Rapid Injection Linear Dichroism for Studying the Kinetics of Biological

Jun 14, 2012 - James Carr-Smith , Raúl Pacheco-Gómez , Haydn A. Little , Matthew R. Hicks , Sandeep Sandhu , Nadja Steinke , David J. Smith , Alison...
3 downloads 0 Views 885KB Size
Download PDF
Article pubs.acs.org/ac

Rapid Injection Linear Dichroism for Studying the Kinetics of Biological Processes Matthew R. Hicks,† Alison Rodger,† Yu-pin Lin,‡ Nykola C. Jones,§ Søren Vrønning Hoffmann,§ and Timothy R. Dafforn*,‡ †

Department of Chemistry and Warwick Centre for Analytical Science, University of Warwick, Coventry, CV4 7AL, United Kingdom School of Biosciences, University of Birmingham, Birmingham, B15 2TT, United Kingdom § Institute for Storage Ring Facilities (ISA), Department of Physics and Astronomy, Aarhus University, DK-8000 Aarhus C, Denmark ‡

ABSTRACT: Linear dichroism is defined as the differential absorbance of linearly polarized light oriented in two orthogonal directions by an aligned sample. The measurement of a linear dichroism (LD) spectrum of a sample provides two key pieces of structural information. First, that the sample and the chromophores within the sample are able to align. Second, given knowledge of the transition polarization directions of the chromophores, the orientation of the chromophores within the aligned sample can be resolved. It has been shown that LD can provide unique information on the structure of some of the more challenging biomolecular complexes. This has included macromolecular protein and peptide fibers such as actin, tubulin, and amyloids as well as protein−membrane complexes and DNA−protein complexes. Much of this work has been enabled by the development of a low volume Couette flow cell that efficiently aligns long molecules in solution. However, the current Couette system is inherently complex to assemble for each experiment and hence not suited to measurement of rapid reactions. In this paper we detail the development of the first rapid injection LD cell. The system utilizes a conventional stopped-flow injection system coupled to a modified low volume Couette cell, where a narrow bore capillary replaces the normal solid central rod. The system is shown to have similar optical characteristics to the conventional LD Couette flow cell but with the added benefit of a much shorter dead time (0.60 s compared to ∼60). The rapid injection Couette cell has been used to measure the degradation of DNA by DNA exonuclease I, providing data that would not be available using a conventional system.

T

when the sample is aligned then provides information on both the shape of the overall particle and the alignment of the chromophores within the system. In the majority of flow LD studies, the biomolecule is aligned by interaction with a shear gradient in a Couette flow cell. A Couette cell is simply an outer cylinder into which a concentrically oriented circular cross-section rod (whose outer diameter is less than the inner diameter of the cylinder) is inserted. The sample is then held in the space between the outer cylinder and inner rod, and the shear force is generated by the rotation of either the inner rod or outer cylinder. The construction of these cells presents many design challenges, and for a long time, designs were specific to each laboratory that used the instrument. However, in the past 10 years we have pioneered the development of a commercially available low volume Couette cell for use in far UV linear dichroism.1 In this instrument the cylinder and rod are constructed of high purity amorphous quartz to allow transit of the spectrometer beam through the sample unhindered. The relatively small size of the cell (typically the inner diameter of the cylinder is 3 mm and the outer diameter of the rod is 2.5 mm, making a total path

he assembly of protein molecules into linear polymers of monomers is a fundamental process that underlies the morphology of biological systems. For example, the polymerization of the protein actin can modify the shape of the cell while also providing pathways for the transport of materials within the cell. It is therefore clear that measuring the kinetic processes of polymerization and depolymerization of biological fibers plays an essential part in our understanding of the cell. Such measurements are often made difficult by the lack of an optical signal that changes because the chemical composition of the chromophores remains the same. We have found that linear dichroism spectroscopy is an attractive method for this purpose, as the signal is inherently dependent on fiber length. However, to date the dead time (the time taken between reagent mixing and measurement) has been at least 30 s, but more typically about 1 min. In this paper we report a new rapid injection system that reduces the dead time by a factor of 100, enabling measurement of the early stages of fiber assembly. Linear dichroism (LD) is defined as the differential absorbance of orthogonally oriented linearly polarized light by a sample. For a sample to exhibit LD, there is a requirement for the chromophores within the sample to be aligned in an ordered fashion. Biomolecules with a high aspect ratio (that is, having one axis much longer than the orthogonal axis) can be aligned using solution shear flow. The resulting LD measured © 2012 American Chemical Society

