SUPPLEMENTAL EXPERIMENTAL PROCEDURES



SUPPLEMENTAL MATERIALS AND METHODS.

Recombinant proteins. N-terminal thioredoxin and his6 tagged Rdh54 and Rdh54 K352R proteins were overexpressed in E. coli Rosetta strain and purified as previously described 1. In brief, cells were grown at 30˚C to an A600 of 0.6 to 0.8. The temperature was then decreased to 16˚C and the cells were induced for 16 hours with 0.1 mM IPTG. The harvested cells were then frozen at –80˚C. The cells were thawed, resuspended into buffer B (20 mM KH2PO4 (pH 7.4), 150 mM KCl, 10% glycerol, 0.5 mM EDTA, 0.01% Igepal) plus 2 mM (-mercaptoethanol and protease inhibitors, lysed by French Press, and the resulting lysate clarified by ultracentrifugation. The soluble fraction was then passed through a Q-sepharose column. The flow through was applied to a SP-sepharose column, which was developed with gradient from 0-325 mM KCl in buffer B. The Rdh54 peak fractions were purified with Ni2+-NTA-agarose and eluted with 200 mM imidazole in buffer B. The resulting fraction was then purified over a Mono-S column (GE Healthcare) and eluted with a 0-325 ml gradient from 0-325 mM KCl. The protein containing fractions from the Mono-S eluate were then combined, concentrated to ~5 mg/ml, divided into small aliquots and stored at –80˚C. Characterization of thioredoxin tagged Rdh54 has been previously published 1, and the tagged protein is functional in vitro and in vivo.

Flowcells, DNA substrates and DNA curtains. The flowcells were assembled from fused silica slides (G. Finkenbeiner, Inc.) on which microscale diffusion barriers were etched using a diamond-tipped scribe 2. Inlet and outlet ports were made by boring through the slide with a high-speed precision drill press equipped with a diamond-tipped bit (1.4 mm O.D.; Kassoy). The slides were cleaned by successive immersion in 2% (v/v) Hellmanex, 1 M NaOH, and 100% MeOH. The slides were rinsed with filtered sterile water between each wash step and stored in 100% MeOH until use. Prior to assembly of the flowcell, the slides were dried under a stream of nitrogen and baked in a vacuum oven for at least 1 hour. A sample chamber was prepared from a borosilicate glass coverslip (Fisher Scientific) and double-sided tape (~25 µm thick, 3M). Inlet and outlet ports (Upchurch Scientific) were attached with adhesive rings and cured at 120˚C under vacuum. The total volume of the sample chambers was ~4 µl. A syringe pump (Kd Scientific) and actuated injection valves (Upchurch Scientific) were used to control sample delivery, buffer selection and flow rate. The flowcell and prism were mounted within a custom-built heater with computer-controlled feedback regulation that could be used to control the temperature of the sample from between 25-37˚C ((0.1˚C).

DNA curtains were constructed essentially as described 2. All lipids were purchased from Avanti Polar Lipids and liposomes were prepared as previously described 2. In brief, a mixture of DOPC (1,2-dioleoyl-sn-glycero-phosphocholine), 0.5% biotinylated-DPPE (1,2-dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-(cap biotinyl)), and 8-10% mPEG 2000-PE (1,2-dioleoyl-sn-glycero-3-phosphoethanolamine-N-[methoxy(polyethylene glycol)-2000]). Liposomes were applied to the sample chamber for 1 hour. Excess liposomes were flushed away with buffer containing 10 mM Tris-HCl (pH 7.8) and 100 mM NaCl. The flowcell was then rinsed with buffer A (40 mM Tris-HCl (pH 7.8), 1 mM DTT, 1 mM MgCl2 plus 0.2 mg/ml BSA. Neutravidin (330 nM) in buffer A was then injected into the sample chamber and incubated for 30 minutes. After rinsing thoroughly with additional buffer A, biotinylated (-DNA (10 pM) pre-stained with 1-2 nM YOYO1 was injected into the sample chamber, incubated for 30 minutes, and unbound DNA was removed by flushing with buffer. Application of buffer flow caused the lipid-tethered DNA molecules to align along the leading edges of the diffusion barriers. Once the DNA curtains were located, 50 µl of 0.5 M NaCl was injected into the sample chamber at a flow rate of 0.1 ml/min to remove all detectable traces of YOYO1.

