Micromixer-Based Time-Resolved NMR: Applications to Ubiquitin

Time-resolved NMR spectroscopy is used to study changes in protein conformation based on the elapsed time after a change in the solvent composition of...
0 downloads 6 Views 87KB Size
Download PDF
Anal. Chem. 2003, 75, 956-960

Micromixer-Based Time-Resolved NMR: Applications to Ubiquitin Protein Conformation Masaya Kakuta,†,‡,§ Dimuthu A. Jayawickrama,| Andrew M. Wolters,| Andreas Manz,*,† and Jonathan V. Sweedler*,|

Department of Chemistry, Imperial College of Science, Technology and Medicine, Exhibition Road, London, SW7 2AY, U.K., and Department of Chemistry and the Beckman Institute, University of Illinois, Urbana, Illinois 61801

Time-resolved NMR spectroscopy is used to study changes in protein conformation based on the elapsed time after a change in the solvent composition of a protein solution. The use of a micromixer and a continuous-flow method is described where the contents of two capillary flows are mixed rapidly, and then the NMR spectra of the combined flow are recorded at precise time points. The distance after mixing the two fluids and flow rates define the solventprotein interaction time; this method allows the measurement of NMR spectra at precise mixing time points independent of spectral acquisition time. Integration of a micromixer and a microcoil NMR probe enables lowmicroliter volumes to be used without losing significant sensitivity in the NMR measurement. Ubiquitin, the model compound, changes its conformation from native to Astate at low pH and in 40% or higher methanol/water solvents. Proton NMR resonances of the His-68 and the Tyr-59 of ubiquitin are used to probe the conformational changes. Mixing ubiquitin and methanol solutions under low pH at microliter per minute flow rates yields both native and A-states. As the flow rate decreases, yielding longer reaction times, the population of the A-state increases. The micromixer-NMR system can probe reaction kinetics on a time scale of seconds. The specific three-dimensional structure of a protein is crucial for its biological activity.1 The amino acid sequence under the influence of other parameters leads the protein to fold to a biologically active conformation through nonrandom pathways.2 The question of how the protein folds to this unique state is a central issue in biology. Much attention has focused on the understanding of a large number of human protein variants encoded by the ∼30 000 genes elucidated by sequencing the human genome. Because many diseases and important physiological activities are related to proteins, folding/unfolding studies * Corresponding authors. A.M.: [email protected]. J.V.S: [email protected]. † Imperial College of Science, Technology and Medicine. ‡ On leave from Chugai Pharmaceutical Co., Ltd., Tokyo, Japan. § Present address: University of Twente, Faculty of Electrical Engineering, P.O. box 217, 7500 AE, Enschede, The Netherlands. | University of Illinois. (1) Creighton, T. E. Biochem. J. 1990, 270, 1-16. (2) Liu, Z.-P.; Rizo, J.; Gierasch, L. M. In Bioorganic Chemistry Peptides and Proteins; Hecht, S. M., Ed.; Oxford University Press: New York, 1998; pp 224-257.