Received: March 27, 2012 Accepted: June 14, 2012 Published: June 14, 2012 6561

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566

Analytical Chemistry

■

length 0.5 mm and a sample volume of only 65 μL) has widened the use of linear dichroism for the study of biological systems and has recently allowed far UV LD measurements to be made using synchrotron light sources.2−4 We and others have used this new low volume Couette cell in combination with LD to provide novel information on the structure and dynamics of a range of biomolecular systems. These systems have encompassed the vast majority of biomolecular interactions including those involving proteins, nucleotides, and phospholipid membranes. Studies of protein fibers using LD have included investigations into the assembly of actin,5,6 tubulin,7 RecA,8,9 amyloid,5,10 filamentous bacteriophage,11 and FtsZ.12−14 Most recently a number of studies involving biological membranes have resulted from the observation that phospholipid vesicles can be shear deformed into elongated ellipsoids that can be aligned along with any chromophores dissolved within the membrane.15 We have shown that LD can also be used to study the folding of membrane proteins without the interference from nonspecific aggregation effects that usually plague these sorts of studies.3,16−18 Linear dichroism data from studies of DNA have provided information on the structure of the DNA itself, including information on the tilts of bases as well as information on the binding modes of small molecules that interact with the DNA.2,19−21 In this latter case the small molecule gains its alignment from the interaction with the aligned DNA molecule. We have also shown that LD can be used to provide one of the few continuous assay methods for studying the kinetics of DNA fragmentation by restriction endonucleases.22 The outcome of all these studies has been an appreciation of the utility of LD measurements but also a realization of the limitations of the instrumentation that accompanies the technique. In particular, although the configuration of the Couette cell is advantageous when used to align biomolecules, it presents a significant hurdle when measurements of rapidly changing biomolecular process are being made. Such rapid changes are often encountered in ligand-induced conformational changes of biomolecules such as in membrane protein folding, fiber formation kinetics, and many enzyme-catalyzed reactions. The studies of such processes using LD have been limited by the time required to assemble and fill the cell. For currently available cells, this process commonly takes in excess of 30 s, meaning that the first minute of any reaction cannot be easily measured using LD. In this work we address this deficiency by developing the first rapid injection Couette flow LD cell. The cell uses the same overall configuration as the previous low volume Couette flow cell but adds a central injection port for sample entry and combines this with a syringe drive and mixing system similar to that used in stopped-flow devices. We have mounted the system on the UV1 beamline at the ASTRID synchrotron radiation source to take advantage of the increased photon flux and hence enhance the signal-to-noise ratio of the measurements; however, the system could also be used on a benchtop instrument. We have tested the performance of the device, demonstrating that it has reduced the dead time by approximately 100-fold with only a minor reduction in optical performance. We illustrate the utility of the device by applying it to the DNA exonuclease I digestion of genomic DNA.

Article

MATERIALS AND METHODS Materials. Deoxyribonuclease I from bovine pancreas and calf thymus DNA were obtained from Sigma-Aldrich as were all other buffers and chemicals. All DNA concentrations are quoted in concentration of bases. Construction of Rapid Injection LD (riLD) Cell. Existing stopped-flow systems are generally based on a configuration that uses a pneumatic or stepper motor-driven ram pushing samples from two drive syringes (containing the two components of the reaction) through a mixer into an optical cell. Measurements are made once the sample enters the cell, an event that is triggered by the sample displacing the plunger in a third syringe that is in line with the cell. In effect this system is a closed hydraulic system with movement of the drive syringes coupled hydraulically to the third stop syringe. Unfortunately, the use of a Couette cell for LD measurements makes the implementation of such a closed system very difficult. We therefore opted for an open injection system where the sample is introduced into the cell, still using the twin drive syringes and mixer arrangement but with the drive system configured to deliver a known amount sufficient to fill the sample chamber. The injection system (see Figure 1) was constructed using 2 2.5

Figure 1. Diagrammatic representation of the riLD system showing how movement of the two syringes leads to mixing of the sample and injection into the LD cell. Reactant valves facilitate the filling of the syringes, while purge valves allow residual reaction solution to be removed from the T-mixer and all pipes downstream.