The intercalating dye YOYO1 is commonly used to label DNA in single-molecule fluorescence assays, but when illuminated, YOYO1 reacts with molecular oxygen to generate free radical species that rapidly cleave DNA 3. This undesirable outcome is normally inhibited by the inclusion of an oxygen scavenging system comprised of glucose oxidase, catalase, glucose and ß-mercaptoethanol. However, preliminary assays revealed that the ATPase activity of Rdh54 was completely abolished in the presence of this oxygen scavenging system. To overcome this problem we used YOYO1 to first stain and locate the DNA curtains (figure 1B). The dye was then completely removed by briefly flushing the sample chamber with 0.5 M NaCl. This was followed by re-equilibration of the sample chamber with reaction buffer that lacked the oxygen scavenging system.

TIRFM. The basic design of the microscope used in this study has been previously described 4. In brief, the system is built around a Nikon TE2000U inverted microscope with a custom-made illumination system. For this study, a 488 nm, 200 mW diode-pumped solid-state laser (Coherent, Sapphire-CDHR) was used as the excitation source. The laser was attenuated with an appropriate neutral density filter, passed through a spatial filter/beam expander, collimated, and focused through a fused silica prism onto the surface of a microfluidic sample chamber (described below). This gave a Gaussian profile with an elliptical illuminated field of approximately 50 x 200 µm, which was centered over the DNA curtain by means of a remotely operated mirror (New Focus) and the intensity at the face of the prism was typically ~5 mW. Images were detected with a back-illuminated EMCCD detector (Photometrics, Cascade 512B). For experiments that required multi-color detection the different emission spectra were separated by a dichroic mirror (630 DCXR, Chroma Technologies) with a Dual-View image-splitting device (Optical Insights).

Quantum dots, Protein Labeling and TIRFM Reaction Conditions. Quantum dots (Invitrogen) coated with short-chain polyethylene glycol with exposed free amines were labeled with affinity purified, reduced anti-thioredoxin (Immunology Consultants Laboratory, Inc.) using SMCC (succinimidyl 4-[N-maleimidomethyl]cyclohexane-1-carboxylate). The resulting conjugates were then purified over a Superdex 200 10/300 GL gel filtration column (GE Healthcare), which yielded a monodisperse peak, and were stored in PBS (pH 7.4) plus 0.1 mg/ml acetylated BSA at 4˚C. According to the manufacturer this protocol yields 2-3 antibodies per quantum dot.

To optimize conditions for the TIRFM experiments the ATPase activity of Rdh54 was assayed at varying NaCl, MgCl2 and KCl concentrations in the presence of 30 mM Tris-Cl (pH 7.5), 1 mM DTT, 50 (g/ml BSA, 15 (M (base pairs) linear (X174 replicative form I, 1.5 mM ATP, 1.2 (M [(-32P]ATP at varying salt concentrations and in the presence or absence of an oxygen scavenging system (glucose/oxidase enzyme, (-mercaptoethanol, glucose) and YOYO1. Reactions were incubated, quenched and analyzed as described above. The hydrolysis reaction was initiated with either 40 or 400 nM Rdh54 and incubated at 25(C, 30(C or 37(C for 30 minutes. The reactions were quenched with 2.5 (l of 0.5 M EDTA and analyzed by polyethyleneimine-cellulose thin layer chromatography in 0.7 M potassium phosphate buffer. These experiments revealed that the oxygen scavenging system completely abolished the ATPase activity.