956 Analytical Chemistry, Vol. 75, No. 4, February 15, 2003

may assist in understanding the causes of cancers and genetic diseases.3-5 The major goals of many protein-folding studies are to obtain structural and kinetic information on intermediate states. The existence of protein-folding pathways with well-defined intermediates is now widely accepted, but there is limited knowledge of the nature of folding intermediates.6 Because of the cooperative nature of folding/unfolding transitions, partially folded forms are rarely observed at equilibrium and under kinetic conditions. These intermediates may be too short-lived for the direct application of high-resolution structural tools such as twodimensional NMR spectroscopy. Sensitive spectroscopic methods based on circular dichroism (CD), fluorescence, and UV/visible are commonly used to study protein folding.7,8 The overall change in conformation during protein folding can be studied using CD spectroscopy. Several fluorophores and chromophores are sensitive to changes in environment and therefore can be used to study protein conformational changes. Other techniques based on calorimetry and electrophoresis have also been used to study protein folding.9,10 NMR spectroscopy has found widespread uses in both academic and industrial laboratories due to its powerful ability to elucidate molecular structures and its broad range of diagnostic capabilities.11 In addition, NMR is a noninvasive technique and does not require special molecular probes. NMR studies of protein folding often involve equilibrium conversion of the protein between the native and the denatured state using heat, pH, and chemical denaturants.12,13 These approaches can be used to extract the kinetics of the initiation steps with stopped-flow and quenchedflow hydrogen-exchange NMR techniques.14 Such techniques have (3) Milner, J.; Medcalf, E. A. Cell 1991, 65, 765-774. (4) Milner, J. Proc. R. Soc., London Ser. B 1991, 245, 139-145. (5) Thomas, P. J.; Ko, Y. H.; Pedersen, P. L. FEBS Lett. 1992, 312, 7-9. (6) Briggs, M. S.; Roder, H. Proc. Natl. Acad. Sci. U.S.A. 1992, 89, 20172021. (7) Baldwin, R. L. Curr. Opin. Struct. Biol. 1993, 3, 84-91. (8) Semisotnov, G. V.; Rodionova, N. A.; Razgulyaev, O. I.; Uversky, V. N.; Gripas, A. F.; Gilmanshin, R. I. Biopolymers 1991, 31, 119-128. (9) Privalov, P. L.; Potekhin, S. A. In Methods in Enzymology; Academic Press Inc: San Diego, 1986; Vol. 131, pp 4-51. (10) Liu, Z. P.; Rizo, J.; Gierasch, L. M. Biochemistry 1994, 33, 134-142. (11) Heinemann, U.; Illing, G.; Oschkinat, H. Curr. Opin. Biotechnol. 2001, 12, 348-354. (12) Roder, H. In Methods in Enzymology; Oppenheimer, N. J., James, T. L., Eds.; Academic Press: San Diego, 1989; Vol. 176, pp 446-473. (13) Zhang, O.; Formankay, J. D. Biochemistry 1995, 34, 6784-6794. (14) Roder, H.; Elove, G. A.; Ramachandra, S., M. C. Mechanisms of Protein Folding, 2nd ed.; Oxford University Press: New York, 2000. 10.1021/ac026076q CCC: $25.00