mL Hamilton glass drive syringes (all tubing was 1/16 in. OD, 0.040 in. ID PEEK tubing) which contained the two components of the reaction to be studied. These were linked to two sets of three-way valves (IDEX Health and Science LLC, Rohnert Park, CA) arranged in series. The first of these valves in each reagent line allowed the filling and purging of the syringes, while the second in the series allowed the system to be washed and dried between each experiment. The mixer was a “T” type mixer of in-house design. The conventional Couette cell sample chamber comprises a quartz outer capillary (stoppered at the bottom) arranged concentrically with respect to a smaller quartz rod. Typically the cell is filled by manually pipetting the sample into the outer capillary, mounting this on a rotating spindle, and then inserting the rod. In the riLD system the inner quartz rod is replaced by another quartz capillary of slightly larger outer diameter than the existing rod (2.8 mm) with an inner bore diameter of 1.2 mm. The sample can then be injected into the cell via the channel in the inner capillary (Figure 1). The third issue that had to be addressed in the design and construction of riLD was a method of cleaning the sample lines downstream of the mixer. This is an important consideration as, 6562

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566

Analytical Chemistry

Article

A nonspinning spectrum was subtracted from this to account for the inherent LD signal of the system originating from the optics and the detector.

after the injection, some of the reagents (now mixed) remain resident in the tubing between the mixer and the Couette cell. This material needs to be ejected between experiments to ensure that there is no contamination between new injections and already reacted reagents. To clean this part of the instrument, a pair of “T” format valves were placed in the sample lines just upstream of the mixer. To clean the system, clean buffer and then compressed air are directed from the vacant ports on the “T” valves through to the capillary. Thus, the cell is dried and filled with air between experiments. The final part of the instrument that needed development was the light source and detection system. Similar injection systems used in rapid circular dichroism, fluorescence, and absorbance measurements tend to use large amounts of sample. It was therefore a priority to design a system where the light source and detection system ensured the best signal-to-noise. This was particularly important in the development stage, where the system is not optimized to reduce the number of data accumulations that have to be averaged for each experiment to produce statistically significant data, which in turn reduces the amount of sample used. Our previous studies have shown that the use of a synchrotron light source for LD experiments significantly improves the signal-to-noise compared with xenon arc light sources.4 To provide such high signal-to-noise, the riLD device was installed on the UV1 beamline of the ASTRID synchrotron radiation source. ASTRID Beamline Configuration. Spectra were first measured between 180 and 350 nm in 1 nm steps using the low-energy grating of the UV1 beamline at a spectral resolution of 0.5 nm. The beamline was set up as described previously.4 In short, the light from UV1 was linearly polarized with a MgF2 Rochon polarizer (B. Halle GmbH, Germany) and converted to alternating horizontal and vertical polarized light with a CaF2 Photo Elastic Modulator (Hinds, Hillsboro, OR). The light was horizontally focused with a lens onto the center of the Couette cell and detected with a UV enhanced photomultiplier tube (ET Enterprises Ltd., UK). The LD signal was extracted from the detector with a lock-in amplifier. The output signal from the lock-in amplifier was calibrated to ΔOD units with a double 45° quartz plate (Jasco, UK). In terms of Δ absorbance units, a lock-in amplifier signal of 1 mV is 3.26 × 10−4 ΔOD.23For rapid kinetic measurements, the speed of collection needs to be increased. To achieve this, the data collection system was reconfigured by fixing the high voltage (HT) supplied to the photomultiplier tube (PMT). This removes the initial optimization carried out by the data collection electronics when running in the more conventional variable HT mode. By using a fixed HT and a short lock-in amplifier time constant of 30 ms, the detector is able to collect data every 150 ms, which is adequate to test the rapid injection system as evidenced in the data presented below. Under these conditions, the noise level in the spectra is in the low 10−5 ΔOD range, well-suited for detecting even small rapid changes. DNA Exonuclease I Kinetics. DNA exonuclease I kinetics were carried out in reaction buffer A, which contained 50 mM Tris-HCl pH 8.0, 1 mM CaCl2, and 1 mM MgSO4. The reaction was initiated by mixing the enzyme with a known concentration of calf thymus DNA (ctDNA) at 27 °C. Conventional Couette Flow Cell. Where rapid injection is not used for LD experiments, the experiments were carried out using our standard microvolume Couette. The Couette requires 65 μL of sample, and for all experiments, spectra were measured while the external cylinder was rotated at 3000 rpm.