ATPase assays were also performed in the presence of anti-thioredoxin antibody or anti-HA tag antibody (ICL Labs) to determine whether antibody binding affected the activity of Rdh54. Reactions were set up in 30mM Tris-Cl pH 7.5, 1 mM DTT, 50 (g/ml BSA, 2 mM MgCl2, 15 (M base pairs cut (X174 replicative form I, 1.5 mM ATP, 0.6 (M [(-32P]ATP and at varying amounts of anti-thioredoxin in 0.6x PBS. The reactions were initiated with either 40 nM Rdh54, quenched and analyzed as described above. These experiments showed no evidence that the ATPase activity of Rdh54 was inhibited in the presence of the anti-thioredoxin antibody or anti-thioredoxin coated quantum dots (Figure S2A and B).

For TIRFM experiments, Rdh54 (10-20 nM monomer) was mixed with anti-thioredoxin quantum dot in a 1:2 ratio (Rdh54 dimer : quantum dot) in reaction buffer containing 40 mM Tris-Cl (pH 7.8), 1 mM MgCl2, 1 mM DTT, and 0.2 mg/ml BSA, in a total volume of 25 µl, and reactions were incubated for 15-20 minutes on ice. Experiments performed at a 1:20 ratio of Rdh54 dimer to Qdot yielded similar results (data not shown). The reactions were then diluted to a final volume of 100 µl immediately prior to injecting the protein into the sample chamber. All TIRFM experiments were done using 40 mM Tris-Cl (pH 7.8), 1mM MgCl2, 1mM DTT, 0.2 mg/ml BSA with or without ATP, as indicated. Importantly, the inclusion of 1 mM DTT was also necessary to minimize “blinking” of the quantum dots 5.

Gel filtration, Analytical Ultracentrifugation, and Rdh54 to Qdot ratio.

Gel filtration was performed to determine the oligomeric state of Rdh54 in solution (Figure 3S). 80 µg of purified Rdh54 was resolved on a Sephacryl S300 column (1 x 44.5 cm; 35 ml bed volume) in buffer T (40 mM Tris-HCl pH 7.8, 1 mM MgCl2, 1 mM DTT, and 150 mM NaCl) with a flow rate of 0.2 ml/min at 4˚C. Fractions (0.4 ml) were collected and analyzed by SDS-PAGE with Coomassie Blue staining. The molecular size markers used for calibrating the column were apoferritin (443 kDa), catalase (232 kDa), aldolase (158 kDa), and albumin (67 kDa). The same buffer and run conditions were used for these standards. Rdh54 (monomer mass of 125 kDa) eluted with an apparent molecular weight of 249 kDa, which was most consistent with a dimer (Figure S4). E. coli thioredoxin is a monomer in solution (Martin, 1995) and therefore does not likely contribute to dimer formation by the thioredoxin tagged Rdh54.

Sedimentation velocity experiments were conducted at 25˚C in a Beckman Optima AXL-1. All experiments were performed in 40 mM Tris (pH 7.8), 1 mM MgCl2, 1 mM DTT, 2.5 nM quantum dot (Qdot 705), and with or without 2.5 nM Rdh54. The sedimentation of the quantum dots ((260nm ( 36x106 M-1cm-1) was monitored by measuring absorbance at 280 nm (Figure 5S). The sedimentation properties of the samples were dominated by Qdots because of their high density, which precluded accurate calculation of the protein:Qdot stoichiometry. However, these data clearly demonstrate that the quantum dots and Rdh54-Qdot conjugates were monodisperse, that conjugating Rdh54 to the quantum dots did not induce aggregation, and that there was no more than one Qdot per dimer of Rdh54.

For the labeling procedure, we assumed that the conjugation reaction was random and that the number of Rdh54 dimers per Qdot followed a Poisson distribution. Therefore we can predict that at a 1:2 ratio of Rdh54 dimer to Qdot 61% of the quantum dots will be unlabeled, 30% of the complexes should be comprised of a single Qdot bound by one dimer of Rdh54, and 9% of the Qdots could have two or more Rdh54 dimers 6; 7. As additional controls, all of the DNA binding and translocation experiments were also performed at a ratio of one Rdh54 dimer per twenty Qdots (1:20). Under these conditions Poisson statistics predict that 95% of the Qdots will be unlabeled, 5% will have a single Rdh54 dimer, and 0% will have two or more dimers of Rdh54. At the 1:20 conjugation ratio we saw similar total amount of protein bound to the DNA, similar translocation rates, and we also confirmed that the DNA-looping activity still occurred at the higher conjugation ratios (Supplementary Video 1 and data not shown). This strongly suggests that the observed behaviors were not due to multiple Rdh54 dimers bound to each quantum dot.

