (pH 8.3), 89 mM boric acid, 2 mM EDTA and run at 5 V/cm at 55 °C for one and half hours. The gel was scanned using a Typhoon scanner set at the Fluorescein wavelength.

Protein Extraction and Purification:

E. coli BL21(DE3) was cultured at 37 °C to late logarithmic phase. The cells were harvested by centrifugation and re-suspended in a buffer containing 50 mM Tris-HCl pH 7.5, 400 mM NaCl, 1 mM EDTA, and 1 mM BME (1 g cell paste per 5 ml buffer) and lysed by sonication in five 1-min intervals using a Vibra-Cell sonicator, set at 40% amplitude. The cell lysates were stirred at 4 °C for 1 hour and the cell debris were pelleted by centrifugation (fraction I). The lysate was then applied to a 10-ml HiTrap Q Sepharose FF column (GE Healthcare Life Sciences) to remove the nucleic acids components and the collected proteins were eluted in the flow-through. The eluent was dialyzed against a buffer containing 50 mM Tris-HCl pH 7.5, 25 mM NaCl, 1 mM EDTA and 1 mM BME (fraction II). Fraction II was applied to two 5-ml HiTrap Q Sepharose FF columns in tandem (GE Healthcare Life Sciences). A 150-ml gradient was run at a rate of 3 ml per minute to a final concentration of NaCl of 500 mM. The fractions were assayed for PX binding via gel shift assay. The fractions containing proteins bound to PX DNA were pooled and dialyzed against a buffer containing 50 mM Tris-HCl pH 7.5, 25 mM NaCl, 1 mM EDTA and 1 mM BME (fraction III). Fraction III was applied to a 1.3 x10 cm column of DNA agarose (GE Healthcare Life


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

Sciences). A 100 ml gradient was run at a rate of 1 ml per minute to a final NaCl concentration of 500 mM. The fractions were again assayed for PX binding via gel shift assay. The fractions containing proteins bound to PX were pooled and dialyzed against a buffer containing 50 mM Tris-HCl pH 7.5, 100 mM NaCl, 1 mM EDTA, and 1mM BME (fraction IV). Fraction IV was concentrated using a Centriprep 10 concentrator (Amicon) Fraction IV was applied to a Superdex 75 gel filtration column (1.6 × 60 cm). The column was eluted at a rate of 1ml per min with a buffer containing 50 mM TrisHCl pH 7.5, 100 mM NaCl, 1 mM EDTA, and 1mM BME. The fractions were assayed for PX binding via gel shift assay and fractions

containing

PX-binding

activity

were

pooled

and

concentrated (fraction V).

Measurement of Pol I concentration:

Pol I concentration was measured by using Pierce BCA Protein Assay Kit (Thermo Scientific). The concentration is around 40 uM.

Binding Reaction and Gel Shift Assay DNA and Pol I were incubated in 4% glycerol, 20 mM Tris-HCl, 2 mM EDTA, 50 mM NaCl and 0.5 mM DTT at room temperature for 1 hour. The protein-DNA samples were run on a native gel containing 6% acrylamide (19:1, acrylamide: bisacrylamide), 40 mM Tris-HCl, 20 mM acetic acid, 2 mM EDTA, and 12.5 mM calcium acetate at 5 V/cm at room temperature and were scanned by a Typhoon scanner set at Cy3 wavelength.


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

Results Strategy: The focus of this study is to identify proteins that bind in a structure-specific manner to PX-DNA. Figure 2 (a) shows the design of this project: Proteins are extracted from E. coli cells and purified by three different kinds of chromatography columns (Qsepharose, DNA cellulose and a sizing column), thereby narrowing the protein fraction that can bind to PX-DNA. After each purification step, gel shift binding assays were performed with PX-DNA, using separated protein fractions; those which bound to PX-DNA were further purified. After protein binding and purification through the three columns, a mass spectrum of the final protein fraction was measured to identify the proteins present in the final fraction. The PX-DNA molecule used in this study (termed PXC) is shown in Figure 2 (b). The four component strands are shown in different colors. The red strand was labeled with fluorescein dye at its 5'-end to locate PXC and PXC-protein complexes. To make sure that the protein-binding seen is structure-specific and not sequence-specific, we designed duplexes to act as competitors. Duplex competitor 1 (DC1) and duplex competitor 2 (DC2) were designed from the PXDNA complex and contain the same sequences as in the two helical domains of the PXC molecule (Figure 2 (b)).