■

RESULTS Comparison of Injection and Noninjection Cell for Wavelength Scans. The format of the riLD cell introduces significant extra optical complexity into the light beam, which has the potential to distort the LD spectrum measured by the spectrometer. To test whether this is the case, LD spectra were measured of calf thymus DNA, a sample with a well-defined LD spectrum. Data from these tests (Figure 2) show that the

Figure 2. Measurements of the LD spectrum of 100 μM ctDNA in a conventional LD Couette containing a solid central rod (thick dashed line) and the Couette containing a capillary in place of the rod (thin line). The experiment was carried out with a 1 nm step size, and a dwell time of 2 s per point.

spectral shape is maintained between the two different cells. The only observable differences are a decrease in LD signal and increase in noise at low wavelengths for the riLD cell compared to the noninjection cell. This increase in noise is an expected result of the increase in overall path length of the riLD cell, the light beam now has to transit the aligned sample as well as the unaligned sample which is in the central capillary. This configuration now has an absorbance higher than that of the conventional Couette (with a solid central rod). This decrease in the photon flux to the detector thus increases the noise at lower wavelengths. The decrease in LD signal in the injection cell was not expected, as in general the decrease in signal caused by the reduction in path length is compensated by the increase in shear rate and hence molecular alignment. The observation that this compensation does not occur in this system could be the result of one of two factors. First, we have observed in other systems that the linear relationship between the increase in alignment and increase in shear does not hold at high shear.24 Such behavior in this system would explain the unexpectedly low signal. A second factor that may also explain this anomaly could be the introduction of some turbulent flow into the system by inaccuracies in the alignment of components in the system, which is exacerbated by the tighter tolerances in the new cell. Alternatively, it is possible that the substitution of an open-ended capillary for a closed rod in the system may be introducing some form of nonoptimal flow regime into the system. Although we have observed this signal difference in the forms of experiments for which the cell is envisaged, the shear flow merely provides the alignment that allows LD to be measured. The kinetics studies for which the system will be used are only interested in the rate of change of that signal, which will not be effected by any of these effects. 6563

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566

Analytical Chemistry

Article

Determination of Dead Time for Rapid Injection. In conventional stopped-flow devices the dead time can be defined simply as the time taken for the sample to transit the mixer, achieve complete mixing, and then enter the detection chamber. However, for the riLD system, the dead time must also include the time taken to establish a shear gradient across the sample in the cell as well as the time taken to induce alignment in the solute molecules. We therefore chose to compare the LD of a ctDNA sample loaded directly into the riLD cell and loaded by injection. We mixed ctDNA (200 μM stock) with water at a ratio of 1:1 and measured the change in LD at 260 nm. The dead time was defined as the time taken from the onset of injection to the establishment of an LD signal at the magnitude expected for the concentration of DNA (Figure 3). These data show that the injection is complete

For this study, the reaction mediated by the nonspecific cleavage of DNA by DNase I was chosen to test the riLD instrument. This reaction offers a number of advantages over that of restriction endonucleases for this purpose, most important of these being the low cost of the enzyme and substrate (calf thymus DNA). It is also an interesting system to study in its own right because riLD gives a direct readout of cleavage in contrast to other methods that have been used. To test the ability of conventional manual LD to detect digestion of ctDNA by DNase I, equal volumes of low concentrations of the enzyme were mixed with the DNA (final concentration of 50 μM ctDNA and 0.65 nMDNase I) and the change in the LD signal of the DNA with respect to time was measured (Figure 4a). These data show a clear negative

Figure 3. The dead time of the riLD system was determined by mixing a known amount of ctDNA (100 μM final concentration) with an equal amount of water using the new apparatus. The change in LD signal at 260 nm shows that 600 ms elapse between the onset of injection and the establishment of a stable LD signal. Data were accumulated every 150 ms with a fixed 561 V supplied to the PMT.

within four data points after injection, which equates to a dead time of approximately 600 ms. It should be noted that the total signal change upon injection is 10 times that which would expected if we were only observing the introduction of a signal from the DNA upon injection. However, this is not the case, as prior to the injection, the sample chamber is entirely empty. This significantly alters optical properties of the cell, leading to a highly distorted baseline LD signal. Upon injection we therefore observe a signal change that includes both the introduction of the LD signal from the DNA and the reestablishment of the correct optical configuration of the cell once filled with sample. Taking this into account, it is clear that the injection is approximately 2 orders of magnitude faster than the manual loading of the microvolume Couette cell. This indicates that injection of the sample is the limiting process as opposed to establishment of shear flow or alignment. Thus, the conventional definition of dead time is appropriate for the riLD cell Measurement of DNA Exonuclease I (DNaseI) Digestion of DNA Using the riLD Cell. Conventional stopped-flow systems employing optical detection are tested using a range of well-characterized chromogenic reactions. Proper testing of the riLD system requires a test reaction where the sample orientation changes. One such reaction is the enzymatic hydrolysis of DNA. In this reaction, DNA, which can be aligned and has a clear LD signal at 260 nm, is degraded to small fragments that have a greatly reduced LD signal. Our previous work22 has shown that LD can be used to follow the kinetics of the hydrolysis of DNA by restriction endonucleases.