Nonradiant Quantum dots

To determine whether all of the quantum dots used in our assays were visible, we fluorescently tagged the quantum dots using as fluorescent dye with an emission spectrum different from the quantum dots and then visualized the individual quantum dots using TIRFM. Briefly, the PEG coated quantum dots (Invitrogen) contain primary amines at a subset of the PEG chain termini. For labeling, the quantum dots were mixed with a 100-fold molar excess of amine reactive fluorescent dye. The green quantum dots (Qdot 565; Invitrogen) were labeled with a red fluorophore (Cy5 N-succinimidly ester; GE Healthcare), and the red quantum dots (Qdot 705; Invitrogen) were labeled with a green fluorophore (Alexa 488 tetrafluorophenyl ester). Reactions were performed at room temperature for 1-hour in 100 mM sodium bicarbonate buffer (pH 8.3). Unreacted dye was then removed by a combination of gel filtration and ultrafiltration, and the concentration of the purified quantum dots was calculated by UV-Vis spectroscopy.

The labeled quantum dots were then immobilized on the surface of a flowcell and the differing emission spectra were separated using a dichroic mirror and imaged to separate halves of the EMCCD. Alexa 488-Qdot705 was imaged using a 488nm laser, and Cy5-Qdot565 was imaged using both a 488 nm and a 633 nm laser. As expected, the signal from the organic dyes bleached rapidly over time, whereas the quantum dot signal did not bleach (data not shown), and these differing photophysical properties were used to verify the identity of all fluorescent signals. The total number of particles in each fluorescent field was then counted and the ratio of Qdot signal to dye signal was used to calculate the fraction of dark quantum dots (data not shown). Based on this analysis, 50.8% of the green quantum dots (Qdot 565) were visible and 49.8% of the red quantum dots were visible (Q705), and these values are in agreement with previously published work 8.

Data Collection, Image Processing, Automated Particle Counting and Particle Tracking with the DNA Curtain Assay

Data collection was done by acquiring streams comprised of 2000-10,000 frames at 8.3 frames per second using a 100 millisecond integration time. All data were collected using Metamorph (Universal Imaging) or NIS-Elements (Nikon) and converted to 8-bit tiff files in NIH Image J.

For particle counting, the processed videos in 8-bit format were imported into Igor Pro (Wavemetrics) for automated image analysis using custom algorithms. An early image in each video time series, [pic], was selected as a threshold to create a binary image, [pic]. To generate this image an intensity value was chosen as the threshold value, [pic], which was initially calculated using:

[pic]

where [pic] was the maximum intensity value of image [pic], and [pic] was the minimum intensity value. The threshold value was then iterated by:

[pic]

where [pic] was the average intensity of the pixels with a value greater than [pic], and [pic] was the average intensity of pixels with values below [pic], and [pic] was the current iteration of [pic]. The current threshold value was compared to the previous value using:

[pic]

The entire process is repeated using [pic] to calculate [pic], which was then used to calculate [pic]. If the threshold value does not change appreciably, [pic] is then chosen from the current iteration. A binary image was then created using:

[pic]

where [pic] is the intensity value of [pic], [pic] is the intensity value of [pic] (the original image), and [pic] is the threshold intensity. The correlation coefficient, [pic], between the binary image and the original image is then calculated to determine the goodness of fit using:

[pic]

where [pic] is the total number of pixels in the [pic]th dimension of [pic], [pic] is the total number of pixels in the [pic]th dimension, [pic] is the average intensity value of the binary image, and [pic] is the average intensity value of the original image that was selected to for the threshold.

