Single molecule turnover of fluorescent ATP by myosin and actomyosin unveil elusive enzymatic mechanisms

Figures

• 

  Fig. 1. Schematics of the study assays and major biochemical transitions. a TIRF microscopy assay with depicted major proteins used: actin, heavy meromyosin (HMM), and subfragment 1 (S1) attached to TMCS modified coverslip surface. Note that the effective evanescence illumination range is comparable to the distance of myosin catalytic site from the surface. b Muscle myosin and actomyosin ATPase cycles. For clarity, only S1 construct is shown. The states with bound nucleotide are marked with a red star. Basal myosin ATPase cycle is rate limited by Pi release, which is significantly accelerated (“activated”) by the presence of actin. Dashed circular arrows represent major ATPase cycle direction.

• 

  Fig. 2. Single molecule ATPase. a Two principles for surface immobilization of myosin motor fragments. b Time-averaged Alexa-ATP fluorescence projections of 15 min videos (50 ms exposure time/frame). (i) Simple HMM deposition at 34.3 pM (left) and 343 pM (right). (ii) Optimized HMM deposition following high [ATP] wash. Only spots defined by previous position of F-actin filament were included in analysis. (iii) Image of optimized actomyosin deposition followed by low [ATP] wash showing co-localization of Alexa-ATP fluorescence (gray spots) and F-actin filaments (yellow). Only spots co-localizing with F-actin filaments were included in analysis. Bars, 5 µm. c Representative time traces of (i) single-molecule Alexa-Ph bleaching, (ii) HMM basal ATPase, and iii, actin-activated HMM ATPase. Bleaching was observed either using standard in vitro motility assay (IVMA) buffer or optimized TIRF microscopy buffer (TIRFM) as described in the text (for details see Supplementary Information sections 1.2 and 1.9), whereas ATPase traces were from experiments using TIRFM buffer. d Cumulative frequency distributions of single-molecule Alexa-Ph until bleach or first blinking event. The distributions are…

• 

  Fig. 3. Analysis of motility in standard IVMA buffer and optimized TIRF buffer. a Actin gliding velocity produced by HMM adsorbed to TMCS derivatized glass surfaces. In each individual experiment, 15 actin filaments were analyzed. b Actin gliding velocity produced by HMM adsorbed to TMCS derivatized glass surfaces. In each individual experiment, 25 (IVMA buffer) and 20 (IVMA buffer with TX/TQ) actin filaments were analyzed. c Fraction of motile filaments from the experiments under a, analyzed from five recordings of, in total, 1004 filaments (IVMA buffer) and 925 filaments (TIRF buffer), respectively. Data are given as mean ± 95% confidence interval. Temperature 26 ± 2 °C. For details about buffer composition please see Supplementary Information sections 1.2 and 1.9).

• 

  Fig. 4. Cumulative dwell time distributions recorded from different individual actomyosin hotspots to illustrate the heterogeneity in time constants. a Cumulative frequency distributions of Alexa-nucleotide dwell time events comparing actin-activated ATPase activity from four different hotspots. Data are fitted with double-exponential functions. b Amplitudes and rate constants obtained from fittings to data in a. Note that from the individual actomyosin hotspots only two processes attributed to actin-activated ATPase and “unexplained” phase can usually be resolved, while basal myosin ATPase remains obscured. For hotspot 2, it seems likely that the binding site of the HMM molecule is unfavorable for actin attachment, explaining a high ratio of basal to actomyosin ATPase. Error estimates refer to 95% confidence intervals derived in the regression analysis. Temperature: 20 °C.

• 

  Fig. 5. Origin of “unexplained” (0.2–0.5 s−1) exponential phase. a Cumulative frequency distributions of Alexa-nucleotide dwell time events on myosin subfragment 1 (S1) surface hotspots (simple deposition of S1, Fig. 1a, i). Data either with (Ndwell = 1964 from 127 S1 molecules in 15 min) or without (Ndwell = 1084 events from 30 molecules in 30 min) nucleotide pocket blocking by nonfluorescent ATP and vanadate (S1 * D·Vi). The data were fitted by triple exponential functions (solid lines) instead of double-exponential function as justified by residual plots (inset; S1 * D·Vi data), difference in Akaike’s information criterion—AICc (4290) and R2 (0.9907 vs 0.9977). b Rate constants obtained from fittings to data in a. c Amplitudes obtained from fitting the data in a. Note that amplitude (AATPase1) of basal ATPase activity (kATPase1 ~0.05 s −1) was significantly reduced for S1 * D·Vi. Unexplained phase with amplitude Aunexplained and rate kunexplained, was not appreciably affected. Note, further that fitting of S1 * D·Vi data set produced an even faster phase (kfast ~3.5 s−1). The extra slow phase with amplitude AATPase2 ~5 % and rate kATPase2 ~0.02 s−1 for the S1 ATPase may be attributed…

• 

  Fig. 6. Effect of AmBleb on single molecule ATPase. a The molecular structure of para-aminoblebbistatin (AmBleb) is shown39. b Concentration–response curve for effect of AmBleb on sliding velocity at 60 mM ionic strength (1 mM MgATP) is shown using HMM. c Representative time traces of single molecule basal HMM ATPase without (i) or with 40 µM AmBleb (ii). Note two long dwell time events as direct consequence of effect of AmBleb on basal ATPase activity. d Cumulative dwell time distributions (i) of basal HMM ATPase in the absence and presence of 40 µM AmBleb were best fitted to double (control, data from 19 HMM molecules, Ndwell = 318) or triple (AmBleb, data from 31 HMM molecules, Ndwell = 280) exponential functions revealing an additional slow phase (ii, iii) with rate constant 0.006 s−1 (~8%) in the presence of AmBleb. Error estimates refer to 95% confidence intervals derived in the regression analysis. Temperature: 20 °C.