Force and velocity measured for single molecules of RNA polymerase

Science

Volumn 282, Issue 5390, 1998, Pages 902-907

  Wang, Michelle D ^a,d   Schnitzer, Mark J ^a   Yin, Hong ^b,e   Landick, Robert ^c   Gelles, Jeff ^b   Block, Steven M ^a  

^a PRINCETON UNIVERSITY   (United States)

^b BRANDEIS UNIVERSITY   (United States)

^c UNIVERSITY OF WISCONSIN MADISON   (United States)

^d Cornell University   (United States)

^e TUFTS UNIVERSITY SCHOOL OF MEDICINE   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

KINESIN; MYOSIN; RNA POLYMERASE;

ARTICLE; BASE PAIRING; DNA TRANSCRIPTION; ESCHERICHIA COLI; FEEDBACK SYSTEM; FORCE; GENE TRANSLOCATION; PRIORITY JOURNAL; VELOCITY;

DNA, BACTERIAL; DNA-DIRECTED RNA POLYMERASES; ESCHERICHIA COLI; MATHEMATICS; MODELS, CHEMICAL; MOLECULAR MOTOR PROTEINS; RNA, BACTERIAL; RNA, MESSENGER; TEMPLATES, GENETIC; THERMODYNAMICS; TRANSCRIPTION, GENETIC;

EID: 0032582494     PISSN: 00368075     EISSN: None     Source Type: Journal    
DOI: 10.1126/science.282.5390.902     Document Type: Article

References (43)