© 2003 American Chemical Society Published on Web 01/11/2003

generated valuable information about the intermediate state/s of proteins. One of the drawbacks is that they require a high-pressure solution delivery system and large sample quantities. NMR has also been used to study nonequilibrium events of protein folding.15 This technique involves recording NMR spectra after initiation of the reaction in the NMR probe. We examine a method to improve the performance of NMR studies directly applied to protein folding using advances in fluidics and detection. A decade has passed since the concept of the micro total analysis systems (µTAS) was proposed.16 Miniaturized systems that incorporate a smaller inner diameter channel and shorter length, for instance, yield better separation performance and shorter transport time. Such systems can improve the performance of separations, chemical synthesis, biochemical applications, and detection techniques.17-22 One example of a µTAS system is the development of a micromixer. At smaller length scales, mixing of two fluids becomes difficult due to the low Reynolds number.23 We have developed a noble micromixer with a glass/silicon/glass sandwich structure that enables the mixing of two nano-/microliter volume fluids at low pressure within the order milliseconds24 and apply this technology to NMR spectroscopic studies. In comparison to other spectroscopic methods of structural characterization, NMR is a relatively insensitive technique. However, recent advances in solenoidal microcoil NMR technology have produced the highest mass sensitivity NMR probes. These probes use NMR active volumes between 5 nL and 1 µL to produce high-resolution NMR spectra.25-31 Capillary-based separation or flow injection analysis can be easily performed with these probes.29 In this work, we use the combination of a micromixer and an NMR probe to follow protein-unfolding kinetics. Of particular interest is the conversion of the reaction time coordinate (time after mixing) to a stable (in time) distance coordinate well suited to capillary-based mass-sensitive NMR microcoils. Obviously, the longer the NMR active observed length, the poorer the resulting (15) Van Nuland, N. A. J.; Forge, V.; Balbach, J.; Dobson, C. M. Acc. Chem. Res. 1998, 31, 773-780. (16) Manz, A.; Graber, N.; Widmer, H. M. Sens. Actuators, B 1990, 1, 244-248. (17) Sanders, G. H. W.; Manz, A. TrAC, Trends Anal. Chem 2000, 19, 364-378. (18) Eijkel, J. C. T.; de Mello, A. J.; Manz, A. Mesoscopic Chemistry; Blackwell Science: Malden, MA, 2000. (19) Kakuta, M.; Bessoth, F. G.; Manz, A. Chem. Rec. 2001, 1, 395-405. (20) Legge, C. H. J. Chem. Educ. 2002, 79, 173-178. (21) Reyes, D. R.; Iossifidis, D.; Auroux, P.; Manz, A. Anal. Chem. 2002, 74, 2623-2636. (22) Auroux, P.; Iossifidis, D.; Reyes, D. R.; Manz, A. Anal. Chem. 2002, 74, 2637-2652. (23) Erbacher, C.; Bessoth, F. G.; Busch, M.; Verpoorte, E.; Manz, A. Mikrochim. Acta 1999, 131, 19-24. (24) Bessoth, F. G.; deMello, A. J.; Manz, A. Anal. Commun. 1999, 36, 213-215. (25) Zhang, X.; Sweedler, J. V.; Webb, A. G. J. Magn. Reson. 2001, 153, 254258. (26) Lacey, M. E.; Webb, A. G.; Sweedler, J. V. Anal. Chem. 2000, 72, 49914998. (27) Subramanian, R.; Kelley, W. P.; Floyd, P. D.; Tan, Z. J.; Webb, A. G.; Sweedler, J. V. Anal. Chem. 1999, 71, 5335-5339. (28) Li, Y.; Wolters, A. M.; Malawey, P. V.; Sweedler, J. V.; Webb, A. G. Anal. Chem. 1999, 71, 4815-4820. (29) Lacey, M. E.; Subramanian, R.; Olson, D. L.; Webb, A. G.; Sweedler, J. V. Chem. Rev. 1999, 99, 3133-3152. (30) Olson, D. L.; Lacey, M. E.; Webb, A. G.; Sweedler, J. V. Anal. Chem. 1999, 71, 3070-3076. (31) Olson, D. L.; Lacey, M. E.; Sweedler, J. V. Anal. Chem. 1998, 70, 645650.