Page 10 of 28

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

Figure 2. Identification of PX-binding proteins. (a) A scheme showing the experimental pathway to identify proteins binding in a structurespecific manner to paranemic crossover (PX) DNA. Protein fractions were extracted from E. coli cells and DNA-binding fractions were isolated by binding assays and concentrated by various HPLC columns. The final purified protein fraction was identified by mass spectrometry. (b) The PX complex used in this study is shown at the top. The major groove and minor groove of this complex contain 6 and 5 nucleotide pairs respectively (denoted as 6:5 PX). The duplex competitors used for studying structure-specificity of the proteins are shown below the PX complex. The duplexes were derived from the top and bottom 'halves' of the PX molecule.


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

Proteins extracted from E. coli were first separated using a Qsepharose column. The different protein fractions were added to PXC to analyze PX-binding ability. Fractions that showed binding with PXC (Figure S1, lanes 6-9) were isolated and run on a DNA cellulose column. In addition, these fractions were tested for structure-specific binding by adding both PXC and the duplex competitors. Figure S2 shows a non-denaturing gel demonstrating the structure-specific binding of these protein fractions with PXC in the presence of duplex competitors. These fractions were then run on a DNA cellulose column (Figure S3). After the Q-sepharose and the DNA cellulose column, it was seen that a higher quantity of the protein fraction bound with PXC (Figure S4). The PXC-binding fractions separated by the DNA cellulose column were further purified by a sizing column. The proteins were separated into fractions, corresponding to their different sizes. Figure S5 (a) shows a non-denaturing gel demonstrating the results of binding PXC to different fractions collected from the sizing column. A gel shift is seen in lanes 1 to 3. These fractions were pooled, concentrated and used for mass spectral analysis. We also did a binding titration for this final fraction to confirm the binding results. As shown in Figure S5 (b), when we increase the amount of protein in the binding assay, more of the PX complex is shifted. We then quantified the percentage of PX in the gel shift and calculated a binding curve for the percentage


Page 12 of 28

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

bound versus the quantity of proteins (Figure S5 (c)). The result appears to be a standard binding curve. An SDS-PAGE gel of the final purified fraction collected from size column is shown in Figure S6. Duplex competitors were again used as controls, making sure that the binding to the PX complex is structure-specific, rather than sequence-specific. Figure 3 shows the binding results in the presence of the duplex competitors. Lane 3 contains 10X concentrated unlabeled DC1/DC2, and lane 4 contains 50X concentrated unlabeled DC1/DC2. The binding efficiency of PXC was not affected, which confirms that the binding is PX-structurespecific, and not specific to the DNA sequence.

Figure 3. PX binding with final purified protein fraction in the presence of duplex competitors. Lane 1 contains only PXC and lane 2 contains PXC and purified proteins. Lanes 3 and 4 contain PXC with protein fractions along with 10X and 50X duplex competitors respectively. There is no evident difference in mobility even in the presence of the duplex competitors. ACS Paragon Plus Environment

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

Identification of the Binding Protein: The mass spectral results had the highest score for ParC.

However, a purified sample

(generous gift of Ken Marians) did not bind to the PX-DNA motif at all (Figure S7). This finding reflects the fact that a score based on primary structure may not be reflected in binding, which is a function of tertiary structure.

DNA Polymerase I (Pol I) had the second

highest score in the mass spectrum results (Table S1), one of only two proteins with especially high scores. Figure 4 (a) shows a nondenaturing gel illustrating the binding results of PXC with Pol I. In lane 2, the binding of Pol I with PXC makes a full gel shift (indicated by the red arrow). Again we used DC1/ DC2 as controls to compete with the binding of PXC. Upon the addition of 25X concentrated unlabeled duplex competitors the complete gel shift remained (lane 3), indicating that the binding of PXC with Pol I is not affected by the presence of the same-sequence duplexes. In (c), we did it the opposite way, using unlabeled PXC as a control to compete with the binding of labeled duplexes. Similar to the binding results of PXC in (b), there is a complete gel shift for labeled duplexes shown in lane 2 (indicated by the blue arrow), but when we added 25X concentrated unlabeled PXC to compete, unlike the results shown in (b), the binding of duplexes is completely blocked, as seen in lane 3.