1. M. J. Chamberlin, Harvey Lect. 88, 1 (1994); S. Uptain, C. Kane, M. Chamberlin, Annu. Rev. Biochem. 66, 117 (1997); R. Mooney, I. Artsimovitch, R. Landick, J. Bacteriol. 180, 3265 (1998).
2. M. J. Chamberlin, Harvey Lect. 88, 1 (1994); S. Uptain, C. Kane, M. Chamberlin, Annu. Rev. Biochem. 66, 117 (1997); R. Mooney, I. Artsimovitch, R. Landick, J. Bacteriol. 180, 3265 (1998).
3. M. J. Chamberlin, Harvey Lect. 88, 1 (1994); S. Uptain, C. Kane, M. Chamberlin, Annu. Rev. Biochem. 66, 117 (1997); R. Mooney, I. Artsimovitch, R. Landick, J. Bacteriol. 180, 3265 (1998).
4. H. Yin et al., Science 270, 1653 (1995).
5. D. A. Schafer, J. Gelles, M. P. Sheetz, R. Landick, Nature 352, 444 (1991); H. Yin, R. Landick, J. Gelles, Biophys. J. 67, 2468 (1994).
6. D. A. Schafer, J. Gelles, M. P. Sheetz, R. Landick, Nature 352, 444 (1991); H. Yin, R. Landick, J. Gelles, Biophys. J. 67, 2468 (1994).
7. H. Kabata et al., Science 262, 1561 (1993); S. Kasas et al., Biochemistry 36, 461 (1997).
8. H. Kabata et al., Science 262, 1561 (1993); S. Kasas et al., Biochemistry 36, 461 (1997).
9. R. Landick, Cell 88, 741 (1997); J. Gelles and R. Landick, ibid. 93, 13 (1998).
10. R. Landick, Cell 88, 741 (1997); J. Gelles and R. Landick, ibid. 93, 13 (1998).
11. A. V. Hill, Proc. R. Soc. London Ser. B 126, 136 (1938).
12. K. Svoboda and S. M. Block, Cell 77, 773 (1994); A. J. Hunt, F. Gittes, J. Howard, Biophys. J. 67, 766 (1994); C. M. Coppin, D. W. Pierce, L. Hsu, R. D. Vale, Proc. Natl. Acad. Sci. U.S.A. 94, 8539 (1997); H. Kojima, E. Muto, H. Higuchi, T. Yanagida, Biophys. J. 73, 2012 (1997).
13. K. Svoboda and S. M. Block, Cell 77, 773 (1994); A. J. Hunt, F. Gittes, J. Howard, Biophys. J. 67, 766 (1994); C. M. Coppin, D. W. Pierce, L. Hsu, R. D. Vale, Proc. Natl. Acad. Sci. U.S.A. 94, 8539 (1997); H. Kojima, E. Muto, H. Higuchi, T. Yanagida, Biophys. J. 73, 2012 (1997).
14. K. Svoboda and S. M. Block, Cell 77, 773 (1994); A. J. Hunt, F. Gittes, J. Howard, Biophys. J. 67, 766 (1994); C. M. Coppin, D. W. Pierce, L. Hsu, R. D. Vale, Proc. Natl. Acad. Sci. U.S.A. 94, 8539 (1997); H. Kojima, E. Muto, H. Higuchi, T. Yanagida, Biophys. J. 73, 2012 (1997).
15. K. Svoboda and S. M. Block, Cell 77, 773 (1994); A. J. Hunt, F. Gittes, J. Howard, Biophys. J. 67, 766 (1994); C. M. Coppin, D. W. Pierce, L. Hsu, R. D. Vale, Proc. Natl. Acad. Sci. U.S.A. 94, 8539 (1997); H. Kojima, E. Muto, H. Higuchi, T. Yanagida, Biophys. J. 73, 2012 (1997).
16. H. C. Berg and L. Turner, Biophys J. 65, 2201 (1993).
17. M. D. Wang, Y. Yin, R. Landick, J. Gelles, S. M. Block, ibid. 72, 1335 (1996).
18. Transcription complexes were prepared as described previously and attached to 0.5-μm-diameter polystyrene beads (2, 3). Voltage signals corresponding to displacement in the interferometer and laser light intensity were low-pass filtered at 1 kHz, digitally sampled by computer at 2 kHz, then averaged and binned at 20 Hz for offline analysis. The interferometer was calibrated and signals were corrected as previously described (9). The corrections adjust for the changes in interferometer sensitivity as beads move vertically away from the trap center and for the geometry as the angle of the DNA changes with respect to the plane of the coverglass during shortening of the tether.
19. Complexes with initial tether lengths >500 nm moving at unloaded speeds >3 nt/s and that reached the clamp limit were selected for analysis. With the clamp active, a stall was operationally defined as beginning at the point where the velocity dropped below ∼1.5 nucleotides per second (the experimental drift limit) for more than 10 s. To minimize any sticking of DNA tethers to the coverglass surface, these experiments were conducted in flow chambers treated with blocking proteins, as described (2, 3, 9).