time resolution. To increase the sensitivity and to keep the microcoil length at a minimum, a 1-mm-long bubble-type flow cell with an active volume of ∼800 nL is used in this study. We demonstrate the use of this simple instrumentation to measure the solvent-induced protein conformation change with low-pressure flows and low-volume sample consumption. The combination of microfabrication technology with information-rich detection has been successfully applied to a protein conformation study interfaced with mass spectroscopy.32 Unfortunately mass spectroscopy does not always provide structural information at the amino acid level. Here we demonstrate time-resolved NMR spectroscopy for studying the real-time methanol-induced conformation changes of ubiquitin. This is a well-characterized, small, single-domain protein with no disulfide bonds, metals, or extrinsic cofactors.6,33-35 Ubiquitin exists in its native state (N-state) at low pH but changes to the so-called A-state with the addition of methanol.36 The feasibility of micromixer-NMR microcoil instrumentation to study the transition kinetics of the native to A-state of ubiquitin is evaluated. Unlike previous real-time NMR studies of protein folding,15 the use of the micromixer allows the observation of kinetics at specific time intervals for a longer time. The limiting factors in observation time become available NMR time and volume of reagents. Therefore, this technique is especially useful to obtain NMR data with high signal-to-noise ratio. EXPERIMENTAL SECTION Materials. Ubiquitin was purchased from Sigma Chemical Co. (St. Louis, MO) and used without further purification. Deuterium oxide (D2O, 99.9% D) was obtained from Sigma Chemical Co. Methanol-d4 and DCl were purchased from Cambridge Isotope Labs (Andover, MA). The matching fluid, MF-1, used in the microcoil NMR probe was from MRM Corporation (Savoy, IL). Apparatus. The micromixer was fabricated by Kymata (Enschede, The Netherlands). Fabrication and design methods are discussed in detail elsewhere.24 The Y-connector mixer, p-773, syringe-capillary connectors, and sleeves were purchased from Upchurch Scientific (Oak Harbor, WA). Gastight 500-µL syringes were purchased from Hamilton (Reno, NV). A syringe pumping system, Harvard 22 (Harvard Apparatus Inc.), was used to deliver solution with high precision. The pH of solutions was measured with a pH meter (Orion model 720A) equipped with pH electrode model 98-26 (Orion Research Inc., Boston, MA). All solutions were degassed for 5 min using a sonicator (Branson 2200, Branson Ultrasonics, Danbury, CT). Interfacing the Fluidics, Micromixer, and NMR Probe. The connections from syringes on the pumping system to the mixer were made with equal-length (∼4 m) fused-silica capillaries (250-µm i.d/350-µm o.d., Polymicro Technologies, Inc., Oak Harbor, WA) and standard Upchurch connectors (Figure 1). The mixer and the NMR microcoil cell were connected using a 72cm-length fused-silica capillary (75-µm i.d./360-µm o.d., Polymicro Technologies) and PTFE (0.012-in. i.d/0.030-in. o.d) tubing (Cole (32) Konermann, L.; Collings, B. A.; Douglas, D. J. Biochemistry 1997, 36, 55545559. (33) Weber, P. L.; Brown, S. C.; Mueller, L. Biochemistry 1987, 26, 7282-7290. (34) Distefano, D. L.; Wand, A. J. Biochemistry 1987, 26, 7272-7281. (35) Lenkinski, R. E.; Chen, D. M.; Glickson, J. D.; Goldstein, G. Biochim. Biophys. Acta 1997, 494, 126-130. (36) Woolfson, D. N.; Cooper, A.; Harding, M. M.; Williams, D. H.; Evans, P. A. J. Mol. Biol. 1993, 229, 502-511.

Analytical Chemistry, Vol. 75, No. 4, February 15, 2003

957

Figure 1. Schematic of the time-resolved NMR spectroscopy system showing (A) syringe pumping system, (B) mixer, (C) microcoil NMR cell, and (D) outlet reservoir. Arrows indicate the flow directions of the solutions from the two syringes. The time to react and incubate ∆t in the reaction capillary (length L, area A) is inversely proportional to the flow rate v.