The binary image was then processed with a seed-fill method to determine the area and pixel location of all particles within the image. The seed-fill method starts with the first pixel of the binary image with an intensity value of 1 and queries all surrounding pixels for intensity values. An object is defined as a “particle” when all surrounding pixels are identified with the same intensity value. The algorithm continues for the remainder of the pixels within the image that have not already been assigned as a “particle” and this continues until all pixels are queried. This yields a map of the particle locations and areas for the entire image. It is important to note that this algorithm can not differentiate between particles with overlapping signals, therefore the returned values represent the minimum number of Rdh54-Qdot complexes bound to the DNA.

Particle tracking was performed using an custom algorithm written in Igor Pro (Wavemetrics, Lake Oswego, OR), which automatically fit the signals from individual quantum dots to a 2D Gaussian function 4. Videos used for particle tracking analysis were processed using NIH Image J () to improve the signal-to-noise ration of the data for particle tracking. A variety of standard processing techniques were employed, including altering the brightness and/or contrast, smoothing, Gaussian convolution, or combinations thereof, as necessary. A map of the particle locations is used to mask the image from [pic] of the data series and a 2D Gaussian fit is then applied:

[pic]

The masking procedure improves the goodness of fit and significantly decreases computational time. The centroid position of each particle was obtained from the 2D Gaussian fit 9 and saved to a position trace. The position trace was then fitted to a line from [pic] to [pic] to use the temporal information to change the location of the mask before the next image within the video data series is analyzed. This algorithm was then applied to all particles that were identified within the binary image. The precision of this method was calibrated by tracking the positions of quantum dots that were immobilized to fixed positions on a fused silica surface, yielding a standard deviation of (16 nm in the y-direction and (17 nm in the x-direction, consistent with previously reported resolution limits 9; 10. Note that the apparent noise in the tracking data is not caused by inaccuracy of the tracking algorithm, but rather is due to the Brownian fluctuations of the DNA molecules.

In all cases, the outcome of each individual particle location and data trace was confirmed by visual inspection of the corresponding video files, and any proteins that were nonspecifically bound the to flowcell surface (determined by transiently pausing buffer flow; See Supplemental Video 1) were excluded from the analysis.

SUPLEMENTAL VIDEO LEGENDS

Supplementary Video 1. “High-throughput” DNA Curtain Assay for Rdh54 activity. (QuickTime, 85 MB). This video shows Qdot labeled Rdh54 (labeled at a ratio of 1 Rdh54 dimer per 20 Qdots) binding to a DNA curtain. The DNA was initially located by staining with YOYO1, but the dye was removed prior to injection of the protein, therefore the DNA itself is not visible. As the protein enters the flow cell (at t=20 seconds) the background signal increases (due to free protein) and binding activity can clearly be seen as fluorescent spots become associated with the individual strands of DNA. The unbound proteins are rinsed from the sample chamber by approximately t=1 minute 30 seconds. Buffer flow is temporarily halted at t=4 minutes, which caused the DNA (and bound proteins) to diffuse away from the surface and out of the evanescent field. This standard control is used in all of our DNA curtain experiments to distinguish between Qdots bound to the DNA and spurious Qdots nonspecifically absorbed to the flow cell surface. A time stamp is shown (seconds : milliseconds), and the frame rate has been increase approximately 10-fold to aid the viewer.

Supplementary Video 2. Rdh54 is a DNA translocase. (QuickTime, 14.3 MB) This video shows a single Rdh54 complex traveling along the DNA in the direction of buffer flow. The graph to the right of the image illustrates the particle-tracking data generated from this particular Rdh54 complex.

Supplementary Video 3. Multicolor labeling of Rdh54 bound to the same DNA molecule. (QuickTime, 1.2 MB) This video shows several different colored complexes of Rdh54 moving along the DNA. As can be seen in the movie, the proteins display significant amounts of movement and highly dynamic behaviors, but they do not appear to bypass one another. The extrusion and release of several DNA loops occurs during the course of this video (refer to the text for further discussion). We encourage the viewer to use the QuickTime video controls to rapidly scroll back and forth through the movie to aid visual detection of the Rdh54 movement.