Page 14 of 28

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

Figure 4. Analysis of structure-specific binding of Pol I to PXC. (a) PXC (lane 1, green arrow) was mixed with 1ul Pol 1 (lane 2). Gel shift indicates PX binding of the protein (red arrow). (b) Lane 1: PX only. Lane 2: PX and Pol 1. Lane 3: PX, Pol 1 and 25X concentrated unlabeled duplex competitors. It can be seen that Pol I binds to PXC (red arrow) even in the presence of duplex competitors. (c) Lane 1: labeled duplex (yellow arrow). Lane 2: labeled duplex and Pol I causing a gel shift (blue arrow). Lane 3: labeled duplex, Pol 1 and 25X concentrated unlabeled PX-DNA. There is no binding of Pol I to duplex competitors in the presence of PXC. (d) Lane 1: labeled PX and Pol 1 (red arrow). Lane 2: labeled duplex and Pol 1 (blue arrow). Lane 3: labeled PX (green arrow) and labeled duplex (yellow arrow). Lane 4: labeled PX, labeled duplex and Pol 1. Lane 5: labeled PX, labeled duplex and 1/3 amount of Pol 1. The bands in lanes 4 and 5 correspond to the PXC-protein complex and with reduced amounts of protein, an increase in the band intensity of PXC can be seen in lane 5 (green arrow).

These results indicate both that Pol I has a preference for binding to the PX structure and that the binding is structure-specific. To confirm this conclusion further, we mixed labeled PXC and duplex of the same concentration to bind with Pol I, and the results are shown in Figure 4 (d). In lane 4, when both structures exist, the largest gel shift corresponds to the PXC-protein complex (indicated by the red arrow). In lanes 4 and 5, when we reduced the amount of Pol I, more duplex (indicated by the yellow arrow) is left unbound than PXC. Thus, the binding of Pol I to PXC is further confirmed to be structure-specific. DNA Polymerase 1 has three domains, corresponding to three different activities as shown in Figure S8 (a). To study further


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

the relationship between binding and different domains, we used Klenow Fragment and Exo- for the binding assays. As seen in Figure S8 (b), Klenow Fragment and Exo- could both bind to duplexes (lane 2 and lane 3 respectively). In a similar experiment with PXC, Klenow Fragment was seen to bind to PXC (Figure S8 (c), lane 2) while Exodid not (lane 3). Thus, the binding of Pol I to PXC is apparently not related to its 5'-3' exonuclease domain. So as to quantitate the binding of Pol I to PX we have measured the concentration of Pol I-PX in a quantitative fashion on a non-denaturing gel (Figure 5 (a)). The resulting curve fitting leads to a Kd of about 220 nM (Figure 5 (b)). As a control even closer to PX structure (with a similar pair of ends to its helices), we have used a DPE double crossover molecule3 (Figure S9) with the same double helical sequences, but lacking the large number of crossovers that characterize PX DNA. The resulting Kd derived from the binding curve is about 13 M Figure S8. Thus, even the closest control still binds more weakly than PX-DNA.

.


Page 16 of 28

Page 17 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

Figure 5. a) Non-denaturing gel showing the binding results of pure Pol I to PX. Lane 1 contains PX only at 2 mM. Lane 2 contains PX at 2 mM mixed with 0.25 mM of Pol I. Lane 3 contains PX at 2 mM mixed with 0.5 mM of Pol I. Lane 4 contains PX at 2 mM mixed with 1 mM of Pol I. Lane 5 contains PX at 2 mM mixed with 2 mM of Pol I. Lane 6 contains PX at 2 mM mixed with 3 mM of Pol I. Lane 7 contains PX at 2 mM mixed with 4 mM of Pol I. Lane 8 contains PX at 2 uM mixed with 5 mM of Pol I. Lane 9 contains PX at 2 mM mixed with 6 mM of Pol I. Lane 10 contains PX at 2 mM mixed with 8 mM of Pol I. b) A graph showing the PX binding with Pol I in different Pol I concentrations. All data from three independent experiments are shown in the graph and fitted to the model B(C) = Bmax*Cn/(C1/2+Cn), in which B(C) is the PX binding efficiency with Pol I; C is the Pol I concentration; Bmax is the maximum binding efficiency and C1/2 is the total Pol I concentration at 50% binding. According to the fitting, Bmax is 1.02; K is 1.22 mM; R-square is 0.9674 and RMSE is 0.06732. The equilibrium dissociation constant (KD) is 0.22 mM (total Pol I concentration minus bound Pol I concentration).


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

Functional Studies:

Figure 6 contains a denaturing gel showing

polymerization results with Pol I, Klenow fragment and Exo fragment with PXC and a duplex competitor. The PXC molecule is stable at 37 ˚C, but when a segment of a single strand is removed to provide a substrate for Pol I, it is no longer stable. Hence, we designed a longer PX complex (Figure S11) which is stable at 37 ˚C even after a single-stranded segment is deleted (Figure S11). The segment deleted for polymerization is shown in black. Pol I, including the Exo fragment, degrades both PXC and duplex, as seen in lane 2 and lane 5, respectively. Klenow fragment and Exo fragment can polymerize the shorter strand of duplex substrates into full length (lane 6 and lane7), but neither of them can polymerize the shorter strand of PXC substrates into full length molecules (lanes 3 and 4). Klenow fragment can bind to PXC, but it cannot polymerize its strands.