20. We reported previously an RNAP stall force of at least 14 pN. Several lines of evidence suggested that this number may underestimate the actual value (2). First, about 20% of beads drawn by RNAP complexes were not stopped at the fixed laser powers used and escaped from the trap, permitting only a lower bound of 15 to 20 pN to be placed on their stall force. Second, long-term exposure to high levels of laser light damaged a significant proportion of complexes, leading to irreversible stalls (that is, movement that did not recover when the laser power was subsequently reduced). Third, many stalls happened abruptly, as the force increased beyond some threshold, whereas a minority of stalls took place more gradually [see figure 2C of (2)]. This suggested that the stall force might be a function of the particular nucleotide sequence being transcribed. Because force mounted slowly as RNAP traversed comparatively long segments of DNA (typically ∼500 base pairs or more), an abrupt stall might correspond to the enzyme reaching a point on the template with a lower intrinsic stall force than its previous positions. Assuming this interpretation to be correct, stall force data acquired for slowly developing loads (at low trap stiffness) would tend to produce biased estimates, preferentially sampling lower values. The feedback system developed here was designed to circumvent these difficulties (Fig. 1D). Once a moving bead reaches a displacement corresponding to some preset limit, the intensity of the laser light is actively servo-controlled (by means of an acousto-optic modulator) to prevent further motion. This clamps the bead at the limit (9), an arrangement that affords several advantages. First, photodamage is minimized; the trap light intensity is kept as low as possible until it becomes necessary to increase it, reducing the overall exposure of RNAP to the laser light by an estimated factor of five relative to earlier work. Higher peak powers are achieved without subjecting molecules to sustained exposure. As a result, all moving beads are now stalled without escaping from the trap. Second, the dynamic response of the system is dramatically improved. Force can be recorded with millisecond resolution (2, 9), and RNAP can be halted within a matter of seconds, 5-to 40-fold faster. Because the DNA tether can be readily pretensioned by the optical trap, the elastic compliance is decreased. With a taut, ∼3-kbp DNA tether, bead equilibrium position can be determined with a precision of ∼2 nm, corresponding to ∼6 base pairs, with 50 ms time resolution (or better). Third, the rapid response of the feedback system stalls the enzyme before it travels as far, leading to a less biased sampling of stall force along the template.
21. Records of bead position were converted to DNA contour length by a variation of the method described (9), taking advantage of the fact that many transcription complexes stopped permanently on the DNA template at the end of experimental runs. Complexes with final tether lengths >300 nm were selected for analysis. With the bead held in the trap, the microscope stage was moved in a preprogrammed fashion, and both the length and elastic properties of the residual tether were measured directly. Using both the final tether length and the force-extension relationship for the DNA molecule, in combination with time records for bead displacement and trap stiffness from the earlier experimental run, the motion of RNAP along the DNA template required to produce such records could be computed. Computed displacements were converted directly to transcript size by assuming that no slippage occurred and that one base pair corresponds to 0.338 nm [W. Saenger, Principles of Nucleic Acid Structure (Springer-Verlag, New York, 1988)].
22. R. R. Reisbig and J. E. Hearst, Biochemistry 20, 1907 (1981); K. M. Arndt and M. J. Chamberlin, J. Mol. Biol. 213, 79 (1990); C. L. Chan and R. Landick, in Transcription: Mechanisms and Regulation, R. C. Conaway and J. W. Conaway, Eds. (Raven Press, New York, 1994), vol. 3, pp. 297-321.
23. R. R. Reisbig and J. E. Hearst, Biochemistry 20, 1907 (1981); K. M. Arndt and M. J. Chamberlin, J. Mol. Biol. 213, 79 (1990); C. L. Chan and R. Landick, in Transcription: Mechanisms and Regulation, R. C. Conaway and J. W. Conaway, Eds. (Raven Press, New York, 1994), vol. 3, pp. 297-321.
24. R. R. Reisbig and J. E. Hearst, Biochemistry 20, 1907 (1981); K. M. Arndt and M. J. Chamberlin, J. Mol. Biol. 213, 79 (1990); C. L. Chan and R. Landick, in Transcription: Mechanisms and Regulation, R. C. Conaway and J. W. Conaway, Eds. (Raven Press, New York, 1994), vol. 3, pp. 297-321.