Supplementary Video 4. Rdh54 catalyzes reversible extrusion of DNA loops. (QuickTime, 14.1 MB) This video shows a pair of Rdh54 complexes traveling in unison along the DNA. The “upstream” complex (red trace) is extruding a DNA loop as it moves along the helical axis and causes the “downstream” protein (green trace) to be pulled along with it. After loop formation, the complex begins to translocate in the reverse direction until both it and the downstream protein reach their original positions. The graph to the right of the image illustrates the particle-tracking data generated from these Rdh54 complexes. The traces of the two complexes parallel one another precisely, even though the individual complexes are separated by ~8 kilobases of DNA.

SUPLEMENTAL FIGURE LEGENDS

Supplementary Figure 1. Illustration of particle counting procedure. This figure demonstrates the automated algorithm that was used to identify the number and locations of the fluorescent complexes bound to the DNA as also shown in Figure 1. Panel (A) shows a single image of the field-of-view taken while buffer was flowing. Variations in particle-to-particle intensity are most likely due to oligomerization of Rdh54 on the DNA (see text for further details). Panel (B) shows the same image after running the particle-counting algorithm (see Supplemental Experimental Procedure for details), and (C) shows the outlines of all identified particles. Note that the algorithm can not distinguish overlapping signals, therefore visual inspection of the data was always used to confirm the number of complexes and to distinguish between closely spaced particles bound to the DNA. Panel (D) shows the same field of view after buffer flow was turned off, and any complexes remaining bound to the surface were discounted from further analysis.

Supplementary Figure 2. Translocation rates at different ATP concentrations with and against buffer flow. Panel (A) a plot of translocation rate as a function of ATP concentration and (B) shows histograms and the corresponding Gaussian fits of the translocation rates at each different ATP concentration. In (C) the translocation rates were segregated into different data sets for complexes that translocated against the direction of buffer flow (upstream, black squares) and those that translocated in the same direction as buffer flow (downstream, red squares). Histograms from each ATP concentration are shown in (D). As illustrated by the graphs, the mean translocation rate did not vary between the two data sets, even though there were more outlying data points found for the complexes traveling in the direction of buffer flow. It is likely that these outlying points represent proteins whose movement was in some manner assisted by the hydrodynamic force exerted from the buffer.

Supplementary Figure 3. ATP hydrolysis activity of Rdh54. (A) Rdh54 ATPase activity in the presence of anti-thioredoxin antibodies and (B) anti-thioredoxin labeled quantum dots. The ratios of Rdh54 (monomer) to either antiboby (AB) or Qdot are indicated. A control reaction with only PBS (phosphate buffer saline) is also shown. The stability of Rdh54 in the presence (C) and absence (D) of dsDNA are also shown. For these experiments, Rdh54 was preincubated at 25˚C under reaction conditions in the presence or absence of DNA. Radiolabeled ATP was then added to aliquots of the reaction mixes at the indicated intervals to determine if the protein remained active. For the minus DNA preincubations, dsDNA was added back along with the ATP.

Supplementary Figure 4. Rdh54 behaves as a dimer in solution. The oligomeric state of Rdh54 in solution was determined with gel filtration on a Sephacryl S300 column by comparison to molecular weight standards. The Rdh54 thioredoxin fusion has a calculated mass of 125 kDa, and the gel filtration data revealed that the protein eluted with an apparent molecular weight of 249 kDa, which was consistent with a dimer.

Supplementary Figure 5. Analytical ultracentrifugation indicates that Rdh54-Qdot conjugates are monodisperse. To determine whether the antibodies or Rdh54 caused the quantum dots to aggregate we determined the sedimentation coefficients for the unlabeled quantum dots (Qdots), quantum dots coupled to the anti-thioredoxin antibody (Qdots-Ab), and quantum dot labeled Rdh54 (Qdots-Ab-Rdh54). (A) Sedimentation velocity profiles and (B) the corresponding transport analysis with the calculated sedimentation coefficients. The solid line was the best fit to the data points. (C) Integral distribution analysis of the different quantum dot samples 11. As illustrated by this data, the sedimentation properties were dominated by the Qdots, there was no more than one Qdot per Rdh54 complex, and there was no evidence that either the antibody conjugation procedure or the binding of Rdh54 led to aggregation of the nanocrystals.