Palmer, Vernon Hills, IL). The volume from the mixing point to the middle of the NMR microcoil was 3.8 µL. Microcoil Probe. A solenoidal rf coil, wrapped around a bubble-type NMR flow cell with an observe volume of ∼800 nL, was used for the NMR measurements. For magnetic susceptibility matching purposes, the microcoil was enclosed by a 10-mL polyethylene bottle filled with a perfluorinated organic liquid MF-1. The microcoil fabrication has been described in detail previously.37 The static line width at half-maximum for 2% methanol in 45% methanol-d4 and 50% D2O was between 1 and 2 Hz. NMR Spectroscopy. 1H NMR spectra were recorded on a 600-MHz (14.1 T) Varian Unity Inova 600 NMR spectrometer (Varian Inc., Palo Alto, CA). A total of 128 FIDs were collected with a 55° flip angle (3 µs), a pulse repetition time of 0.5 s, an acquisition time of 1 s, a spectral width of 8000 Hz, and 8000 complex points (NP). The NMR spectra were processed on a PC platform using the NUTS software package (Acorn NMR Inc., Fremont, CA). For processing, NMR data were zero filled to 16 000 points and exponentially multiplied by a line broadening (LB) of 3 Hz. The baseline was corrected by a seventh-order polynominal fit of selected spectral regions. The integral values or intensities of selected NMR resonance were measured in the fully processed spectra. For the continuous-flow experiments, the acquisition of NMR data started after appropriate time intervals determined by the flow rates. Mixing Procedures with Continuous Flow. Mixing was performed at ambient temperature using either a micromixer or a Y-connector mixer. The pH meter recording of ubiquitin in 20% methanol-d4/D2O (uncorrected pH electrode reading) was adjusted to pH 2.0 with DCl. The concentration of ubiquitin was 7 mM. The native to A-state transition was initiated by rapid mixing of ubiquitin in 20% methanol-d4/ 80% D2O with an 80% methanold4/20% D2O solution (pD ) 2.4) using both the micromixer and the Y-connector, resulting in a final ubiquitin concentration at the NMR detection cell of 3.5 mM. In the micromixer, the two solutions are split into 32 partial flows reducing the diffusion length of the two solutions to 5 µm, which are then combined in a similar pattern to provide a single mixed flow. The time, t, for diffusion of a molecule is proportional to the square of the distance, L, and inversely proportional to the diffusion coefficient, D, of the molecule: t ∝ L2/D.23 Thus, considering small molecules with D of ∼10-9 m2 s-1, the mixing time is on the order of 25 ms for the micromixer. On (37) Lacey, M. E.; Sweedler, J. V.; Larive, C. K.; Pipe, A. J.; Farrant, R. D. J. Magn. Reson. 2001, 153, 215-222.

958 Analytical Chemistry, Vol. 75, No. 4, February 15, 2003

the other hand, the distance, L, is estimated to be ∼37 µm for the Y-connector mixer. Therefore, the diffusional mixing time for a similar-sized molecule is ∼1.4 s for the Y-connector. In this study, the reaction time after mixing for protein folding is controlled by changing the flow rate and not the capillary length. Flow rates ranging from 60 to 2 µL/min, corresponding to reaction times between 3.8 and 114 s, respectively, have been explored. The curve-fitting program, Origin version 6 (Microcal Software Inc., Northampton, MA) was used to calculate time constants. RESULTS AND DISCUSSION The overall goal of this study is to explore the combination of a micromixer and a microcoil NMR probe and to evaluate the application of this new instrumentation to study protein transition kinetics. The schematic of the mixer and the NMR microcoil probe is illustrated in Figure 1. Two different mixers have been explored, one made with microfabrication technology and the other a simple Y-type capillary connector. Ubiquitin is stable in a variety of extreme conditions such as high temperature and low pH.16,35,38-41 However, under acidic conditions, methanol induces a transition from a native to a partially unfolded state, the A-state.42 Certain amino acids in ubiquitin, His-68 and Tyr-59, are well characterized as proton NMR probes to follow such conformational changes.39,40 All the experiments were performed by mixing two solutions, one filled with 20% CD3OD/80% D2O and 7 mM ubiquitin (pD ) 2.4) and the other filled with 80% CD3OD/20% D2O (pD ) 2.4). After mixing, the composition of the resultant solution should reach 50% CD3OD and 50% D2O at pD ) 2.4. The transition reaction occurs from the point of solution confluence in the mixer and continues until the solution reaches the NMR microcoil cell, with NMR spectra acquired with the solution flowing through the NMR flow cell. To study different solution mixing times, one must change the time allowed for mixing either by changing the flow rates while keeping a fixed length between the mixer and the NMR coil or by changing the position of the NMR cell relative to the mixing point with fixed flow rates. The relative position of the NMR probe can be changed using sleeve probe technology, which allows the capillary to be pulled through the microcoil NMR probe.43 However, in this study, we observe the ubiquitin transitions at different time intervals controlled by differing flow rates. The flow rates ranged from 2 to 60 µL/min, corresponding to times of 1143.8 s (with the times defined as the time for the solution to move from the point of confluence to the midpoint of NMR microcoil cell). Selected NMR spectra recorded at three different flow rates using the Y-connector are shown in Figure 2. The peaks at 8.75 and 8.65 ppm (a, b) are assigned to ubiquitin His-68 (C4 proton) in the native and the A-state, respectively.39,40 Similarly, the peaks at 6.83 and 6.70 ppm (c, d) are assigned to native and A-state of (38) Jenson, J.; Goldstein, G.; Breslow, E. Biochim. Biophys. Acta 1980, 624, 378-385. (39) Jourdan, M.; Searle, M. S. Biochemistry 2000, 39, 12355-12364. (40) Jourdan, M.; Searle, M. S. Biochemistry 2001, 40, 10317-10325. (41) Peti, W.; Smith, L. J.; Redfield, C.; Schwalbe, H. J. Biomol. NMR 2001, 19, 153-165. (42) Brutscher, B.; Bruschweiler, R.; Ernst, R. R. Biochemistry 1997, 36, 1304313053. (43) Kautz, R. A.; Lacey, M. E.; Wolters, A. M.; Foret, F.; Webb, A. G.; Karger, B. L.; Sweedler, J. V. J. Am. Chem. Soc. 2001, 123, 3159-3160.