25. Pauses were removed by applying a threshold template to the velocity data, V(t), and deleting those intervals during which average speed (computed as described) fell below 50% of the template value at any given time. The remaining portions of the record were then concatenated to form a new record. The threshold template function was fixed in advance for each run and consisted of two sequential linear segments: one constant, the next tapering linearly from this constant value to zero at some fixed point in the record. The amplitude of the constant, the slope of the linear taper, and the fixed intercept point were chosen empirically. Taken together, the two line segments roughly approximate the shape of V(t) curves. In principle, it is impossible to distinguish between transient transcriptional pausing and mechanical stalling in the limit that complexes are halted by an external load, based solely on velocity data. Furthermore, uncertainties due to noise and drift make it difficult to ascertain exactly when stall is achieved (11). For these reasons, the shapes of F-V curves are poorly determined near their x-intercepts (zero velocity).
26. The calculated efficiencies neglect energetic contributions from RNA folding.
27. S. M. Block, Trends Cell Biol. 5, 169 (1995).
28. O, in nt/s.
29. The large heterogeneity in stall force may reflect several contributions. (i) Complexes were oriented randomly on the surface with respect to the direction of the applied load, and the dependence upon angle may be significant; (ii) Stall forces may depend on the template sequence; (iii) Heterogeneity may exist in the properties of nominally identical RNAP molecules.
30. In principle, elongation by RNAP can be arrested by any of a variety of physical mechanisms. (i) The applied load could reversibly halt progress, via mechanochemical coupling in the enzyme; (ii) the load could place the enzyme in a transcriptional pause state; (iii) the load could temporarily inhibit the enzyme, placing it in a state that is biochemically distinct from a pause; or (iv) the enzyme could become irreversibly damaged, either directly by the applied load or by exposure to the laser light. In these and earlier experiments (2), both reversible and irreversible stalls were found, and reversible stalls required the same average force within error as irreversible stalls, suggesting that most damage occurred subsequent to stall. Distinguishing among mechanisms (i) through (iii) is nontrivial and will require additional biochemical, as well as mechanical, data.
31. D. A. Erie, O. Hajiseyedjavadi, M. C. Young, P. H. von Hippel, Science 262, 867 (1993); H. Matsuzaki, G. A. Kassavetis, E. P. Geiduschek, J. Mol. Biol. 235, 1173 (1994); C. Chan, D. Wang, R. Landick, ibid. 268, 54 (1997).
32. D. A. Erie, O. Hajiseyedjavadi, M. C. Young, P. H. von Hippel, Science 262, 867 (1993); H. Matsuzaki, G. A. Kassavetis, E. P. Geiduschek, J. Mol. Biol. 235, 1173 (1994); C. Chan, D. Wang, R. Landick, ibid. 268, 54 (1997).
33. D. A. Erie, O. Hajiseyedjavadi, M. C. Young, P. H. von Hippel, Science 262, 867 (1993); H. Matsuzaki, G. A. Kassavetis, E. P. Geiduschek, J. Mol. Biol. 235, 1173 (1994); C. Chan, D. Wang, R. Landick, ibid. 268, 54 (1997).
34. N. Komissarova and M. Kashlev, J. Biol. Chem. 272, 15329 (1997); E. Nudler, A. Mustaev, E. Lukhtanov, A. Goldfarb, Cell 89, 33 (1997).
35. N. Komissarova and M. Kashlev, J. Biol. Chem. 272, 15329 (1997); E. Nudler, A. Mustaev, E. Lukhtanov, A. Goldfarb, Cell 89, 33 (1997).
36. M. Palangat, T. Meier, R. Keene, R. Landick, Mol. Cell 1, 1033 (1998).
37. More complex models for movement can be entertained, including those where the tight-coupling assumption is violated. Such loosely coupled enzymatic schemes are not parsimonious, however, as they involve greater numbers of free parameters.
38. H. Y. Wang, T. Elston, A. Mogilner, G. Oster, Biophys. J. 74, 1186 (1998).
39. F. Jülicher and R. Bruinsma, ibid., p. 1169.
40. M. D. Wang, H. Yin, R. Landick, J. Gelles, S. M. Block, ibid. 70, A44 (1996).
41. K. A. Johnson, Annu. Rev. Biochem. 62, 685 (1993).
42. eff/F. Normalization through Eq. 10 permits F-V curves with different apparent δ to be compared, yielding an average value.
43. Supported by grants from the National Institute of General Medical Sciences (R.L., J.G., and S.M.B), and from NSF and the W. M. Keck Foundation (S.M.B). M.D.W. was supported by a Damon Runyon-Walter Winchell Cancer Research fund postdoctoral fellowship. M.J.S. was supported by an American Heart Association predoctoral fellowship and a Proctor Honorific Fellowship from Princeton University. S.M.B. and M.D.W. thank S. Gross, L. Satterwhite, and K. Visscher for helpful discussions and advice, and K. Neuman for comments on an early draft of the manuscript. A movie of the tether-particle assay can be viewed at www.rose.brandeis.edu/users/gelles/ stall/