Page 18 of 28

Page 19 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

Figure 6. Denaturing gel of polymerization results of PX-DNA complex and duplex with Pol I, Klenow and Exo-. Lane 1 contains the labeled fulllength strand. Lanes 2, 3 and 4 contain the PX substrates with Pol I, Klenow and Exo- respectively. Lanes 5, 6 and 7 contain the duplex substrates with Pol I, Klenow and Exo- respectively. Lane 8 contains the short strand in the substrate.

Structural Modeling: A docking model in Figure 7 shows how two domains of Klenow Fragment might bind to PXC.

The basis of this model is

alignment of the dyad axis of the PX-DNA molecule with a dyad axis relating two Klenow Fragments.

This initial arrangement was followed by MD

simulations. A fully-atomistic model of the PX DNA was constructed using Nucleic Acid Builder (NAB),11 part of AmberTools. Each duplex was made independently using NAB, and then they were connected manually using Cerius2 (a graphical molecular modeling package previously sold by Accelrys, Inc). The resulting structure was minimized in LAMMPS (a molecular dynamics program from Sandia National Laboratories) using a steepest descent method. The sequences for each of the four strands are included in the supplementary information. The fully atomistic model for the


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

Klenow fragment for DNA Pol I was taken from RSCB Protein Databank (1KLN).12

Consideration of Klenow features together with experimental

results for the stoichiometry of Klenow PX binding led us to examine binding orientations that would allow Klenow to bind to the sides of the PX DNA. We examined both canonical and non-canonical DNA binding positions and identified a groove that constituted a possible binding site for PX. The poor binding of Pol I to the DPE molecule serves as a control to exclude its binding to the blunt ends of the PX molecule. Starting from symmetry considerations, we positioned two Klenow fragments in symmetric positions around the PX using Chimera.13 To the resulting complex was solvated with TIP3P waters and neutralized with Mg2+ and Na+ ions. In consideration of the tendency for divalent ions to condense around highly charged polyvalent ions in solution, half the total charge were neutralized with Mg2+, and the other half with Na+. Ions were placed at the loci that would minimize the electrostatic potential energy of the solvated system. Na+ and Cl- ions were then added to a concentration of 150 mM.


Page 20 of 28

Page 21 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

Figure 7. Molecular docking of Klenow Fragment binding to PXC. PXC is indicated by green and the two domains (3’-5’ exonuclease domain and polymerization domain) composing of Klenow Fragment are indicated by purple and red. The left panel shows the binding viewed from the side of PXC and the right panel shows the binding viewed from the top of PXC.


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

The resulting atomistic model was then minimized using steepest descents methods followed by conjugate gradient methods, and the temperature of the system was brought up to 298 K using NVT (constant-temperature, constant volume ensemble) integration. To relax the side chains of the Klenow fragments in proximity to the PX DNA, all Klenow atoms within 4.5 nm of the central axis of the PX DNA were equilibrated for 0.5 ns while the rest of the system was held fixed. After this relaxation, a harmonic potential was placed between the center of mass of each Klenow fragment and the center of mass of the PX DNA. The harmonic potential had a spring constant of 0.25 Kcal/mol*A^2 and an r0 of 20 Ångstroms. This was run using NVT integration at 298 K for 1 ns to bring Klenow fragments into contact with the PX DNA.

Discussion We have found that Pol I binds to the PX motif in a structuredependent fashion, independent of the sequence of the molecule. However, the enzyme is incapable of extending a shortened strand in the PX motif to full-length. Thus, the binding of Pol I to PX-DNA is either a coincidence, or is a consequence of an activity or feature of Pol I that is unknown at this time. Candidates for such activities include binding to parallel DNA duplex molecules, but do not include binding that is associated with strand extension.


Page 22 of 28

Page 23 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

Binding to parallel duplex molecules may very well be associated with the repair and recombination functions noted above. In addition, we have noted earlier9 that the PX structure is a signature of the presence of either complete or interrupted homology, at least in supercoiled molecules. The presence of Pol I, particularly the Klenow portion, in the vicinity of homologous molecules may facilitate homology-dependent processes, even though the molecule is incapable of extending the PX portion itself. Further experimentation clearly will be necessary to elucidate the ultimate importance of our unexpected findings.