Figure 4. Digestion of ctDNA (100 μM) by the addition of DNase I (1 nM) is monitored by measurement of (a) a decrease in the amplitude of the linear dichroism spectrum. Completion of the reaction is observed when the LD signal throughout the spectral range has reached zero. The time between the initiation of each spectrum is 340 s. (b) To provide a higher density of data points and hence temporal accuracy, the change in LD at 260 nm was measured with respect to time, showing an initial linear phase followed by a progressive decrease in the slope of the data to zero.

minimum at 260 nm in the first spectrum, which indicates the presence of DNA bases arranged perpendicular to the alignment axis of the Couette cell. Subsequent scans show that this signal diminishes, agreeing with the expectation that the action of DNase I leads to the generation of progressively shorter DNA fragments which are less able to align and hence produce a less intense LD signal. The end-point of the experiment is a complete disappearance of the LD signal, indicating that the DNA has fragmented to a level at which it can no longer be aligned. 6564

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566

Analytical Chemistry

Article

The same experiment monitored at a fixed wavelength gave the data shown in Figure 4b. This shows that after an initial linear decrease in LD signal, the reaction decelerates. Note also that for this manual experiment the dead time was measured at 70 s. Measurement of DNase I Digestion of ctDNA Using riLD. To establish the performance of the riLD system, experiments were carried out using increased amounts of DNase I in the reaction mixture (27 nM, 54 nM, and 108 nM before mixing). The increase in DNase I concentration significantly increases the rate of ctDNA digestion, rendering it impossible to monitor using the manual injection system. The riLD data (Figure 5) show that at the higher enzyme

(and hence reaction rate) than was previously possible using the conventional Couette system Now that the riLD system is established it stands ready to be applied to a number of biological systems where rapid measurements are essential to understanding mechanistic details. It is expected that these studies will address some fundamental processes, from the assembly of proteins in the cellular cytoskeleton through the folding of membrane proteins to the measurement of conformational changes induced by DNA binding proteins.

■

AUTHOR INFORMATION

Corresponding Author

*E-mail: T.R.Daff[email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS We acknowledge funding from the European Community′s Seventh Framework Programme (FP7/2007−2013) under grant agreement n° 226716 and the ELISA project. S.V.H. acknowledges financial support from the Lundbeck Foundation and The Danish Council for Independent Research | Natural Sciences (FNU). T.D. and A.R. acknowledges funding from the EPSRC GR/T09224/01. The manuscript was written through contributions from all authors.