Supplementary Figure 6. Effect of ATP(S on Rdh54 translocation. To confirm that the long-term movement observed for Rdh54 was ATP-dependent an experiment was performed identically to that presented in figure 7. Immediately after the proteins were injected ATP was replaced with nonhydrolyzable ATP(S, and the long-term activity of the proteins was monitored by collecting 5-minute videos (presented as kymograms) at the indicated intervals. As illustrated by this experiment, Rdh54 does not actively translocate along the DNA in the presence of ATP(S.

Supplemental Figure 7. Models for loop extrusion by Rdh54. Models are presented that would allow for translocation-dependent DNA loop extrusion by Rdh54. In each example, the free (F) and tethered (T) ends of the DNA are indicated. The translocase motors are shown as blue circles and the stationary domains are shown as blue squares. (A) Bipartite DNA binding with a separate motor and anchor domain contained within the same polypeptide. A single protein is presented for clarity, but similar models can be invoked with a multimeric translocase. The relative movement of the DNA end and the bound protein would depend upon the proteins orientation on the DNA. (B) A partially active multimer with several translocase subunits. Alternating the use of motor subunits with differing orientations would provide a mechanism allowing reverse translocation. (C) A fully active multimer with motors oriented in opposite directions. This organization yields a complex that would always appear to move toward the tethered end of the DNA and the free end would move at twice the rate of the translocating protein. Additional details are presented in the text.

[pic]

[pic]

[pic]

SUPPLEMENTAL REFERENCES

1. Chi, P., Kwon, Y., Seong, C., Epshtein, A., Lam, I., Sung, P. & Klein, H. L. (2006). Yeast recombination factor Rdh54 functionally interacts with the Rad51 recombinase and catalyzes Rad51 removal from DNA. The Journal of Biological Chemistry 281, 26268-26279.

2. Graneli, A., Yeykal, C. C., Prasad, T. K. & Greene, E. C. (2006). Organized arrays of individual DNA molecules tethered to supported lipid bilayers. Langmuir 22, 292-299.

3. Åkerman, B. & Tuite, E. (1996). Single- and double-strand photocleavage of DNA by YO, YOYO, and TOTO. Nucleic Acids Research 24, 1080-1090.

4. Graneli, A., Yeykal, C. C., Robertson, R. B. & Greene, E. C. (2006). Long-distance lateral diffusion of human Rad51 on double-stranded DNA. Proceedings of the National Academy of Sciences (USA) 103, 1221-1226.

5. Hohng, S. & Ha, T. (2004). Near-complete suppression of quantum dot blinking in ambient conditions. J Am Chem Soc 126, 1324 - 5.

6. Block, S. M., Goldstein, L. S. & Schnapp, B. J. (1990). Bead movement by single kinesin molecules studied with optical tweezers. Nature 348, 348 - 52.

7. Seitz, A. & Surrey, T. (2006). Processive movement of single kinesins on crowded microtubules visualized using quantum dots. EMBO J 25, 267 - 77.

8. Yao, J., Larson, D. R., Vishwasrao, H. D., Zipfel, W. R. & Webb, W. W. (2005). Blinking and nonradiant dark fraction of water-soluble quantum dots in aqueous solution. Proc Natl Acad Sci U S A 102, 14284 - 9.

9. Cheezum, M. K., Walker, W. F. & Guilford, W. H. (2001). Quantitative comparison of algorithms for tracking single fluorescent particles. Biophys J 81, 2378 - 88.

10. Qian, H., Sheetz, M. P. & Elson, E. L. (1991). Single particle tracking. Analysis of diffusion and flow in two-dimensional systems. Biophys J 60, 910 - 21.

11. Van Holde, K. E. & Weischet, W. O. (1978). Boundary analysis of sedimentation-velocity experiments with monodisperse and paucidisperse solutes. . Biopolymers 17, 1387-1403.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download