Supporting Information: PX-binding of protein fractions collected from Q-sepharose columns (Figure S1), PX-binding of protein fractions in the presence of duplex competitors (Figure S2), PX-binding of protein fractions collected from DNA cellulose columns (Figure S3), Binding titration of protein fractions with PXC (Figure S4), PX-binding of protein fractions collected from size columns (Figure S5), Final purified protein fraction collected from size column (Figure S6), PX binding results with ParC (Figure S7), Duplex and PX binding with Klenow and Exoproteins (Figure S8), Sequences and the schematic drawing of DPE control molecule (Figure S9), Binding results of Pol I to DPE control molecule (Figure S10), PX molecule used for polymerization reaction


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 S11), Short and long PX molecules (Figure S12), Mass spectrum scores (Table S1)

Acknowledgements We thank Prof. Kenneth J. Marians for his generous gift of ParC. We also thank Prof. Paramjit Arora for valuable discussions. This work was supported by the following grants to NCS: GM-29554 from the National Institute of General Medical Sciences, grants CMMI1120890, EFRI-1332411 (also to WAG), and CCF-1526650 from the National Science Foundation, MURI W911NF-11-1-0024 from the Army Research Office, grants N000141110729 and N000140911118 from the Office of Naval Research, DE-SC0007991 from the Department of Energy for DNA synthesis and partial salary support, grant GBMF3849 from the Gordon and Betty Moore Foundation and grant RGP0010/2017 from the Human Frontiers Science Program and by NIH CO6RR0154464 to RPC.


Page 24 of 28

Page 25 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

References 1.

Holliday, R. (1964) Mechanism for gene conversion in fungi, Genet. Res. 5, 282-304.

2.

Wang, X.; Seeman, N. C. (2007) The assembly and characterization of 8-arm and 12-arm DNA branched junctions, J. Am. Chem. Soc., 129, 8169-8176.

3.

Fu, T. J.; Seeman, N. C. (1993) DNA Double crossover structures, Biochem. 32, 3211-3220.

4.

Mathieu, F.; Liao, S.; Mao, C.; Kopatsch, J.; Wang, T.; Seeman, N. C. (2005) Six-helix bundles designed from DNA, NanoLett. 5, 661665.

5.

Rothemund, P. W. K. (2006) Scaffolded DNA origami for nanoscale shapes and patterns. Nature 440, 297-302.

6.

Liu, D.; Wang, W.; Deng, Z.; Walulu, R.; Mao, C. (2004) Tensegrity: Construction of rigid DNA triangles with flexible four-arm junctions, J. Am. Chem. Soc. 126, 2324-2325.

7.

Shen, Z.; Yan, H.; Wang, T.; Seeman, N. C. (2004) Paranemic crossover DNA: A generalized Holliday structure with applications in nanotechnology J. Am. Chem. Soc. 126, 1666-1674.

8.

Seeman, N. C. (2010) Nanomaterials based on DNA, Ann. Rev. Biochem. 79, 65-87.


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

9.

Page 26 of 28

Wang, X.; Zhang, X.; Mao, C.; Seeman, N. C. (2010) Doublestranded DNA homology produces a physical signature, Proc. Nat. Acad. Sci. (USA) 107, 12547-12552.

10.

Gladyshev, E.; Kleckner, N. (2014) Direct recognition of homology between double helices of Neurospora crassa, Nature Comm. 5, article # 3509.

11.

Macke, T.; Case, D. A. (1998) Modeling unusual nucleic acid structures. In Molecular Modeling of Nucleic Acids, Leontes, N. B.; SantaLucia, Jr. J. eds. (Washington, DC: American Chemical Society, pp. 379-393.

12.

Burley, S. K.; Berman, H. M.; Christie, C.; Duarte, J. M.; Feng, Z. K.; Westbrook, J.; Young, J.; Zardecki, C. RSCB Protein Data Bank: Sustaining

a

Living

Digital

Data

Resource

that

Enables

Breakthroughs in Scientific Research and Biomedical Education, Faseb J. 32(1) Supplement S, Abstract 674.3. 13.

Pettersen, E. F.; Goddard, T. D.; Huang, C. C.; Couch, G. S.; Greenblatt, D. M.; Meng, E. C.; Ferrin, T. E. (2004) UCSF Chimera— A visualization system for exploratory research and analysis. J. Comput. Chem., 25, 1605-1612.

14.

Seeman, N. C. (1990) De Novo Design of Sequences for Nucleic Acid Structure Engineering, J. Biomol. Struct. Dyns. 8, 573-581.


Page 27 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


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

Isolation protocol for PX-DNA binding proteins. 57x44mm (300 x 300 DPI)


Page 28 of 28

The PX-Motif of DNA Binds Specifically to E. coli DNA Polymerase I

Recommend Documents