■

REFERENCES

(1) Marrington, R.; Dafforn, T. R.; Halsall, D. J.; Rodger, A. Biophys. J. 2004, 87 (3), 2002−12. (2) Rittman, M.; Hoffmann, S. V.; Gilroy, E.; Hicks, M. R.; Finkenstadt, B.; Rodger, A. Phys. Chem. Chem. Phys. 2012, 14 (1), 353−66. (3) Hicks, M. R. ; Dafforn, T. R.; Damianoglou, A.; Wormell, P.; Rodger, A.; Hoffmann, S. V. The Analyst 2009, 134 (8), 1623−8. (4) Dicko, C.; Hicks, M. R.; Dafforn, T. R.; Vollrath, F.; Rodger, A.; Hoffmann, S. V. Biophys. J. 2008, 95 (12), 5974−7. (5) Dafforn, T. R.; Rajendra, J.; Halsall, D. J.; Serpell, L. C.; Rodger, A. Biophys. J. 2004, 86 (1), 404−10. (6) Szczesna, D.; Lehrer, S. S. Biophys. J. 1992, 61 (4), 993−1000. (7) Marrington, R.; Seymour, M.; Rodger, A. Chirality 2006, 18 (9), 680−90. (8) Takahashi, M.; Kubista, M.; Norden, B. Biochimie 1991, 73 (2− 3), 219−26. (9) Norden, B.; Wittung-Stafshede, P.; Ellouze, C.; Kim, H. K.; Mortensen, K.; Takahashi, M. J. Biol. Chem. 1998, 273 (25), 15682−6. (10) Marshall, K. E.; Hicks, M. R.; Williams, T. L.; Hoffmann, S. V.; Rodger, A.; Dafforn, T. R.; Serpell, L. C. Biophys. J. 2010, 98 (2), 330− 338. (11) Clack, B. A.; Gray, D. M. Biopolymers 1992, 32 (7), 795−810. (12) Marrington, R.; Small, E.; Rodger, A.; Dafforn, T. R.; Addinall, S. G. J. Biol. Chem. 2004, 279 (47), 48821−9. (13) Pacheco-Gomez, R.; Roper, D. I.; Dafforn, T. R.; Rodger, A. PloS One 2011, 6 (6), e19369. (14) Small, E.; Marrington, R.; Rodger, A.; Scott, D. J.; Sloan, K.; Roper, D.; Dafforn, T. R.; Addinall, S. G. J. Mol. Biol. 2007, 369 (1), 210−21. (15) Ardhammar, M.; Mikati, N.; Norden, B. J. Am. Chem. Soc. 1998, 120, 9957−9958. (16) Rodger, A.; et al. Phys. Chem. Chem. Phys. 2002, 4, 4051−4057. (17) Hicks, M. R.; Damianoglou, A.; Rodger, A.; Dafforn, T. R. J. Mol. Biol. 2008, 383 (2), 358−66. (18) Damianoglou, A.; Rodger, A.; Pridmore, C.; Dafforn, T. R.; Mosely, J. A.; Sanderson, J. M.; Hicks, M. R. Protein Pept. Lett. 2010, 17 (11), 1351−1362.

Figure 5. Rapid kinetics of DNase I digestion of the ctDNA measured using the riLD cell. DNase I (27 nM, 54 nM, and 108 nM) was mixed with 200 μM ctDNA in a 1:1 ratio using the riLD system. LD was measured at 260 nm with a 0.5 nm bandwidth. Data were zeroed and scaled to be 0 LD at T = 0 and 1 at 300 s. HT supplied to the PMT was held at 561 V.

concentrations the overall form of the kinetics is similar to that measured for the lower enzyme concentrations using the manual injection system. The experiment also shows that the rate of the reaction increases with enzyme concentration in a way that would be expected for such a reaction.

■

CONCLUSION Over the past 10 years, we have pioneered the application of LD spectroscopy to biological systems by a concerted improvement in the performance of the instrumentation used for data collection. This development process has particularly concentrated on reducing sample volume while increasing data quality, both aspects being aimed at increasing the number of systems that can be studied. In this study we report a significantly improved time scale of measurement. We have shown that sample can be injected into the cell through a narrow bore quartz capillary rod that is substituted for the normal central quartz rod and reliable LD data collected despite the introduction of two additional interfaces in the light beam. We then showed that by coupling the new central capillary to a modified stopped-flow injection system, sample can be introduced into the Couette flow and aligned in less than a second. Finally we have demonstrated that the system can be effectively used to provide data on the digestion of DNA by DNase I. Using the riLD system, we are able to measure the kinetics of this reaction at much higher enzyme concentrations 6565

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566

Analytical Chemistry

Article

(19) Chou, P. J.; Johnson, W. C., Jr. J. Am. Chem. Soc. 1993, 115, 1205−1214. (20) Rodger, A.; Blagbrough, I. S.; Adlam, G.; Carpenter, M. L. Biopolymers 1994, 34 (12), 1583−93. (21) Patel, K. K.; Plummer, E. A.; Darwish, M.; Rodger, A.; Hannon, M. J. J. Inorg. Biochem. 2002, 91 (1), 220−9. (22) Hicks, M. R.; Rodger, A.; Thomas, C. M.; Batt, S. M.; Dafforn, T. R. Biochemistry 2006, 45 (29), 8912−7. (23) Gilroy, E. L.; Hoffmann, S. V.; Jones, N. C.; Rodger, A. Eur. Biophys. J. 2011, 40 (10), 1121−9. (24) Simonson, T.; Kubista, M. Biopolymers 1993, 33, 1225−1235.

6566

dx.doi.org/10.1021/ac300842h | Anal. Chem. 2012, 84, 6561−6566