Figure 2. Spectra from the Y-connector with stopped flow and at various flow rates (mixing times): stopped flow, 2 (114 s), 20 (11.4 s), and 60 µL/min (3.8 s). The labeled peaks correspond to (a) native state His-68, (b) A-state His-68, (c) native state Tyr-59, and (d) A-state Tyr-59.

Tyr-59 (meta protons), respectively. One spectrum in Figure 2 was recorded under stopped flow, and the other three with three different reaction times. As the time interval increased, the intensity/area of the native state His-68 (a) and Tyr-59 (c) decreased. In reverse, the intensity/ area of the A-state His-68 (b) and Tyr-59 (d) increased, demonstrating that ubiquitin changed its conformation from the native to the A-state with time. The time-dependent behavior of these two states was quantified as the population ratio of the A-state to the native state and is shown in Figure 3. The population ratios, A-state/native (A/N) of Tyr-59 and His-68, show a similar change with both mixerssthe Y-connector and micromixer. This is likely because the time for complete diffusional mixing is much shorter for both mixers than even the shortest reaction time observed here. The diffusional mixing times for a small molecule in the fabricated mixer and the Y-connector are ∼25 ms and 1.4 s, respectively. Because of the upper limit of the flow rate and the minimum capillary length between the mixer and NMR microcoil, the shortest reaction time recorded with the current instrumentation is ∼3.8 s. As a result, the mixing capabilities of two mixers cannot be distinguished. Indistinguishable results were obtained from experiments carried out on different days, indicating the reproducibility of the integrated system. Interestingly, until 40 s, the data follow an exponential behavior and reach a plateau at ∼80 s. After this point, the population ratios of A/N increase again. These phenomena suggest that ubiquitin changes its conformation in at least second order and may have more than two states during its structural transition. The concentration-dependent aggregation of ubiquitin at low pH in the range of 1-6 mM in 40% CD3OD has been reported.44 Therefore, the (44) Harding, M. M.; Williams, D. H.; Woolfson, D. N. Biochemistry 1991, 30, 3120-3128.

