Studies on the impellers generating force in muscle

This is intended to be a tribute to our friend, Cristobal dos Remedios, on his 80th birthday. I briefly describe a personal retrospection and historical perspective of studies on chemo-mechanical coupling in muscle and myosin motors. It was probably 1982 when I first met Cris. At that time, I was a postdoc in the laboratory of Manuel Morales at UC San Francisco (UCSF). Cris also worked in the same lab from December 1969 through March 1972 as an American Heart Association (AHA) Research Fellow. In this period, Cris developed a fast apparatus, based on the PDP-12 computer (an unusual tool in biophysics laboratories at that time), to record changes of polarized tryptophan fluorescence from single striated muscle fibers while simultaneously recording the tension and length changes of the fibers. By then, the cross-bridge orientation had been studied only by X-ray diffraction and electron microscopy, which had no time resolution in practice. Using the fast apparatus, Cris succeeded in finding that the cross-bridge orientation dynamically changes in response to the alteration of the physiological state of muscle fibers (Dos Remedios et al. 1972). For this achievement, he received the Louis N. Katz Basic Science Research Prize 1971 from the AHA. This study was selected as one of fifty important papers in the history of muscle contraction and myosin motility (Ostap et al. 2004). Then, Cris went back to Sydney, where he started his own laboratory to study actin and its binding proteins.

Around the period when I worked at UCSF (October 1979–March 1986), the following researchers of UCSF were closely studying muscle, myosin, and actin: Roger Cooke, Robert Mendelson, Jean Botts, Julian Borejdo, and Paul Wagner as group leaders; John Shiner, Reiji Takashi, Deborah Stone, Tetsu Hozumi, Hiroshi Kagawa, Sudhir Srivastava, Edward Pate, Kunihiko Konno, Peter Torgerson, Michael Wilson, Thomas Burghardt, Katalin Ajtai, Andrzej Kasprzak, Dominique Mornet, and Paul Curmi as assistant biophysicists or postdocs; Edward Giniger as a gap year fellow (Fig. 1). We often had visitors. To name a few, Setsuro Ebashi, Yuji Tonomura, Fumio Oosawa, Andras Muhlrad, Ridha Kassab, and Watt Webb. Cris also occasionally visited UCSF. At that time, all studies in the team were mostly focused on the orientational changes of cross-bridges in muscle fibers, and the structure and conformational changes of myosin head coupled with the ATPase reaction.

Fig. 1.

Snapshots of members of the UCSF muscle research team gathered at a potluck party in 1980

Various techniques were used in the UCSF team. The Cooke group used electron paramagnetic resonance spectroscopy and a thermodynamics approach. They found that the spin probe–introduced domain within the myosin head does not change orientation during the power stroke in muscle contraction or force application to the rigor bond (Cooke 1981; Cooke et al. 1982). This finding suggests that the location of elastic and force-generating elements within the cross-bridge is restricted to another domain (i.e., light chain–binding domain or S2). The Mendelson group mainly used small-angle X-ray scattering (Mendelson 1982) and neutron scattering for myosin head (Curmi et al. 1987) to determine the shape of S1 and its change upon binding to nucleotides or actin. The Wagner group took biochemical approaches to study the role of myosin light chains in the ATPase and interaction with actin (e.g., Wagner & Giniger 1981). The Botts group used the fluorescence resonance energy transfer method to determine the distances between multiple sites within the myosin head and between sites on myosin and actin, and their dependence on the nucleotide state (e.g., Botts et al. 1984). The Borejdo group used a linear dichroism method to measure the orientation of cross-bridges in different states (e.g., Burghardt et al. 1983).

These methods were all based on ensemble averaging, while single-molecule techniques were emerging at that time, and also some members of the myosin superfamily were being discovered. One of the dreams of Manuel Morales was to reveal the chemo-mechanical coupling between the cross-bridge motion and the ATPase reaction. He conceived a plan of measuring the mechanical and chemical signals simultaneously, from which their cross-correlation (i.e., coupling) could be obtained. In equilibrium or steady states, a system containing a small number of molecules in a small volume exhibits concentration fluctuations of species by translational diffusion, orientational changes, and chemical reactions. The correlation function of such time series of fluctuation data can provide rate constants for these dynamic processes. When fluorescence signals are used to detect such motions and reactions, this technique is called fluorescence correlation spectroscopy (FCS) (Magde et al. 1972; Ehrenberg and Rigler 1974). My own project was to create fluorescent signals whose intensity could change in a large extent upon nucleotide binding, hydrolysis, and release, while Julian Borejdo’s was to measure fluctuations of polarized fluorescence from muscle fibers. After having struggled for 2 years, I created a chemical signal based on dynamic fluorescence quenching of an ATP analogue (fluorescent ε-ATP) by acrylamide; the fluorescence intensity of ε-ATP increased 4 times upon binding to myosin (Ando et al. 1982). Importantly, muscle fibers contract by hydrolysis of ε-ATP in the presence of acrylamide. Then, I attempted to perform FCS experiments using single muscle fibers immersed in a solution containing ε-ATP and acrylamide. But it was not successful due to an insufficient photon collection efficiency in the fluorescence microscope, photobleaching occurring even in the presence of anti-bleach reagents that I tested, and an inefficiency of the 1-bit digital correlator. We eventually gave up this FCS project, which was too pioneering in that era.

The dream of Manuel Morales eventually came true more than 20 years later, which largely owed to the advent of both single-molecule measurement techniques and processive myosin motors. Single molecules of processive myosin V (M5) can move on actin filaments over a long distance (Mehta et al. 1999; Sakamoto et al. 2000), unlike muscle myosin. Takeshi Sakamoto at NIH, a former student in my lab at Kanazawa University, directly showed that M5 stepped forward every ATP-binding/dissociation event, using single-molecule fluorescence microscopy and a fluorescent ATP analogue (Sakamoto et al. 2008). In my lab, the swinging lever-arm motion was directly visualized with unprecedented clearness by imaging of single M5 molecules walking on actin filaments with high-speed atomic force microscopy (HS-AFM) (Kodera et al. 2010) (for a historical view on the path to this direct observation, see the review of Kodera and Ando 2014).

Seen in the historical perspective described above, questions about what really occurs in muscle contraction and myosin motors have impelled researchers to develop new measurement tools and methods. Time-resolved measurements of cross-bridge orientation in muscle was pioneered by Cris using tryptophan fluorescence, which was followed by the introduction of extrinsic fluorophores or spin probes to SH1 and later to myosin light chains in muscle fibers (e.g., Hopkins et al. 1998). Although the lever-arm swinging was clearly visualized for M5 using HS-AFM, no one has yet succeeded in directly visualizing this motion in muscle fibers in real space and in real time. I would foresee that a technique enabling this observation will emerge in the near future. This technique will surely bring a tremendous impact in life sciences.