Figure 3. Population ratio of A-state/native state as a function of time using (A) the Y-connector and (B) the micromixer.

role of aggregation in our observed ubiquitin transition kinetics cannot be ruled out. Detailed analyses of the behavior of the A-state and the native state populations within the first 40 s after mixing are shown in Figure 4. The population ratios, A/N obtained with the Y-connector and micromixer can be fitted to a single exponential. The time constants obtained with the Y-connector and micromixer for Tyr-59 and His-68 are 0.25 ((0.07, R2 ) 0.92) and 0.21 s-1 ((0.06, R2 ) 0.92) respectively, an insignificant difference between the two rate constants. The observed line width of an NMR signal can be related to the effective transverse relaxation time, T2* (eq 1), where fwhm is the full width at half-maximum of the NMR line. However, in a

fwhm ) 1/πT2*

(1)

flowing system, the line width is the convolution of the molecule’s T2* decay rate and the observe time. Therefore, the resonance time of a molecule in the flow cell affects the line width of NMR45

1/T2 flow ) 1/T2* + 1/τ

(2)

where T2 flow is the transverse relaxation time under flowing conditions, and τ is the residence time (ratio of the NMR observe volume to flow rate) of a nucleus within the NMR coil. Basically, (45) Albert, K.; Bayer, E. In HPLC Detection Newer Methods; VCH Publishers: New York, 1992; pp 197-227.

Analytical Chemistry, Vol. 75, No. 4, February 15, 2003

959

increase in line width may decrease the signal-to-noise ratio, which could cause overlap of NMR peaks. Taking the highest practical flow rate as 60 µL/min, and assuming the mixer is located adjacent to the NMR flow cell (not an easy implementation), then the shortest observation time point corresponds to ∆t ∼ 0.4 s. Using 0.6 µL/min as the slowest practical flow rate and a longer capillary, ∆t ) 400 s, thus defining a range of observation times for practical observation. Shorter flow cells and faster flow rates (even with a decrease in NMR line width) will expand this range.

Figure 4. Kinetics data using (A) the Y-connector and (B) the micromixer. Each curve was fitted to a single exponential.

the NMR signal can be truncated because the spins move out of the flow cell. This is especially critical with faster flow rates. Instead of Lorentzian, the NMR line shape becomes Gaussian in the flow-limited case. For a Gaussian peak, the fwhm is given by (4π ln 2)1/2/T2*.46 Because the half-width of NMR signal is inversely proportional to the transverse relaxation time, the increase in flow rate will increase the line width. The maximum flow rate employed in this study is 60 µL/min; under this flow rate, the average residence time of a molecule in the flow cell of 800 nL is 0.8 s. The contribution to the line width at this flow rate is 1.2 Hz. Although higher flow rates will allow the study of faster kinetic events, the (46) Pastirk, I.; Lozovoy, V. V.; Dantus, M. Chem. Phys. Lett. 2001, 333, 76-82.

960

Analytical Chemistry, Vol. 75, No. 4, February 15, 2003

CONCLUSIONS In this study, we have coupled a small-volume mixer and an NMR microcoil to produce a simple instrument that operates at low pressures and consumes low sample volumes. The compatible sizes of the microcoil and the mixers allow easy integration. The time-resolved NMR spectroscopy described in this work presents information about ubiquitin conformation changes at the amino acid level. Although the 1H NMR spectrum does not allow us to probe every amino acid in ubiquitin, it provides enough information to follow such conformational changes. The behavior of ubiquitin does not fit into a two-state equilibrium. To understand this behavior, further research is needed using other spectroscopic methods such as multidimensional NMR, CD, FT-IR, or MS spectroscopy. Future research will focus on faster kinetic approaches and the development of multidimensional NMR measurements. To observe subsecond kinetics, a micromixer is required as opposed to the Y-connector because of the much shorter time needed for diffusional mixing. Another exciting area of instrumental integration is the combination of multiple microcoil probes28 which would allow multiple detection points (time points) to be monitored simultaneously. Micromixermicrocoil NMR instrumentation may also be employed to study reaction intermediates and molecular interactions. ACKNOWLEDGMENT M.K. thanks Chugai Pharmaceutical Co. Ltd. for financial support. We acknowledge financial support from the National Institute of Health (GM53030). Received for review August 25, 2002. Accepted November 14, 2002. AC026076Q